38020 Commits

Author SHA1 Message Date
Beat Küng
dc87f8a7a4 events: unescape message to match the behavior of the C++ compiler 2021-09-11 14:08:41 -04:00
Beat Küng
e1b8611323 px4/fmu-v5/optimized: disable some modules to reduce flash 2021-09-11 14:08:41 -04:00
Beat Küng
9bba13f3e2 px4/fmu-v2/multicopter: disable some modules to reduce flash 2021-09-11 14:08:41 -04:00
Beat Küng
1f73294ad7 navigator: switch to events 2021-09-11 14:08:41 -04:00
Beat Küng
5ac43e7236 commander: switch to events 2021-09-11 14:08:41 -04:00
Daniel Agar
8866215d1d sensors/vehicle_imu: add protections for bad sensor data
- check for inconsistent timestamps, zero samples, and numerical issues in online mean
2021-09-11 13:04:46 -04:00
Matthias Grob
3c57c67ce3 Make arch setup script executable
I usually called it using `source arch.sh`
but if it's executable it could also be called using `./arch.sh`
and that's also what's done for the ubuntu setup script.
2021-09-10 07:50:23 +02:00
m4v3r
a238965228
boards: modalai_fc-v1 update default.cmake (#18202)
- makes TELEM/SERIAL-4 (VOXL-Flight J1009) work
2021-09-09 09:03:21 -04:00
Jukka Laitinen
28dbd8dc1d NuttX at 10.1.0+
NuttX backport fixing ARM protected build
2021-09-09 04:43:22 -07:00
Thomas Schneider
2e36fff31d Implement option to reboot px4 using protocol_splitter format when flashing. 2021-09-09 10:12:11 +02:00
Jukka Laitinen
ffa5abbd65 ci: update container versions to 2021-09-08
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
2021-09-09 09:43:15 +02:00
Beat Küng
c5c80e31b6 boards: remove DIRECT_INPUT_TIMER_CHANNELS from board_config.h 2021-09-08 16:10:24 -04:00
Beat Küng
ae8733c2d4 pwm_input: remove 'test' command and use 'status' instead 2021-09-08 16:10:24 -04:00
Beat Küng
f1aafe83b3 pwm_out: cleanup, remove _output_mask & only update owned channels
_pwm_mask reflects the actually owned channels.
2021-09-08 16:10:24 -04:00
Beat Küng
a8e75d174c boards: remove BOARD_HAS_PWM and use DIRECT_PWM_OUTPUT_CHANNELS 2021-09-08 16:10:24 -04:00
Beat Küng
1ee423e3c8 pwm_out, dshot: remove mode_* handling and capture ioctl's
They now use all unused pins, whereas camera_capture, camera_trigger and
pwm_input modules start before to reserve their pins.
2021-09-08 16:10:24 -04:00
Beat Küng
062dd28f4d camera_capture: use up_input_capture_set directly
It reserves the channel and pwm_out will not use it
2021-09-08 16:10:24 -04:00
Beat Küng
78b5cdae4c s32k1xx: io_timer: reserve pins & timers on first use 2021-09-08 16:10:24 -04:00
Beat Küng
ab828b8e94 kinetis: io_timer: reserve pins & timers on first use 2021-09-08 16:10:24 -04:00
Beat Küng
7d9a3bf29f imxrt: io_timer: reserve pins & timers on first use 2021-09-08 16:10:24 -04:00
Beat Küng
9a74c6f3c6 stm32: io_timer: lock around io_timer_allocate_channel and io_timer_get_channel_mode
This is to avoid potential race conditions during startup.
All startup code runs sequentially atm, so this is just for robustness
for later (e.g. concurrent pwm_out and dshot start).
2021-09-08 16:10:24 -04:00
Beat Küng
847bd120fa io_timer: reserve pins & timers on first use
This allows modules to do a first-come-first-serve pin/timer reservation
on bootup.
E.g. camera trigger reserves any of the pins, and then PWM/DShot output
will just use the rest of the available pins.
2021-09-08 16:10:24 -04:00
Beat Küng
0d7176b766 boards: remove camera_capture from sitl and linux targets
It uses io_timer lib
2021-09-08 16:10:24 -04:00
Daniel Agar
422a0f7103 boards: px4_fmu-v5_debug enable more debug errors and warnings 2021-09-08 16:04:45 -04:00
bresch
aad11ecc65 PreFltCheck: do not force to report ekf2 failures on GCS connection
EKF2 has a grace period of 10 seconds after boot where it doesn't need
to warn the user while the sensors (especially GNSS) are still
converging.
A connection to a GCS shouldn't skip this grace period but
an arming request should.
2021-09-08 14:41:26 -04:00
Matthias Grob
58ea97a699 MulticopterPositionControl: add horizontal margin for saturation cases 2021-09-08 19:55:27 +02:00
benjinne
06a91ec752
Improve offboard failsafe (#18160)
* state_machine_helper: improve offboard failsafe

