- MPC_VEL_LP: new velocity first order low pass filter (off by default)
- MPC_VEL_NF_FRQ/MPC_VEL_NF_BW: new velocity notch filter (off by default)
- MPC_VELD_LP: existing velocity derivative low pass filter, but I've dropped the remaining controllib usage
The case where the airframe maintainer wants to enforce a reset to airframe is
not covered anymore with the `param set-default` mechanism. For products based
on PX4 this is still required to ensure proper functionality after a major update.
Having a generic interface over the GPS drivers makes dedicated
functionality for each driver harder. Move the Septentrio driver into
its own module under the `gnss` driver directory, and let it have its
own parameters for only the functionality it requires. This also helps
with adding new features because they only need to be implemented for
the driver that wants it, simplifying testing.
- these are old RC switch configuration parameters that haven't been used for
years, but were hardcoded into old versions of QGC
- RC_MAP_RATT_SW, RC_MAP_POSCTL_SW, RC_MAP_ACRO_SW, RC_MAP_STAB_SW, RC_MAP_MAN_SW deleted
-
* ist8310: do reset before WHOAMI call
Apparently, the IST8310's WHOAMI register is writeable. Presumably, this
can get corrupted by bus noise. It is only reset if powered off for 30s.
Therefore, we do a reset before doing the WHOAMI.
* ist8310: improve comment
Co-authored-by: Beat Küng <beat-kueng@gmx.net>
---------
Co-authored-by: Beat Küng <beat-kueng@gmx.net>
Compared to GNSS, alternate position observation methods are less accurate
and thus generally not good enough to do airspeed validation with.
Airspeed validation is thus disabled if no GNSS fusion is happening.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* AirspeedValidator: remove additional one second of hysteresis for triggering
innovation checks
- this check already uses an integrator and so adding more delay just makes
log analysis more difficult and does not really add any value
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* removed unnecessary conditions
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* AirspeedValidator: only disable innov checks if ASPD_FS_INTEG is negative
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* get rid of unnecessary check on innovation threshold parameter
Signed-off-by: RomanBapst <bapstroman@gmail.com>
---------
Signed-off-by: RomanBapst <bapstroman@gmail.com>
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: Silvan Fuhrer <silvan@auterion.com>
This fixes various edge cases when input is set to both: RC and
MAVLink gimbal protocol v2.
Specifically:
- We no longer immediately reset primary control if there is no update,
otherwise this will reset immediately after an commands.
- Talking of commands: GIMBAL_MANAGER_CONFIGURE no longer switches
control to MAVLink but only an actual incoming setpoint command does.
- And incoming setpoint command only triggers UpdatedActiveOnce which
means we will check RC again afterwards and if there is big movement
switch back to RC. That's the intuitive thing to do until we have
better reporting about who/what is in control.
- Also, with RC we no longer always set us to be in control but only on
major movement.
192.168.0.x is often used by routers for WIFI/ethernet networks, and thus
can create conflicts.
This can happen for example if a companion is connected to the FMU via
ethernet and at the same time connects to a WIFI network as DHCP client.
VTOLs consume a lot more power in hover copared to fixed-wing fligt.
The remaining flight time thus should reset if one switches from MC to FW,
as otherwise it takes several minutes until the estimate goes down.
The argparse module has been builtin to Python since
Python 3.2, released in 2011 (see
https://docs.python.org/3/whatsnew/3.2.html). Further,
the argparse pip module has not been released or updated
since 2015, and lacks some of the features of the modern,
built-in argparse. Drop the pip installed version in
favor of the built-in version.
Signed-off-by: Chris Lalancette <clalancette@gmail.com>
- upate to NOAA grid API to build WMM table in one pass
- refactor declination/inclination/totalintensity table printing to
shared method
- compute scaling factor to maximize resolution
fix: ControlAllocator float and int comparison bug
There was an incorrect comparison between a float variable `dt` and a `uint64_t` value representing 5 milliseconds (`5_ms`). As a result, `do_update` could never become true even if the last torque setpoint was received more than 5 milliseconds before.
To solve this, the `5_ms` value has been converted to seconds (0.005f) for the comparison with `dt`.
In this mode, the mag is used to inilialize the heading.
During flight, the heading is predicted using gyros and corrected
by GNSS measurements if available.
Previously, if a mag was not required (not index 0 and not used by ekf),
it was not counted in num_enabled_and_valid_calibration.
If a user set SYS_HAS_MAG to e.g. 3, it would then trigger a preflight
failure, even if there were 3 calibrated and enabled mags.
The EKF can recover from an initial bad earth mag field estimate.
Constraining the field is not necessary and can lead to an unpredicted
behavior of the filter.
Declination fusion is safe to run even when the horizontal field is 0
On slowly moving vehicles (e.g.: boats, rovers), the yaw estimator has
worse convergence than the main EKF. Resetting the mag states using the
yaw estimator as reference can lead to poor heading. Also, the EKF can
recover really well from initially incorrect mag states.
The previous default of 900m leads to many warnings if left
unchanged, especially if the vehicle is already in-air when
the Mission is started.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Includes:
- Remove some of the outdated Python2 checks and compatibility.
- Try not catch all exceptions but only the expected ones. Otherwise,
this makes it really hard to debug if anything unexpected actually
goes wrong.
- Make use of fstrings.
- Make output slightly prettier.
Signed-off-by: Julian Oes <julian@oes.ch>
This adds a new protocol extension which allows to get the bootloader
version.
The bootloader version is different from the bootloader protocol
revision which has stabilized at 5 and is not easy to update unless a
bootloader is actually breaking the protocol. The reason being that both
the Python script as well as the uploader used in QGC will not attempt
to load firmware if they don't know the bootloader version, so it could
basically be considered a "breaking" protocol revision.
Signed-off-by: Julian Oes <julian@oes.ch>
If the STM32H7 fails to program or erase a full chunk of 256 bytes, the
ECC check will trigger a busfault when trying to read from it.
To speed up erasing and optimize wear, we read before erasing to check
if it actually needs erasing. That's when a busfault happens and the
erase time outs.
The workaround is to add an option to do a full erase without check.
Credit goes to:
https://github.com/ArduPilot/ardupilot/pull/22090
And the protocol option added to the bootloader is the same as for
ArduPilot, so compatible.
Signed-off-by: Julian Oes <julian@oes.ch>
* usb: Added parameter to enable always starting mavlink on USB.
Refactored cdcacm_init into a module and added a paramter to allow always starting mavlink on
USB, also added a paramter to choose the mode. The current default behavior is to wait and listen
for data on USB and auto-detect the protocol (mavlink, nsh, ublox). This results in the mavlink
stream not starting until something else on the mavlink network sends a packet first. The new
default behavior is to always start mavlink.
Added parameters
MAV_USB_ENABLE -- default 1 (always start mavlink on USB)
MAV_USE_MODE -- default 3 (onboard)
* added 3 retries for opening serial port in mavlink, removed sleep before sercon
* added DRIVERS_CDCACM_AUTOSTART to ark-v6x default.px4board
* added CONFIG_DRIVERS_CDCACM_AUTOSTART=y to default.px4board for boards with CONFIG_CDCACM in their nsh/defconfig
* format
* remove PGA460 from COMMON_DISTANCE_SENSOR to save flash
* remove LIS2MDL from COMMON_MAGNETOMETER to save flash
* disable CONFIG_DRIVERS_CDCACM_AUTOSTART for fmu-v5 protected.px4board
* moved and renamed parameters, removed mode logic in mavlink
* changed parameter names, added mode none
* remove parameters from mavlink
- increase symforce CppConfig zero_initialization_sparsity_threshold so
that a Matrix setZero() call is performed instead of individually zeroing
Co-authored-by: bresch <brescianimathieu@gmail.com>
The setting wasn't consistent with the one used in the Visual Studio
Code settings, which caused different newline formatting depending on
whether the user uses Visual Studio Code or another editor that uses
EditorConfig.
This notes the reference yaw angle for the Septentrio Mosaic-H.
It's unfortunately a bit tricky in that Unicore has the main antenna
in front by default while Septentrio decided to put the aux antenna in
front.
Signed-off-by: Julian Oes <julian@oes.ch>
Regression from https://github.com/PX4/PX4-Autopilot/pull/23043
Also avoids a race condition by making sure the command ack is handled
before sending out the mavlink message (in case an external component
reacts immediately to the mavlink message).
- this warning was to catch any potential errors in sensor selection
relative to what's actually available, we don't need to complain
about initial selection before the EKF selector is available
When North-East (e.g.: GNSS pos/vel) aiding is active, the heading
estimate is constrained and consistent with the vel/pos aiding. Reset to
GNSS heading should only occur if no N-E aiding is active or if the
filter is not yes aligned. Otherwise, just wait for the consistency
check to pass again (will pass at some point if the heading uncertainty
of the filter is getting too high).
- topic update was checked twice in the same loop and thus
the second time the topic would never indicate to have updated
Co-authored-by: RomanBapst <bapstroman@gmail.com>
- warn about full traffic conflict buffer at 1/60hz.
- add conflict expiry for buffer.
- use only events for buffer full warning. mavlink_log_critical no longer needed.
- use icao address for conflict warnings id, stop using uas_id. UTM_GLOBAL_POSITION assumed deprecated.
- stop spamming when buffer is full
- fix warning wording if buffer is full.
- remove UTM_GLOBAL_POSITION
Fixes failing unit test:
* [adsbTest] Reduce conflict timestamps - not enough time has passed in ci
- failed ci output - (passes locally with make tests TESTFILTER=AdsbConflict)
- Timestamp: 6000000000
- Time now: 457720038
- Time since timestamp: 0
- Old Conflict warning expired: 0
- --------------------
- adsb_conflict._traffic_state 0
- ../../src/lib/adsb/AdsbConflictTest.cpp:244: Failure
- Value of: adsb_conflict._traffic_state == TRAFFIC_STATE::REMIND_CONFLICT
- Actual: false
- Expected: true
* Made Serial API open the UART in NON BLOCKING mode
* Updated voxl_esc driver to latest from ModalAI fork
* Ported voxl_esc driver over to new Serial UART API
* Removed voxl_esc serial abstraction since new Serial API is already a serial abstraction
- with the addition of the navsat plugin in PR#22638, the callback would
reassign the previous timestamp used in the calculations of the
angular_velocity causing derivative type noise in the groundtruth
measurements
This is a minimal change to make it harder to crash a
vehicle with an empty battery if the capacity was set wrong.
The disadvantage is that the state of charge estimate will
fluctuate more under load.
We need better documentation and improvements to the estimation.
The existing disarmed logic already handles disabled outputs
it makes sense to reuse it and not have lockdown handled
differently resulting in unexpeced corner cases.
This removes the duplication with unexpected differences
and allows to consistently handle the output instead of
overriding the output for some specific cases which
leads to unexpected corner cases. E.g. disabled outputs
suddenly outputing PWM in lockdown.
In HITL the actuators should not be mapped
and they are in lockdown.
We should not reconfigure disarmed, min, max PWM values
without updating the actual output values because the IO
will consider the last outputs before the FMU was rebooted
with the configuration of the new boot. This can result
in spinning motors when switching to SIH.
For Septententrino we seem to sometimes fill the buffer pretty full.
If we ask for too much, readAtLeast will fail completely and make the
GPS discovery logic fall over. Therefore, let's not ask for too much and
just read what we can given the available buffer.
Signed-off-by: Julian Oes <julian@oes.ch>
* Added decoy termios support to Qurt so that ghst parser in RC library can be used. No termios is actually needed but has to be there for the parser to work
- Added an empty constructor, setPort, and validatePort functions for Serial API
- Changed GPS to not allocate Serial object dynamically
- Moved access check on serial port name into the Serial API
- Improved the Qurt platform validatePort Serial function to implement a more rigorous check. Added safety check
to the setPort Serial function to make sure it isn't called after the port has been already opened.
This matters for PWM when the last output gets disabled on either FMU or IO
it would just keep on running.
Also when rebooting with a parameters reset or new airframe with no mapped outputs
it would previously keep outputting PWM with the disarmed value of the new airframe
e.g. 1000us which is a safety hazard because servos could break the physical limit of the
model or miscalibrated ESCs spinning motors.
If the output is set to 0 then the FMU had this channel disabled/no function mapped
to it. In that case we do not want to suddenly start outputing failsafe or disarmed
signals.
- Added offset to Posix hrt time to account for synchronization with Qurt hrt time
- Added new Kconfig to configure synchronization of HRT timestamps on VOXL2
- Moved voxl2 libfc sensor library submodule from muorb module to boards directory
- Added check to make sure hrt_elapsed_time can never be negative
EditorConfig is a well-known convention to share style settings across
different editors. Adding one will make it easier for new contributors
or people who like to use a different editor to contribute.
The duration of a unit test had to be increased because the incorrect
covariance matrix update, was making the unit test passing faster
(over-optimistic variance).
Using full 3D acceleration provides better horizontal acceleration
tracking but also creates a sometimes unwanted behavior because the tilt
is directly coupled with the vertical acceleration setpoint.
* Commander: make low remaining flight time configurable and do not clear
- add _ACT param to disable/warning/RTL this feature
- publish rtl flight time estimate also in RTL, and thus fix re-validation
- make failure message clearer, distinguish from battery low
* battery check: add hysteresis for declaring battery_low_remaining_time false again
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: KonradRudin <98741601+KonradRudin@users.noreply.github.com>
The voxl2 has a split architecture. PX4 runs on a posix platform and a Qurt platform. The two communicate uorb topics back and forth with the muorb module. But each has it's own parameters database and they need to stay in sync with each other. This PR adds support to keep the 2 parameter databases in sync. The main parameters database running on Linux has file system support while the Qurt one does not. The Linux side is considered the primary and the Qurt side is considered the remote.
We need to make sure that when the RTL is triggered, it should not reevaluate it, as when it was valid but evaluated to false on activation, it can't do a RTL.
Previously uORB queue size was an awkward mix of runtime configurable (at advertise or IOCTL before allocate), but effectively static with all queue size settings (outside of test code) actually coming from the topic declaration (presently ORB_QUEUE_LENGTH in the .msg). This change finally resolves the inconsistency making the queue size fully static.
Additionally there were some corner cases that the muorb and orb communicator implementation were not correctly handling. This PR provides fixes for those issues. Also correctly sets remote queue lengths now based on the topic definitions.
* Made setting of uORB topic queue size in based on topic definition only
* Fixes to the ModalAI muorb implementation
* Removed libfc sensor from format checks
* msg/TransponderReport.msg ORB_QUEUE_LENGTH 8->16 (was set to higher in AdsbConflict.h
---------
Co-authored-by: Eric Katzfey <eric.katzfey@modalai.com>
Co-authored-by: Daniel Agar <daniel@agar.ca>
Remove option to track from previous wp to reduce complexity and fix
case where prev=current point and the line following broke down.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
We had a special handling for RTL triggered in vtol_takeoff state.
The idea is to wait until the VTOL Takeoff is completed and only
then switch to RTL. On a second thought this special handling isn't
really necessary and for the sake of simplicity should be removed.
This also removes the side effect of the indicated flight mode
after RTL being set to VTOL_Takeoff again.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Position Control: some cosmetical changes
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Position Control: disable roll constraining warning in VTOL transition
In transitions it is expected that the roll is constrained, and
instead of defining an aribitrary threshold let's rather disable
the user warning in that case.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Pos C: define magic numbers for roll constraining warning as constants
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
The gyro bias estimate from EFK2 is really good when at rest and should
be used by the yaw estimator to prevent heading drifts due to poor
heading observability.
ENOENT returns if the file doesn't exist yet, when using mtd /fs/mtd_net always exist.
On a filesystem you've to generate the file so if ENOENT returns we've to regenerate the default config as well.
The main problem was that during initial negotiation the client would
request the gimbal_manager_information from px4 but px4 would never send
it because in this configuration the device_compid was set to 0.
`estimator_baro_bias` requires to have GNSS and baro hgt active and GNSS as the reference. This is quite restrictive. Instead, we can simply use a high-passed version of the baro error.
* Added Lawnmower airframe
* Update 5005_gz_lawnmower
Works all right
* Update 5005_gz_lawnmower
RDD_WHL_SPEED has new name: RDD_WHEEL_SPEED
* Update ROMFS/px4fmu_common/init.d-posix/airframes/5005_gz_lawnmower
Co-authored-by: Per Frivik <94360401+PerFrivik@users.noreply.github.com>
* Update ROMFS/px4fmu_common/init.d-posix/airframes/CMakeLists.txt
Co-authored-by: Per Frivik <94360401+PerFrivik@users.noreply.github.com>
* Renamed 5005_gz_lawnmower to 4011_gz_lawnmower
also pulled latest GZ models hash
---------
Co-authored-by: Per Frivik <94360401+PerFrivik@users.noreply.github.com>
This is a workaround to hotfix CI but the root cause is #22792
(MAVSDK test failing after EKF change, accelerometer simulation issues not learned anymore?)
MissionBase did not initialize its mission data, thus could enter an
infinite loop in updateDatamanCache() if the initMission() failed to
read the mission off, for example, due to the SDCard storage task taking
longer than the timeout to respond.
This change constrains the loading loop and resets the mission data even
if the data write failed.
The setpoint.yaw can be NAN, and this made the calculated land point NAN
as well. Looking at the current yaw is anyway a better way to approximate
the course over ground that fundamentally should be used.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
ADd a new misison feasiblity checker option to check if a proper landing approach is defined when in air. There must be at least a mission landing or a VTOL approach defined in order for the mission to be accepted. Else, use the same logic as in MIS_TKO_LAND_REQ=4
* publish the global groundtruth from the navsat callback and rearrange the local groundtruth as the altitude reference now has a dependency on the global groundtruth being initialized
---------
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
-remove RTL_HDG_MD
-only set heading setpoint in Navigator::RTL once above landing point,
or when RTL is triggered close to it
-never set a heading during RTL if weather vane is enabled
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
After smoothing the linera velocity setpoint, the EKF has trouble initializing, becuase the acceleration is too smooth, to combat this issue, there is a 1 second delay when initializing the mission mode
Instead of directly setting the attitude setpoint for usage inside the same module
only publish it to the uorb topic, which is subscribed to in the same module.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* battery: make time remaining estimation dependent on level flight characteristis for FW
* battery: fix that FW flight is also correctly detected when vehicle_status is not updated
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FixedwingPositionControl: Move constant to header file
* flight phase estimation: use tecs height rate reference to check for level flight
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: Silvan Fuhrer <silvan@auterion.com>
When flow control is used together with DMA, we need to add a pulldown
to CTS. Without it, it assumes flow control and gets stuck when
CTS is not connected.
Signed-off-by: Julian Oes <julian@oes.ch>
When the CPU load monitor is started while already running, then the
idle thread last_times[0] is reset to the last 1 second, rather than
since when the CPU load monitor was last started. The remaining threads
are not impacted, since their last_times[i] is reset to zero here.
This results in the idle thread having a lower than real CPU load, with
the remaining CPU time being wrongly attributed as scheduler load.
The ADC peripheral can only support up to
50MHz on rev V silicon and 36MHz on Y silicon.
The existing driver always used no prescaler
and kept boost setting at 0.
* refactor driver to dynamically check registers and do reset if register does not match desired value
* have seen various times where shunt calibration was reset in air
---------
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* avoid waypoint following during backtransition
- if vehicle overshoots transition, make controller track the prev/current
waypoint line, this avoids large banking maneuvers that delay the transition further
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* navigator: use SETPOINT_TYPE_LAND as setpoint type for VTOL_LAND waypoints
such that fixed wing position controller can distinguish from other types
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* FixedWingPositionControl: split vtol backtransition logic into separate method
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* review changes
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* prevent loiter when approaching vtol backtransition and invalidate previous waypoint during mission landing
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* removed unused parameter
Signed-off-by: RomanBapst <bapstroman@gmail.com>
* small renaming of transition mode
Signed-off-by: RomanBapst <bapstroman@gmail.com>
---------
Signed-off-by: RomanBapst <bapstroman@gmail.com>
When a state stops to be estimated it becomes a "consider state". Its value and variance are frozen but its covariance with other states continue to evolve normally.
- removes conservative accel bias variance limiting
- force symmetry is skipped after fusion of NED vel/pos (a direct measurement)
---------
Co-authored-by: bresch <brescianimathieu@gmail.com>
When handling multiple commands, it could happen that the first command
updates _reposition_triplet. Normally this would then get handled after
getting the mode change from commander through vehicle_status.
But if the next command is handled before an update from commander, it
could overwrite the triplet.
This patch ensures that navigator waits for an update from commander (and
therefore process the _reposition_triplet) before handling the next
command.
This happened specifically when pressing 'Pause' from QGC during a mission.
QGC sends VEHICLE_CMD_DO_REPOSITION twice, first for pausing, then changing
the altitude.
The result was that the vehicle would not stop at the current location but
continue to the next mission waypoint and stop there.
- fake_pos only if at rest or tilt variances becomes large
- fake pos: don't run when grav fusion is enabled
- gravity fusion enabled by default
- gravity: only fuse when accel norm and lpf norm are consistent
Co-authored-by: bresch <brescianimathieu@gmail.com>
A board can optionaly define RC_SERIAL_SWAP_USING_SINGLEWIRE
If the board is wired board with TX to the input (Swapped) and
the SoC does not support U[S]ART level RX-TX swapping to allow
useing onewire to do the swap if and only if:
RC_SERIAL_SWAP_USING_SINGLEWIRE is defined
RC_SERIAL_SWAP_RXTX is defined
TIOCSSWAP is defined and retuns !OK
TIOCSSINGLEWIRE is defined
* TECS: reduce default of FW_T_I_GAIN_THR
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* TECS: reduce default of FW_T_THR_DAMP
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* TECS: improve param descriptions and meta data of some params
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Pos C: rename FW_T_THR_DAMP to FW_T_THR_DAMPING
Rename wihtout param translation as the interpretation of this param has changed
recently (a bug was fixed that caused the damping to have a much lower effect than
it should). We want to avoid that users keep the previously tuned value (with the bug)
and do not update the tuning when using the fixed version (v1.14.0 and later).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Pos C: rename FW_T_I_GAIN_THR to FW_T_THR_INTEG
Rename wihtout param translation as the interpretation of this param has changed
recently (a bug was fixed that caused the damping to have a much lower effect than
it should). We want to avoid that users keep the previously tuned value (with the bug)
and do not update the tuning when using the fixed version (v1.14.0 and later).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This check gave a warning when one or more mission items were below the
Home altitude. As it is a valid usecase to plan missions below Home,
those warnings were too often seen as a inconvenience and are thus
removed here.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
That enables us to update the pitch setpoint in Stabilized mode
in the VTOL controller (pitch ramp for transition).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Separate offboard path setpoints as fixedwing pos control state
This commit separates offboard path following as a separate state inside the FW Poscontrol module.
