- this fixes a minor edge case when you only have a single accelerometer and haven't made any parameter changes, the sensor sample rate (used for filtering) might not have been updated from default
- this error is primarily useful when significant gaps in magnetometer are present, otherwise the occasional gaps at startup, etc are distracting in regular usage
- ecl in PX4/Firmware (17da6293aa8e32a0e41c8fe10b8ffbf4e3e922b3): e57af44d71
- ecl current upstream: 71fc1b8161
- Changes: e57af44d71...71fc1b8161
71fc1b8 2021-05-20 bresch - airspeed fusion: use yaw from 312 sequence when pitch is close to 90deg
- when the mpu9250 mag is active it disrupts IMU sampling
- this change modifies the simple mpu9250 FIFO integrity checks (comparing duplicate accel samples) to tolerate these samples that don't match if the mag is active
- fixes#17615
- libcanard in PX4/Firmware (e2ba4858a3285fe7ff0f2888d8b1ce5dbda589a1): 38796831f9
- libcanard current upstream: 2a11617028
- Changes: 38796831f9...2a11617028
2a11617 2021-05-20 Pavel Kirienko - Update the changelog
a1efa3a 2021-05-20 Peter van der Perk - Expose CanardRxSubscription fields for read-only use (#167)
- ecl in PX4/Firmware (2a18b8feeb24dcb07c2dd141e6097e211f83e6bd): 85f5a935fe
- ecl current upstream: e57af44d71
- Changes: 85f5a935fe...e57af44d71
e57af44 2021-05-18 Bazooka Joe - set parameters that doesn't change with const attribute, for clearification
- sitl_gazebo in PX4/Firmware (61e972ba350e4bbd14963ed6a3680da5281e7b1d): f10ccb7ec3
- sitl_gazebo current upstream: 5c24889852
- Changes: f10ccb7ec3...5c24889852
5c24889 2021-05-20 Andrew Voznytsa - Use cmake to find and link to gstreamer-app-1.0
This includes the following:
- Remove warning/throw_error flag as it is always !home_alt_valid.
- Remove impossible code path when home_alt_valid = false and
throw_error = false.
- Add home_alt_valid check in second check before using home_alt.
- Only use warning for the second check to allow waypoints below home.
- Don't return early when only warning.
This adds support to handle INT32_MAX for COMMAND_INT.x/y by converting
it to NAN internally.
It also adds paranoid checks to prevent:
- NAN being used sent by a COMMAND_INT by mistake.
- INT32_MAX being used sent by a COMMAND_LONG by mistake.
- disable noise suppression filter
- select more appropriate dynamic range
- carefully set and clear all ODR, DR, and OSRCNTL bits
- retry in probe multiple times
- increase SPI stall time slightly
- tolerate mag self test failure (could be due to local magnetic field)
- register configuration compatible with older ADIS16448AMLZ
- don't publish duplicate accel/gyro
- only allocate CRC perf counter if using CRC
- sitl_gazebo in PX4/Firmware (c3884b5bc1949a2bbdf61c40deef9d0799035152): 402897d225
- sitl_gazebo current upstream: f10ccb7ec3
- Changes: 402897d225...f10ccb7ec3
f10ccb7 2021-05-17 grahamjamesaddis - Fix non unique name warning in tailsitter.sdf.jinja (#756)
05b8625 2021-05-13 Andrew Voznytsa - Fix gstreamer timestamps; Use push mode all the way; Avoid duplicated frame encoding; Cleanup a bit there and here
9ec04cc 2021-05-16 Graham James Addis - Tidy rotor colours Grey/Blue.
493c5dc 2021-05-16 Graham James Addis - Minimal conversion to jinja format.
- ecl in PX4/Firmware (4668ea5c32075aca28fd813881a349f765ad8d2a): 29243ac5cb
- ecl current upstream: 85f5a935fe
- Changes: 29243ac5cb...85f5a935fe
85f5a93 2021-05-11 bresch - [AUTO COMMIT] update change indication
f7c687e 2021-05-11 bresch - GNSS yaw unit test: test fallback to non yaw aiding mode
a4fe2b1 2021-05-11 bresch - yaw_align: let the mag control logic do the alignement and resets
2cc0230 2021-05-11 bresch - mag control: select mag mode before doing the reset
fe943cf 2021-05-11 bresch - mag reset: force align in MC custom mode (aka. indoor mode)
cadcc58 2021-05-11 bresch - mag reset: do not stop the fusion from inside the reset function
6248646 2021-05-11 bresch - EV yaw: move EV yaw reset out of resetMagHeading function
48722a2 2021-05-07 bresch - control: inhibit EV yaw from restarting when GPS aid wants to start
a685987 2021-05-07 bresch - GNSS yaw: unset yaw_align if fusion stops before takeoff
7ee6970 2021-05-04 bresch - GNSS yaw: add test case when measurement jumps
7a11871 2021-05-04 bresch - GNSS yaw tests: adjust timing
51197a5 2021-05-03 bresch - GNSS yaw selection logic refactor
e.g.:
WARN [mavlink] stream ADSB_VEHICLE not found
WARN [mavlink] stream GIMBAL_DEVICE_ATTITUDE_STATUS not found
WARN [mavlink] stream GIMBAL_MANAGER_STATUS not found
WARN [mavlink] stream GIMBAL_DEVICE_SET_ATTITUDE not found
WARN [mavlink] stream GPS2_RAW not found
WARN [mavlink] stream UTM_GLOBAL_POSITION not found
ERROR [mavlink] configure_streams_to_default() failed
This avoids incremental build errors when switching between branches with
a different set of airframes.
E.g:
Aborting due to missing @type tag in file: 'Firmware/build/px4_fmu-v5_default/etc/init.d/airframes/13030_generic_vtol_quad_tiltrotor'
- update clean to recurse all build directories and use build system clean
- git clean is used to remove submodule generated build artifacts that
are left in the source tree
- distclean now discards all build directories and any gitignored
filess that were generated in source directories (but not top level)
The calibration is not found in HITL mode. Therefore, I suggest to
ignore this step and assume the calibration is fine.
This mostly fixes the preflight check indicator in QGC, arming was (for
some reason?) already possible.
