Commit Graph

29189 Commits

Author SHA1 Message Date
misswhile 37fe090225 fix bmp280 driver init faild. 2019-04-25 10:19:12 +02:00
Julian Oes c95394f57f navigator: fix uninitalized param values
By declaring the params above the navigator modes inheriting from
mission block we make sure the params in navigator are set before
read in the constructor of mission block.
2019-04-24 11:15:43 -04:00
Julian Oes 90ee26dd2d mavlink_mission: properly initialize padding
This prevents uninitialized padding from being written to dataman.
2019-04-24 11:15:43 -04:00
Julian Oes 4c56994d7a mavlink_system: set update_counter for safe points
This variable was not set and lead to uninitialized memory being written
to dataman.
2019-04-24 11:15:43 -04:00
Julian Oes bf32ff32f8 dataman: make _file_write more readable
This should not be any functional change. The only difference are the
variable scopes, and early return versus nested ifs.
2019-04-24 11:15:43 -04:00
Julian Kent dc3341db59 Fix uninitialized memory found using Valgrind 2019-04-24 11:15:43 -04:00
Julian Oes a9c3bce20d px_uploader.py: fix version check
Presumably older versions of pyserial do not implement __version__ but
only VERSION, so we need to check for that as well.
2019-04-24 14:47:10 +02:00
Roman 42d0522cdd FixedWingAttitudeControl: removed check for negative airspeed and slew rate
on airspeed scaling

Signed-off-by: Roman <bapstroman@gmail.com>
2019-04-24 14:27:37 +02:00
Daniel Agar 36aeb9defc WIP: FW attitude control limit airspeed scaling changes per iteration 2019-04-24 14:27:37 +02:00
Martina Rivizzigno eba0bb389a ObstacleAvoidance: fix comment, update failsafe position if one axis is NAN 2019-04-24 08:54:50 +02:00
Martina Rivizzigno e037edd2cc ObstacleAvoidance: once the commadand loiter has been sent, keep using
the first position when the oa fails as setpoint to avoid jumps
2019-04-24 08:54:50 +02:00
Paul Riseborough 0f346af226 ekf2: Improve parameter description
Inhibiting bias estimation using the EKF2_AID_MASK parameter only applies to the delta velocity bias states.
2019-04-24 06:55:54 +02:00
Daniel Agar d8a1a79f69 Mavlink: SYS_STATUS trivial style check fix 2019-04-23 08:48:11 -04:00
jdzerve 2e9d0f0751 Mavlink: Get full sensor status even if only subset has changed when composing SYS_STATUS message (#11886) 2019-04-23 08:31:13 -04:00
Arkadiusz Niemiec c50c44cc58 Add a boolean to cleanly exit sender thread v1.9.0-beta3 2019-04-21 10:40:25 +01:00
Julian Oes c78aaeb0f7 ak09916: fail if device is not found
This should fix the case where the driver initializes even though the
device is not found. The change changes the behavior to return ERROR
if the whoami call fails several times instead of returning OK.

Also, the reset() and thus probe() calls are moved before initializing
the ringbuffer and device name.
2019-04-20 12:52:56 +02:00
liubaoxu db37cd8c71 BMP280:fix definition misspelling
from BPM280 to BMP280
2019-04-20 08:52:06 +02:00
David Sidrane 2b68076663 Update Nuttx with F7 i2C fixes 2019-04-18 08:55:50 +02:00
baumanta c0a1f5b72f Collision prevention improvements (#11866)
* match max vel in col-prev to regular pos ctrl
* change warning criterium to avoid float error
* disable push back from obstacles
* use floats consistently
2019-04-17 13:20:29 -04:00
Roman 463b8a7528 vtol_att_control: respect min pwm values
Signed-off-by: Roman <bapstroman@gmail.com>
2019-04-17 09:13:27 +02:00
Dennis Millard 70681f990f BMI055 IMU driver fix temperature reporting 2019-04-16 22:53:16 -04:00
sfuhrer 3d36dbfba3 added babyshark VTOL config file and mixer 2019-04-16 11:05:35 -04:00
Mark Sauder 74bc6870ed simulator: uorb and initialization cleanup (#11825)
* Cherry pick the vehicle_imu PR #9756 src/modules/simulator directory work to submit as standalone PR.

* Modify the accelsim init method mag_report usage to match usage in measure() and mag_measure() methods in the class.

* Incorporate review comments in the accelsim.cpp init() method and also make the same modifications in gyrosim.cpp.

* Delete unneeded mag_report initialization from accelsim init() method.

* Deprecate unneeded measure() call from accelsim.cpp init() and gyrosim.cpp init().
2019-04-16 08:29:53 +02:00
Gabriel Moreno f9324fb76a mavlink: set correct distance_sensor timestamp. Fixes #11840
When PX4FLOW is connected to PX4 through MAVLink (e.g., through a USB
port), the timestamp assigned to the distance_sensor was wrong. This fix
uses the same timestamp assigned to the optical_flow message created from
the same OPTICAL_FLOW_RAD MAVLink message.