This is a cleanup on clearly defining fw pos control behaviors
* Update src/modules/fw_pos_control/FixedwingPositionControl.cpp
Co-authored-by: Silvan Fuhrer <silvan@auterion.com>
* Fix format
---------
Co-authored-by: Silvan Fuhrer <silvan@auterion.com>
The IMU topic is not zeroed, and the clipping information was not
copied, so this field ended up being garbage.
Signed-off-by: Julian Oes <julian@oes.ch>
With this margin it is made sure that if the loiter is not perfectly tracked,
(vehicle outside of path setpoint) a wp just at the border of the loiter
is still reachable.
It should though be as small as necessary, as otherwise, with good
loiter tracking, waypoints that are close but not right on the loiter
radius are not enforcing the exit course neither.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* PositionSetpoint: remove yaw_valid field
* Navigator: set yaw setpoint to NAN for Takeoff
Don't set a yaw setpoint for takeoff, as Navigator doesn't handle the yaw reset.
The yaw setpoint generation is handled by FlightTaskAuto.
* PositionSetpoint.msg: remove disable_weather_vane and instead only use the yaw field
Strictly follow the concept that if the position_setpoint.yaw is set, then
follow it the controller, and otherwise let the controller set it as it
thinks it's best.
* Navigator: remove logic that sets yaw to be accepted in TAKEOFF
No longer needed as during Takeoff we anyway don't set a yaw setpoint.
* PositionSetpoint.msg: remove yawspeed_valid
* PositionSetpoint.msg: remove yawspeed
* Navigator: set yaw setpoint to NAN instead of current
In set_takeoff and set_land_item, as well as for VTOL transition.
The flight tasks then set the yaw corresponding to the current yaw.
* Navigator: change get_yaw_acceptance into a bool
* PositionSetpoint.msg: improve comment for yaw
* MissionBlock: remove unnecessary code from set_vtol_transition_item
* Navigator: clean up calculate_breaking_stop(), set yaw to NAN
* Navigator: set yaw to NAN in variouls places where not specifc setpoint is desired
* Navigator: set yaw to NAN in reset_position_setpoint()
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: Matthias Grob <maetugr@gmail.com>
1s gives some false positives at boot up, as the airspeed selector only
starts publishing 2s after its startup.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
-remove dedicated vtol transition airspeed init logic
-init airspeed setpoint on first usage of tecs
-init to max of current airspeed and min airspeed
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
simulation gazebo: move the gazebo models to submodule, allow for operation with external gazebo instance, independent of startup order. Allows drag an drop of models from gazebo fuel.
* rolled back updates
Signed-off-by: frederik <frederik@auterion.com>
* fixing empy
Signed-off-by: frederik <frederik@auterion.com>
* Update GZBridge.cpp to lower drop position
Dropping from 1m leads to movement in the rc_cessna. Dropping from 0.5m leads to no movement.
* Update STANDALONE env variable.
* Update STANDALONE env_variable on GZBridge
* Update src/modules/simulation/gz_bridge/GZBridge.cpp
Co-authored-by: Daniel Agar <daniel@agar.ca>
* test removal of x500
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* removed all models and reworked logic
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* remove model path in set_sdf_filename
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* filter resource path for world sdf
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* updated structure to keep old make px4_sitl
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* remove gz tools
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* import gz as submodule and reverse rc simulator logic
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
* [gz-sim]: source GZ_SIM_RESOURCE_PATH only if PX4 starts gz server
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@gmail.com>
* Typo fix
---------
Signed-off-by: frederik <frederik@auterion.com>
Signed-off-by: frederik <frederik.anilmarkus@gmail.com>
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@gmail.com>
Co-authored-by: Daniel Agar <daniel@agar.ca>
Co-authored-by: Beniamino Pozzan <beniamino.pozzan@gmail.com>
* events: pass relative paths plus base path to source parser script
to work around maximum Makefile command lenght limits.
* events: correct cmake comment typo
Co-authored-by: Beat Küng <beat-kueng@gmx.net>
---------
Co-authored-by: Beat Küng <beat-kueng@gmx.net>
The _control_mode_current wasn't updated otherwise, and only done so by luck
because we already set the current type to SETPOINT_TYPE_POSITION.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- double required max consecutive counts
- don't continuously complain about round trip time (RTT) unless
there's been at least one acceptable round trip (latency < 100 ms)
Before this fix, the time_utc_usec output from blending was always 0.
This means that you wouldn't get a valid vehicle_gps_position/time_utc_usec
With this commit, UTC timestamps are blended according to weights for
all GPSes with a nonzero UTC timestamp value.
It would be possible to simply use the first valid UTC timestamp instead
of blending, but since the system timestamps are blended, it seems
suitable to blend UTC timestamps as well.
- This fixes the force call on SensorSelectionUpdate
- In contrary to the rest of the codebase, this method also takes a timestamp: When you call SensorSelectionUpdate(true), time_now_us is actually set to 1 and force stays false, as this is the default value for in the method.
wait_until_altitude() checks for absolute altitude being
close so checking for 1m below the setpoint can fail
if the speedup results in no sample inside the altitude
window being checked.
Ideally the test could check if the takeoff is done directly
instead of comparing altitudes in the first place.
-remove external init, and instead always (but only) init when dt is too large
-init the controller params correctly
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
The control params (eg min/max pitch) are used before they are
correctly set by TECS::update(). While this is an issue we should fix,
it also doesn't hurt to set them to more reasobale values (eg 30° limit).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
As discussed in the maintainer call we should adhere to the
parameter naming scheme that makes it clear what vehicle type the
configuration is good for.
by only applying the sqrt linear drag when brakeing.
It was originally not done this way to avoid discontinuities and
the exact speed bewlo 1m/s didn't matter. With the position slow mode
the exact slow speeds now matter. And the discontinuities are avoided by
reusing the brake boost filter.
It is generally preferred to have the camera pointing forward on pause,
e.g. to use the camera for Navigation.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
The velocity smoothing library constrains the maximum vellocity.
I set the default constraint to 0 to find these exact issues.
The heading smoothing did not initialize the maximum velocity
which in this use case is the maximum heading.
So heading was always constrained to 0 -> north until this change.
The defintion of the custom symbol M_PI_PRECISE
was not included in one of the only places it's used.
Looks like a mistake that happened in
34c852255e
possibly because a lot of things are included
almost everywhere and if the include order
ligns up there are no compile errors.
Note that the unit test also passes without
thechange. But the VelocitySmoothing's
local_time would get NAN. This would leads to
wrong trajectory calculations.
Problem: Zero maximum jerk and/or zero
direction would lead to divisions by zero in
various cases depending on the exact configuration and code path.
Solution: There was already one check for the
direction being zero in one path and I
summarized to check in both
updateDurations...() functions the product
of direction * max_jerk to not be zero
because that's exactly the value calculations
devide by.
goto control class handles smoothing of goto setpoints, outputs trajectory setpoint for mc pos control
some minor encapsulation done in mc pos control for readability
new param MPC_YAWAAUTO_MAX limiting heading accelerations in heading smoother
The size input argument for monocypher crypto_xchacha20_ctr should be the
plaintext message length.
The promise of the interface is, that the call to encrypt_data updates the
ciphertext message length after the call succeeds.
The crypto should check that the output buffer length (cipher length) is
large enough to contain the encrypted data.
Fix these issues; these have gone unnoticed for a long time since the interface
has been only used by logger, and passing the same size for both in and out.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
The manufacturer uses both naming schemes, RoboClaw more than Roboclaw
but it's always one word and hence I think it's more consistent to name
it the latter.
This adds the ICM42686p as a possible alternative to the existing IMUs.
Apparently, the ICM20689 was never actually used in these boards.
Signed-off-by: Julian Oes <julian@oes.ch>
Otherwise if is set to Offtrack, which in turn leads to weird behavior.
E.g. when triggering Land while flying fast forward, the vehcile doesn't
descned to the land point it keeps getting a velocity setpoint from the
smoother that pushes it away from the land point.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This switches all SENS_EN_INAXXX params to 0 by default and instead uses
the i2c_launcher command.
However, if any one the SENS_EN_INAXXX param is set, it no longer tries
to auto-configure it.
Signed-off-by: Julian Oes <julian@oes.ch>
actuator_motors and actuator_servos subscribers for offboard control
Being subscribers, this does not increase the computational load
over the uxrce_dds brigde
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@gmail.com>
New thrust and torque control mode added which replaces the previous
actuator mode,
in this mode the rate controllers are disables, the control allocator
is enabled and the used externally provices
vehicle_thrust_setpoints and vehicle_torque_setpoints.
New direct_actuator mode
In direct_actuator mode the control allocator module does not publish
actuator_motors and actuator_servos messages which must instead be
provided extrernally by the user.
Removed old direct mode.
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@gmail.com>
* Added parameter UXRCE_DDS_SYNCC to enable system clock synchronization. Refactored and cleaned up. Only set system time if it's off by more than 5 seconds (same as mavlink and gps).
Only re-evaluate if no longer in GF when flight mode is changed or
Loiter position was changed (eg through goto).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- check if Home would breach it (we do not want to upload GF that
do not allow to return to Home)
- check if the current vehicle position isn't contained in the GF
if armed (as then it would immediately trigger)
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- do reporting of breaching in-air only in geofenceCheck
- remove geofence_violation_reason_t
- replace geofence_breached field in GeofenceResult.msg with 3 fields
(one for each GF type: max dist, max alt, custom geofence)
- the warning message after breaching a GF is only done by Commander,
and it's specific to the GF type failure
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Use flow rates instead of integrals in backend. This allows us to delay
the data to the mitpoint integration time and simplifies the code in
general.
Gyro compensation can still be done in EKF2 if needed, but the
flow module normally already appends the correct gyro data to the flow
message.
* created a Performance Model for fixed wing vehicle
- added compensation for maximum climbrate, minimum sinkrate, minimum airspeed and trim airspeed based on weight ratio and air density
- added atmosphere lib to standard atmosphere calculations
---------
Signed-off-by: RomanBapst <bapstroman@gmail.com>
Co-authored-by: Thomas Stastny <thomas.stastny@auterion.com>
- in theory these are helpful to ensure EKF2_HGT_MODE configuration is
consistent with the relevant aid source (GPS, baro, etc), but it can
be a little awkward with users having to fight manual parameter
changes in the right order
nxp/rt117x:Fix Pin IRQ
nxp/rt117x:Support 4 i2c busses
nxp/rt117x:Add px4io_serial support
nxp/imxrt:Expand ToneAlarmInterface to GPT 3 & 4
px4_fmu-6xrt:Using imxrt_flexspi_nor_octal
px4_fmu-6xrt:Entry is start
px4_fmu-6xrt:Add Proper MTD
px4_fmu-6xrt:Set I2C Buses
px4_fmu-6xrt:Proper SPI usage
px4_fmu-6xrt:Adjust memory Map to use the 2 MB
px4_fmu-6xrt:Bring in ROMAPI
px4_fmu-6xrt:Push FLASH to 200Mhz
px4_fmu-6xrt:Use BOARD_I2C_LATEINIT
px4_fmu-6xrt:Clock Config remove unused devices
px4_fmu-6xrt:Remove EVK SDRAM IO
px4_fmu-6xrt:Enable SE550 using HW_VER_REV_DRIVE
px4_fmu-6xrt:Use MTD to mount FRAM on Flex SPI
px4_fmu-6xrt:Manifest
px4_fmu-6xrt:Restore board_peripheral_reset
px4_fmu-6xrt:Set I2C buss Interna/Externa and startup
nxp/rt117x:Set 6 I2C busses
px4_fmu-6xrt:Correct Clock Sources and Freqency Settings
px4_fmu-6xrt:Correct ADC Settings
px4_fmu-6xrt:Tune FlexSPI config and sync header with debug variant Linker prep for rodata ahb partitioning
px4_fmu-6xrt:FlexSPI prefetch partition split .text and .rodata
Current config
1KB Prefetch .rodata
3KB Prefetch .text
px4_fmu-6xrt:Run imxrt_flash_setup_prefetch_partition from ram with barriers
px4_fmu-6xrt:Use All OCTL setting from FLASH g_flash_config SANS lookupTable
px4_fmu-6xrt:Octal spi boot/debug problem bypass
px4_fmu-6xrt:Add PWM test
px4_fmu-6xrt:Fix clockconfig and USB vbus sense
px4_fmu-6xrt: Use TCM
px4_fmu-6xrt: Ethernet bringup
imxrt: use unique_id register for board_identity
px4_fmu-6xrt: update ITCM mapping, todo proper trap on pc hitting 0x0
px4_fmu-6xrt:correct rotation icm42688p onboard imu
rt117x: Add SSARC HP RAM driver for memory dumps
px4_fmu-6xrt: Enable hardfault_log
px4_fmu-6xrt: Enable DMA pool
px4_fmu-6xrt: fix uart mapping
px4_fmu-6xrt: enable SocketCAN & DroneCAN
px4_fmu-6xrt:Command line history TAB completion
px4_fmu-6xrt:Fix pinning duplication
px4_fmu-6xrt:Support conditional PHY address based on selected PHY
px4_fmu-6xrt:Add Pull Downs on CTS, use GPIO for RTS
px4_fmu-6xrt:Set TelemN TX Slew rate and Drive Strenth to max
px4_fmu-6xrt::Set TELEM Buffers add HW HS
px4_fmu-6xrt:Turn off DMA poll
px4_fmu-6xrt:RC_SERIAL_PORT needed to be px4io to disable rc_input using TELEM2!
px4_fmu-6xrt: bootloader (#22228)
* imxrt:Add bootloader support
* bootloader:imxrt clear BOOT_RTC_SIGNATURE
* px4_fmu-6xrt:Add bootloader
* px4_fmu-6xrt:bootloader removed ADC
* px4_fmu-6xrt:bootloader base bootloader script off of script.ld
* px4_fmu-6xrt:add _bootdelay_signature & change entry from 0x30000000 to 0x30040000
* px4_fmu-6xrt:hw_config Bootloader has to have 12 bytes
px4_fmu-6xrt:Default to use LAN8742A PHY
px4_fmu-v6xrt:VID Set to Drone Code
board_reset:Enable ability to write RTC GP regs
px4_fmu-6xrt:Fix CMP0079 error
rt117x:micro_hal Add a PX4_MAKE_GPIO_PULLED_INPUT
px4_fmu-v6xrt:Set CTS High before VDD_5V applided to ports to avoid radios fro entering bootloaders
fmu-v6xrt: increase 5v down time
fmu-v6xrt:Ready for Release DEBUGASSERTS off and Console 57600,
Bootloder updated.
imxrt:board_hw_rev_ver Rework for 3.893V Ref
px4_fmu-v6xrt:Move ADC to Port3
imxrt:117x Reuse all but io_timer_hw_description and imxrt_pinirq.c
imxrt:ADC & LPADC bifurcation and restructuring
imxrt:hrt support Up to GPT6
nxp/rt117x:adc Corrected
It was failing occationally with:
14:12:43 CMake Error at CMakeLists.txt:129 (list):
14:12:43 list index: 1 out of range (-1, 0)
14:12:43
14:12:43
14:12:43 CMake Error at CMakeLists.txt:131 (list):
14:12:43 list index: 2 out of range (-1, 0)
14:12:43
14:12:43
14:12:43 CMake Error at CMakeLists.txt:132 (string):
14:12:43 string sub-command REPLACE requires at least four arguments.
14:12:43
14:12:43
14:12:43 -- PX4 version: f513c40d62 (f513c40d62..NOTFOUND)
During a round corner the L1 distance calculation
was only done in 2D and the z-axis coordinate
was directly coming from the next waypoint.
This lead to an unpredictable altitude profile
between two waypoints. Sometimes almost all
altitude difference was already covered during
the turn instead of going diagonally.
The param is not really required anymore with the actuator
configuration. Also, when it is set to 0, RC doesn't work for some
boards which would be nice to avoid.
Signed-off-by: Julian Oes <julian@oes.ch>
The UART7 TXDMA services TELEM1 with flow control. If CTS is high, the
transmitting thread will wait on a semaphore, which may block other
threads from acquiring necessary resources to make progress, for
example, preventing MAVLINK instances from transmitting.
This change in NuttX makes the TXDMA acquire the semaphore in a
non-blocking way, preventing this issue.
- add functionality to specify world name for simulation in case name
- add test for triggering an airspeed invalidation in case of pitot blockage
- add test for high wind (ramped up wind over short period) to NOT invalidate airspeed in this case
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Increase wind process noise default (ASPD_WIND_NSD) and gate
(ASPD_TAS_GATE) to be able to catch rapid wind increases with
the internal wind estimator of the airspeed validator.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Trust the beta innovations more compated to the TAS innovations.
That should help with detecting real airspeed failures even with
a dynamic wind estimate (as long as vehicle doesn't fly straight)
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Increasing the wind process noise results in a more dynamic
wind estimation, which is capable of catching fast-varying
winds. As wind is used in the lateral guidance it's important
that we don't filter it too much.
Furher the gate of the airspeed fusion is increased, to
reduce the likelihood of airspeed fusion stopping due to
dynamic wind conditions. The airspeed is validated in
the airspeed validator (EKF consumes the validated one).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Reduces flash usage by ~16KB.
- compress formats at build-time into a single string with all formats
- then at runtime iteratively decompress using
https://github.com/atomicobject/heatshrink
- if not in air the accel noise is doubled
- if landed don't init unless GPS velocity is non-negligible
- when inactive continue seeding with EKF gyro bias
- reset yaw estimator if GPS fusion is stopped
Has options *None where the check is disabled, and *Warning, where only a warning is
published (which replaces the high wind warning from the COM_WIND_WARN limit).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* started tiltrotor port
* added advanced plane and changed some parameters on the tiltrotor
* added advanced plane
* removed tiltrotor for clean push
* removed standard vtol old model file
* removing the standard vtol changes from this PR, since it is not part of the advanced plane
* removed advanced plane meshes as they are already found in the rc_cessna
* updating and improving airframe parameters
* updated mesh paths
Signed-off-by: frederik <frederik@auterion.com>
---------
Signed-off-by: frederik <frederik@auterion.com>
Co-authored-by: frederik <frederik@auterion.com>
There is no reason to keep an uncertainty on the origin as it is then
already contained in the local position estimate when GNSS data is fused
in the filter.
The command is sent by a dedicated mavlink command and forwarded to the fixed wing position controller.
The pattern is defined by the radius of the major axis, the radius of the minor axis and the orientation. The pattern is then defined by:
The upper part of the pattern consist of a clockwise circle with radius defined by the minor axis. The center of the circle is defined by the major axis minus the minor axis away from the pattern center.
The lower part of the pattern consist of a counter-clockwise circle with the same definitions as above.
In between, the circles are connected with straight lines in a cross configuration. The lines are always tangetial to the circles.
The orientation rotates the major axis around the NED down axis.
The loitering logic is defined inside its own class used by the fixed wing position control module. It defines which segment (one of the circles or lines) is active and uses the path controller (npfg or l1-control) to determine the desired roll angle.
A feedback mavlink message is send with the executed pattern parameters.
- make it work again
- also supports Kconfig based clk source selection
- adapt to recent changes of default PWM limits
- support outputting in duty-cycle mode
- i2c addr use use hex representation
- revert back to common min/max value & move duty-cycle mode to advanced
We provide a latency measurement in the input capture handler.
However, since the timer was not enabled, none of the counter were
running therefore all counters were zero, thus latency was also zero.
Since the HRT is used to provide a timestamp, the lack of the running
timer was never noticed. After enabling the timer, latency now correctly
shows 9-10 counts.
* Fixed an issue where if the GCC_VER_STR would not contain the right NUTTX_GCC_VERSION, the grep -c command would throw a failure, silently exiting the entire ubuntu.sh setup script
This is a tool that can be used to generate advanced lift drag plugin parameters automatically using AVL. Rather than having to create .avl files yourself, pass them to AVL, read out the correct parameters and place them in the Advanced Lift Drag plugin, this tool will do all that for you and generate a complete advanced_lift_drag plugin sdf containing all necessary parameters for any vehicle. All that is required is to specify what the physical geometries of the vehicle are. The scripts are adaptable enough to support a self-selected number of control surfaces.
---------
Co-authored-by: frederik <frederik@auterion.com>
This switches from the horribly intertwined ringbuffer implementation to
the new VariableLengthRingbuffer implementation.
By ditching the previous implementation, we fix MAVLink message
forwarding, which didn't work reliably. The reason it didn't work is
that multiple mavlink messages could be added but only one of them was
sent out because the buffer didn't keep track of the messages properly
and only read the first one.
Signed-off-by: Julian Oes <julian@oes.ch>
This adds a reusable class for a FIFO ringbuffer that accepts variable
length packets. It is using the Ringbuffer class internally.
Signed-off-by: Julian Oes <julian@oes.ch>
* mission_base: reset inactivation index when user set a new mission index, or mission is reset.
* mission_base: check Climb required always on current mission item
This prevents accidental misconfiguration via I2C if there are
multiple SPI devices on the same bus. The device may hear it's i2c address
and write some register, while there is data transfer ongoing with another
device.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
WorkItemSingleShot::_sem is a signaling semaphore, disable PI for it.
Set CONFIG_DEBUG_ASSERTIONS=y and the kernel panics due to the semaphore
having no holder, disabling PI fixes this.
The head/tail semaphores are not used as lock but rather as resource
counters and thus relate more as signaling semaphores. Disable PI for
them.
I run my code with CONFIG_DEBUG_ASSERTIONS=y and the kernel panics due
to the semaphore having no holder, disabling PI fixes this.
This prevents the mavlink transmit loop from waiting on the module mutex
thus not stopping transmissions when the mutex is already taken.
This can happen when calling `mavlink status` from the mavlink shell,
where `Mavlink::get_status_all_instances()` takes the mutex and then
prints the status via pipes to the mavlink transmit buffer.
If that pipe cannot be emptied a deadlock happens.
Since the MavlinkReceiver thread also waits on the module mutex, both
reception and transmission of Mavlink packets are then prevented thus
disabling communications entirely.