[BACKPORT] binnfmt:Fix return before close ELF fd
stm32h7: serial: use dma tx semaphore as resource holder
[BACKPORT] stm32h7:Serial Add RX and TX DMA
[BACKPORT] drivers/serial: fix Rx interrupt enable for cdcacm
[BACKPORT] stm32h7:Allow for reuse of the OTG_ID GPIO
[BACKPORT] stm32f7:Allow for reuse of the OTG_ID GPIO
The both results of ?: should be of same type, and some compilers give error
on this:
" implicit conversion from 'float' to 'double' to match other result of conditional"
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
This was only handled outside because FlightTaks lived in the
multicopter position controller which produces the data that was
needed but now it doesn't make sense anymore to handle this
subscription separately.
It's better to have it inside the base task to have the data available
on task activation sucht that e.g. Altitude mode can take over smoothly
from Position mode.
For any normal use case where a downwards velocity setpoint is set
this works exactly the same as before.
E.g. autonomous landing, landing in Altitude or Position mode
The advantage is that the very common case where a vehicle tries
to hold a constant altitude but fails to do so e.g. during a hard brake
with too much lift the resulting downwards velocity was interpreted
as descend intent and since the vehicle already struggled to hold altitude
with low thrust and was not moving fast anymore because it was braking
this lead to a lot more false positives on certain vehicle types.
The disadvantage is that not setting a downwards velocity setpoint but
just moving the position setpoint into the ground does not result in
land detection anymore. We do not use this method of landing anymore for
quite a while. It's not recommended and I wonder if there's some rare use
case like offboard where this is done.
We could add an additional case for the specific case to land with a
position setpoint only.
The tree stages used arbitrary 350, 250 300ms totally 900ms
So this changes it to each stage to a third of the parameter.
Default it is 1 second -> 333ms per stage.
Script failed for me with following error
PX4-Autopilot/Tools/setup/arch.sh: line 159: syntax error near unexpected token `else'
PX4-Autopilot/Tools/setup/arch.sh: line 159: ` else'
Seems like there is nothing to do in case of positive if case.
Changed code should maintain logic
It was used to make the vehicle needing to accept the waypoint after a VTOL transition in the new VTOL mode
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
If the GPS data is passed from companion computer, there may be inaccuracies
with timesync. This may cause time deltas to overflow.
Make the time delta calculation using floats directly, wich results in negative
numbers in case there are such inaccuracies.
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
Instead of only keeping track of the sequence ID of specific "supported"
components, we now keep track of any sysid/compid of an incoming
message. Before this change, unknown components (such as jMAVSim) would
completely screw up the mavlink message stats and create confusion (at
least in my case).
With this change we currently keep track of up to 8 other components.
Once we reach the limit, we will print a warning.
Without these fields the pre-arm check would complain and fail.
Also, the voltage is adjusted to be at around 70% rather than 30% which
would almost start to trigger warnings.
- ecl in PX4/Firmware (cb999f37d4891ebfbbc21b4ce9b3851888b39ad4): 5d34d7a24e
- ecl current upstream: a7b8afe420
- Changes: 5d34d7a24e...a7b8afe420
a7b8afe 2021-04-30 Eike - Allow rangefinder fusion in vision height mode (Fix for #994) (#999)
4ac57d3 2021-04-25 Daniel Agar - EKF: increase fault flags value size to fit current flag bits (> 16)
based on our experience with VTOL it makes sense to change the yaw mode default. You always want your VTOL to yaw in transition direction before starting the transition.
Starting with MAVSDK 0.39.0 we can specify the version required. This
way we can fail at configure time instead of later during compilation or
linking.
- if distance to the ground is available then hysteresis times will be increased
by a factor of 3 if vehicle is higher than 1m above ground
Signed-off-by: RomanBapst <bapstroman@gmail.com>
Now sets the GDB serial device to the path of a DroneCode Probe by
default, with option command-line option override.
A few usage instructions added to the top of the file.
- no warning if accel is disabled
- threshold increased 100 -> 1000 (only warn if severe ongoing clipping)
- more generic warning message (vehicle isn't necessarily in air flying)
For the sake of efficiency (at 8 kHz) all filtering is performed on the raw data before the calibration and rotation is applied to only the final output. As a result we have to be a bit more careful when switching between sensors or in cases where the gyro scale factor changes (eg icm42688p 20 bit data rescaled to fit in int16 output).
to suppress the warning message:
`xargs: warning: options --max-args and --replace/-I/-i are mutually exclusive, ignoring previous --max-args value`
when running make format.
This entire feature only has an impact if the last mode set
a huge acceleration and we have to take over as smooth as possible.
But it's stil lworth fixing.
- matrix in PX4/Firmware (d9ebf2a842f611976d7ba5078d3f31ad8cde55a7): 3679f7fd51
- matrix current upstream: 1d0e7f1ca1
- Changes: 3679f7fd51...1d0e7f1ca1
1d0e7f1 2021-04-14 Daniel Agar - Matrix: use naive per element copy instead of memcpy call
1344ce0 2021-04-14 Daniel Agar - Euler: simplify DCM constructor
Fix a potential crash caused by calling erase_sector with a null
sector_descriptor (current_sector == 0).
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
There are settings that have to be the same for a split output across multiple instances, like for example the PWM mask and rate configurations. This change collects them for all outputs of the same underlying driver structure and applies the complete set.
- public_regulated_data_types in PX4/Firmware (9101f5707a251c33048289249bd565eba5d3a848): 53a7dbbf85
- public_regulated_data_types current upstream: 309b251a7e
- Changes: 53a7dbbf85...309b251a7e
309b251 2021-04-11 Peter van der Perk - reg.drone Battery parameters add nominal voltage (#113)
This change reflects the medium-term experience with users using USB as data link or power source in R&D setups. While we still discourage this for volume products, not allowing it seems like a major hurdle and blocker in R&D setups.
- public_regulated_data_types in PX4/Firmware (9f25ac4175fd81259b11cfe79f0f875a21562d3d): 53a7dbbf85
- public_regulated_data_types current upstream: 309b251a7e
- Changes: 53a7dbbf85...309b251a7e
309b251 2021-04-11 Peter van der Perk - reg.drone Battery parameters add nominal voltage (#113)
- public_regulated_data_types in PX4/Firmware (1557f3a23461a144dd75edae4595665e5684597d): 7f5489e6e9
- public_regulated_data_types current upstream: 53a7dbbf85
- Changes: 7f5489e6e9...53a7dbbf85
53a7dbb 2021-03-31 Pavel Kirienko - Change wording in uavcan.pnp.NodeIDAllocationData to reflect the less static nature of UAVCAN v1 (#111)
c3aa82f 2021-03-23 Pavel Kirienko - Update Readiness as discussed at the DS-015 call
fcc1062 2021-03-16 Pavel Kirienko - Doc amendment for uavcan.register.Access: mapping between registers and environment variables (#109)
342f358 2021-02-22 Pavel Kirienko - Remove misleading comment in battery.Status.0.2
1baa9cb 2021-02-22 Pavel Kirienko - Add link to Nunaweb and synchronize the description with the front page (#108)
- include icm20948 on most boards by default
- create more test variants for default boards near flash limit (cuav_nora_test, cuav_x7pro_test, holybro_durandal-v1_test)
Tailsitter VTOLs very occasionally gets stuck with zero roll and pitch angle in multicopter mode after
a forward transition command is issued.