Signed-off-by: Gabriel Moreno <gabrielm@cs.cmu.edu>
2019-04-15 20:31:15 +02:00
bresch 93d52581ef Auto - Fix cruise speed. Tested with and without SDK. Before that commit, the drone files at max speed instead of cruise speed if not precised by the SDK 2019-04-15 16:32:26 +02:00
Beat Küng 226f3c0999 rc.board_sensors: probe for external qmc5883 2019-04-15 10:37:06 +02:00
Daniel Agar d36b06f779 fw_pos_control_l1: FW_LND_EARLYCFG disable by default
- change to boolean param
2019-04-13 17:35:12 +02:00
Daniel Agar 6d30b13afa update Tools/setup/ dev environment setup scripts
- Ubuntu install NuttX toolchain, gazebo, and switch to python3
 - OSX switch to python3
2019-04-12 10:43:21 -04:00
Julian Oes e88f1b33b2 FlightTasks: fix mission DO_CHANGE_SPEED
This fixes the issue where the  DO_CHANGE_SPEED command was ignored and
the drone always travelled at the MPC_XY_CRUISE velocity.
2019-04-12 15:47:36 +02:00
Matthias Grob 1454694bdd FlightTaskAuto: separate default speed and limit
It wasn't possible to fly faster than cruise speed even if planned
in the mission.

Limiting the planned cruise speed is necessary because
the smoothed trajectory mission plans to the _mc_cruise_speed and
if that's higher than the maximum it gets capped for safety by the
position controller and the result is a jerky flight.
2019-04-12 15:47:36 +02:00
Beat Küng 3135f9f0d2 sih: avoid static variable + style fixes 2019-04-12 09:25:07 +02:00
romain 65f623bd73 acceleration added when hitting the ground 2019-04-12 09:25:07 +02:00
romain 7be1cc389e removed arguments from constructor 2019-04-12 09:25:07 +02:00
Beat Küng 744b50b478 send MAVLink GROUND_TRUTH at 25 Hz, only in SIH mode. And minor cleanup 2019-04-12 09:25:07 +02:00
romain 9adb4410bd newline added at the end of files 2019-04-12 09:25:07 +02:00
romain 8b6480c1cf sih.msg removed, serial port communication removed 2019-04-12 09:25:07 +02:00
romain e8c5d85525 sih param names cleanup 2019-04-12 09:25:07 +02:00
Beat Küng f542736140 'simulation' category added in srcparser.py 2019-04-12 09:25:07 +02:00
romain e1d9438f79 sih: remove SYS_SIH parameter and extend SYS_HITL 2019-04-12 09:25:07 +02:00
romain a9660d9da0 white noise generator updated 2019-04-12 09:25:07 +02:00
romain cdd7e95a92 added a call to px4_sem_destroy() 2019-04-12 09:25:07 +02:00
romain 6ff228c2ff sih module implemented with hrt_call_every and semaphore_wait 2019-04-12 09:25:07 +02:00
Beat Küng 914a9b78b6 new airframe for sih, HIL_STATE_QUATERION sent through MAVLink 2019-04-12 09:25:07 +02:00
romain c09e9ec97f simulator in hardware, new module added that allows to run a simulator in the hardware autopilot, for more documentation visit https://github.com/romain-chiap/PX4_SIH_QuadX 2019-04-12 09:25:07 +02:00
Daniel Agar 4fe9ac9993 vscode add uavcan.dsdl to recommended extensions
- this extension adds support for Data Structure Description Language (DSDL)
2019-04-11 21:04:59 -04:00
Andreas Daniel Antener 95a0803b14 Adding AK09916 driver to support the Here2 GNSS emulated mag (I2C) (#11837)
* drivers, magnetometer: added ak09916 driver to support the mag driver emulated by the Here2 gps
* boards: set yaw 270 rotation for external ak09916 driver and only start it for the PH21
* drivers: only start ak09916 when icm20948 is not available on a ph21 setup
2019-04-11 12:43:50 -04:00
Daniel Agar 5a50f96bcf move icm20948 (Here GPS compass) to Cube sensors start (#11838) 2019-04-11 11:15:53 -04:00
Julian Oes cd9b3d6b27 CI: move no-ninja builds to Jenkinsfile 2019-04-11 10:07:00 +02:00
Julian Oes 471d05381b CI: also build PX4 SITL without ninja 2019-04-11 10:07:00 +02:00
Julian Oes f74aab33ad CI: also build two targets without ninja
This should prevent future regressions when compiling with "Unix
Makefiles" or others instead of ninja as the generator behind cmake.
2019-04-11 10:07:00 +02:00