IMUs were replaced in V3 with 2x ICM42688P. This configuration should
work with all revisions of Matek F743 Slim board, including V1, V1.5
and interim variant of V3 (ICM42688P + ICM42605).
Signed-off-by: Andrei Korigodskii <akorigod@gmail.com>
In the lightware_parser function, LW_PARSE_STATE2_GOT_DIGIT0 state can be repeated unexpectedly without proper parserbuf_index or state checking. This behavior will trigger a heap buffer overflow vulnerability by allowing to write some data. And the writable size is sizeof(unsigned).
* ekf2_derivation: use single source of state definition
The state is defined as an ordered dictionary of group elements and
everything else is generated using that state definition
* ekf2: generated state sample add const reference getter
---------
Co-authored-by: bresch <[brescianimathieu@gmail.com](mailto:brescianimathieu@gmail.com)>
Co-authored-by: Daniel Agar <daniel@agar.ca>
The length check for unknown packets did not include PACKET_SIZE_TYPE_SIZE
and CRC_SIZE, and hence working_index could overflow CRSF_MAX_PACKET_LEN,
triggering invalid memory access further down in QueueBuffer_PeekBuffer.
Also the working_segment_size was wrong for unknown packets.
Credits for finding this go to @Pwn9uin.
This allows to inject an updated ext_autostart folder with the name
ext_autostart_new, and then PX4 takes care of renaming it to
ext_autostart during bootup.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
The sparse vector template requires to know which states are non-zero in
the observation jacobian. This complicates the modularity of the code
when the state vector or the derivation is changed.
The computation cost difference is almost negligible for this size.
For FW takeoffs we need to keep the course straight towards the first
waypoint, and not climb in loiter patterns first.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Current threshhold has no margins of errors with the set acceptance radius. Increase corridor radius by 5m.
Increase time to disarm for tailsitter test to pass.
Rverse mission needs check tracks test disabled since mavsdk does not really work with mission has reverse order.
mavsdk-test: added a vtol mission without a landing at the end (used for reversed RTL)
math_helpers: added function to compute local position from a raw mission item
Signed-off-by: RomanBapst <bapstroman@gmail.com>
Dataman: Add write function to dataman cache.
RTL and mission have a new common base class mission_base. Both inherit from them and mission, RTL mission, and rtl reverse mission inherit from them and implement their desired functionalities. This simplifies the logic in mission as well as make the logic in rtl mission reverse and mission more readable.
Rtl mission reverse now functional again for VTOL flying back the mission and transitioning to MC at the home position.
Dataman cache has new write functionality to write to dataman while updating write item in its cache if necessary.
Dataman cache is now only updated when the respective module is active. Leads to a higher computation time once on activation, but decreases unnecessary cache updates when inactive.
This is required to support fixed pitch propeller helicopters that have
no collective but only cyclic pitch with two degrees of freedom and
hence only two servos.
The amount of thrust in the body z axis is then controlled using the
motor speed which makes particularly sense on coaxial helicopters
that need to control yaw with changing motor speeds already.
Events have a global, system-wide sequence number, which must be handled
atomically, (fetching and incrementing the sequence AND sending the event
to uORB must be atomic). Currently in FLAT mode, only one instance of this
sequence number exists, so it is OK to have it in px4_platform.
However, in PROTECTED mode px4_platform is instantiated both in kernel-
and user spaces, which makes two instances of this sequence number, which
causes problems in the mavlink event handling logic.
When mavlink receives and handles events, it expects that:
- The sequence numbers arrive in order (seq n is followed by n+1 etc)
- It increments by 1
- There is only one instance of the sequence number
In PROTECTED mode this is violated, as the kernel and user sequence
numbers run freely on their own. This patch fixes the issue by moving
the event backend to the kernel and by providing user access to it via
ioctl.
Even if there is some horizontal motion, a passing check should be
accepted as the terrain can be flat. However, the vehicle must not be
moving horizontally to invalidate the consistency as a change in terrain
can make the kinematic check temporarily fail.
A setpoint of type IDLE can be published by Navigator without a valid position,
and should be handled in the auto function in FW_POSCTRL_MODE_AUTO. If it wouldn't be
handled there the controller would switch to mode FW_POSCTRL_MODE_OTHER, in which case
no attitude setpoint is published and the lower controllers would be stuck with the
last published value (incl. thrust).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
but this time compared to
2fbb70d9ca
the minimum thrust is ramping to not produce a thrust jump.
The better long term solution will be to always have airmode but
with the effect limited to a magnitude of MPC_MANTHR_MIN.
high rates and (super) expo are necessary for
- acrobatic flying
- fpv
- racing
but they are not useful for:
- a typical vehicle's rate control tuning
- beginners
The defaults that I set in #8036 were put with the assumption Acro mode
is mainly used on racing drones for acrobatic manouvers but instead
users including me use it most of the time to tune any drone.
When tuning the rates are lowered and expo disabled.
My suggested strategy is to make the beginner friendly rates without
expo which are good for tuning the default and document more clearly
how this can be raised for acrobatic flying in documentation and later
on hopefully also directly in the ground station UI
on an Acro mode specific page.
* FW Positon Controller: set altitude_ref to 0 if not provided by GPS
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Positon Controller: set lat/lon reference to 0 if not provided in local_position
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
If no altitude setpoint is specified then takeoff to the default
altitude above the current global position, not above Home.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
c23b72dffe [BACKPORT] sched/semaphore: Remove restriction to use nxsem_trywait from ISR
fd47cd20a2 [BACKPORT] imxrt:Serial Preserve all but W1C bit in SR
c55f0fd3ac [BACKPORT] imxrt: lpspi dma invalidate cache after exchange
198c7caecb [BACKPORT] imxrt:lpi2c fix status handeling & race
cbd2e44c10 [BACKPORT] s32k3xx: lpspi dma invalidate cache after exchange
e71618d60e [BACKPORT] s32k3xx:lpi2c fix status handeling & race
6f59cc3659 [BACKPORT] s32k1xx:lpi2c fix status handeling & race
1e316d7e32 [BACKPORT] imxrt: flexcan use hpwork for receiving frames
67c1c59865 [BACKPORT] net/can can_readahead_timestamp always free iob
8be831a4ff [BACKPORT] imxrt: fix txdeadline add ecc/fd support
00a68b7668 [BACKPORT] fs/cromfs: Fix faulty DEBUGASSERT() check
d5cf545d6e [BACKPORT] S32K3XX EMAC MCAST support Fix compile warning when ioctl is not enabled
4265c830fa [BACKPORT] imxrt:edma {s|d}last needs to be total xfer size
24b4d44896 [BACKPORT] s32k3xx:edma {s|d}last needs to be total xfer size
eed0482f64 [BACKPORT] s32k1xx:edma {s|d}last needs to be total xfer size
36aab4146a [BACKPORT] kinetis:edma {s|d}last needs to be total xfer size
a0faf31f6f [BACKPORT] arch/stm32f7: fixes for pinmap
eb8255121d [BACKPORT] stm32h7:sdmmc It is not an error if no wait was needed
062044fe41 [BACKPORT] board nucleo-h743zi:Rework board.h not use CONFIG_STM32_USE_LEGACY_PINMAP
e03f9d3917 [BACKPORT] board olimexino-stm32:Rework board.h not use CONFIG_STM32_USE_LEGACY_PINMAP
4c3a467415 [BACKPORT] stm32l5:pinmap Add suffix to all pins and add legacy pinmap
153069ed40 [BACKPORT] stm32wb:pinmap Add suffix to all pins and add legacy pinmap
d84d737f89 [BACKPORT] stm32f0l0g0:stm32f0{3|5|7|9}x_pinmap & stm32g0_pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
5fc7071ac1 [BACKPORT] stm32l4:stm32l4x{3|4|5|6|r}xx_pinmap pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
20061c2aab [BACKPORT] stm32:stm32f10{0|2|3{c|r|v|z}|5{r|v}|7v}_pinmap refactor
4d1f83d484 [BACKPORT] stm32:stm32l15xxx pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
2dfa3f2601 [BACKPORT] stm32:stm32g4xx{c|k|r|m|v|q} pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
d206327809 [BACKPORT] stm32:stm32f3{0|3|7}xxx pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
95e66ab508 [BACKPORT] stm32:stm32f20xxx pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
d2fd9178ad [BACKPORT] stm32:f4/f412 pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
a9df45166d [BACKPORT] stm32f7:pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
07dd2b424e [BACKPORT] stm32h7:pinmap Remove GPIO_SPEED_xxx and add legacy pinmap
1e3065344f [BACKPORT] stm32u5:stm32u585xx_pinmap Fix typo
0a05365a90 [BACKPORT] stm32wl5:pinmap Fix typo
e3834138dc [BACKPORT] tools:Add STM32 Pin migration tool
df851a8768 [BACKPORT] stm32h7/rcc: make VOS0 configurable from board.h also for stm32h7x7xx
d75dfcf1e9 [BACKPORT] stm32h7/rcc: make VOS0 configurable from board.h
963f35f4fc [BACKPORT] {stm32,stm32f7,stm32h7,stm32l4,efm32}/otg: rasie an assertion if IN request is not possible to transfer
de2fcc6668 [BACKPORT] {stm32f7,stm32h7,stm32l4}/sdmmc: callback support requires HPWORK
6929144fc2 [BACKPORT] stm32h7/otgdev: FS transceiver must be enabled if OTGFS enabled
a2078afaea [BACKPORT] stm32h7/otg: add support for external ULPI
26e1246c86 [BACKPORT] stm32h7/rcc: OTGHS ULPI works only in VOS0
cd6daa185e [BACKPORT] stm32h7: update ULPI pins
c73c261ae3 [BACKPORT] arch/boards: fix stm32f411-mininum:nsh compilation failure after enabling IRQMONITOR
8078f134ef [BACKPORT] arch/stm32/stm32.h: do not include stm32_usbdev.h if not supported
60e884fa92 [BACKPORT] {stm32,stm32l4,stm32f0l0g0}/otg: move STM32_NENDPOINTS definitions to header files
dda297cb78 [BACKPORT] arch/arm/src/stm32/hardware: Fix register define
362b976b0e [BACKPORT] arch/arm/src/stm32/hardware: Add stm32g4 rcc apb1 timer enable compatibility
434fd71f2c [BACKPORT] {stm32/stm32l4/stm32f7/stm32h7/efm32}/otgdev: remove invalid use of the priv field for EP
2476d24e8c [BACKPORT] {stm32f7,stm32h7}/otg: fix compilation for USBDEV when USB_DEBUG=y
eb43c582ea [BACKPORT] drivers/mtd/ramtron: change nsectors size to uint32
20f61ff0d5 [BACKPORT] fs/littlefs: add full support for LittleFS block device cfg in Kconfig
60471fbf8c [BACKPORT] drivers/mtd: add Kconfig options for RAMTRON emulated page & sector size
- for FW the primary range finder usage is estimating the distance to
the ground (terrain estimate) to guide the landing approach and initiate flare
- increase EKF2_RNG_QLTY_T default 1->3 seconds
- force initialize takeoff airspeed setpoint at start of takeoff modes
- force set airspeed constraints if slewed value is out of bounds
- always slew airspeed setpoints as long as inside constraints
- move target airspeed setpoint calculation into mode specific logic regions (hand vs runway)
when we're in a takeoff situation, we only want to adapt the airspeed to
avoid accelerated stall due to load factor changes. Disable othre logic
like minimum ground speed, wind based adaption and airspeed slew rating.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
From looking at the history the BMM150 rotation was initially 0. Then,
this was changed to 6 when the intent was to only change it for Skynode.
A bit later, the rotation was changed back to 0, but only for Skynode.
This tells me that rotation 0 was correct for all 6X including Skynode
all along.
Signed-off-by: Julian Oes <julian@oes.ch>
- ekf2 heading first initializes to 0 degrees, then immediately resets to mag heading once a few samples are accumulated
- the yaw standard deviation check could be adjusted to exclude this brief (<1s) initial period
- on NuttX if the SD card was just cleared (quick format) it's possible for the newly created uninitialized dataman file to still have the previous compatibility key, which causes the initialization to be skipped
* added a new driver ASP5033 for measuring the differential pressure and airspeed
---------
Co-authored-by: nano <nanobotzhe@abv.bg>
Co-authored-by: Denislav Petrov <you@example.com>
When the yaw estimate is converging, the controller makes the drone yaw
in order to follow the current setpoint. This is unintuitive for the
pilot and it is preferable if the drone continues to fly towards the
same physical direction.
This estimate doesn't converge to the true yaw but can be used as a
more consistent but drifting heading source.
It can be used by a setpoint generator to adjust its heading setpoint
while the true yaw estimate is converging in order to keep a constant
course over ground.
This updates the gps/devices submodule which includes a fix that
requests the Unicore HEADINGA message in case the message is not enabled
by default.
Signed-off-by: Julian Oes <julian@oes.ch>
If the measured voltage is more than 65v we need to split the voltage
over multiple cells in order to avoid overflowing the uint16. This is
according to the MAVLink spec.
Signed-off-by: Julian Oes <julian@oes.ch>
The extension fields need to be 0 by default according to the MAVLink
spec. This is because extensions are 0 by default and need to be 0 when
unknown/unused for backwards compatibility.
The patch also simplifies the flow slightly in that it doesn't create a
temporary array but just fills in the cell voltages directly.
Signed-off-by: Julian Oes <julian@oes.ch>
Fixes the clang error:
/__w/PX4-Autopilot/PX4-Autopilot/src/drivers/drv_hrt.h:297:35: fatal error: identifier '_s' preceded by whitespace in a literal operator declaration is deprecated [-Wdeprecated-literal-operator]
297 | constexpr hrt_abstime operator "" _s(unsigned long long seconds)
- Enable arbitrary euler angle for Mag rotation
- new CUSTOM rotation enum out of the normal enum range
- mag_rot: automatically change to custom if euler rot is set
- sensor_calibration: Magnetometer save custom rotation parameters
- mag_cal: cross mention rotation parameters
- This allows the user to see the RPY options when searching for the rotation parameter
---------
Co-authored-by: Junwoo Hwang <junwoo@auterion.com>
Co-authored-by: bresch <[brescianimathieu@gmail.com](mailto:brescianimathieu@gmail.com)>
Co-authored-by: Daniel Agar <daniel@agar.ca>
During transition to FF, only allow update thrust axis up to 45° as with a high tilt angle
the effectiveness of the thrust axis in z is apporaching 0, and by that is increasing
the motor output to max.
Transition to HF: disable thrust axis tilting, and assume motors are vertical. This is to avoid
a thrust spike when the transition is initiated (as then the tilt is fully forward).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
EffectivenessTiltrotor: link time when to tilt motors to MC position to COM_SPOOLUP_TIME
- remove VT_TILT_SPINUP and special spin up tilt handling form the VTOL module
- now handle the spoolup in the allocation, directly linked to COM_SPOOLUP_TIME
- leave tilts at disarmed value during spoolup
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- when GNSS is used require low mag heading innovations during
horizontal acceleration (yaw observable) to validate the mag
- only fuse mag heading just enough to constrain the yaw estimate
variance to a sane value. Leave enough uncertainty to allow for a
correction when the yaw is observable through GNSS fusion
In the mavlink_receiver code, after a while it will try to resend some
parameter update through the MAVLink instance. But for Iridium links
those are not a good idea. So this adds a condition that prevent the
sending if the MAVLink instance is in Iridium mode.
Related to issue #21496
sem_wait() can be interrupted if the task receives a signal, however
the blockinglist implementation depends on blocking until the semaphore
can be obtained.
Fix a memory overflow in case SENS_BOARD_ROT is set to Rotation::ROTATION_MAX (41) which is not a valid value
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
* Navigator: DO_CHANGE_SPEED: only store sinlge cruising_speed_current_mode
This stored cruising speed setpoint is reset on mode change and
after a VTOL transition.
* Navigator Mission: replay DO_CHANGE_SPEED items when resuming mission
* Navigator: remove cruising_speed_sp_update()
Speed changes in a mission are handled directly in the position controllers,
and no longer in Navigator.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Increased size for ORB_ID from uint8_t to uint16_t
Created a type: orb_id_size_t = uint16_t.
There are still a couple of places where the size
of the ORB_ID is assumed to be less than 16-bits.
The places that I have found are commented regarding
this and can be found with a search on orb_id_size_t.
- split mag_3d into new standalone mag fusion and mag fusion allowed to update all states (full mag_3d)
- new dedicated control logic for mag/mag_3d fusion and standalone mag heading fusion
- if WMM available use for mag_I and mag_B init
- mag states reset if external yaw reset (yaw estimator, GPS yaw, etc)
- mag reset if declination changed (eliminate _mag_yaw_reset_req)
- mag fusion (but not mag_hdg or mag_3d) can be active during gps_yaw or ev_yaw (if yaw aligned north)
Co-authored-by: bresch <brescianimathieu@gmail.com>
- simplify vehicle_status.arming_state down to just armed and disarmed
- ARMING_STATE_INIT doesn't matter
- ARMING_STATE_STANDBY is effectively pre_flight_checks_pass
- ARMING_STATE_STANDBY_ERROR not needed
- ARMING_STATE_SHUTDOWN effectively not used (all the poweroff/shutdown calls loop forever in place)
- ARMING_STATE_IN_AIR_RESTORE doesn't exist anymore
- collapse ArmStateMachine into commander
- all requests already go through Commander::arm() and Commander::dismarm()
- other minor changes
- VEHICLE_CMD_DO_FLIGHTTERMINATION undocumented (unused?) test command (param1 > 1.5f) removed
- switching to NAVIGATION_STATE_TERMINATION triggers parachute command centrally (only if armed)
---------
Co-authored-by: Matthias Grob <maetugr@gmail.com>
* ekf2-test: remove outdated codegen comparison
The definition of states changed so the comparison with the old
derivation cannot work anymore.
---------
Co-authored-by: bresch <brescianimathieu@gmail.com>
Some system are able to dead-reckon for a while after losing GPS or
other sources providing positional feedback. If the estimated position
error grows above the failsafe threshold, the system enters a failsafe
mode. As the position error estimate is growing linerly over time, and
it is recommended to take action before entering the failsafe, we here
warn the user about the imminent failsafe and propose to take manual
control.
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* [not-working] add a gz-omnicopter model
* Fix axis directions on omnicopter model
The omnicopter joint axis directions had to be adapted for sdf 1.9 as it has different conventions for joint axis definitions.
* include model from gz-fuel & remove mesh files
* Fix omnicopter model using fuel
---------
Co-authored-by: Jaeyoung Lim <jalim@ethz.ch>
If at the last powercycle one mission was uploaded, the counter in dataman was 1. On the next powercycle the mavlink mission counter was reset to zero and on first mission upload updated to 1 again. Other modules check, if the mission was changed based on the counter, like the mission.cpp loaded the mission counter from the dataman. On a new mission, the comparison of the counters failed, because both were the same value even if the mission was completely different.
This can be useful if using a full EV + GNSS setup and you start indoors, then fly outside. Once GPS is good and the only missing requirement is yaw alignment the yaw estimator reset is performed and EV yaw will automatically stop itself.
There is no reason to contrsain the position uncertainty, estpecially
when flying with velocitiy aiding only.
Note that all the variances are already contrained to sane values at the
end of the covariance prediction
- set vision attitude error filter uninitialized if vision data stops
- ev error filter is only compiled when ev config is selected
Co-authored-by: bresch <brescianimathieu@gmail.com>
A version update is needed since the dataman is showing errors if data doesn't exist or if it is wrongly stored. This will force default data to be initialized.
There is some actual changes because the earth mag field states are now
reset using the WMM when available. In the replay log, the measured mag
field does not match the WMM and this is why there is a large diff. It
is however more correct now than before.
because sprintf is deprecated on MacOS
and CI fails with warning -> error:
'sprintf' is deprecated: This function is
provided for compatibility reasons only.
Due to security concerns inherent in the
design of sprintf(3), it is highly
recommended that you use snprintf(3) instead.
I updated all versions to the newest one that's used anywhere.
Then at least the straing can be found with full text search.
It's another step towards understanding and unifying the CI environment.
Since we changed the threshold for allowing arming from TRIM/2 to AIRSPEED_MAX
is is very unlikely that one needs to disable this check.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Experience from tuning different VTOL backtransitions showed only having a
I controller is beneficial over a combined FF/I breaking controller.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Case: A vehicle is already operating but has no stick input or another
source than RC. When RC stick input is switched to either because it gets
first time available or as a fallback to joystick then the mode was
immediately changed to the switch position. This can lead to loss of
control e.g. when the vehicle is flying far away and the
mode switch of the RC is in some fully manual piloted mode.
I added tests to cover the cases where RC mode initialization is expected
and also unexpceted because the vehicle is already armed.
modules__ instead of module__ prefix
The module depends on the hysteresis library and probably because
it compiles with the still prevalent global includes the dependency
is not declared.
also for the other linux target builds. It's a follow up to
e5503480e3
I wasn't aware that there are multiple different
container versions used for almost the same build.
in air bias estimation is usually really accurate and should be weighted
more heavily compared to the calibration parameters that are often
more approximate given the worse magnetic environment near the ground.
- since last_us is set to 0 every time the bias is not observable, the
total time was also reset -> needed 30 consecutive seconds in mag 3D
to be declared "stable"
- after landing, the mag_aligned_in_flight flag is reset. Using this for
bias validity makes it invalid before we have a chance to save it to
the calibration.
Opt flow raw innovations can be really large on ground due to the small
distance to the ground (vel = flow / dist). To make the pre-flight check
more meaningful, scale it with the current distance.
Channel values stay over one unit test but some tests assumed they are
reset each time. Reset the channel after these mode button tests.
Parameters survive between unit tests presumably as long as
the bianry runs. Reset them if a test requires that.
This allows PWM and all other output methods to configure
stopped, idling and full thrust points and use them consistently.
The fact that a fixed wing motor can be stopped when zero thrust
is demanded is explicit and could in principle even be disabled.
The mechanism is the same as for a standard VTOL stopping the
multicopter motors in the fixed wing flight phase.
This allows to consistently define:
Motor stopped - disarmed PWM
Motor idling - minimum PWM
Motor at full thrust - maximum PWM
Any allocation can then distinctly decide if
a motor should be running or not depending
on the context and also explicitly command that.
This is required to process data from the ADS1115 ADC and enables the
params BATx_I_CHANNEL and BATx_V_CHANNEL.