This very rare behavior is triggered by the following events:
1> a forward transition is triggered either in auto or manual mode.
2> in the vtol_att_control main loop, if multicopter and fixed wing attitude setpoints are not updated, transition state is not updated
3> the commander changes the vehicle status to transition mode.
4> multicopter pos controller initiated Transition flight task. This results in zero roll and pitch setpoint due to zero acceleration setpoint
5> now vtol_att_control executes and updates the transition state. Specifically, _q_trans_start and _q_trans_sp are set with zero roll and pitch sp
6> tilt is evaluated to be NaN, despite _q_trans_sp being normalized. This happens for 25% of all yaw angles when using float datatype. This can be
verified using the matrix library
7> once tilt is evaluated to be NaN, _q_trans_sp is never updated again and is stuck in this state for ever.
This has been fixed by constraining the cos(tilt) within +1 to -1 range
Further, _q_trans_start and _q_trans_sp are immedietly initialized after a transition event is triggered.
This changes the default mavlink message set from onboard to
onboard_low_bandwidth, which drastically reduces the bandwidth required to get a
usable connection.
- this is to help more users get the benefit (by default) and perhaps circumvent the common mistake of setting EKF2_HGT_MODE to range sensor
- this should be safe to enable as the range aid defaults are fairly conservative (max horizontal velocity 1 m/s, and range aid gate 1 SD)
statfs accesses the file-system and can be blocking for an extended period. Since the SD card check is part of the preflight checks in the main thread of commander, it could block its execution and cause various issues. The SD card is only mounted in rcS during boot so the state will not change after the first check.
- Source processing now happens on original source files:
- processing to line-by-line
- required overhaul of regex match patterns + processing
- pros:
- enable tracing of ambiguous parsing sites -- reports (module, file, line-number, line-contents)
- simplifies code
- reduces computational complexity
- cons:
- certain declarations are harder to parse: multiline declarations
- refactors:
- added specific subclasses for each: Publications, Subscriptions, Ambiguities
- added a "Scope" class to represent either a module ('ModuleScope') or a library ('LibraryScope')
- regexes:
- added cases for C++-style classes
- expanded C++-style regex cases to accommodate templates
- `ORB_ID::` is accepted wherever `ORB_ID(` is accepted
- adds 'orb_copy' regex to the subscription cases
- emit ambiguous-line warning for declarations with `ORB_ID` on the same line
- emit ambiguous-line warning for `ORB_ID` with a declaration on the same line
- changed 'module whitelist' to 'scope-whitelist'
- whitelist may now apply to libraries
- libraries are optionally merged with their depending modules (but not by default)
- may be merged with their depending modules with the `--merge-depends` cli flag
- eliminates some redundant 'special-case' handling code
- debug output
- raises exception and aborts if a topic is found outside of a scope
- debug output is now printed & filtered with the python 'logging' standard module
- alphabetizes topic output in debug logging
- fixes debug output if package dependencies are missing
- now warns on ambiguous matches
- prints a list of ambiguous source sites (aka warnings) on completion
- (still) emits a warning if we find ORB_ID outside of a scope
- adds warnings if any of the source paths are invalid
- do not emit debug output for modules outside of the module/scope whitelist
- Expand script's CLI parameters
- added 'none' output options: undocumented debugging option to silence file output while debugging
- added the `--merge-depends` cli flag -- merges output of modules & their dependee libraries
- the ekf2 frontend typically runs in the background for up to 30 seconds waiting for all instances to appear, but this isn't supported by the legacy posix launcher
This is to avoid race condition with the yaw emergency estimator having
the same trigger delay of 1 second. Commander will now give more time to
EKF2 to reset itself before switching to altitude mode.
px4_fmu-v4_uavcanv1 fails to build:
../../src/drivers/uavcan_v1/Uavcan.cpp: In static member function 'static int UavcanNode::start(uint32_t, uint32_t)':
../../src/drivers/uavcan_v1/Uavcan.cpp:140:29: error: 'interface' was not declared in this scope
140 | _instance = new UavcanNode(interface, node_id);
- ecl in PX4/Firmware (a730f0f5ce7a2cc909cb3d0dfab5f0106a9b3aeb): bb950a1550
- ecl current upstream: e3d1ade660
- Changes: bb950a1550...e3d1ade660
e3d1ade 2021-03-12 Daniel Agar - EKF: use flow for vel test ratio if only active source of horizontal aiding
I don't think we should be broadcasting by default as we haven't done
that in the past. This suddenly spams the network with a lot of
messages, and leads to confusing situations in offices where there are
multiple PX4 SITL and QGC intances are open.