Testing is required whether this actually works on Pixhawk 6X though.
Signed-off-by: Julian Oes <julian@oes.ch>
When switching into Hold mode establish a Loiter around current position,
even if we were before already loitering (eg in Mission mode).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Having newline between I think didn't apply the PR stale days setting of
30 properly (it was using 45)
And it seemd that unless I set the days-before-close to -1
intentionally, it would still close the Issue/PRs, as the default value
is 7 already.
Also updated version of the stale action to v8
* Remove old stale bot yaml file
* Remove slack svg file (unused)
* Stale action: Only apply stale label, and no other actions
Respecting the opinion on
https://github.com/PX4/PX4-Autopilot/commit/fa9ac6ecf651232f3105ca124a2d2c54ab8620d0,
it seems reasonable to disable commenting feature, and just keep the
action only applying the `stale` label. This will reduce the noise /
email spams / ping pong fight with the stale bot (Action)
- Remove unnecessary text rendering, which made clicking log links
harder
- Re adjust markdown elements to make it easier to read / intuitive when
creating an issue
Instead of hard-coding the tilting duration (from FW to MC tilt),
expose it as a parameter (VT_BT_TILT_DUR). The default is the same
as the hard-coded value previously (1s).
Slower tilting mechanisms need a higher value here, while for smaller
ones a too high value results in unnecessary delays till the motors
are in hover configuration.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This affects how soon after a backtransition the vehicle has the
full hover controller running again. Specifically it reduces the
duration of the ramp down of the motors prior to tilting them.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
COM_ARMABLE is set to "Disabled" will prevent arming.
This allows to set the parameter when ground demoing a drone or
if it's in maintenance for safety reasons.
const char *data = "www\r\n";
Defines a cstring of 6 bytes: 'w', 'w', 'w', '\r', '\n', '\0'
type of data: char const*
type of &data: char const**
So when we call
write(_fd, &data, strlen(data));
then strlen(data) == 5
and we send the 4 byte memory address of data
+ some additional random byte.
Correct is
write(_fd, data, strlen(data));
where char const* gets casted to const void * and we pass
the pointer to the content of data.
The fundamental problem here is write() not being typesafe.
Before this the ESC calibration aborts if battery detection doesn't work.
The problem is if the user still connects the battery as he gets instructed
and the calibration aborts then the ESCs are in calibration mode and
after abortion calibrate to the wrong value.
Also I realized there's no additional safety by aborting the calibration
if the battery cannot be detected during the timeout because a pixhawk
board without power module will report a battery status from the
ADC driverand in it that no battery is connected which is the best
it can do. In this situation the motors will still spin if the
ESCs are powered.
Some ESCs e.g. Gaui enter the menu relatively quickly if the
signal is high for too long. To solve that it's kept high shorter.
Also all tested ESCs detect the low signal within a shorter time
so no need to wait longer.
- Change timings for a more reliable calibration.
- Make sure there's an error message when battery measurement is not
available also when it gets disabled after system boot in the power
just above the calibration button.
- Safety check if measured electrical current goes up after issuing the
high calibration value for the case the user did not unplug power
and the detection either fails or is not enforced.
Before:
When the mixed throttle for the motor was exactly zero the ramp went
from the disarmed PWM value to the minimum PWM value.
When the throttle was even just slightly different from zero the ramp
made a jump up to the commanded throttle scaled between
disarmed PWM and maximum PWM, then ramped between
disarmed PWM and minimum PWM and at the end jumped up again to
the commanded throttle scaled between minimum PWM and maximum PWM.
After:
The ramp goes from disarmed PWM value to the the
commanded throttle scaled between minimum PWM and maximum PWM.
If the commanded throttle changes during the ramp then the scale and
hence also end value of the ramp changes.
When at rest, directly fuse the gyro data as an observation of its bias.
This allows to strongly observe the gyro biases without having to fuse a
constant heading that makes the ekf too confident about its heading.
The entire logic did not work for the case when the throttle channel is
reversed because then QGC sets trim = max for that channel and
the result is only half the throttle range.
This adds support for the TI LP5562 RGB LED driver.
Things to note:
- The driver is initialized in simple PWM mode using its internal clock,
for R,G,B, but not for W(hite).
- The chip doesn't have a WHO_AM_I or DEVICE_ID register to check.
Instead we read the W_CURRENT register that we're generally not using
and therefore doesn't get changed.
- The current is left at the default 17.5 mA but could be changed using
the command line argument.
Datasheet:
https://www.ti.com/lit/ds/symlink/lp5562.pdf
Signed-off-by: Julian Oes <julian@oes.ch>
* Removed obsolete voxl2-io directory
* Updated support for ModalAI FC v2 board
* Added UAVCAN back in and removed local position estimator and attitude estimator Q that are no longer supported.
* Removed unneeded IMU drivers
set_vtol_transition_item sets the params of the mission item directly
to values that make sense for NAV_CMD_DO_VTOL_TRANSITION, but don't
for other NAV_CMDs. So make sure that whenever we use it, we then in
the next step reset the touched mission_item fields.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Our serial_test showed only ~84kB/s with the default 256 RX buffer size
with significant ~2.5ms periods of the flow control RTS pin being
asserted. Increasing size to 600 (same as FMU-v5x) brings the throughput
only to ~190kB/s, while a size of >1500 achieves ~350kB/s. Larger RX
buffers do not increase throughput anymore, while the theoretical
maximum is 375kB/s.
Transmit buffer size is increased to 10kB same as on FMUv5x to prevent
any future differences in queue behavior and throughput. serial_test
showed ~350kB/s throughput at 3kB TX buffer size, so this is just a
precaution.
* refactored uncommanded descend quadchute
- use fixed altitude error threshold
- compute error relative to higest altitude the vehicle has achieved
since it has flown below the altitude reference (setpoint)
* disabled altitude loss quadchute by default
* altitude loss quadchute: added protection against invalid local z
---------
Signed-off-by: RomanBapst <bapstroman@gmail.com>
- ArmingStatus DroneCAN message STATUS field is only set to true based on
arming_status.armed
- Cannode prearm state is set to true always when ArmingStatus DroneCan
message is received
- Instead we disable interrups on Nuttx where needed
- No lock is held during param export. Params can be changed
concurrently and we rely on the fact that another export will be
triggered in that case.
5s is a more reasobale time for tailsitters, which rely differently on this param
than other VTOL types. Tailsitters will ramp the pitch up withing this time,
while for other VTOLS types its only the max transitiont time.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This is now using the advanced lift drag plugin.
The important step was to enable airmode for yaw, otherwise yaw gets
saturated at low throttle and we can barely roll.
The other trick was to raise airspeed a little bit to avoid operating
too much at the lower end of throttle where control authority is low.
Signed-off-by: Julian Oes <julian@oes.ch>
* RateControl: rename setGains to setPidGains to be more precise
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FW Rate controller: only allow to disable Yaw in Acro, not roll and pitch
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- re-enable once the estimator selector respects configured mag
priority (at least initially) or is otherwise able to automatically
prefer an external mag over internal
- for SITL disabled because the full matrix of esitmator instances
(IMUs X mags) was too many topics for logger currently
Transitions in Stabilized mode are done manually, the pilot controls the pitch angle
and if it's above the threshold the transition is declared finished (plus airspeed
check for front transition). Thus we can't have fixed thresholds but need to link
them to the actual max pitch angle in Stabilized mode.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
We currently fuse 0 as airspeed rate measurement, and thus simply low-pass
filter the airspeed measurements. Testing has shown that the current default
on the airspeed rate measurement noise is set to low, and thus the airspeed
mesurement is filtered too much.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
A VTOL plane in MC mode has no yaw setpoint during takeoff because of
weather-vane. To align for the front transition, the yaw target jumps
and caused a step in the controller, making it reach saturation.
With this commit, the previous yaw setpoint is set to the current yaw
when no yaw setpoint is sent in order to create a smooth yaw trajectory
starting at the current orientation when yaw target is suddenly finite.
The yawspeed filter also now contains the yaw speed instead of dyaw in
order to prevent chattering due to dt jitter.
The API is cleaner if the control_data is const reference and the device
compid is an explicit output argument.
Signed-off-by: Julian Oes <julian@oes.ch>
The resets in the modes (eg Loiter mode) are not active yet, so manually
set rep->current.cruising_speed = -1.f if not already in the same flight mode.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
When flow control is used together with DMA, we need to add a pulldown
to CTS. Without it, it assumes flow control and gets stuck when
CTS is not connected.
Signed-off-by: Julian Oes <julian@oes.ch>
When we fall back to another link, we are already doing a uORB copy when
checking the data. Therefore, we should further down use/send that data
instead of overwriting it immediately.
Signed-off-by: Julian Oes <julian@oes.ch>
I don't think it makes sense to slow down switching RTCM injection to
once every 5 seconds. If we don't have corrections, we should check and
use whatever we get as soon as possible.
Signed-off-by: Julian Oes <julian@oes.ch>
We need to reset the instance after looping through all instances.
Otherwise, we are left with the last instance if none is valid but have
not updated the _selected_rtcm_instance which is what is logged.
Therefore, this change fixes the logged RTCM instance.
Signed-off-by: Julian Oes <julian@oes.ch>
It turns out that I had omitted implementing the gimbal_device_id which
is the component ID of the gimbal device that the gimbal manager (in
this case PX4) is responsible for.
Signed-off-by: Julian Oes <julian@oes.ch>
Once passed the transition and in FW mode, it takes some ms unitl the FW att sp is updated
by the FW att controller. During this time the last published attitude sp is kept bein used,
which is the one that was published during transition. So let's fill the thrust[0] of it.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This adds the ability to manually set who has primary control over a
gimbal and also adds printing it as part of the status.
This is helpful, especially while working on the QGC gimbal v2
implementation.
Signed-off-by: Julian Oes <julian@oes.ch>
A default of 10s makes more sense compared to the old 4s, as this is only the
max time of a back transition and it's also decleared completed if the
speed drops below the hover cruising speed (MPC_XY_CRUISE).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
There have been substantial TECS changes lately, and the config files
for custom VTOLs have not been updated since a long time. I would thus
rather use the default TECS gains for lightly maintained configs.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
with the goal to make it more clear and the error only appear when
armed but every time the running task doesn't actually match
the mapping inside the start_flight_task() function.
* Navigator: on_mission_landing() only can return true if currently in mission mode
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* RTL: reset RTL state when not in RTL nav_state
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Navigator: fix mission vs. normal RTL
- remove extra state _should_engange_mission_for_landing from rtl and have
this logic outside of RTL where Navigator decides on running mission RTL or normal RTL
- fix logic in Navigator to decide mission RTL vs normal RTL
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Mission: land_start(): fix decision if already on mission landing
Simply checking landing() is not enough, as that is not reset until
set_current_mission_index(get_land_start_index()) later in the function.
Instead ask Naviator about it (on_mission_landing()).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Navigator: only update _shouldEngangeMissionForLanding once, to not set it to false after VTOL backtansition
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Port of the PX4 Vision gazebo-classic model to Gazebo. With Gazebo Garden, everything seems functionnal.
* Port of the PX4 Vision gazebo-classic model to Gazebo.
With Gazebo Garden, everything seems functionnal.
* Added airframe for command make px4_sitl gz_px4vision
* revert author field
* Import model from Ignition Fuel instead of defining it directly in the .sdf
Tests showed that sideslip fusion often starts just before airspeed
fusion, resetting the wind states to 0 instead of using the airspeed
data. A quick look at the wind uncertainty allows to know if the current
wind estimate is meaningful or if we should rather reset it using the
airspeed data.
- if the height rate input into TECS is finite, use that one to
update a velocity reference generator
- if the velocity reference generator reports position locked or
height rate input is not finite, then run altidue reference generator
- run altitude controller only if altitue is controlled
if height_rate_setpoint is set, only use that one to update altitdue trajectory
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* FWPositionController: add linear airspeed-to-trim-throttle mapping
Scale trim throttle in a linear way with airspeed.
Different gradients above/below trim airspeed.
The airspeed- and airs density-corrected trim throttle is required for an accurate
throttle prediction in TECS, while the default trim throttle is still needed
as well, as the throttle delta from height rate setpoint delta is linked to it.
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Navigator: change way of telling logic if RLT was started just now
* Navigator: change logic around when to engage Mission mode for RTL
To find out if we're currently on a mission landing, check if we either are
past the land start marker OR currently land start marker is current WP
and vehicle is already in LOITER mode.
* Navigator: do not engage RTL at all if already on mission landing
* Navigator: consider to be on mission landing if the LOITER_TO_ALT and dist small
To find out if we're currently on a mission landing, check if we either are
past the land start marker OR currently land start marker is current WP,
the type is LOITER_TO_ALT and the vehicle is inside loiter.
---------
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This fixes the case where the hardware detection fails on CubeBlack when
a CAN device is connected to CAN1 and talking, and therefore preventing
the check from getting a clear result.
Signed-off-by: Julian Oes <julian@oes.ch>
This consolidates the version/revision detection function.
This should allow for actual changes in a follow up commit.
Signed-off-by: Julian Oes <julian@oes.ch>
- multi-instances support is removed from the parameter definitions.
- XRCE_DDS_AG_IP allows to define the agent IP when udp transport is used.
The parameter is used by default if the module is started without the -h argument.
- XRCE_DDS_PRT allows to define the agent listning udp port when the
udp transport is uded. The parameter is used by default if the module is started
without the -p argument.
- Tools/convert_ip.py assists in converting ips in decimal dot notation into int32
notation.
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
To prevent mission rejections when vehicle is currently far away from Mission start,
but planned home is close to it.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* MAVSDK_Test_Runner: Place PX4 instance runner after Gazebo server runner
- This was a nasty bug where starting PX4 instance first, then starting
Gazebo server was causing PX4 instance' EKF to freak out, probably
because it doesn't like getting data a while after it is started
- Detailed breakdown is given here: https://github.com/PX4/PX4-Autopilot/issues/21229#issuecomment-1450761542
- This now guarantees that such edge case won't occur and MAVSDK test
will run as it should
* MAVSDK Test Runner: Retain comment within 79 character limit
- To pass flake8 python style check
* Update test/mavsdk_tests/mavsdk_test_runner.py
Co-authored-by: Julian Oes <julian@oes.ch>
---------
Co-authored-by: Julian Oes <julian@oes.ch>
Testing has shown that 10m is a bit too tight, most of all as this check is also run in not height-controlled
flight modes (eg Stabilized), and there 10m altitude loss during transitions can happen easily.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This adds the Cube Orange+ to the list, and also changes the Hex naming
to CubePilot as that is how it is sold/marketed now.
Signed-off-by: Julian Oes <julian@oes.ch>
The attitude frame is wrong for tailsitters doing side slip fusion for wind estimation.
It doesn't take into account that the frames is 90deg tilted in FW flight.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
-set roll/pitch used for failure detection during transition to 0
-rotate estimated attitude 90° in FW flight
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
to enable proper automatic output in gtest unit tests
that compare two `Matrix`, `Vector` or two `Dual` objects.
Credits to @jwidauer for showing me this trick.
Maximum plaintext length was wrong, it is just k - 2 * hLen -2, where k is RSA key modulus length and hLen is hash length
Also minimum block that can be encrypted didn't make sense for RSA-OAEP, it is just 1 byte, the rest will be padded.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
Bootloader needs to have a mechanism to de-initialize crypto, in case some HW accelerator
is being used. This adds the needed function for it
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
Camera controls should not happen through the flight controller, and
the control allocation has no means of controlling the camera zoom.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- add unit tests for adsb conflict detection
- move adsb conflict detection to lib/adsb and adsb conflict class
- use containers/Array.hpp for buffer array
- expand fake_traffic
This adds a parser for Unicore sentences sent in-between NMEA sentences.
This brings support for the Unicore UM982 module which publishes heading
information based on the two antennas.
Signed-off-by: Julian Oes <julian@oes.ch>
If NuttX is built without support for SMPS it can brick the hardware.
Therefore, I suggest that we add this additional compile-time check.
Signed-off-by: Julian Oes <julian@oes.ch>
It turns out that when you rotate by 45 degrees, as required on the
CubeOrange+, then you can easily get into clipping because the vector
components are constrained after the rotation. In order to avoid that,
we have to avoid getting close to the int16 range and switch from 20 bit
resolution to 16bit resolution earlier.
Signed-off-by: Julian Oes <julian@oes.ch>
This switches from attitude_estimator_q to EKF2 which should now work
without mag when the params are set to SYS_HAS_MAG = 0 and
EKF2_IMU_CTRL = 7 to enable gravity fusion.
Signed-off-by: Julian Oes <julian@oes.ch>
PX4_SIM model need the simulator (gz_) prefix
Fix post debug task
Add x500_depth, rc_cessna, standard_vtol
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
- Add the possibility in the parser to replace the defines made in the current file with their argument (includes are not supported)
- Add the possibility for the parser to parse int argument with bitwise shift operators
- remove deprecated actuator_controls[INDEX_FLAPS/SPOILERS/AIRBRAKES]
- use new topic normalized_unsigned_setpoint.msg (with instances flaps_setpoint
and spoilers_setpoint) to pass into control allocation
- remove flaps/spoiler related fields from attitude_setpoint topic
- CA: add possibility to map flaps/spoilers to any control surface
- move flaps/spoiler pitch trimming to CA (previously called DTRIM_FLAPS/SPOILER)
- move manual flaps/spoiler handling from rate to attitude controller
FW Position controller: change how negative switch readings are intepreted
for flaps/spoilers (considered negative as 0).
VTOL: Rework spoiler publishing in hover
- pushlish spoiler_setpoint.msg in the VTOL module if in hover
- also set spoilers to land configuration if in Descend mode
Allocation: add slew rate limit of 0.5 to flaps/spoilers configuration change
Instead of doing the flaps/spoilers slew rate limiting in the FW Position Controller
(which then only is applied in Auto flight), do it consistently over all flight
modes, so also for manual modes.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- remove deprecated actuator_controls[INDEX_LANDING_GEAR]
- remove dead code in mc rate controller that used to prevent it from being retracted
on the ground (anyway had no effect as it only affected the actuator_control[LANDING_GEAR]
which wasn't sent to the control allocation)
- for VTOLs handle deployment/retraction of landing gear in AUTO as a MC (retract if
more than 2m above ground, deploy if WP is a landing WP), plus additionally when transition
flight task is called (ALTITUDE mode and higher)
- for FW in AUTO: add logic in FW Position Controller, depending on waypoint type mainly
- manual landing gear settings always come through (a manual command overrides a previous
auto command, and vice-versa)
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Tiltrotor_extra_controls also contains collective thrust beside collective tilt, as passing a 3D
thrust setpoint vector beside the tilt is not feasible.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Changed the method of checking and setting the server file lock on Posix to avoid conditions where the server can indicate that it is running but still hasn't finished it's initialization
- always publish esc_status
- when enabled via MODAL_IO_VLOG param, enable actuator debug output
- for modal_io commands, use ESC HW ID values instead of motor number for easier use
- publish esc_status message for command line commands
- Uncommented the code that fills in the cmdcount and power fields in the esc_status topic
---------
Co-authored-by: Travis Bottalico <travis@modalai.com>
- if timestamp (or timestamp_sample) unpopulated fill it with current hrt_absolute_time()
- if using provided timestamp don't allow it to exceed current hrt_absolute_time()
- Switching to the first order filter that was previously
only in FlightTaskManualAltitude.
- Moving the scaling of full stick deflection to
radians per second into the class.
* RTL: only do calculations in is-inactive if global position is recent
* RTL: refactor calcRtlTimeEstimate to only calc and not pub
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
When requesting a message from a stream that is not active we start the
stream with interval=0 and call the request method once. For all streams
this works fine except the gps_global_origin. For this one the request method
is actually overidden to throttle down the rate and not just send out the message.
This will cause this message to never being sent on request if the stream
is not active by default.
* Update README: Maintainers, Boards, Roadmap
- First step after the community coordination call from January 30th
* README: Fromat list & remove discontinued boards & add others
- Addressed comments
* README: Add Simulation, remove QGC
- Only leave the PX4 specific categories (QGC is not)
* Add Beniamino as ROS2 maintainer
* README: Add note that README is main source of truth for maintainers
- We need to have a source of truth, we can use Github README for that.
- vectornav library (libvnc) fixed for NuttX
- open serial port O_NONBLOCK (like __APPLE__)
- set serial port baud rate with cfsetspeed (like __APPLE__)
- vectornav backend thread increase stack and run at higher priority (SCHED_FIFO)
Block Device driver uses a buffer so we need to ensure data is written or read to the device and not to the buffer so we can be sure if the device works properly
* Cyphal: fix comparing floating-point issue
* Cyphal: fix setpoint serialization
* Cyphal: fix bug with wrong comparasion of param name and pub/sub name: remove prefix from UavcanPublisher::updateParam and UavcanDynamicPortSubscriber::updateParam and PublicationManager::updateDynamicPublications
* Cyphal: integrate UavcanEscController with PublicationManager, remove second instance of UavcanEscController from CyphalNode
* Cyphal: publish readiness with minimal frequency because according to UDRAL The drive shall enter STANDBY state automatically if the readiness subject is not updated for CONTROL_TIMEOUT
* Cyphal: increase setpoint publish rate from ~75 to 200 by removing PX4_INFO (it really significantly react on the the output rate) and changing the mixing output rate and the shedule interval
* Cyphal: restore prefix because we need it for uorb over uavcan/cyphal and add udral prefix for non uorb pub/sub
* Cyphal: fix DynamicPortSubscriber subscription: if it has multiple subscribers, it should call subscription only after updating of all port subscribers port identifiers
* Cyphal: fix SubscriptionManager: we should take care about prefix
* Cyphal: fix readiness for test motor mode
* [Cyphal] Fix dynamicsubscription, improve printinfo, enable MR-CANHUBK3 config
---------
Co-authored-by: Peter van der Perk <peter.vanderperk@nxp.com>
- pass new airspeed sample around when available
- can't completely eliminate _airspeed_sample_delayed until resetWind()
called from sideslip fusion is updated
- reduce boost priority to PX4_WQ_HP_BASE - 6
- add cli command 'trigger_watchdog' to manually trigger watchdog
- add perf counters when triggering watchdog
- reduce top measurement to 300ms
- restore priorities after 1.5s
There are precautions in case the SD card code itself has a busy-loop.