- rotates accel & gyro FIFO data before publication both to simplify downstream usage (including log review) and fix other issues
- to best handle int16_t data rotations are now either performed with swaps if possible, otherwise promoted to float, rotated using the full rotation matrix, then rounded back to int16_t
- fix sensors/vehicle_angular_velocity filter reset both with proper rotation and new calibration uncorrect helper
- in FIFO case filtering is done before calibration is applied, but we need to handle a possible reset from a completely different sensor (vehicle body angular velocity -> sensor frame uncorrected data)
- sitl_gazebo in PX4/Firmware (6e9f745809c0c99e7c73efa3e40b87ad39f144e8): b195315b86
- sitl_gazebo current upstream: 1f339cdf5c
- Changes: b195315b86...1f339cdf5c
1f339cd 2021-03-20 Dongoo Lee - Pass ros_distro in CMakeLists.txt instaed of checking it in jinja_gen.py (#712)
1b1afca 2021-03-18 David Jablonski - gst camera: add RTMP streaming and nvidia encoding (#727)
- debug output is now printed & filtered with the python 'logging' standard module
- changed 'module whitelist' to 'scope-whitelist'
- whitelist may now apply to libraries
- libraries are not included by default
- may be merged with their depending modules with the `--merge-depends` cli flag
- eliminates redundant 'special-case' handling code
- greatly expands debugging output
- fixes debug output if package dependencies are missing
- still crashes on error matches
- now warns on ambiguous matches
- prints a list of ambiguous source sites (aka warnings) on completion
- adds warnings if any of the source paths are invalid
- do not emit debug output for modules outside of the module/scope whitelist
- Expand script's CLI parameters
- added 'none' output options: undocumented debugging option to silence file output while debugging
- added the `--merge-depends` cli flag -- merges output of modules & their dependee libraries
- Source processing now happens on original source files:
- processing to line-by-line
- required overhaul of regex match patterns + processing
- pros:
- enable tracing of ambiguous parsing sites -- reports (module, file, line-number, line-contents)
- simplifies code
- reduces computational complexity
- cons:
- certain declarations are harder to parse (multiline arrays)
- refactors:
- added specific subclasses for each: Publications, Subscriptions, Ambiguities
- added a "Scope" class to represent either a module ('ModuleScope') or a library ('LibraryScope')
With this change we prevent the case where arming silently fails within
the first 10 seconds after boot.
Also, we now set the sensors as healthy as soon as the ekf is healthy,
and don't wait 10 seconds without actually checking.
The user-defined literals for milli- and microseconds
should have argument names matching their units. The
current argument names 'seconds' is probably an oversight.
Refering to the refernece manual:
Tx queue operation is configured by programming FDCAN_TXBC.TFQM to 1. Messages
stored in the Tx queue are transmitted starting with the message with the lowest message
ID (highest priority). **In case that multiple queue buffers are configured with the same
message ID, the queue buffer with the lowest buffer number is transmitted first**
Tx FIFO operation is configured by programming FDCAN_TXBC.TFQM to 0. Messages
stored in the Tx FIFO are transmitted starting with the message referenced by the get index
FDCAN_TXFQS.TFGI. After each transmission the get index is incremented cyclically until
the Tx FIFO is empty. The Tx FIFO enables transmission of messages with the same
message ID from different Tx buffers in the order these messages have been written to the
Tx FIFO
The issue will be cancelation:
The FDCAN supports transmit cancellation. To cancel a requested transmission from a
dedicated Tx buffer or a Tx queue buffer the Host has to write a 1 to the corresponding bit
position (= number of Tx buffer) of register FDCAN_TXBCR. Transmit cancellation is not
intended for Tx FIFO operation.
But there is nothing preventing it. This seems to indicate it will
work. When a transmission request for the Tx buffer referenced by the get index is canceled, the
get index is incremented to the next Tx buffer with pending transmission request and the Tx
FIFO free level is recalculated. When transmission cancellation is applied to any other Tx
buffer, the get index and the FIFO free level remain unchanged.
PX4 uses banks of 8 outputs as a logical structure. Boards that have
more outputs than 8 get multiple instances. This is an arbitrary choice
that helps with overall structure and enables the mixing of different
device classes (like FMU, IO or UAVCAN).
Without reverting this a constant boot loop occurs.
Prior to merging, why this occurs on some F7 boards and not this one should be looked at.
With LazyFPU enabled, v1.12.0 beta1 will not boot but without it removed it works fine.
- fully respect datasheet quality and shutter metrics for mode changes
- use MOTION pin for scheduling if available
- log light mode
- refactor common enable LED code
- respect read and write time delays
Send battery status (ghst_telemetry). Apply factors to show correct values of volts, amps and mAh. Change ghost protocol code to follow more MISRA C++ guidelines.
- adds new ulog message & compatibility flag
- separately add airframe and system-wide defaults
- log only non-volatile defaults that are different from the current value
- additional size is ~3KB for 100 params
This was an idea to be able to reinitialize on mode change e.g. from
Holde mode to Land mode which are currently all still handled by
FlightTaskAuto and don't require a task switch.
But I found out it leads to issues because the last setpoint and the
ekf reset counter state from the previous task are lost and as a result
the setpoint transition cannot be handled consistently anymore.
Stall speed check would otherwise trigger during landing if airspeed falls below
stall speed before landing is detected.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
subsscription for services responses and request and helps the usage of fixed
port subscribers Furthermore move register autconfigure logic from Uavcan.cpp
to NodeManager
* msg: Add estinator information and warning events message (estimator_event_flags)
* ekf2: publish information and warning events
* logger: log estimator_event_flags
* update ecl submodule to latest
Co-authored-by: Daniel Agar <daniel@agar.ca>
This commit adds support for specifying the spawn location of vehicles
in the Gazebo multi-vehicle simulator script (frame:number:x:y).
Behavior when x and y are not specified remains the same as before.
that prevents the vehicle from crashing with invalid setpoints or
states.
This broke with #16869 when the scheduling of the position control module
and the setpoint generation got independent. The failsafe mechanism assumed
the setpoint is overwirtten by the possibly infeasible input on every loop
iteration which is not the case anymore. As a result the failsafe reset its
histeresis based on the failsafe setpoint from the last loop iteration.
Keeping the failsafe_setpoint separate solves this issue. Note that
these setpoints to the bare minimum to keep the vehicle safely in the air
and do not suffer from sideffects ignoring the EKF reset.
If you're flying in manual position control mode and lose position the state machine will put you in altitude control mode. Extending the reliant on optical flow relaxed position validity thresholds allows you to potential get back into position control mode with flow alone.
- handle SET_POSITION_TARGET_LOCAL_NED and SET_POSITION_TARGET_GLOBAL_INT with ORB_ID(trajectory_setpoint)
- FlightTaskOffboard not needed at all
- bypass position_setpoint_triplet entirely (start removing extraneous fields)
- simplify offboard_control_mode to map to supported control modes
Adds support for using the MAVLink command MAV_CMD_DO_SET_ACTUATOR to
update the actuator values on control group 3 aux{1, 2, 3}. A simple
deconfliction with rc_update is implemented: when a MAVLink command is
sent, RC is disabled for that channel until a major stick movement is
detected.
- in multi-EKF mode the EKF selector becomes repsonsible for sensor
selector rather than the sensors module
- this updates the sensors module to still make the initial primary IMU
selection on startup before the EKF selector (including if the
estimators never fully initialize)
Also add commented-out code for use with PR-16808
(MixingOutput + output_control)
Bench-tested PWM output on a Pixracer via UAVCANv1 ESC commands from a
Pixhawk 4.
Now running into issues with running out of stack frame memory
For now I'm going to leave the relevant code in so it's at least
readable, but in its current state it will not compile
nxp_ucans32k146:Relocation for Bootloader
nxp_ucans32k146:can_boot enable CAN
nxp_ucans32k146:Save Space use Non Optimize memcpy
nxp_ucans32k146:Increase to 24K
nxp/ucans32k146:Canbootloader LED Driver
nxp_ucans32k146:Can bootloader shut down CAN
nxp_ucans32k146:Use NVMEEPROM for Paramaters
nxp_ucans32k146:Use bootloader AppDescriptor
px4 mtd:Support onchip emulated eeprom
nxp/s32k14x:board_identity: Return length of mfguid
nxp/s32k14x:CAN driver
nxp/s32k14x:Drver Added ABORT on error
canbootloader:Use N words for first word
canbootloader:Ensure the up_progmem API always defined
The drag is based on max_acc/max_vel, which means that increasing the
maximum velocity leads to slower braking (at the same starting speed).
Especially a combination of small max_acc (slow responsiveness) with high
max_vel led to an exceedingly high braking distance.
This improves that while still being smooth.
Previous horizontal position noise was a white Gaussian noise with std=0.8m
It results in a noise with high frequencies too high making some ekf
position tests fail (test ratio to allow arming).
The new noise values are below real GPS errors but as theses errors are
generally low frequency, so they cannot be modeled with a white noise.
These changes remove the two code paths to updates (forceupdate and update) and try to reboot and bootload IO independent of its state, wether it is in bootloader mode already, safety switch is off or if it is in "nominal state" (running and safety on). This ensures that there are no conditions where user error or boot sequence can prevent a successful upgrade. The only state where an upgrade will not be done is if the system is fully armed.
- publish wind estimate only from EKF, and wind speeds from airspeed selector to new separate topic (airspeed_wind)
- rename message wind_estimate to wind
- publish wind from currently used ekf instance (ekf2selector)
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
It was for example possible that DSM got parsed, and in the next iteration
SBUS got parsed, leading to multiple PX4IO_P_STATUS_FLAGS_RC_ flags set.
input_rc::input_source was then incorrectly set to DSM.
Partially guarding was already done: if SBUS got parsed, DSM and others
were skipped.
The IO will still clear all PX4IO_P_STATUS_FLAGS_RC_* flags on RC loss.
batt_smbus for BQ40Z80 transfers up to 34 bytes (32 byte block + 2 byte
address), but this was overflowing and failing the PEC check.
So increase the buffer size
Signed-off-by: Alex Mikhalev <alex@corvus-robotics.com>
Supporting direct down loads from ROMFS with preferece give to the
files fould on the SD card first. This will allow a user to provide
an updated uavcan firware on the SD card, and there is no overhead
of coping files from the ROM FS to the SD card.
Previously, we did not set a remote port which meant that the default
remote port 14550 was used. This meant that the mavlink instance
talking to the gimbal was interfering with the connection to the ground
station (on 14550).
We use the stabilize param to set the input. Like this the input flags
can be changed using the configure message later, and then eventually
used in the output.
This includes several changes to fix POI when used with MAVLink gimbal
v2 input:
- Correctly set capability flag that POI is supported.
- Keep lat/lon and calculate attitude on each cycle, instead of only
once on init.
- Always publish gimbal manager information, with or withoug gimbal v2
device.
We should be using gimbal_manager_information and not
gimbal_device_information. Plus, this updates the fields and flags
according to the MAVLink changes.
This should fix the case where AUX gimbals don't work anymore when
gimbal input is mavlink gimbal v2.
The fix is mostly to take care against NAN inputs.
This is a hotfix to restore gimbal functionality for SITL set-ups that
are set to MNT_MODE_IN 4 (gimbal protocol v2) but still need to accept
gimbal v1 commands that might be part of a mission, e.g. uploaded by
MAVSDK.
This is a hack to make sure the messages from the gimbal arrive at other
links (e.g. the ground station). It means that the gimbal does not get
flooded with all other messages that would get forwarded but messages
from the gimbal will still make it through.
- add command to request a message
- add gimbal attitude message
mavlink_receiver handle GIMBAL_MANAGER_SET_ATTITUDE
first implementation of new vmount input MavlinkGimbalV2
- setup class
- decode gimbal_manager_set_attitude in ControlData
add gimbal information message
add gimbal manager information and vehicle command ack
mavlink messages: add stream for GIMBAL_MANAGER_INFORMATION
mavlink_receiver: handle GIMBAL_DEVICE_INFORMATION
remove mavlink cmd handling from vmount input MavlinkGimbalV2
complete gimbal manager:
- send out fake gimbal_device_information for dummy gimbals
- complete ROI handling with nudging
small fixes
fix typos
cleanup
- gimbal device information
- flags lock
- check sanity of string
add support for CMD_DO_GIMBAL_MANAGER_ATTITUDE
stream GimbalDeviceAttitudeStatus for dummy gimbals
- add uROB gimbal_attitude_status
- fill status in vmount output_rc for dummy gimbals not able to send the
status themselves
- stream mavlink GimbalDeviceAttitudeStatus
better handle the request for gimbal infomation request
clean up
bring gimbal information back on vmount init
add new gimbal message to mavlink normal stream
fix publication of gimbal device information
rename gimbal_attitude_status to gimbal_device_attitude_status
stream gimbal_manager_status at 5Hz
mavlink: send information only on request
Sending the information message once on request should now work and we
don't need to keep publishing it.
mavlink: debug output for now
make sure to copy over control data
mavlink: add missing copyright header, pragma once
mavlink: address review comments
mavlink: handle stream not updated
Our answer does not just depend on whether the stream was found but
whether we actually were able to send out an update.
mavlink: remove outdated comment
vmount: add option for yaw stabilization only
The stabilize flag is used for gimbals which do not have an internal IMU
and need the autopilot's attitude in order to do stabilization. These
gimbals are probably quite rare by now but it makes sense to keep the
functionality given it can e.g. be used by simple servo gimbals for
sensors other than highres cameras.
The stabilize flag can also be re-used for gimbals which are capable of
stabilizing pitch and roll but not absolute yaw (e.g. locked to North).
For such gimbals we can now set the param MNT_DO_STAB to 2.