Enable automatic landing abort on timed out distance sensor reading also for
the circular landing. Do not enable the no-terrain timeout check, as, opposed
to the straight landing, we here don't know when to expect the distance sensor
to get valid.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
As we don't know the landing point altitude in non-mission landings, assume
the worst case (abort right before touchdown) and thus always climb
MIS_LND_ABRT_ALT on triggering an abort.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Add method for circular landing, that is used instead of the straight fixed-wing
landings in case the landing is not part of a mission landing.
Use straight landing if previous WP is valid, and the ciruclar otherwise.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
_land_start_index is used to to start the mission from this item index, and to
avoid to publish a triplet.current.type=IDLE, we need to fill it with the actual
position setpoint that the vehicle should go to at the start of a mission landing.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Prior commit added opening of /dev/null as 0, 1 and/or 2 file
descriptors, if they where not present. However, if the temporary
file descriptor used to open /dev/null matched the target file
descriptor, it would be immediately closed again. This commit fixes that,
and does not duplicate and close the temporary file descriptor if it is
already at the correct number.
- new FIFO_RESET state used to give the sensor more time after successful configuration before sampling begins
Co-authored-by: Mathieu Bresciani <brescianimathieu@gmail.com>
* rtl: remove unconditional transition to land after descent
This was a bug, as it renders the above code lines useless.
This would cause a undesired FW landing for VTOL vehicles if
RTL_LAND_DELAY is above 0.
* rtl: head to center after loiter in VTOL FW
To get the same behavior for RTL with and without loiter before land for
VTOL drones.
* rtl: always go to descend state after return
Previously, the state would change directly to land if in MR and
RTL_LAND_DELAY was 0.0, but we will still wish to descent to
RTL_DESCEND_ALT at descent speed, instead of using landing speeds.
* rtl: mark head to center state as part of vtol transition
The next step in the sequence is transition to MC. By setting
vtol_back_transition we ensure that the acceptance radius is adapted to
the expected transition distance.
If 0, 1 and/or 2 file descriptors are not open when mavlink module
starts (as might be the case for USB auto-start), use default /dev/null
so that these numbers are not used by other other files.
Instead of interpret a request for "more logs than currently exists" as
a new request, use a request for index 0, which is more likely to be
the first request.
* [npfg]: Remove the guideToPoint function and replace with guideToPath
* [npfg]: remove unused navigateXXX functions
* [npfg]: Move navigateXXX Function into FWPoscontrol
* [FixedwingPositionControl]: Set default flaps and spoilers in attitude setpoint topic, and only change if necessary.
- precision landing works incorrectly, target position is not updated during the descent above target
- _prepareLandSetpoints needs to update _land_position continuously
Co-authored-by: kapacheuski <kapacheuski@gmail.com>
- the thermal offsets are an optional correction applied to the raw data, so when updating an existing calibration offset with new learned bias we don't want this incorporated
- EKF isTimedOut(), isRecent(), and isNewestSampleRecent() need to handle the case where the timestamp has never been set
- reset() more thoroughly reset fields (mainly impacts unit tests)
Set this flag to true if local position is valid but accuracy low, such that
the operator can be warned before system switches to position-failure failsafe.
Additionally, switch to RTL if currently in Mission or Loiter to try to reach home
or fly out of GNSS-denied area.
Set low accuracy threshold to 50m by default for FW and VTOL.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Otherwise the flight time restriction flag gets cleared too early, before
disarming (which puts the vehicle into the previous mode and it might
take off again).
Fixes the following case:
- user intention set to X
- failsafe triggers, mode = Y
- can_run for X becomes false
- user tries to switch to X
-> need to re-evaluate can_run
Previously, it was not possible to enable forwarding of messages to/from
teh USB instance because it does not have a param for it, like the
serial instances have.
With this commit, we change the default to always set forwarding on for
the USB instance as that is likely desired.
Signed-off-by: Julian Oes <julian@oes.ch>
- from the C standard, "All identifiers that begin with an underscore and either an uppercase letter or another underscore are always reserved for any use"
- from the C standard, "All identifiers that begin with an underscore and either an uppercase letter or another underscore are always reserved for any use"
The static object is destroyed on at_exit while threads might still be
inside a CS. This can lead to a hanging process.
Cleaner would be to gracefully stop the threads.
According to https://linux.die.net/man/3/pthread_cond_destroy:
Attempting to destroy a condition variable upon which other threads are
currently blocked results in undefined behavior.
Instead check if system has previously switched into LOITER to acheive the current
WP of type POSITION, and in that case stay in LOITER until altitude is reached.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Do the same as DO_REPOSITION wit only the altitude field populated
and MAV_DO_REPOSITION_FLAGS set, which means:
- switch to Loiter mode if not already in it
- set the current altitude to what is specified in the altitdue field,
keep current altitude setpoint otherwise
- keep current position setpoint
- fall back to current estimated position in case a position setpoint
is not finite
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Previously the condition was based on hard coded height rate estimate and
setpoint values and an altitude error threshold. That showed to be leading
to false positives when the vehicle doesn't tightly follow the altitdue
ramp given by TECS to achieve a new altitude setpoint, and has become
completely infeasibly with the latest TECS rework that leads to non-ramped
altitude setpoint changes in the tecs_status message.
The new check no longer checks the altitude error but only the height rate
instead. It begins to integrate the height rate error once it detects an
uncommanded descend condition (height rate negative while setpoint is
positive). Integral threshold can be tuned by user (VT_QC_HR_ERROR_I).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
In rare occasions asking for the protocol values after setting it returned
[0, 0]. I did not see any documentation for having to wait, but adding a
short wait period fixes it.
- refactor all EKF backend output predictor pieces into new OutputPredictor class
- output states are now calculated immediately with new high rate IMU rather than after EKF update
- IMU delayed sample is passed as around as control data to avoid storing an extra copy and make the requirement clear
- update motor mapping to use new UART_ESC_FUNC* auto generated params
- add support for actuator_test msg to support Actuator Testing in QGC
- modalai_fc-vX targets start driver if configured
- keep track of ESC spin direction in own param
- set ramp up param in MixerOutput to false
- use `gazebo-classic` everywhere consistently referring to the original Gazebo (eg version 9,10,11)
- additional `gazebo_*` helper targets added for compatibility, but warn about deprecation and tell you the new target naming
- use `gz` everywhere when referring to Gazebo (aka Ignition Gazebo or new Gazebo)
For the esc_calibration code, the actuator test is published for each motor
in a row. If the orb queue size is too small, messages are lost and not
received in mixer_module.
Set the default orb queue size of ActuatorTest high enough to keep the commands
for all motors in the queue.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
Strictly follow the following convention for tailsitter:
FW Attitude and FW rate controller always operate in the FW frame, meaning that roll is
roll in FW, which for tailsitter means around the yaw axis in the body frame. The interfaces
between modules is though always in body frame.
That enables us to do the axis transformations for tailsitter, that are currently distributed
all over the controller (attitude, rate, vtol module), only at the input and output data of modules.
Side effect is that the FW rate control tuning gains meanings change: while before the roll gains
where meant for the body axis, they are now always applied for the FW roll axis (also in hover). So
the naming now is correct for FW, while before it was for Hover.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This failsafe flag is currently used for not allowing to re-transition to FW, as well
as disabling pusher assist in hover. Till now it was only possible to reset it with
a commanded transition to MC, which many ground station interfaces didn't allow
as the system, after a quad-chute, already was in MC mode.
Hereby it is changed to reset when a new transition to FW is triggered (either via
RC switch or MAVLink command). It is the users responsibility to assess the situation
after a quad-chute happened to try to transition to FW again, manually proceed/land
the vehicle in MC, or let it finish the defined behavior after a quad-chute.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- during casual testing on default configs the stack was penetration was reaching ~90% which is a bit too close for comfort
- increasing by 50% to be conservative
When we receive a command_long or command_int message to
answer, it arrives with a source sysid/compid, so this means we can send
the command_ack back on the appropriate MAVLink instances instead of all
of them.
This commit filters outgoing command_ack messages, so they are only sent
on the MAVLink instances where the sysid/compid has been seen in the
past.
This means that a command_ack is likely still sent on multiple links for
a setup with redundant links.
This should also prevent command_acks from being blasted on Iridium links
when it's not required.
Signed-off-by: Julian Oes <julian@oes.ch>
* Fixedwing rate control into a separate module
* Start fw_rate_control for vtol
* Move over airspeed related parameters to fw_rate_control
Co-authored-by: Silvan Fuhrer <silvan@auterion.com>
On CubeOrange where no console is configured by default, starting
MAVLink shell just stalls, and doesn't work.
Also, logfile download has been reported not to work, and again, seems
to work with this change.
Signed-off-by: Julian Oes <julian@oes.ch>
- Do not pull in PWM parameters when DISABLE_PARAMS_MODULE_SCOPING is TRUE since VOXL2 has no PWM nor any of the required timer_config files that go along with that
- Replace non-standard M_PI constants with PX4 defined M_PI_F constants
- Include missing header file for function hrt_absolute_time declaration
- Add new PX4_SOC_ARCH_ID for the VOXL2 board
- move EV yaw and EV position to new state machines
- EV yaw and EV pos now configured via EKF2_EV_CTRL (migrated from EKF2_AID_MASK)
- new EV position offset estimator to enable EV position while GPS position is active (no more EV pos delta fusion)
- yaw_align now strictly means north (no more rotate external vision aid mask)
- automatic switching between EV yaw, and yaw align north based on GPS quality
* AirspeedSelector: use vehicle_air_data.rho for calculating groundspeed-wind CAS
Previously the vehicle_air_data.temperature and pressure was used, instead of the
density field directly.
Only makes a difference if there is an airspeed sensor connected to provide
the air temperature.
* AirspeedSelector: only safe estimated scale in param if airspeed is valid
* AirspeedSelector: remove 0.01 cliff for saving learned scale to param
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This checks was introduced to catch the (unlikely) case of the driver publishing
a stale value that is not actually from a current measurement right after boot.
This was done by declaring it invalid if there is no update of the measurement
within the first 10s after boot.
Practice though has shown that around 0, many airspeed sensors have such a low
resolution that the reported airspeed value is often at the exact same value
for longer periods of time on totally healthy sensors, and thus we trigger
some false positive failure detections.
Given that this failure mode (driver publishing stale data) is quite rare (if
it doesn't receive new data it should stop publishing), I remove this check
here again.
When in aerodynamic flight mode and armed, there is still the data stuck
check that can trigger if there is no update for 2s.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* GPS_RAW_INT: Only send no gps messages if gps has ever been present
* GPS2_RAW: Keep in sync with GPS_RAW_INT
Signed-off-by: Marcell Rausch <marcell@auterion.com>
- respect new EKF2_EV_CTRL parameter for VPOS usage
- EV hgt rotate EV position before usage (there's often a small offset in frames)
- EV hgt reset use proper EV velocity body frame
- try to keep EV hgt and EV vel state machines consistent
- small incremental piece of https://github.com/PX4/PX4-Autopilot/pull/19128
- first version of IMU driver for the VOXL 2 platform (Qurt)
- this is a customized version of the Invensense ICM42688P driver, it is currently in the VOXL 2 board directory
- don't store unnecessary IMU copy in class, pass it through where required
- remove custom pi constants
- remove unused fields from ahrs_ekf_gsf (vel_NE, fuse_gps, accel_dt)
- explicitly initialize everything in header
- apply PX4 code style
- set GPS velocity before yaw estimator update, not after
- new ROS2 platform in PX4 intended for creating configs that build and run entirely in ROS2
- PX4_CONFIG defaults to px4_ros2_default if no config specified and in a colcon workspace with ROS_VERSION=2
- currently doesn't do much other than allow you to build px4 msgs interface package
Multiple agents can connect to the same client
For sitl builds, if the intance number is different from zero,
XRCE_DDS_KEY is set to the instance number and
the microdds_client is automatically started
with namespace
px4_"instance_number"
and udp port 8888
If the instance number is equal to zero
XRCE_DDS_KEY is left untouched and
the microdds_client is automatically started
without namespace
and udp port 8888
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
The partecipant name is modified into
"client_namespace"/px4_micro_xrce_dds
For sitl builds the microdds_client is automatically started
with namespace
px4_"instance_number"
and udp port
8888+"intance_number"
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
With openjdk 17.0.5 2022-10-18, jmavsim fails to start due to missing
exports. This was previously a warning and turned into an error now.
Caught AppContextInfo(Bug 1004) InaccessibleObjectException: Unable to make public static sun.awt.AppContext sun.awt.AppContext.getAppContext() accessible: module java.desktop does not "exports sun.awt" to unnamed module @2c767a52 on thread J3D-Renderer-1
[0]: java.base/java.lang.reflect.AccessibleObject.checkCanSetAccessible(AccessibleObject.java:354)
[1]: java.base/java.lang.reflect.AccessibleObject.checkCanSetAccessible(AccessibleObject.java:297)
[2]: java.base/java.lang.reflect.Method.checkCanSetAccessible(Method.java:199)
[3]: java.base/java.lang.reflect.Method.setAccessible(Method.java:193)
[4]: com.jogamp.nativewindow.awt.AppContextInfo$1$1.run(AppContextInfo.java:40)
[5]: com.jogamp.common.util.UnsafeUtil.doWithoutIllegalAccessLogger(UnsafeUtil.java:219)
[6]: com.jogamp.nativewindow.awt.AppContextInfo$1.run(AppContextInfo.java:34)
[7]: java.base/java.security.AccessController.doPrivileged(AccessController.java:318)
[8]: com.jogamp.nativewindow.awt.AppContextInfo.<clinit>(AppContextInfo.java:31)
[9]: com.jogamp.nativewindow.awt.JAWTWindow.<init>(JAWTWindow.java:128)
[10]: jogamp.nativewindow.jawt.x11.X11JAWTWindow.<init>(X11JAWTWindow.java:60)
[11]: java.base/jdk.internal.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
[12]: java.base/jdk.internal.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:77)
[13]: java.base/jdk.internal.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
[14]: java.base/java.lang.reflect.Constructor.newInstanceWithCaller(Constructor.java:499)
[15]: java.base/java.lang.reflect.Constructor.newInstance(Constructor.java:480)
[16]: jogamp.nativewindow.NativeWindowFactoryImpl.getAWTNativeWindow(NativeWindowFactoryImpl.java:105)
[17]: jogamp.nativewindow.NativeWindowFactoryImpl.getNativeWindowImpl(NativeWindowFactoryImpl.java:66)
[18]: com.jogamp.nativewindow.NativeWindowFactory.getNativeWindow(NativeWindowFactory.java:654)
[19]: javax.media.j3d.JoglPipeline$QueryCanvas.addNotify(JoglPipeline.java:8604)
[20]: java.desktop/java.awt.Container.addNotify(Container.java:2804)
[21]: java.desktop/java.awt.Window.addNotify(Window.java:791)
[22]: java.desktop/java.awt.Frame.addNotify(Frame.java:495)
[23]: java.desktop/java.awt.Window.show(Window.java:1053)
[24]: java.desktop/java.awt.Component.show(Component.java:1728)
[25]: java.desktop/java.awt.Component.setVisible(Component.java:1675)
[26]: java.desktop/java.awt.Window.setVisible(Window.java:1036)
[27]: javax.media.j3d.JoglPipeline.getBestConfiguration(JoglPipeline.java:8379)
[28]: javax.media.j3d.Renderer.doWork(Renderer.java:496)
[29]: javax.media.j3d.J3dThread.run(J3dThread.java:271)
Caught AppContextInfo(Bug 1004) IllegalAccessException: class com.jogamp.nativewindow.awt.AppContextInfo cannot access class sun.awt.AppContext (in module java.desktop) because module java.desktop does not export sun.awt to unnamed module @2c767a52 on thread J3D-Renderer-1
[0]: java.base/jdk.internal.reflect.Reflection.newIllegalAccessException(Reflection.java:392)
[1]: java.base/java.lang.reflect.AccessibleObject.checkAccess(AccessibleObject.java:674)
[2]: java.base/java.lang.reflect.Method.invoke(Method.java:560)
[3]: com.jogamp.nativewindow.awt.AppContextInfo.fetchAppContext(AppContextInfo.java:191)
[4]: com.jogamp.nativewindow.awt.AppContextInfo.update(AppContextInfo.java:135)
[5]: com.jogamp.nativewindow.awt.AppContextInfo.<init>(AppContextInfo.java:50)
[6]: com.jogamp.nativewindow.awt.JAWTWindow.<init>(JAWTWindow.java:128)
[7]: jogamp.nativewindow.jawt.x11.X11JAWTWindow.<init>(X11JAWTWindow.java:60)
[8]: java.base/jdk.internal.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
[9]: java.base/jdk.internal.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:77)
[10]: java.base/jdk.internal.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
[11]: java.base/java.lang.reflect.Constructor.newInstanceWithCaller(Constructor.java:499)
[12]: java.base/java.lang.reflect.Constructor.newInstance(Constructor.java:480)
[13]: jogamp.nativewindow.NativeWindowFactoryImpl.getAWTNativeWindow(NativeWindowFactoryImpl.java:105)
[14]: jogamp.nativewindow.NativeWindowFactoryImpl.getNativeWindowImpl(NativeWindowFactoryImpl.java:66)
[15]: com.jogamp.nativewindow.NativeWindowFactory.getNativeWindow(NativeWindowFactory.java:654)
[16]: javax.media.j3d.JoglPipeline$QueryCanvas.addNotify(JoglPipeline.java:8604)
[17]: java.desktop/java.awt.Container.addNotify(Container.java:2804)
[18]: java.desktop/java.awt.Window.addNotify(Window.java:791)
[19]: java.desktop/java.awt.Frame.addNotify(Frame.java:495)
[20]: java.desktop/java.awt.Window.show(Window.java:1053)
[21]: java.desktop/java.awt.Component.show(Component.java:1728)
[22]: java.desktop/java.awt.Component.setVisible(Component.java:1675)
[23]: java.desktop/java.awt.Window.setVisible(Window.java:1036)
[24]: javax.media.j3d.JoglPipeline.getBestConfiguration(JoglPipeline.java:8379)
[25]: javax.media.j3d.Renderer.doWork(Renderer.java:496)
[26]: javax.media.j3d.J3dThread.run(J3dThread.java:271)
Exception in thread "main" java.lang.reflect.InvocationTargetException
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:568)
at org.eclipse.jdt.internal.jarinjarloader.JarRsrcLoader.main(JarRsrcLoader.java:61)
Caused by: java.lang.RuntimeException: java.lang.IllegalAccessException: class javax.media.j3d.JoglPipeline cannot access class sun.awt.X11GraphicsDevice (in module java.desktop) because module java.desktop does not export sun.awt to unnamed module @2c767a52
at javax.media.j3d.JoglPipeline.getScreen(JoglPipeline.java:8553)
at javax.media.j3d.Screen3D.<init>(Screen3D.java:354)
at javax.media.j3d.Canvas3D.<init>(Canvas3D.java:1124)
at javax.media.j3d.Canvas3D.<init>(Canvas3D.java:1026)
at javax.media.j3d.Canvas3D.<init>(Canvas3D.java:990)
at me.drton.jmavsim.Visualizer3D$CustomCanvas3D.<init>(Visualizer3D.java:909)
at me.drton.jmavsim.Visualizer3D.<init>(Visualizer3D.java:196)
at me.drton.jmavsim.Simulator.<init>(Simulator.java:193)
at me.drton.jmavsim.Simulator.main(Simulator.java:944)
... 5 more
Caused by: java.lang.IllegalAccessException: class javax.media.j3d.JoglPipeline cannot access class sun.awt.X11GraphicsDevice (in module java.desktop) because module java.desktop does not export sun.awt to unnamed module @2c767a52
at java.base/jdk.internal.reflect.Reflection.newIllegalAccessException(Reflection.java:392)
at java.base/java.lang.reflect.AccessibleObject.checkAccess(AccessibleObject.java:674)
at java.base/java.lang.reflect.Method.invoke(Method.java:560)
at javax.media.j3d.JoglPipeline.getScreen(JoglPipeline.java:8551)
... 13 more
* VTOL: take home_position into account for ground clearance
For approximating distance to ground in case there is no valid distance sensor
data, subtract the home position altitude from the local position altitude.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: Matthias Grob <maetugr@gmail.com>
Instead of always starting with instance 0 (potentially an internal
mag), first take the current sensor as reference to compare the other
ones against it.
The issue is that otherwise we can end up in a
situation where a switch occurs because the currently used sensor isn't
much better than a sensor with a lower index: because the selected one
isn't much better, we cannot "fail-over" to it and get stuck on another
sensor, triggering a fail-over.
- WelfordMeanVector now computes covariance
- use Kahan summation for Welford mean (but continue using float32 for actual mean, etc)
- WelfordMean and WelfordMeanVector handle initial value and count roll over
- Welford mean count rollover at 16 bit max to prevent numerical issues and shift weight to newer samples
- sensors/vehicle_imu: update Welford mean usage (now simplified with resets removed)
- fix vehicle_imu_status accel var, now properly rotated with full covariance matrix
- gyro_calibration: update Welford mean usage
This is fully backwards compatible: If the throttle trim is set to
the minimum then it's the legacy calibration and gets
interpreted such that there is no trim and behavior remains as before.
If the trim is set to a different value than the minimum then it gets
used like with all other channels which was unsupported before.
In review it was requested to have a different name for
manual_control_setpoint.z because of the adjusted range.
I started to investigate what naming is most intuitive and found
that most people recognize the stick axes as roll, pitch, yaw, throttle.
It comes at no surprise because other autopilots
and APIs seem to share this convention.
While changing the code I realized that even within the code base
the axes are usually assigned to a variable with that name or
have comments next to the assignment clarifying the axes
using these names.
To be consistent with all other axes of stick input and avoid future
rescaling confusion.
Note: for the MAVLink message 69 MANUAL_CONTROL it's using the full range
according to the message specs now [-1000,1000].
Add ARK_FMU_V6X to RCS netman
Remove arkv6x rc single wire
Fix arkv6x mtd
arkv6x bootloader init all pins to prevent power cycling peripherals on boot
arkv6x don't power cycle sd card on boot
arkv6x add UART4 Telem 4
Also applies to Loiters that are started due to the previous mode being over (Takeoff,
VTOL_Takeoff, Mission).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
landing airspeed was previously defined by a scale factor multiplied by minimum airspeed. this commit changes this parameter to an explicit speed, and when unspecified, defaults to the minimum airspeed
Use kClearanceAltitudeBuffer for it, which is also used to ensure that during takeoff an
altitude setpoint above the clearance altitdue is set.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
MIS_LTRMIN_ALt was used to limit the go-to altitude of a LOITER_TO_ALT (not the exit altitude,
but the altitude that the vehicle went to to fly to WP), and during landing abort to climb to
at least this altitude. The min entry altitude of LOITER_TO_ALT I remove with this commit, while
for the min alt during abort I added the new parameter MIS_LND_ABRT_ALT.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
previously, the next position setpoint in the triplet was left unchanged during an abort, which meant that the abort loiter current point combined with land next point triggered the early landing config logic. this commit is only a hack to make things work temporarily.. this needs to be handled better.
- explicitly defined takeoff airspeed setpoint
- dont use climbout mode
- allow max climb on takeoff
- dont handle post clearance altitude case (navigator will switch anyway)
Completely detach the steering wheel logic from the yaw controller (beside using the
same manual stick input in a manual flight mode).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- also put flaring internal states into a struct to organize a bit
- one concern with blending the throttle setpoint like this with the flare time param is that folding prop belly landing airframes may want to have a separate param for shorter throttle kill and still use the flare time ramps for everything else
abruptly changing to a heading setpoint on flare can cause the aircraft to roll and deviate from the runway, this commit
- maintains path following control during the flare not to disrupt the tracking just before touchdown
- (unfortunately for crosswind landing) removes the body axis alignment for runway bearing - this is a compromise
to achieve both runway bearing body axis alignment AND a specific touchdown point, either
1. the wind would need to be considered, and an appropriate diagonal approach (obstructions allowing) defined to the runway
2. slip control added, keeping path following outputs only commanding roll (controlling airspeed vector) and using yaw-rate command (only actuated by e.g. rudder) to align body axis with the runway
- consolidate takeoff rotation transition times for pitch constraints and throttle setpoint with a single param
- consolidate pitch takeoff constraint parameters (remove rwto_max_pitch, use nominal max)
- input correct units to rwto pitch constraint getters
- encapsulate absolute time interpolator method for transitions
- start runway ops from idle throttle
TECS climbout mode was used for takeoff climbout, which puts throttle to full and does not regulate a specific airspeed.
This commit sets the desired takeoff airspeed explicitly and allows max climb rate to track the ascent.
previously a scale factor param on min airspeed was used to define the climbout airspeed for runway takeoff
additionally, the rotation speed was defined by another hardcoded scale on top of the previously scaled min airspeed
this commit explicitly defines a takeoff speed and rotation speed for runway takeoff in params, with option to disable
Before, the logic to update disarm and rate values also triggered during
bootup on the px4io, because the output functions are only set in
updateSubscriptions().
Therefore change the check to prevent updating during the first cycle.
When using ignition SITL simulation with NO_LOCKSTEP, the SITL PX4 fails to update the IMU data from Ignition Gazebo.
The timestamp for the IMU data is taken from the ignition message:
- In LOCKSTEP mode the clock from the ignition simulation and the one from PX4 SITL are synchronized, hence everything works fine
- In NO_LOCKSTEP mode, those clocks are not synchronized anymore, so the timestamp for the IMU data should not be the one from Ignition but the current time in PX4 SITL when receiving the message.
The update function of InputMavlinkGimbalV2 returns UpdatedNotActive when receiving control commands from a system or component that doesn't match the primary control. This can happen if a component just sends commands without being in control or when transitioning to a different primary control.
If the input is marked as already_active, it will reset last_input_active which in turn resets the primary control in the next iteration. According to the MAVLink gimbal protocol v2, a component needs to send MAV_CMD_DO_GIMBAL_MANAGER_CONFIGURE to start controlling the gimbal and to remove control. However, with the current implementation there are several other commands that would reset the primary control.
With this PR, the primary control remains with the component that requested it last if updates from a not active component are received.
* Log position setpoint reference of npfg
This commit logs the local position setpoint reference when using NPFG
* Address review comments
This commit address review comments from @tstastny
Replace the existing check for the availability of a takeoff mission item with a combined
check for takeoff and landing item (or landing pattern). New param MIS_TKO_LAND_REQ
can be set to require only a takeoff, only a landing, both takeoff and landing, and
both or none. The latter is meant to be set if is e.g. deemed unsafe to start a flight
through a Takeoff WP without though defining a Landing - as then in case of a RTL the
vehicle doesn't follow a pre-defined path but instead only can do default RTL that especially
for FW and VTOL isn't always safe.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Previously the minimum airspeed setpoint was adjusted to the load_factor compensated
stall speed, which, when the stall speed was set without margin, gave the controller
no room for error (the vehicle would stall if the controller has even a small airspeed
error).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- Include board_config.h for BOARD_GET_EXTERNAL_LOCKOUT_STATE etc. macros
- Include fcntl.h for "open"
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
* LandDetectorMC: enforce that LNDMC_Z_VEL_MAX * 1.2 is below *MPC_LAND_CRWL/MPC_LAND_SPEED
Otherwise the _in_descend flag doesn't get set correctly during the last part
of the landing, where the descend speed is at MPC_LAND_CRWL or LAND_SPEED.
The _in_descend flag is set it the velocity setpoint is >1.1*LNDMC_Z_VEL_MAX.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: Matthias Grob <maetugr@gmail.com>
* Made voxl2 apps processor and slpi dsp processor builds into separate board builds so that they can
more easily be configured independently.
* Removed board specific link library command from px4_config.cmake and moved it to a more generic
board specific solution that can be used by any board that needs custom link libraries.
* Removed redundant cmake command for Qurt
* Removed unused definition from Qurt cmake file
* Removed unnecessary QURT_LIB cmake function
* Reorganized the voxl2 build structure to avoid 4 level board directories.
* Reverted cmake files to remove 4 level board naming code
* Updated documentation
- in a lot of cases this won't really matter, although you can see it in logs (especially with things like vehicle_at_rest), but it's something we might as well do properly
Co-authored-by: Mathieu Bresciani <brescianimathieu@gmail.com>
Fixes a regression from 8bae4e5c0e, where
the orbit flight task wasn't an extra task (flight_tasks_to_add) anymore
and therefore the command handling wasn't generated.
There was a race condition that could cause several outcomes. The most severe
was that flight_mode_manager gets the command, switches to orbit and then
in the next iteration switches back because commander did not change
nav_state yet. When commander then switches, flight_mode_manager would still
be in the old mode.
This is prevented by storing the command (allowing it to arrive before or
after mode switch), and then apply it after the switch happens.
- vehicle_angular_velocity and vehicle_angular_acceleration are produced together from the same input data, consumed together, and share the the same metadata (timestamps)
- individually these topics each have 16 bytes of metadata (2 timestamps) for 12 bytes of data (x,y,z float32)
During the whole flight, if the difference between the yaw estimate from
EKF2 and the emergency estimator is large and that the GNSS velocity
fusion is failing continuously, immediately reset to the emergency yaw
estimate.
- with Ekf::initialiseFilter() resetting _time_last_hagl_fuse the terrain estimate was technically valid regardless of any range or flow data availability and as a result optical flow fusion is able to start and stay active
- only consider terrain estimate valid based on control status flags and recent fusion
The CameraFeedback module used only the vehicle attitude for the camera orientation so far. With this change, the gimbal_device_attitude_status is used to compute the global camera orientation when a gimbal is used.
In this case it did not cause any problems.
Fixes a compiler warning:
/src/modules/commander/HealthAndArmingChecks/HealthAndArmingChecks.cpp:39:21: error: member ‘HealthAndArmingChecks::_failsafe_flags’ is used uninitialized [-Werror=uninitialized]
39 | _reporter(_failsafe_flags)
| ^~~~~~~~~~~~~~~
* simulator_mavlink: Add basic vio failure injection
Signed-off-by: Paul Frivold <paul@kefrobotics.com>
* simulator_mavlink: Rm failure not supported warning
Failures commands are also handled in other files,
so warning here could be confusing.
Signed-off-by: Paul Frivold <paul@kefrobotics.com>
Signed-off-by: Paul Frivold <paul@kefrobotics.com>
previous change in logic to hold after mission clear also broke rtl, as non-mission takeoff still published a mission result which allowed entering the mission finished condition and always changing state to loiter (ignoring rtl). new logic only switches navigation states if mission is finish and the nav state is explicitly in takeoff state, or in mission state
The srcparser.py is specific to each use case (e.g. Airframes, Parameters, px4events, etc as in Tool/* folders).
Therefore it is confusing to have the px_process_airframes.py script handle concept of airframes under the generic name 'params'.
This improves readability and sets the baseground for implementing more specific vehicle type supports, as mentioned in https://github.com/PX4/PX4-user_guide/pull/1858#discussion_r876554728
once the rampup starts. The rampup requires a valid vertical velocity setpoint.
The corner case is:
- We are before takeoff and amending the setpoint to be 0,0,100 acceleration
in order to idle
- The rampup starts BUT the setpoint is not yet overwritten by the trajectory
setpoint topic
- The idle setpoint gets amended to not contain a feed-forward vertical
acceleration because the rampup is velocity based
- The result is a brief invalid 0,0,NAN acceleration setpoint
- That invalid setpoint gets overridden by a failsafe that holds zero velocity
- Zero velocity leads to applying ~hover thrust briefly
- buffer at least 2 samples for the IMU output predictor buffers
- dropping below 2 becomes problematic for the minimum observation
interval calculation and the vertical output buffer trapezoidal
integration
arm
- Previously, due to the way MAVSDK's `health_all_ok` was implemented,
vehicle often didn't have a valid global position estimate (although
function returned true), and it wouldn't arm, and the SITL would fail
- Also sometimes as vehicle didn't have manual control, it entered weird
states where it wasn't able to arm as well
- This adds a check to make sure vehicle is able to arm, directly from
the Health struct
- preflight mag bias estimate is in the vehicle body frame and removed from the raw mag data after it's rotated, calibrated, etc
- in flight mag bias (from ekf2) is in the vehicle body frame, but stored as a sensor frame offset immediately
* Enable motor controls for fixed wing mode in tailsitters
This commit enable motor controls in fixed wing mode for tailsitters
This is needed for enabling quad tailsitters
* VTOL: differential thrust in FW: adapt params to be generic for all axes
Until now only suppoted on yaw axis. Not to be supported also on Roll and Pitch.
- VT_FW_DIFTHR_EN: make bitmask for all three axes independently. First bit is Yaw,
sucht that existing meaning of VT_FW_DIFTHR_EN doesn't change.
- VT_FW_DIFTHR_SC: rename to VT_FW_DIFTHR_S_Y and add same params for roll (_R) and
pitch (_P).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Integrate differential control bits to three axis control
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
It's enough that the setpoints and the unallocated values are logged, from these
the allocated values can be calculated if required.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
It's very important that the in descend detection is always
at a strictly higher velocity than the vertical movement check.
This combination is basically used to check for vertical downwards
velocity tracking. Desired descend, no movement -> ground
If in descend threshold is lower than vertical movement it is
by definition even with perfect tracking the case that with
any velocity between the two thresholds there is no movement
even though a descend is commanded.
See first fix of this problem #7831e39b38ba96
I don't see where this is necessary. During takeoff, the maybe landed flag should
only get cleared once system is about to takeoff, and thus well after the spool up
is complete (for which the higher thresholds are meant in this case).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Don't consider these params for vertical speed threshold,
and instead increase the default for LNDMC_Z_VEL_MAX and
use solely that one. Makes the land detector outcome more
predictable and its interal logic simpler, while the new
default tuning is resulting in about the same vz threshold
as before.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
* Changed M_PI to M_PI_F in the matrix library since M_PI is non-standard.
* Added a new M_PI_PRECISE constant definition to px4_platform_common/defines.h to be
used in places when M_PI is desired but shouldn't be used because it is not C standard.
* Added the px4_platform_common/defines.h include to the matrix library math.hpp header to pull
in some non-standard M_PI constants and updated the test files to use those constants.
* Fixed PI constants in matrix helper test to prevent test failure
before there was a corner case where if in an auto mode that sets a reset command on the attitude setpoint message (e.g. auto takeoff), if the mode was then switched stabilized, this reset bool would never be changed back to false and the integrators would reset every cycle
* Removed dprintf from perf library since it is only ever used with fd=1 (STDOUT) so moved to PX4_INFO_RAW instead. This helps with some platforms (e.g. Qurt) which have some Posix support but not full Posix support.
This was previously required to reset the flight speed after a VTOL transition,
but is now no longer required as the DO_CHANGE_SPEED commands are handles directly
in the controllers.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This fixes the issue where the init happended in the initializer list, at which point
the params were not yet initialized and thus resulted in random values for the init
values of _mission_item.loiter_radius and .acceptance_radius.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Refactor
- Require reboot for PD_GRIPPER_EN parameter change
- Define gripper ACTION_NONE for readability. This makes implicit assumption that -1 equals no-action commanded more explicit
- Tidy the scattered vcmd_ack struct handling cases into a single function
- Refactor to remove return in the middle of function: avoids future complications where a programmer may expect the logic at the end of the function to be executed, but isn't
Vehicle Command Handling
- Cancel the previous running vehicle command if we receive a different vehicle command
- Reject vehicle command if we get a same one that is getting executed
- Save the source system & component of currently running vehicle command
- Added note about new discovered edge case of having same entity sending different gripper commands consequently, where an unexpected ack result may be received
Gripper:
- Don't command gripper (via uORB `gripper` topic, which maps into an
actuator via Control Allocation) if we are already at the state we want
(e.g. Grabbed / Released) or in the intermediate state to the state we
want -> This prevents spamming on `gripper` topic
Payload Deliverer:
- Add read-once function for Gripper's released / grabbed state
- Send vehicle_command_ack for both release/grab actions.
TODO: target_system & target_component for the released/grabbed vcmd_ack
is incomplete, since we are not keeping track of the vehicle_command
that corresponds to this. This needs to be dealt with in the future, not
sure what the best solution it is for now.
Possible solutions:
- Queue-ing the vehicle command?
- Tying the gripper's action to specific vehicle command one-on-one, to make sure if we send multiple vehicle commands, we know
which command resulted in the action exactly?)
Only command Gripper grab when we are actually initializing gripper
- Previously, on every parameter update, gripper grab was being
commanded
- This commit narrows that scope to only when we are actually
initializing the gripper
Handle gripper de-initialization upon parameter change
- Also added some local state initialization code to init() function of
Gripper
- This will now make init / de-init more graceful & controlled compared
to before
- This hopefully then alerts the GCS that the command is getting
processed
- Referenced commander's `handle_command` function to implement this. As
it seems like GCS needs the acknowledgement of the command being
processed to execute such commands properly
- Also send FAILED command ack if we can't actuate the gripper
Fix wrong GRIPPER_ACTION conversion from floating point to int32_t
- Due to the MAVLink spec, we actually just convert enums into floating
point, so in PX4 we need to convert the float directly into integer as
well (although there can be precision issues on large numbers)
- This is a limitation in MAVLink spec, and should hopefully be
changed in MAVLink v2
Affects the states AUTO_ALTITUDE and AUTO_CLIMBRATE. Those modes should only be entered
if armed (as they are pure failsafe modes). Also allow though to enter them even if
the position setpoint(s) are invalid, as they are not needed.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Instead of checking the .valid flag of position_setpoint, check for ISFINITE() of lat, lon, alt
when pulling the position_setpoint triplet. This fixes problems where the .valid flag didn't
reflect the proper state of the setpoint (e.g. .valid was true, .lat though NAN)
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
the nuttx and posix specific options files since this option cannot be used with
the qurt platform. There are header files in the hexagon sdk that fail this check.
as Tools/update_px4_ros2_bridge.sh as been deleted
update_ros2_bridge, update_px4_ros_com and update_px4_msgs
are no more needed
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
- vehicle_angular_velocity: ESC RPM notch filters minimize filter resets
- only allow one filter init per axis per cycle
- "park" ESC notch filters at min frequency instead of full disable
- relax timeout before a notch filter is disabled
- add new parameter IMU_GYRO_DNF_MIN for configuring the minimum notch filter frequency
- avoid the possibility of unsigned underflow from subtracting two HRT timestamps (uint64_t)
- most of these aren't problematic, but people tend to replicate the pattern, so it's better to be safe
- likely wasn't a problem when people were using hrt_absolute_time() in place, but if using an existing timestamp there's the possibility it's older than a more recent topic update
- update all msgs to be directly compatible with ROS2
- microdds_client improvements
- timesync
- reduced code size
- add to most default builds if we can afford it
- lots of other little changes
- purge fastrtps (I tried to save this multiple times, but kept hitting roadblocks)
- working towards keeping all height source (baro/ev/gnss/rng) handling as consistent as possible, possibly refactoring these out into separate classes later
This fixes a bug where by accident the vtol_status was considered instead of the
vehicle_status, preventing it from running on planes.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This commit fixes feedforward acceleration setpoints for fixedwing offboard position control.
Previously when acceleration feedforward inputs were sent, negative curvature accelerations were not being computed properly
ekf2: merge mag 3d innov var, Hx and Kx computation to reduce flash
Slightly less code produced, almost no performance change
ekf2_mag3D: do not pre-compute Kalman gains
The vector of Kalaman gains is not too expensive to compute using
matrix-vector multiplication. Pre-generating it using CSE takes a lot of
flash space for little benefit.
- this is a precaution to eliminate the possibility of getting stuck in
a loop trying to keep up with a high rate publication that could be
coming from a higher priority task
- in the worst case scenario printing the error message can take longer than the next gyro publication, so combined with an infinite loop you could get stuck here
The previous default for yaw and roll was 0.2, which is very low, and for wheel was
1, which is very high. A value of 0.4 makes sense to me for all axes.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
I guess this was a left over from times where the controller had a completely different structure,
a value of 50 is well outside of reasonable range (50x the default).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
The previous max of 10 for the FF gains seems a bit very high for me, well outside
of reasonable reange.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- Fixed clearing arming word if flightmode unknown
- Added power and cell voltage elements
- New OSD layout
- Fixed home direction/distance are now correct
Co-authored-by: Chris Seto <chris1seto@gmail.com>
Moves the arming checks before the command handling.
This reduces the chance of race conditions. However it does not prevent
them! The SDK will need to check when offboard is ready to run/arm to fix
this.
Specifically this is for sitl offboard tests, where offboard_control_mode
is updated and immediately after a mode switch into offboard is commanded.
- trajectory_setpoint and vehicle_local_position_setpoint used to be
the same data type
- we got extremely lucky here that this didn't cause any issues due to
all the fields still aligning
Starting the fake position as soon as all the aiding sources stop makes
the local position immediately invalid while we could continue to navigate
for a couple of second with inertial dead-reckoning
- vehicle_imu_status can publish immediately on any measured sample
rate change or sensor error increment, but the windowed mean/variance
shouldn't necessarily reset that often
This is to prevent that that data from non-recent publications of the virtual
attitude setpoints are used in the transition code.
It removes the previous implementation where the update_transition_state()
method was only run when both mc and fw virtual att sp where recent, independetly
of whether both are used or not.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- for accel/gyro/mag estimated bias only consider them stable (valid
for calibration updates) if the value isn't changing (10% of limit) over
the validity period
* Changed sitl_gazebo to the advanced liftdrag plugin fork.
* Added advanced plane and its associated parameters to PX4. Also tweaked one of the plugins to fix a compilation error (upcasting from float to double).
* Switched gazebo back to main branch, to avoid merge conflicts.
* Change simulator bridge back to what it was in main branch
* Changed sitl_gazebo to match the PX4 main branch's (commit hash b968405)
* Changed SimulatorIgnitionBridge to match most recent main.
* Removed exit after dsp signature generation.
* First full test suite
* Cleaned up the muorb tests
* Improved VOXL 2 self test trigger in startup file
* Removed unneeded include file
This removes the odd px4_i2c_bus_external override which was confusing
me and lead to odd and inconsistent results.
The function is now only available with an int as the argument.
- Previously it was hard to check which state the gripper was in
(GRAB/GRABBED/ etc), not even whether it was correctly initialized, this
commit adds that functionality
- When the gripper was uninitialized, the error message popped up
everytime the module had 'any' vehicle command received. But we should
be spamming only when we do actually get 'DO_GRIPPER' command received,
this commit adds that feature
- Also converted error messages for CLI when gripper test functinality
fails, so that user would get a feedback (originally, the PX4_DEBUG was
of course not outputting everything)
Also containst a mini refactor by removing the unnecessary function
get_default_altitude_acceptance_radius().
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This new check is inteded to trigger if there is no data variation published by
the airspeed driver for the first 10s after the first data is published.
This is to capture malfunctioning sensors/drivers that do publish a value, this
value though does not come from real sensor measurements.
Previously this was captured by the data stuck check, but it has shown that
some drivers can publish a stuck value without being actually malfunctioning
(e.g. when the airspeed is outside of their measurement range). Checking for
any data variation is the more conservative check that still catches the above
described failure case.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
When a mission starts capturing images using the
MAV_CMD_IMAGE_START_CAPTURE command, it should also stop it again, in
case it is stopped early, e.g. with RTL.
Issue addressed: ROS2 is built from source and
no system-wide version of fastrtps is installed
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
Such that the loiter (orbit) status can be displayed on groundstation
and the WP is accepted once within loiter radius
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- To check stack smashing (buffer overflow)
- To check if the buffer without enough length gets appropriate values
filled
- To check if the format has expected length
- Previously, not having a proper boundary check caused overflows in the
buffer (wrong memory access)
- Moreoever, negative size values getting introduced to snprintf &
strncat were also being truncated to unsigned value, hence causing
overflow, so index check needed to be added before both functions
- Fixed typo in board_common.h
If the current yaw setpoint is valid we should use it to make the transition in this direction.
For a VTOL_TAKEOFF the yaw_setpoint is used to specify the transition direction (the
vehicle is aligned towards it and then transition is started). The current yaw can be
a bit off (e.g. because MIS_YAW_ERR is large), and it is better to track the actual setpoint.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- for convenience merge px4_sitl_ign into px4_sitl_default, but allow simulator_ignition_bridge to quietly skip inclusion if ignition-transport isn't available
- simulator_ignition_bridge only try setting the system clock in
lockstep builds
- this simplifies usage and CI system dependencies
- rcS parameter backup try to directly restore param (FRAM) from backup (in case SD card is removed before successful export)
- rcS parameter backup logging rearrange to capture more logging output (param_import_fail.txt)
- posix rcS try to keep param backup and restore roughly in sync with NuttX rcS
- tinybson fix debug printf format
- param_export_internal ensure file descriptor positioned at 0 (precaution)
This state is added to give extra time between FIFO flush command. Some icm42688p IMUs need more time between config -> FIFO reset -> FIFO read. More about the issue #20181
As I understand it, only Rev 3 and Rev 4 were shipped by HB for Mini and
CM4, and are likely to be used for it.