We still support configuring which axes are stabilized by the
MAVLink command DO_MOUNT_CONFIGURE, however, this is generally not
recommended anymore.
vmount: fix incorrect check for bit flag
mavlink_messages: remove debug message
Signed-off-by: Claudio Micheli <claudio@auterion.com>
use device id
remove debug print
gimbal attitude fix mistake
clang tidy fix
split:
- gimbal_attitude -> gimbal_device_set_attitude, gimbal_manager_set_attitude
- gimbal_information -> gimbal_device_informatio, gimbal_manager_information
add gimbal protocol messages to rtps msg ids
support set attitude for gimbal directly speaking mavlink
clean up gimbal urob messages
vmount: address a few small review comments
vmount: split output into v1 and v2 protocol
This way we can continue to support the MAVLink v1 protocol. Also, we
don't send the old vehicle commands when actually using the new v2
protocol.
vmount: config via ctor instead of duplicate param
vmount: use loop to poll all topics
Otherwise we might give up too soon and miss some data, or run too fast
based on commands that have nothing to do with the gimbal.
typhoon_h480: use gimbal v2 protocol, use yaw stab
Let's by default use the v2 protocol with typhoon_h480 and enable yaw
lock mode by stabilizing yaw.
- ecl in PX4/Firmware (a5a1d4caa0d04cb79ba29e98ba0af04e20c53de9): d4258cc66d
- ecl current upstream: 310f415175
- Changes: d4258cc66d...310f415175
310f415 2021-02-16 Daniel Agar - EKF add const state reset status access
0c5291d 2021-02-16 isidroas - Heading reset to magnetometer from GPS or EV (#969)
- new heater_status logging message
- run continously at low rate until configured sensor is found
- fix px4io fd bugs (fd open/close/ioctl must be same thread)
- new IMU driver structure with state machine (no sleeps in bus thread)
- verify all configured registers and trigger reset on failure
- detect if DIO1 or DIO2 are actually connected for data ready interrupt usage
- don't use CRC-16 on burst transfers except for verified lots
* Commit for the Integration of a position controller for the a Underwater vehicle.
This module is an extension of the uuv_att_control to control an Underwater vehicle to any position, given by the SET_POSITION_TARGET_LOCAL_NED which includes x y z yaw.
Since the position control is designed for a 6DOF Robot, the roll and pitch angle are controlled to be 0.
Additionally there is a stabilization control, which holds the robot at a defined depth, and not move in any direction.
In general the idea is to have this position module to control the position of the uuv. The position module reseives the desired position of the uuv and sends appropriate attitude setpoints to the uuv_attitude_control module.
Additionally the mixer file is adapted, to include the 6 different inputs(x y z roll pitch yaw).
* Commit for the Integration of a position controller for the a Underwater vehicle.
This module is an extension of the uuv_att_control to control an Underwater vehicle to any position, given by the SET_POSITION_TARGET_LOCAL_NED which includes x y z yaw.
Since the position control is designed for a 6DOF Robot, the roll and pitch angle are controlled to be 0.
Additionally there is a stabilization control, which holds the robot at a defined depth, and not move in any direction.
In general the idea is to have this position module to control the position of the uuv. The position module receives the desired position of the uuv and sends appropriate attitude setpoints to the uuv_attitude_control module.
Additionally the mixer file is adapted, to include the 6 different inputs(x y z roll pitch yaw).
Currently not solved/missing:
- Problem with gazebo model(propeller moving chaotically).
- Mixer correct gazebo vs real life (has to be tested in the future)
- correct integration in uuv.apps (when choose which module)
- very basic controller chosen (could be improved a lot in the future)
* Remove error caused by unused variables and a different build error
* added better description of the parameter. Additionally the group is changed.
* added better description of the parameter. Additionally the group is changed.
Fixed bug about parameter
* Added EOF to the files.
* Removed parameter for direct position control for safety reasons.
* small bugfix
- keep `vehicle_control_mode` last state in commander and use appropriate flags in place of various main_state and nav_state checks
- consolidate scattered arming requirements in `arm_disarm()`
- there were a number of different requirements from arming via RC vs Mavlink that don't make any sense
- if geofence enabled require valid home before arming
- throttle requirements for manual modes
- remove unnecessary mavlink feedback that differs between arming interfaces (mavlink vs RC)
- let the preflight/prearm checks respond directly in most cases
Co-authored-by: Matthias Grob <maetugr@gmail.com>
It is more intuitive to use the roll stick as lateral control input given that drones defined this category, plus this is how consumer car / rover radio controls have been working already.
This commit moves the steering output from yaw to the roll channel to better reflect the lateral control input / reaction mapping. It also removes old unused parameters and modernizes the mainloop to remove unnecessary polling.
90% of all real-world vehicle configs default to this and it is something that users stumble over if they configure a new system. There are valid cases where this would not be desired - for these it can be still switched off.
- store parameter type and if volatile separately (saves kilobytes of flash)
- use Bitset for tracking active and changed parameters
- use atomic for autosave_enabled flag
- compile at ${MAX_CUSTOM_OPT_LEVEL} (-O2 on non flash constrained boards)
- ecl in PX4/Firmware (2292cf3ed4f8c4391131870936881331bc870260): 602e65146c
- ecl current upstream: ffab483504
- Changes: 602e65146c...ffab483504
ffab483 2021-02-11 Loic Dubois - Waiting for the filter to be initialized to set global origin
In case of a diverge of the nav filter, both position and velocity test
ratios are large. If only one of them is too large, the nav filter is
most likely not diverging but is encountering a GNSS position reset and
it is still safe to fly in position mode.
It seems like we are often missing updates from topics like flight mode
or in air state, both topics that are sent out infrequenctly e.g. at 1
Hz. Therefore, instead of polling for that data we should probably
subscribe to the updates and that way get notified of each an every
update.
For instance this should prevent the case where we miss the mode change
from mission to descend and back to mission once landed and disarmed.
The mapping itself was seprated out into a calls because it was reused
for the experimental nudging implementation.
The position resets which were handled correctly before now
change the wrong setpoints and I adjusted.
The nudging has to be before any filtering, then these member setpoints
which are essentially copies are not needed anymore.