It would be nice to have all combinations but it requires quite some
flash in the current implementation.
For targets that define the SPI buses via px4_spi_buses_all_hw, a call
to px4_set_spi_buses_from_hw_version() is needed. Otherwise a NULL
de-reference will occur when trying to access px4_spi_buses.
Fixes a system crash in px4_fmu-v5_protected:
up_assert: Assertion failed at file:armv7-m/arm_memfault.c line: 101 task: wq:lp_default
This feature allows user to use a Gripper type pacakge delivery
mechanism on a drone to trigger the delivery during a mission via the
mission item `DO_GRIPPER`.
This is a minimal change that is intended to have simplest pacakge
delivery feature on PX4, however the future scope would extend this
feature out of Navigator, and rather move towards a federated PX4
(flight-mode flexibility) architecture. But until then, this will serve
the purpose.
Update Tools/sitl_gazebo submodule to remove sdf file overwrite error
- There was an error happening due to .sdf file being overwritten, it
was caused by a wrongfully added. sdf file.
- This update pulls in the PR commit: https://github.com/Auterion/sitl_gazebo/pull/147
Initial cut on supporing PAYLOAD_PLACE mission item
Tidy and comment on navigation.h to clarify mission item definition
- Convert vehicle command ack subscription data type to
SubscriptionData, to not care about having a dedicated struct for
copying the latest data
- Tidy and comment on navigation.h to clarify the definition of
mission_item_s, which is confusing as it is an intergration of MAVLink
Standard into PX4's internal Mission Item structure
Rename mission_block's mission item reached function & cleanup navigator
- Isolated Handle Vehicle Commands function inside the Navigator
- Rename mission_block's mission item reached function to 'reached or
completed', as the navigation command can also be an action (e.g.
DO_SET_SERVO, which doesn't make sense to refer to as 'reached' when we
have successfully done executed the command)
Include MAVLink PR commit to include payload_drop message
More changes to add payload_drop MAVLink message support
- Comitting for testing purposes
Add mission item payload_drop to vehicle command payload drop link
- Now with a mission item with the nav_cmd set to 'payload drop', the
appropriate 'payload drop' vehicle command will be issued
Make Payload drop executable via Mission Plan
Implement payload_drop module to simulate payload delivery
- Simple module that acknowledges the payload drop vehicle command after
certain time, to simulate a successful delivery
Additional changes - payload drop module not working yet
- Need to do more thread stuff to make it work :(
Fix Payload Drop enum mismatch in vehicle_command enums
- First functional Payload Drop Implementation MVP
- Simple Ack & resuming mission from Navigator tested successfully
Hold the position while executing payload drop mission item
- Still the position hold is not solid, maybe I am missing something in
the position setpoint part and all the internal implications of
Navigator :(
Add DO_WINCH command support
Some fixes after rebase on develop branch
- Some missed brackets
- Some comment edits, etc
Add DO_WINCH command support
- Still has a problem of flying away from the waypoint while the
DO_WINCH is being executed, probably position setpoint related stuff :(
Apply braking of the vehicle for DO_WINCH command
- Copies the behavior of NAV_CMD_DELAY, which executes a smooth, braking
behavior when executing the delay because of the braking condition in
`set_mission_items` function
- This will not apply to Fixed wings
- The payload deploy getting triggered may be too early, as right now as
soon as the vehicle approaches the waypoint within the acceptance
threshold, the payload gets deployed
Add DO_GRIPPER support
Implement Gripper actual Hardware triggering support
- Currently not working, possibly in the mixer there's a bug
- Implemented the publishing of actuator_controls_1 uORB topic
- Implemented the test command for the payload_drop module, to test the
grpiper functionality
- Edited px4board file to include the payload_drop module
- Added Holybro X500 V2 airframe file, to enable testing on X500 V2
- Created new Quad X Payload Delivery mixer, which maps the actuator
controls 1 topic's data into the MAIN pin 5 output
Make Payload Drop Gripper Work
- Initialization of the Gripper position to CLOSED on Constructor of the
payload_drop module
- Setting the OPEN and CLOSED value to the appropriate actuator controls
input
Set vehicle_command_ack message's timestamp correctly
- By not setting the timestamp, the ack commands were not correctly
graphed in PlotJuggler!
Rename payload drop module to payload deliverer
- I think it's a more complex name (harder to type), but more generic
Add Gripper class (WIP)
Add Gripper class functionalities
- Add gripper uORB message
- Add gripper state machine
Use Gripper class as main interface in payload_deliverer
- Utilizes Gripper class functions for doing Gripper functionality
Remove mixer based package delivery trigger logic
- Remove custom mixer files that mapped actuator controls to outputs
statically
Additional improvements of the payload_deliverer
Fix payload_deliverer module not starting
- _task_id wasn't geting set appropriately in task_spawn function, which
led to runtime failure
Add Gripper Function to mixer_module
- Still not showing up as function mapping in QGC, needs fix
Add parameters to control gripper behavior
- Now user can enable / disable gripper
- Also select which type of gripper to use
Applying review from nuno
Remove timeout fetching from mission item and use gripper's timeout
- Previously, it was planned to use a custom DO_GRIPPER and DO_WINCH
MAVLink message definitions with information on timeout, but since now
we are using original message definition, only relevant timeout
information is defined in the payload_deliverer class
- This change brings in the timeout parameter to the Navigator, which
then sets the timeout in the mission_block class level, which then
processes the timeout logic
Make payload deployment work for Allmend test :P
Support gripper open/close test commands in payload_deliverer
Move enum definition for GRIPPER_ACTION to vehicle_command.msg
Remove double call for ` ${R}etc/init.d/rc.vehicle_setup`
- Was introduced during the rebase
- Was causing module already running & uORB topic can't be advertised
errors
Fix format via `make format` command
Modify S500 airframe file to use for control allocation usage
- Added Control allocation related parameters as default to not have it
reset every time the airframe is selected
Implement mission specific payload deploy timeout and more changes
Switch payload_deliverer to run on work queue
Remove unnecessary files
- Airframe changes from enabling control allocation are removed
Address review comments
- Remove debug messages
- Remove unnecessary or verbose comments & code
- Properly call parameter_update() function
Switch payload_deliverer to scheduled interval work item & refactor
- Switch to Schedeuled on Interval Work Item, as previous vehicle
command subscription callback based behavior led to vehicle comamnd ack
not being sent accordingly (since the Run() wouldn't be called unless
there's a new vehicle command), leading to ack command not being sent
out
- Also, old vehicle commands were getting fetched due to the
subscription callback as well, which was removed with this patch
- Fix the wrong population of floating point param2 field of vehicle
command by int8_t type gripper action by creating dedicated function
- Refactor and add comments to increase readability
Add gripper::grabbing() method and handle this in parameter update
- Previously, the intermediate state 'grabbing' was not considered, and
when the parameter update was called after the first initialization of
the gripper, the grab() function was being called again, which would
produce unnecessary duplicate vehicle command.
- Also replaced direct .grab() access to sending vehicle comamnd, which
unifies the gripper actuation mechanism through vehicle commands.
Navigator: Change SubscriptionData to Subscription to reduce memory usage
- Also removed unused vehicle command ack sub
PayloadDeliverer: Remove unnecessary changes & Bring back vehicle_command sub cb
When hitting zero thrust by stick there is 0 torque authority
without airmode. So 0% minimum manual thrust should never be the default
without airmode.
Uses COM_SPOOLUP_TIME to slowly ramp the throttle and allow the
tailrotor to compensate in a coordinated way based on the yaw
compesation from throttle, see CA_HELI_YAW_TH_S.
This coordinated spoolup is necessary to avoid unsafe yaw twitches
because of the heli rotating until the correct compensation kicks in
through the feedback controller.
Move logic implemented in the header files to source files, this way
it will be simpler to define which is compiled to kernel space and
which to user space
Also allows to remove some headers that pull in half the universe
from the board definitions (which should just be pin definitions and
no functionality)
Fixes the compile error on llvm:
mixture of designated and non-designated initializers in the same initializer list is a C99 extension [-Wc99-designator]
- a growing number of samples come into the backend with the time
already delayed (sensor's interrupt setting timestamp sample)
- if the incoming timestamp is already delayed then the new data checks
(relative to latest IMU) can be slightly wrong
- handle almost all timestamps and checks on delayed time horizon,
except for explicit checks of new samples
- isRecent() and isTimedOut() helpers use delayed time
- add new isNewestSampleRecent() used for checking the incoming
timestamp of the incoming (adjusted) data
- print list of events whenever they change if
CONSOLE_PRINT_ARMING_CHECK_EVENT is set, instead of on each update call
- reduce flash usage by moving out templated code into non-template method
For SITL it's important that the GCS receives the first events messages
that reset the sequence number, in case the user does not press 'Disconnect'
when restarting SITL.
- new modules/simulation directory to collect all simulators and related modules
- new Tools/simulation directory to collect and organize scattered simulation submodules, scripts, etc
- simulation module renamed to simulator_mavlink
- sih renamed to simulator_sih (not a great name, but I wanted to be clear it was a simulator)
- ignition_simulator renamed to simulator_ignition_bridge
- large sitl_target.cmake split by simulation option and in some cases pushed to appropriate modules
- sitl targets broken down to what's actually available (eg jmavsim only has 1 model and 1 world)
- new Gazebo consistently referred to as Ignition for now (probably the least confusing thing until we fully drop Gazebo classic support someday)
Also remove the legacy "range aid" than can be achieved by setting the
height reference to range finder and the range finder control parameter
to "conditional".
Conditional range aiding cal also be set when the height reference isn't
the range finder. This prevents the ratchetting effect due to switching
between references.
Instead of having a single height source fused into the EKF and the
other ones "waiting" for a failure or the primary sensor, fuse all
sources in EKF2 at the same time. To prevent the sources from fighting against each
other, the "primary" source is set as reference and the other ones are
running a bias estimator in order to make all the secondary height
sources converge to the primary one.
If the reference isn't available, another one is automatically selected
from a priority list. This secondary reference keeps its current bias
estimate but stops updating it in order to be the new reference as close
as possible to the primary one.
- much simpler direct interface using Ignition Transport
- in tree models and worlds
- control allocation output configuration, no more magic actuator mapping to mavlink and back
- currently requires no custom Gazebo plugins (keeping things as lightweight and simple as possible)
Co-authored-by: Jaeyoung-Lim <jalim@ethz.ch>
-always reset roll/pitch/yaw integrators at the same time
-reset them while waiting for launch or during FW Takeoff before Climbout
-reset wheel rate integrator only when disarmed
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Since the horizontal and vertical velocity controllers are now
decoupled, it can be that the horizontal acceleration produced by the
controller is actually greater than the desired one (by design). This
condition would actually make the ARW run "backwards", degrading the
controller performance.
Directly use yaw controls for it, and don't add it to the allocation matrix,
as that would have an effect on rudder scaling if the wheel also would have
a yaw effectiveness.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
This is the control surface type for airframes that have only a
single aileron servo or have the ailerons on a single output channel.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
2fbb70d9ca made the lowest possible
throttle value commanded by stick in Stabilized mode before taking off 0.
The real world problem with this is that when takeoff is detected then
the entire throttle scaling range jumps from
[0, MPC_MANTHR_MIN]
to [MPC_MANTHR_MIN, MPC_MANTHR_MIN].
As a result whenever MPC_MANTHR_MIN is not zero there is a thrust jump
on every takeoff just at the moment takeoff is detected even when the
stick is moved continuously.
Because of this I suggest to revert to having a higher throttle value
from the beginning on because it's less complicated and there's
no obvious value in starting out with zero thrust if anyways not
possible to go back to zero for safety once takeoff is detected.
Using the control status flags isn't robust as this part of the code
runs at the EKF update rate while the in_air transition is don at the
prediction rate. It was then likely to miss the transition
Fixes the error
googletest-src/googletest/src/gtest-death-test.cc:1283:24: error: ‘dummy’ may be used uninitialized [-Werror=maybe-uninitialized]
with GCC 11
For most VTOLs the param defaults for the agressiveness of the MC attitude controller
are too high, as VTOLs usually have high intertia and lot af drag due to wings and
can thus not rotate as fast as MCs.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
I think most vehicle can safely decend with at least 1.5m/s, and having this
value too low makes Descents/Landings/RTLs unnecessary long.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
The VTOL default was set to 1, while the param default is 1.5.
I don't see why it shuold be a different default for VTOLs and thus remove it.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- replace float32[21] URT covariances with smaller dedicated position/velocity/orientation variances (the crossterms are unused, awkward, and relatively costly)
- these are easier to casually inspect and more representative of what's actually being used currently and reduces the size of vehicle_odometry_s quite a bit
- ekf2: add new helper to get roll/pitch/yaw covariances
- mavlink: receiver ODOMETRY handle more frame types for both pose (MAV_FRAME_LOCAL_NED, MAV_FRAME_LOCAL_ENU, MAV_FRAME_LOCAL_FRD, MAV_FRAME_LOCAL_FLU) and velocity (MAV_FRAME_LOCAL_NED, MAV_FRAME_LOCAL_ENU, MAV_FRAME_LOCAL_FRD, MAV_FRAME_LOCAL_FLU, MAV_FRAME_BODY_FRD)
- mavlink: delete unused ATT_POS_MOCAP stream (this is just a passthrough)
Co-authored-by: Mathieu Bresciani <brescianimathieu@gmail.com>
fix test by reducing the distance to the new origin: the maximum size of
the local position origin is a cube of 1e6m. If the origin is moved
further than this, the state is clipped to that maximum value
Navigator sets the cruise_speed to -1 if the controller shouldn't listen to
it and instead use the default speed (for MC: MPC_XY_CRUISE). This is for
example for RTL the case, where we want to return at the default speed,
independetly of what the mission speed before was.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Add support for the basic GPS telemetry values when using the GHST protocol.
* Fix formatting in GHST GPS telemetry changes
* GHST GPS Telemetry formatting cleanup
* GHST GPS Telemetry, Last formatting change
- synthetic airspeed will only be declared valid as soon as the wind variance
has dropped below a parameterized threshold. This is useful for vehicles without
an airspeed sensor which rely on synthetic airspeed but only once the vehicle
has turned sufficiently for the wind estimates to be reliable.
Signed-off-by: RomanBapst <bapstroman@gmail.com>
- msg constant names now comply with ROS conventions:
uppercase alphanumeric characters with underscores for separating words
partially fix#19917
Signed-off-by: Beniamino Pozzan <beniamino.pozzan@phd.unipd.it>
As per the discussion in #15331, fixed issue where stm32h7 chips
use hardware ECC bits in program memory that disallow overwriting
32-byte flash line that has already been written. As such,
this change allows for a variant implementation of the flashfs system
that uses more space in the flash entry header in order to
allow an entire line to be reserved for erasing an entry.
Signed-off-by: Taylor Nelms <tnelms@roboticresearch.com>
rtl state was getting reset on inactive, which meant that the state which triggered resuming e.g. mission landing would be overwritten, and the navigator mode would switch back and forth between rtl and mission. this commit:
1. moves the reset of rtl state to the on activation function (removing it from the on inactive function)
2. functionalizes the rtl state input to the rtl time estimator so that rtl time can still be calculated from state=none while inactive
This commit removes the additional airspeed check (airspeed for VTOLs in
hover below LNDFW_AIRSPD_MAX), as it is not a required condition in the
landed state (headwind blowing into the airspeed sensor won't stop
once on the gruond). In FW mode the check would make more sense, but there
the land detector is currently simply disabled.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- if using multi-EKF across all magnetometers then an instance of
vehicle_magnetometer is advertised immediately for every sensor_mag
instance
- this can become problematic if EKF2 multi-mag is enabled, but with
only 1 IMU (EKF2_MULTI_MAG) because you will be stuck with no magnetometer data
FW_LND_USETER defaulted to 1 and FW_LND_ABORT terrain based bits all enabled. why? because using a distance sensor is critical to detecting when to flare, and we want to force the user to actively disable these safety settings if they so choose, so that they understand the implications.
- new param, FW_LND_TER_REL
- fixing the glide slope helps keep the landing glide behavior steady (avoiding bumps in the altitude setpoint from e.g. trees)
- flare is still triggered via the distance sensor, if enabled by FW_LND_USETERR
when the vehicle did not track the slope well (e.g. at an offset above the track) and the altitude setpoint flattening on intersection with terrain, the throttle would spool up to smoothly intersect the newly flattened altitude setpoint, this could happen before the flare altitude was reached, which is bad. now the steady state glide behavior will be maintained, and flare can trigger at the appropriate time
- the target_sink_rate param could possibly constrain the maximum commanded sink rate to something less than that of the landing glide slope, which would make it impossible to track. this commit allows opening up the desired max sink rate up to the performance limits of the aircraft, if necessary, for the landing case
- landing slope/curve library removed
- flare curve removed (the position setpoints will not be tracked during a flare, and were being ignored by open-loop maneuvers anyway)
- flare curve replaced by simply commanding a constant glide slope to the ground from the approach entrance, and commanding a sink rate once below flaring alt
- flare is now time-to-touchdown -based to account for differing descent rates (e.g. due to wind)
- flare pitch limits and height rate commands are ramped in from the previous iteration's values at flare onset to avoid jumpy commands
- TECS controls all aspects of the auto landing airspeed and altitude/height rate, and is only constrained by pitch and throttle limits (lessening unintuitive open loop manuever overrides)
- throttle is killed on flare
- flare is the singular point of no return during landing
- lateral manual nudging of the touchdown point is configurable via parameter, allowing the operator to nudge (via remote) either the touchdown point itself (adjusting approach vector) or shifting the entire approach path to the left or right. this helps when GCS map or GNSS uncertainties set the aircraft on a slightly offset approach"
- refactor updateQuaternion() to compute the yaw jacobian directly (respecting the rotation sequence determination)
- fuseHeading()/fuseYaw321()/fuseYaw312() helpers are eliminated and now mag heading fusion and EV yaw fusion compute the innovation in place
- clear up logic for performing zero innovation heading fusion when quaternion variance exceeds threshold (no more _is_yaw_fusion_inhibited flag manipulation)
- when at rest continue fusing last static heading with very low variance even if other heading sources are active
We need to ensure the Dronecode logo is prominently displayed
and linked to the PX4 brand for trademark protection of PX4 and Dronecode.
If you have any questions about this, please feel free to reach out directly to me.
- post takeoff, the aircraft follows the infinite line sourced from the launch point in the direction of the takeoff waypoint
- takeoff waypoint altitude is used as a clearance altitude, set such that once above, the aircraft has cleared all ground occlusions and may proceed with the mission
- runway takeoff state machine simplified to throttle ramp, clamped to runway, climbout, and fly
- throttle ramp must complete before switching to next state to avoid a jump in throttle setpoint just after takeoff if the takeoff airspeed is reached before the ramp is complete
- roll constraints near ground post takeoff removed from runway takeoff class (handled externally now)
- minimum airspeed in TECS is reduced to takeoff speed (if necessary) to lower the underspeed detection bound
- lateral-directional guidance uses a different period parameter during ground roll
- apply constraint only for takeoff and landing modes
- add two params, wing span and wing height, to calculate a reasonable height at which roll limits can be opened
- takeoff situational knowledge removed from all other modes except manual (or actual takeoff mode)
- manual takeoff is marked complete if at a controllable airspeed
- minimum pitch bounds TECS until manual takeoff complete
- remove individual roll constraints during manual takeoff (ground proximity constraints coming in subsequent commit)
- underspeed condition only determined by true airspeed undershoot
- change binary underspeed condition to a continuous percent undersped
- ramp-in max throttle, pitch speed weight, and TAS setpoint reduction during underspeed to avoid jumpy commands at the true airspeed error boundary
- let true airspeed filter reach zero airspeed
- create integral and trajectory generator reset methods
- always run TECS unless in rotary-wing mode (or in transition)
- constantly reset TECS integrals and trajectory generators when landed
- simplify takeoff reset method
- removes _last_manual variable in favor of _skipping_takeoff_detection, which is handled in the control mode setter
- takeoff detection (both launch and runway) is skipped if entering takeoff mode from any other mode while having already been in the air
- added method to runway takeoff class for force setting the fly state when we want to skip the takeoff detection
- array length of data was increased without changing the data type of
the variable holding the length
Signed-off-by: RomanBapst <bapstroman@gmail.com>
As they are always moving horizontally, the check doesn't make sense
for fixed-wings.
Also don't run the check while on ground to prevent getting a failed
check during pre-takeoff manipulation.
- trim throttle is handled entirely by the position controller
- navigator should use speed setpoints
- added flag gliding_enabled in position setpoint to stills support gliding
Signed-off-by: RomanBapst <bapstroman@gmail.com>
- allow compensation based on vehicle weight (parameterized)
- use density for calculating trim throttle compensation instead of pressure
- removed parameter FW_THR_ALT_SCL
Signed-off-by: RomanBapst <bapstroman@gmail.com>
Make Power Off tune not interruptable
This solves the case of Low Battery warning tune overriding the power
off tune, as now the Power off tune is not interruptable by any other
tune unless override flag is specified
commander_helper: resolve "redundant boolean literal in ternary expression result"
- Existing code was hard to read and quite ambiguous
- Also it allowed constantly re-sending the tune_control request for a
fixed duration tunes like "TUNE_ID_BATTERY_WARNING_SLOW", while not
respecting the tune duration
When wind is already estimated, we don't reset the states using airspeed
data, so it could be that the fusion fails if the airspeed isn't
consistent with the filter (test ratio > 1). In this case, don't start
the fusion.
When wind isn't already estimated, the wind states are reset using
airspeed so the fusion can start regardless of the current test ratio.
- As it is always only used for the vehicle command ack message
- It was a duplicate, hence making it error prone for maintainment
- The uORB message comments were updated to make the relationship with
the MAVLink message / enum definitions clear
split the fusion process into:
1. updateAirspeed: computes innov, innov_var, obs_var, ...