- ecl in PX4/Firmware (afb692ecffccc3ba7ade39c5906c8c5660066121): cd38621dd6
- ecl current upstream: 602e65146c
- Changes: cd38621dd6...602e65146c
602e651 2021-02-05 Paul Riseborough - EKF: Remove unnecessary initialisation
6f664ab 2020-07-20 Paul Riseborough - EKF: Non functional changes arising from review
b87778e 2020-06-30 Paul Riseborough - EKF: Make clip detection logic easier to follow
9aa8d93 2020-06-30 Paul Riseborough - EKF: Use isRecent function
e948b35 2020-06-29 Paul Riseborough - EKF: Don't reject vertical aiding data if inertial solution is bad
c566318 2020-06-23 Paul Riseborough - EKF: misc improvements to handling of accel clipping
9c89fa3 2020-06-02 Paul Riseborough - EKF: Use IMU clipping to adjudicate bad accel data check
The velocity setpoint of the position controller
does a jump when unlocking position with a non-zero position error.
This is solved by using the velocity setpoint feedback to smoothly
take over.
Improves compression, e.g. current params file: 62KB to 51KB
There's also a PRESET_EXTREME option, which reduces by another 2KB.
We can revisit that once needed, as it increases mem usage as well.
- implement COMP_METADATA_TYPE_VERSION & add component_version.json.gz to
ROMFS for all targets.
- implement COMP_METADATA_TYPE_PARAMETER (only enabled for SITL currently)
With a higher responsiveness, after centering the stick, the velocity and
acceleration setpoints could oscillate around 0 and never reach 0, due to
discretization.
This also prevented position lock engagement.
- libcanard in PX4/Firmware (0c6cf6cb99914925cd09f0e991031b78654e6e3e): cde6703474
- libcanard current upstream: 55938c51cf
- Changes: cde6703474...55938c51cf
55938c5 2020-12-29 Pavel Kirienko - Merge pull request #160 from UAVCAN/readme-fix
1c91c49 2020-12-29 Pavel Kirienko - Improve the example in README, fix#159
Fixed UAVCAN146 linker script that caused CXX initalization to wrong
Enabled uORB listener command
Updated UAVCANv1 msg definition
Known isssue: PX4 params are not working on the UAVCAN146
- same pattern as other new drivers (state machine, checked register mechanism, no sleeps, etc)
- self test has been rolled into regular startup sequence
- still I2C only, but will be expanded in the future
- configure to high accuracy preset
- ecl in PX4/Firmware (1e079d0142b813c3c65c6796576e79d5662a217b): b919f3ddf4
- ecl current upstream: 46251db4a1
- Changes: b919f3ddf4...46251db4a1
46251db 2021-02-05 Daniel Agar - geo: explicitly link libm
In some cases e.g.: (VTOL in wind) a good tracking cannot be
achieved. This condition then needs to be relaxed, otherwise the
drone cannot land properly.
This allows to still fly missions completely without RC
but reacts if RC is lost during the mission because
the safety pilot expects to be able to take over.
Instead of directly doing the link loss reaction which by default is RTL a delay
can be configured such that the drone first switches to hold and waits
for the link to be regained.
As hardware buttons are not particularly reliable and the user flow is to disable safety then arm, then disarm via software / remote, it makes sense to make the button safety state itself sticky and require it to be reset via software.
As hardware buttons are not particularly reliable and the user flow is to disable safety then arm, then disarm via software / remote, it makes sense to make the button safety state itself sticky and require it to be reset via software.
For some usecases like starting video recording on a companion computer or triggering a ROS based (offboard control) mission control execution the status of RC channel is needed in ROS(2). This allows the user to start/stop such functionalities from the RC transmitter.
Therefore InputRc to be changed accordingly.
add 'Set send: true' for id: 30
- vehicle_attitude_setpoint thrust_body is a vector, but mavlink ATTITUDE_TARGET thrust is only a magnitude
- this allows the stream to be correct for both MC & FW use cases
This will ensure that the CRCs will match with deployed units to avoid unnecessary incompatibilities while we are figuring out how to best standardize these between stakeholders.
- currently the main change is that it reduces the max number of ORB multi instances to 4, but usage will be expanded as needed
- limits number of EKF2 multi instances to 2.
- enabled on all cortex m4 boards
The overhead of the MAVROS setup means that most developers are not using it, leading to tests that are not suitable for day-to-day workflows. We are replacing these with MAVSDK tests that can be run locally pre-commit.
- sitl_gazebo in PX4/Firmware (bb7dd0cf0016ac1998a79814ec52b15079a74c1a): 46aef29718
- sitl_gazebo current upstream: e580bbcd1e
- Changes: 46aef29718...e580bbcd1e
e580bbc 2021-01-30 JaeyoungLim - Use include model syntax for bluerov (#702)
663a553 2021-01-30 Tim - Added GPS and changed the Thruster config of the bluerov2 SDF file (#700)
4674826 2021-01-29 Thies Lennart Alff - increased ODE solver iterations to avoid deflection of the thrusters (#701)
This system command will display, set and save the network
settings.
netman show - Displays the current settings.
netman update - Will check for a net.cfg file on the SD card.
If present, it will update the paramaters,
delete the file, and reboot. Using the new settings.
netman save - Saves the current settings to net.cfg on the SD card.
This file shoulf be renamed to preserver it across
reboots or editited to chech networkin paramates.
File format is name<space>value:
echo DEVICE=eth0 > /fs/microsd/net.cfg
echo BOOTPROTO=fallback >> /fs/microsd/net.cfg
echo IPADDR=192.168.0.4 >> /fs/microsd/net.cfg
echo NETMASK=255.255.255.0 >>/fs/microsd/net.cfg
echo ROUTER=192.168.0.254 >>/fs/microsd/net.cfg
echo DNS=192.168.0.254 >>/fs/microsd/net.cfg
Valid values for `proto` are `dhcp`, `static`, `falback`
Both will try dhcp for CONFIG_NETINIT_FALLBACK times
and fall back to the static address.
NETMASK - is the network mask.
IPADDR - this nodes ip address for static or fall back.
ROUTER - The default route.
DNS - The address of the dns server.
* stm32_common: Add support for STM32F42x/STM32F43x rev. 5/B
According to ST, both 5 and B share the same REV_ID.
Signed-off-by: Alexey Rogachevskiy <sfalexrog@gmail.com>
* stm32_common: Change comment to match code, change enum names to match revisions
- generate PWM_MAIN 1-14
- generate PWM_AUX 1-8
- generate PWM_EXTRA 1-8
- px4io and pwm_out directly read configuration parameters
- only available and active physical outputs are actually shown for configuration
- overall saves flash despite adding many new parameters
The number of arguments was increased by one, see: 1719ff9892
Because the above commit was merged before https://github.com/PX4/PX4-Autopilot/pull/15443 . It broke support for the remote host option.