2. fuseAirspeed: uses data computed in 1. to generate K, H and fuse the
observation
from 300 ms to 60 seconds, to give enough time for the user to configure
the vehicle in the mean time.
This is needed especially when the battery cell count setting is wrong
(when it should be 3, but set to 4 for example), since then whenever you
boot the vehicle, it will shutdown after 300 ms, which leaves the user
puzzled as to exactly what's happening. And it also prevented the user
from changing the Parameter since it's shutting down so quickly.
60 second window is intended to be a reasonable time that will allow the
user to figure out what's going on (via checking the battery level on
QGC, etc) but also not deep discharge the battery to a dangerous level.
Fixes the error:
Aborting due to missing @type tag in file: 'build/px4_fmu-v5_test/etc/init.d/airframes/11001_hexa_cox'
This can happen due to a change to e.g. board_serial_ports, which changes
the CLI command and triggers a re-execution of the airframe processing.
- this allows landing gear to retract automatically when doing a takeoff
and the vehicle is considered high enough
Signed-off-by: RomanBapst <bapstroman@gmail.com>
- 4096 of 3 hex digits each for rev and ver is enough.
#defines used in SPI versions do not be long format, use use the macro
- Board provides a prefix and the formatting is sized and built in
- No need for funky board_get_base_eeprom_mtd_manifest interface
Original mft is used where the abstraction is done with the MFT interface
Co-authored-by: David Sidrane <David.Sidrane@Nscdg.com>
If the capture GPIO is exposed to a signal with high frequency changes, a lot of
interrupts are scheduled and the handling of these call can worst-case
starve flight critical processes leading to a loss of control. Since camera capture
is not flight critical, we now give up the capture
functionality and stop the interrupts to prevent the starvation of other processes.
If the PPS GPIO is exposed to a signal with high frequency changes, a lot of
interrupts are scheduled and the handling of these calls can worst-case
starve flight critical processes leading to a loss of
control. Since PPS is not flight critical, we now give up the PPS
functionality and stop the interrupts to prevent the starvation of other processes.
Remove the entire external yaw handler, dynamic memory allocation,
pointer passing logic. Directly instanciate the weather vane instance
in the flight tasks that support it.
- remove internal accumulation and publish every valid raw sample synchronized with sensor
- store timestamp_sample from motion interrupt
- improve timing requirements from datasheet (minimum delays after register read/write)
- remove internal accumulation and publish every valid raw sample synchronized with sensor
- store timestamp_sample from motion interrupt
- improve timing requirements from datasheet (minimum delays after register read/write)
- all sources of optical flow publish sensor_optical_flow
- sensor_optical_flow is aggregated by the sensors module, aligned with integrated gyro, and published as vehicle_optical_flow
Co-authored-by: alexklimaj <alex@arkelectron.com>
Follow me : tidied second order filter implementation
Added velocity filtered info to uORB follow target status message, and rebase to potaito/new-follow-me-task
FollowMe : Rebasing and missing definition fixes on target position second order filter
Follow Me : Remove Alpha filter delay compensation code, since second order filter is used for pose filtering now
Followme : Remove unused target pose estimation update function
Follow Target : Added Target orientation estimation logic
Follow Target : Replaced offset vector based setpoint to rate controlled orbital angle control
Follow Target : Bug fixes and first working version of rate based control logic, still buggy
Follow Target : Added target orientation, follow angle, orbit angle value into follow_target_status uORB message for debugging
Follow Target : Fix orbit angle step calculation typo bug
Follow Target : Few more fixes
Follow Target : Few fixes and follow angle value logging bug fix
Follow Target : Added lowpass alpha filter for yaw setpoint filtering
Follow Target : Remove unused filter delay compensation param
Follow Target : Add Yaw setpoint filter initialization logic and bufix for when unwrap had an input of NAN angle value
Follow Target : Add Yaw setpoint filtering enabler parameter
Follow Target : Change Target Velocity Deadzone to 1.0 m/s, to accomodate walking speed of 1.5 m/s
Follow Target : Add Orbit Tangential Velocity calculation for velocity setpoint and logging uORB topics
Follow target : Fix indentation in yaw setpoint filtering logic
Follow Target : Fix Follow Target Estimator timeout logic bug that was making the 2nd order pose filter reset to raw value every loop
Follow Target : Remove debug printf statement for target pose filter reset check
Follow Target : Add pose filter natural frequency parameter for filter testing
Follow Target : Make target following side param selectable and add target pose filter natural frequency param description
Follow Target : Add Terrain following altitude mode and make 2D altitude mode keep altitude relative to the home position, instead of raw local position origin
Follow Target : Log follow target estimator & status at full rate for filter characteristics test
Follow Target : Implementing RC control user input for Follow Target control
Follow Target : edit to conform to updated unwrap_pi function
Follow Target : Make follow angle, distance and height RC command input configurable
Follow Target : Make Follow Target not interruptable by moving joystick in Commander
Follow Target : reconfigure yaw, pitch and roll for better user experience in RC adjusting configurations, and add angular rate limit taking target distance into account
Follow Target : Change RC channels used for adjustments and re-order header file for clarity
Follow Target : Fix Parameters custom parent macro, since using DEFINE_PARAMETERS alone misses the parameter updates of the parent class, FlightTask
Follow Target : Fix Orbit Tangential speed actually being angular rate bug, which was causing a phenomenon of drnoe getting 'dragged' towards the target velocity direction
Follow Target : Final tidying and refactoring for master merge - step 1
Add more comments on header file
Follow Target : tidy, remove unnecessary debug uORB elements from follow_target_status message
Follow Target : Turn off Yaw filtering by default
Follow Target : Tidy maximum orbital velocity calculation
Follow Target : add yaw setpoint filter time constant parameter for testing and fix NAV_FT_HT title
Follow Target : Add RC adjustment window logic to prevent drone not catching up the change of follow target parameters
Follow Target : fixes
Follow Target: PR tidy few edits remove, and update comments
Follow Target : apply comments and reviews
Follow Target : Edit according to review comments part 2
Follow Target : Split RC adjustment code and other refactors
- Splitted the RC adjustment into follow angle, height and distance
- Added Parameter change detection to reset the follow properties
- Added comments and removed yaw setpoint filter enabler logic
Follow Target : Modify orbit angle error bufferzone bug that was causing excessive velocity setpoints when setpoint catched up with raw orbit setpoint
Follow Target : Remove buffer zone velocity ramp down logic and add acceleration and rate limited Trajectory generation library for orbit angle and velocity setpoint
Follow Target : Remove internally tracked data from local scope function's parameters to simplify code
Follow Target : Fix to track unwrapped orbit angle, with no wrapping
Follow Target : Apply user adjustment deadzone to reduce sensitivity
Follow Target : Apply suggestions from PR review round 2 by @potaito
Revert submodule update changes, fall back to potaito/new-followme-task
Follow Target : [Debug] Expose max vel and acceleration settings as parameters, instead of using Multicopter Position Controller
's settings
Follow Target : Use matrix::isEqualF() function to compare floats
Follow Target : Add Acceleration feedback enabler parameter and Velocity ramp in initial commit for overshoot phenomenon improvement
Follow Target : Implement Velocity feed forward limit and debug logging values
Follow Target : Apply Velocity Ramp-in for Acceleration as well & Apply it to total velocity setpoint and not just orbit tangential velocity component
Follow Target : Don't set Acceleration setpoint if not commanded
Follow Target : Use Jerk limited orbit angle control. Add orbit angle tracking related uORB values"
Follow Target : Add Orbit Angle Setpoint Rate Tracking filter, to take into consideration for calculating velocity setpoint for trajectory generator for orbit angle
Revert "Follow Target : Add Orbit Angle Setpoint Rate Tracking filter, to take into consideration for calculating velocity setpoint for trajectory generator for orbit angle"
This reverts commit a3f48ac7652adb70baf3a2fed3ea34d77cbd6a70.
Follow Target : Take Unfiltered target velocity into acount for target course calculation to fix overshoot orbit angle 180 deg flip problem
Follow Target : Remove Yaw Filter since it doesn't make a big difference in yaw jitterness
Follow Target : Remove velocity ramp in control & remove debug values from follow_target_status.msg
Follow Target : Tidy Follow Target Status message logging code
Follow Target : Remove jerk and acceleration settings from Follow Target orbit trajectory generation
Follow Target : Change PublicationMulti into Publication, since topics published are single instances
Follow Target : Edit comments to reflect changes in the final revision of Follow Target
Follow Target : Apply incorrectly merged conflicts during rebase & update Sticks function usage for getThrottled()
Follow Target : Apply final review comments before merge into Alessandro's PR
Apply further changes from the PR review, like units
Use RC Sticks' Expo() function for user adjustments to decrease sensitivity around the center (0 value)
Update Function styles to lowerCamelCase
And make functions const & return the params, rather than modifying them
internally via pointer / reference
Specify kFollowPerspective enum as uint8_t, so that it can't be set to negative value when converted from the parameter 'FLW_TGT_FP'
Fix bug in updateParams() to reset internally tracked params if they actually changed.
Remove unnecessary comments
Fix format of the Follow Target code
Fix Follow Perspective Param metadata
follow-me: use new trajectory_setpoint msg
Convert FollowPerspective enum into a Follow Angle float value
1. Increases flexibility in user's side, to set any arbitrary follow
angle [deg]
2. Removes the need to have a dedicated Enum, which can be a hassle to
make it match MAVSDK's side
3. A step in the direction of adding a proper Follow Mode (Perspective)
mode support, where we can support kite mode (drone behaves as if it is
hovering & getting dragged by the target with a leash) or a constant
orbit angle mode (Drone always on the East / North / etc. side, for
cinematic shots)
Continue fixing Follow Target MAVSDK code to match MAVSDK changes
- Support Follow Angle configuration instead of Follow Direction
- Change follow position tolerance logic to use the follow angle
*Still work in progress!
Update Follow Me MAVSDK Test Code to match MAVSDK v2 spec
- Add RC Adjustment Test case
- Change follow direction logic to follow angle based logic completely
- Cleanup on variable names and comment on code
follow-me: disable SITL test
Need to update MAVSDK with the following PR:
https://github.com/mavlink/MAVSDK/pull/1770
SITL is failing now because the follow-me
perspectives are no longer defined the
same way in MAVSDK and in the flight task.
update copyright year
follow-me: mark uORB topics optional
Apply review comments
more copyright years
follow-me sitl test: simpler "state machine"
flight_mode_manager: exclude AutoFollowTarget and Orbit on flash contrained boards
Remove unnecessary follow_target_status message properties
- As it eats up FLASH and consumes uLog bandwidth
rename follow_me_status to follow_target_status
enable follow_target_estimator on skynode
implement the responsiveness parameter:
The responsiveness parameter should behave similarly to the previous
follow-me implementation in navigator. The difference here is that
there are now two separate gains for position and velocity fusion.
The previous implemenation in navigator had no velocity fusion.
Allow follow-me to be flown without RC
SITL tests for follow-me flight task
This includes:
- Testing the setting for the follow-me angle
- Testing that streaming position only or position
and velocity measurements both work
- Testing that RC override works
Most of these tests are done with a simulated model
of a point object that moves on a straight line. So
nothing too spectacular. But it makes the test checks
much easier.
Since the estimator for the target actually checks new
measurements and compares them to old ones, I also added
random gausian noise to the measurements with a fixed seed
for deterministic randomness. So repeated runs produce
exactly the same results over and over.
Half of the angles are still missing in MAVSDK. Need to create
an upstream PR to add center left/right and rear left/right options.
These and the corresponding SITL tests need to be implemented
later.
sitl: Increase position tolerance during follow-me
Astro seems to barely exceed the current tolerance (4.3 !< 4.0)
causing CI to fail. The point of the CI test is not to check
the accuracy of the flight behaviour, but only the fact that the
drone is doing the expected thing. So the exact value of this
tolerance is not really important.
follow-me: gimbal control in follow-me
follow-me: create sub-routines in flight task class
follow-me: use ground-dist for emergency ascent
dist_bottom is only defined when a ground facing distance sensor exist.
It's therefore better to use dist_ground instead, which has the distance
to the home altitude if no distance sensor is available.
As a consequence it will only be possible to use follow-me in a valley
when the drone has a distance sensor.
follow-me: point gimbal to the ground in 2D mode
follow-me: another fuzzy msg handling for the estimator
follow-me: bugfix in acceleration saturation limit
follow-me: parameter for filter delay compensation
mantis: dont use flow for terrain estimation
follow-me: default responsiveness 0.5 -> 0.1
0.5 is way too jerky in real and simulated tests.
flight_task: clarify comments for bottom distance
follow-me: minor comment improvement
follow-me: [debug] log emergency_ascent
follow-me: [debug] log gimbal pitch
follow-me: [debug] status values for follow-me estimator
follow-me: setting for gimbal tracking mode
follow-me: release gimbal control at destruction
mavsdk: cosmetics 💄
Symmetric buffers allow a much more reliable QGC Wifi telemetry connection especially when (virtual) joysticks are used. (this board does not provide RX DMA on UART 4 as its timer does DSHOT).
The noise spectral density, NSD, (square root of power spectral density) is a
continuous-time parameter that makes the tuning independent from the EKF
prediction rate.
NSD corresponds to the rate at which the state uncertainty increases
when no measurements are fused into the filter.
Given that the current prediction rate of EKF2 is 100Hz, the
same tuning is obtained by dividing the std_dev legacy parameter by 10:
nsd = sqrt(std_dev^2 / 100Hz)
The noise spectral density, NSD, (square root of power spectral density) is a
continuous-time parameter that makes the tuning independent from the EKF
prediction rate.
NSD corresponds to the rate at which the state uncertainty increases
when no measurements are fused into the filter.
Given that the current prediction rate of the wind estimator is 1Hz, the
same tuning is obtained with the same values as before.
1. Rename *_header_s structs to *_s, since the _header postfix is not
helpful
2. Rename the "key" string variables in the message structs to
"key_value_str" as the string actually contains not just the key but the
key and value pair information
3. Add comments on other uLog messages to clarify use (need more
improvement / can be even more simplified)
* sih: Move sih out of work queue
This reverts commit bb7dd0cf00.
* sih-sim: Enable sih in sitl, together with lockstep
* sih-sim: new files for sih: quadx and airplane
* sih: Added tailsitter for sih-sitl simulation
* sitl_target: Added seperate target loop for sih
* jmavsim_run: Allow jmavsim to run in UDP mode
* lockstep: Post semaphore on last lockstep component removed
* sih-sim: Added display for effectively achieved speed
* sih: increase stack size
* sih-sim: Improved sleep time computation, fixes bug of running too fast
* sitl_target: place omnicopter in alphabethic order
Co-authored-by: romain-chiap <romain.chiap@gmail.com>
Co-authored-by: Matthias Grob <maetugr@gmail.com>
- documentation of units, params, returns, and descriptions for variables and methods
- rename ambiguous or erroneous state variables
- remove unused or unecessary input arguments to functions
- remove ugly header white space
The safetyButtonHandler() reports that the safety statatus
changed on the first loop iteration when safety is disabled which makes
sense to inform the system that safety is off but the tune for the user
should not be played because it interrupts the startup tune.
They cause problems with building px4_msgs in ROS2 Humble Hawksbill and removing them fixes the issue.
Co-authored-by: Agata Barcis <agata.barcis@tii.ae>
This corrects the board definition to use the proper polarity
for the brick power valid signal, thus allowing the board to
detect the battery and monitor it properly.
internal PX4IO safety_off state is removed and replaced with a normal safety button event. From this 'commit' commander is taking care of the PX4IO safety.
- on common IMUs like the mpu6000, mpu9250, icm20602, etc each FIFO
sample is only 12 bytes so this is still more than large enough for the
worst case transfer
- free memory is getting tight on these older boards (depending on
configuratoin) and the pixhawk 2 cube still has 2 other superior IMUs, so this is just
dropping dead weight
H7 Only supports 2 not 3 CAN interfaces.
CanInitHelper passes in in the run time configuration of
the number of interfaces. The code was ignoring these!
## Please note that feature requests are not 'fire and forget'
It is a lot more likely that the feature you would like to have will be implemented if you keep watching your feature request, and provide more details to developers looking into implementing your feature, and help them with testing.
- type:textarea
attributes:
label:Describe problem solved by the proposed feature
description:A clear and concise description of the problem, if any, this feature will solve. E.g. I'm always frustrated when ...
validations:
required:true
- type:textarea
attributes:
label:Describe your preferred solution
description:A clear and concise description of what you want to happen.
validations:
required:true
- type:textarea
attributes:
label:Describe possible alternatives
description:A clear and concise description of any alternative solutions or features you've considered.
validations:
required:true
- type:textarea
attributes:
label:Additional context
description:Add any other context or screenshots for the feature request here.
Please use [PX4 Discuss](http://discuss.px4.io/) or [Slack](http://slack.px4.io/) to align on pull requests if necessary. You can then open draft pull requests to get early feedback.
<!--
## Describe problem solved by this pull request
A clear and concise description of the problem this proposed change will solve. Or, what it will improve.
E.g. For this use case I ran into...
Thank you for your contribution!
## Describe your solution
A clear and concise description of what you have implemented.
Get early feedback through
- Dronecode Discord: https://discord.gg/dronecode
- PX4 Discuss: http://discuss.px4.io/
- opening a draft pr and sharing the link
## Describe possible alternatives
A clear and concise description of alternative solutions or features you've considered.
-->
## Test data / coverage
How was it tested? What cases were covered? Logs uploaded to https://review.px4.io/ and screenshots of the important plot parts.
### Solved Problem
When ... I found that ...
## Additional context
Add any other related context or media.
Fixes #{Github issue ID}
### Solution
- Add ... for ...
- Refactor ...
### Changelog Entry
For release notes:
```
Feature/Bugfix XYZ
New parameter: XYZ_Z
Documentation: Need to clarify page ... / done, read docs.px4.io/...
The [developer guide](http://dev.px4.io/) explains how to set up the development environment on Mac OS, Linux or Windows. Please take note of our [coding style](https://dev.px4.io/master/en/contribute/code.html) when editing files.
The [developer guide](https://docs.px4.io/main/en/development/development.html) explains how to set up the development environment on Mac OS, Linux or Windows. Please take note of our [coding style](https://docs.px4.io/main/en/contribute/code.html) when editing files.
This repository holds the [PX4](http://px4.io) flight control solution for drones, with the main applications located in the [src/modules](https://github.com/PX4/PX4-Autopilot/tree/master/src/modules) directory. It also contains the PX4 Drone Middleware Platform, which provides drivers and middleware to run drones.
This repository holds the [PX4](http://px4.io) flight control solution for drones, with the main applications located in the [src/modules](https://github.com/PX4/PX4-Autopilot/tree/main/src/modules) directory. It also contains the PX4 Drone Middleware Platform, which provides drivers and middleware to run drones.
PX4 is highly portable, OS-independent and supports Linux, NuttX and MacOS out of the box.
* Official Website: http://px4.io (License: BSD 3-clause, [LICENSE](https://github.com/PX4/PX4-Autopilot/blob/master/LICENSE))
## Building a PX4 based drone, rover, boat or robot
The [PX4 User Guide](https://docs.px4.io/master/en/) explains how to assemble [supported vehicles](https://docs.px4.io/master/en/airframes/airframe_reference.html) and fly drones with PX4.
See the [forum and chat](https://docs.px4.io/master/en/#support) if you need help!
The [PX4 User Guide](https://docs.px4.io/main/en/) explains how to assemble [supported vehicles](https://docs.px4.io/main/en/airframes/airframe_reference.html) and fly drones with PX4.
See the [forum and chat](https://docs.px4.io/main/en/#getting-help) if you need help!
## Changing code and contributing
This [Developer Guide](https://docs.px4.io/master/en/development/development.html) is for software developers who want to modify the flight stack and middleware (e.g. to add new flight modes), hardware integrators who want to support new flight controller boards and peripherals, and anyone who wants to get PX4 working on a new (unsupported) airframe/vehicle.
This [Developer Guide](https://docs.px4.io/main/en/development/development.html) is for software developers who want to modify the flight stack and middleware (e.g. to add new flight modes), hardware integrators who want to support new flight controller boards and peripherals, and anyone who wants to get PX4 working on a new (unsupported) airframe/vehicle.
Developers should read the [Guide for Contributions](https://docs.px4.io/master/en/contribute/).
See the [forum and chat](https://dev.px4.io/master/en/#support) if you need help!
Developers should read the [Guide for Contributions](https://docs.px4.io/main/en/contribute/).
See the [forum and chat](https://docs.px4.io/main/en/#getting-help) if you need help!
### Weekly Dev Call
The PX4 Dev Team syncs up on a [weekly dev call](https://dev.px4.io/master/en/contribute/#dev_call).
The PX4 Dev Team syncs up on a [weekly dev call](https://docs.px4.io/main/en/contribute/).
> **Note** The dev call is open to all interested developers (not just the core dev team). This is a great opportunity to meet the team and contribute to the ongoing development of the platform. It includes a QA session for newcomers. All regular calls are listed in the [Dronecode calendar](https://www.dronecode.org/calendar/).
Note: This is the source of truth for the active maintainers of PX4 ecosystem.
See also [maintainers list](https://px4.io/community/maintainers/) (px4.io) and the [contributors list](https://github.com/PX4/PX4-Autopilot/graphs/contributors) (Github).
See also [maintainers list](https://px4.io/community/maintainers/) (px4.io) and the [contributors list](https://github.com/PX4/PX4-Autopilot/graphs/contributors) (Github). However it may be not up to date.
## Supported Hardware
This repository contains code supporting Pixhawk standard boards (best supported, best tested, recommended choice) and proprietary boards.
Pixhawk standard boards and proprietary boards are shown below (discontinued boards aren't listed).
For the most up to date information, please visit [PX4 user Guide > Autopilot Hardware](https://docs.px4.io/main/en/flight_controller/).
### Pixhawk Standard Boards
* FMUv6X and FMUv6U (STM32H7, 2021)
* Various vendors will provide FMUv6X and FMUv6U based designs Q3/2021
* [Raspberry PI with Navio 2](https://docs.px4.io/master/en/flight_controller/raspberry_pi_navio2.html)
These boards fully comply with Pixhawk Standard, and are maintained by the PX4-Autopilot maintainers and Dronecode team
Additional information about supported hardware can be found in [PX4 user Guide > Autopilot Hardware](https://docs.px4.io/master/en/flight_controller/).
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Daniel Agar