This has been fixed in this commit by increasing all argv's by one.
Signed-off-by: Peter Blom <peterblom.mail@gmail.com>
Add new DeviceBusType_SERIAL to Device::DeviceId union
Add DRV_DIST_DEVTYPE's for all distance sensors
Change distance_sensor_s.id to distance_sensor_s.device_id
Modify all distance_sensor drivers to apply 'proper' device_id
* extend uuv_att_control
* add feedthrough thrust_y + thrust_z direction
* extend uuv_att_control
* add feedthrough thrust_y + thrust_z direction
* update formatting
* fix submarine sitl: indicate motor channel range 0..1 or -1..1 in simulation_mavlink.cpp (#16637)
change motor_count variable to 'pos_thrust_motor_count'
This is more specific to what is actually happening in the code.
'pos_thrust_motors_count' indicates number of motor channels which are configured with 0..1 range (positive thrust) all other motors are configured for -1..1 range
submarine only have motors with -1..1 range.
Thus, pos_thrust_motor_count = 0
Co-authored-by: Thies Lennart Alff <33184858+lennartalff@users.noreply.github.com>
* extend uuv_att_control
* add feedthrough thrust_y + thrust_z direction
* update formatting
Co-authored-by: Thies Lennart Alff <33184858+lennartalff@users.noreply.github.com>
When an environment variable "SIGNING_TOOL" is defined, the make will
call this with two filename arguments: <input> and <output>.
The SIGNING_TOOL will read in the binary from input, and append
signature to the output. For example:
SIGNING_TOOL=${PWD}/Tools/cryptotools.py make px4_fmu-v5_default
Will use the example "cryptotools.py" to sign the px4_fmu-v5.bin
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
- Append signature in the end of the signee
- Add parameters for output file and rd certificate
- Add a default test key
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
* drivers/uavcannode: Add a simple parameter server
Added a simple parameter server using the libuavcan ParamServer class.
The parameter server exposes a selection of PX4 parameters via UAVCAN.
Also, Increased the stack size of the uavcan work queue in order to
prevent a stack overflow.
* uavcannode: fetch all active PX4 parameters
Co-authored-by: Kenneth Thompson <ken@flyvoly.com>
change motor_count variable to 'pos_thrust_motor_count'
This is more specific to what is actually happening in the code.
'pos_thrust_motors_count' indicates number of motor channels which are configured with 0..1 range (positive thrust) all other motors are configured for -1..1 range
submarine only have motors with -1..1 range.
Thus, pos_thrust_motor_count = 0
Co-authored-by: Thies Lennart Alff <33184858+lennartalff@users.noreply.github.com>
- matrix in PX4/Firmware (92f2043d8ba1320ff185bdb13274332b649a0ffa): 054f8b12f4
- matrix current upstream: 977cf52322
- Changes: 054f8b12f4...977cf52322
977cf52 2021-01-22 Julian Kent - Ignore debug line in code coverage
4837316 2021-01-22 Julian Kent - Fix small inconsistencies and compiler warnings
d540ca5 2021-01-13 Julian Kent - Coerce default epsilon values to Type
That enables the sensor checks by default
The default for the innovation check integral of 5 is rather conservative, thus
makeing it less likely to be triggered by accidnet (e.g. due to bad wind estimate).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
Previously, the logic was not consistent with the description in the params.
It now, if the integral is enabled, declares innovation check failed if
integral is above threshold.
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- ecl in PX4/Firmware (18bac6aff57c9c32a5effcc086954264680f7978): 99575d73e1
- ecl current upstream: b919f3ddf4
- Changes: 99575d73e1...b919f3ddf4
b919f3d 2021-01-21 Paul Riseborough - EKF: Restore protection from extreme innovation values
efb78de 2021-01-21 Paul Riseborough - EKF: Set position fusion gate option false by default
556a195 2021-01-19 Paul Riseborough - EKF: Improve handling of non position mode large position innovations
In some cases, the fitness of the optimizer does not shrink at each
iteration, causing the "decreasing fitness check" to fail. This stops
the optimization and returns with sub-optimal, or unrealistic results.
This fixes RSSI for the Control Zero F7 but I have noticed that while this works perfectly for SBUS receivers, for PPM receivers it does not decrease the RSSI visual value in QGC when removing the RC transmitter connection.
When a PPM receiver is connected and the connection is lost the autopilot goes into RC Scan Mode (in the RC Update Module) to determine what is connected (even though something already is connected).
The main issue with this is that PPM receivers don't go into RC Failsafe but I don't think it is an issue with this autopilot. It looks to be an issue with the RC Update Module and how it is handled at the module level for non I/O coprocessor autopilots.
Tested with an X8R (SBUS) and a Dragonlink (PPM) as well as a Dragonlink set to SBUS as the output. SBUS worked as intended. See screenshots below.
in workspace settings. This comes down to user preference
and there's no particular reason to change this based on the workspace.
I found it confusing to have this non-default behavior
just for PX4 editing.
- control allocation module with multirotor, VTOL standard, and tiltrotor support
- angular_velocity_controller
- See https://github.com/PX4/PX4-Autopilot/pull/13351 for details
Co-authored-by: Silvan Fuhrer <silvan@auterion.com>
Co-authored-by: Roman Bapst <bapstroman@gmail.com>
Two sources:
1. global to local conversion was sometimes giving issues, so do everything in global
2. on startup the RTL didn't check if the home position was valid before processing it
Moving the command handling to a separate function that gets called
whenever a vehicle command is available to always react on commands
and not just when already a task is running.
This solves e.g. commanding an Orbit when in Staibilized.
It is important that setpoints get initialized with NAN and not overwritten
if specifically set by a successful flight task execution. It's then
clear if any setpoints were intentionally and successfully set.
Crucial for the position controller's emergency failsafe and the
seamless setpoint handover to the next flight task.
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.
PX4 is highly portable, OS-independent and supports Linux, NuttX and QuRT 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!
## PX4 Developers
## Changing code and contributing
This [Developer Guide](https://dev.px4.io/) 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/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.
Developers should read the [Guide for Contributions](https://dev.px4.io/master/en/contribute/).
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!
@ -69,10 +73,8 @@ The PX4 Dev Team syncs up on a [weekly dev call](https://dev.px4.io/master/en/co
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.