* state_machine_helper: add missing parameter to set_link_loss_nav_state

* state_machine_helper: fix no rc and offboard reason

* Fix offboard test by enabling rcl_except

* mavros_test fix offboard_posctl_test with rcl_except

* autopilot_tester make RcLossException bits explicit

Co-authored-by: Julian Oes <julian@oes.ch>

* autopilot_tester change rcl_except to rc_loss_exception

Co-authored-by: Julian Oes <julian@oes.ch>

* autopilot_tester fix rc_loss_exception renaming errors

Co-authored-by: Julian Oes <julian@oes.ch>
2021-09-08 18:19:53 +02:00
ShiauweiZhao
686bcff8a7
new TDK ICM42670p IMU driver support (#18141)
Co-authored-by: zhaoxiaowei <zhaoxiaowei@efy-tech.com>
2021-09-08 10:24:22 -04:00
Daniel Agar
000765e9f0 NuttX mount procfs and binfs in px4 init 2021-09-08 10:21:24 -04:00
Daniel Agar
29a91306b5 lib/led: error if dropping ORB_ID(led_control) messages 2021-09-08 10:21:09 -04:00
Daniel Agar
5c7882b787 icm42688p: properly disable anti-aliasing and notch filter 2021-09-08 10:20:57 -04:00
Daniel Agar
4b5a91bf7d
experimental driver for the Tattu 12S 16000mAh Smart Battery on /dev/can0 (#18164)
-  This driver is highly specific to this model of Tattu battery and cannot work with other models without code changes. 
 - The driver simply reads CAN frames using the NuttX CAN character device interface and converts the data into the proper format as specified in the Tattu datasheet

Co-authored-by: Jacob Dahl <dahl.jakejacob@gmail.com>
2021-09-08 10:20:14 -04:00
Daniel Agar
5eddf32cd3
boards: cubepilot_cubeorange only start ADSB mavlink if console not present
- cubepilot_cubeorange_test has the console enabled (used for test rack and bench debugging)
2021-09-06 11:52:58 -04:00
Sander Swart
dbb0974b0c Removed TEL3 from cube orange default.cmake as it is used for the built in ADS-B receiver 2021-09-06 09:51:38 +02:00
Sander Swart
bce1c06174 Added new line at the end of the file as per code style 2021-09-06 09:51:38 +02:00
Sander Swart
223410e17a Enable Cube Orange built in ADS-B receiver by default 2021-09-06 09:51:38 +02:00
Daniel Agar
8c47a190ac Jenkins: hardware stop sensors module before mirobenchmarks
- ignore irrelevant sensor timeouts during microbenchmarks
2021-09-05 23:18:28 -04:00
Daniel Agar
cf0a4e06db boards: px4_fmu-v5x_test disable unused drivers to save flash 2021-09-05 23:18:28 -04:00
Daniel Agar
560ff42923 boards: px4_fmu-v2_test disable ms4525 to save flash 2021-09-05 23:18:28 -04:00
Daniel Agar
56a8f5f73e boards: px4_fmu-v2_test disable mc_hover_thrust_estimator to save flash 2021-09-05 23:18:28 -04:00
Daniel Agar
3dda360fac boards: px4_fmu-v5_optmized disable all additional barometers to save flash 2021-09-05 23:18:28 -04:00
Daniel Agar
476e050000 boards: cubepilot_cubeorange_test disable unused modules to save flash 2021-09-05 23:18:28 -04:00
Beat Küng
291ca246d2 matrix: update submodule (improves matrix inversion) 2021-09-05 23:18:28 -04:00
Beat Küng
73ab153fe0 fix test_microbench_math: don't try to measure single instructions
The clock is simply not accurate enough to do that.
Plus the measuring overhead is much higher than the executed instruction.

Remaining issue: memory transfers (due to volatile) add non-negligible
overhead and distort the result. Could be solved by using inline assembly.
2021-09-05 23:18:28 -04:00
Daniel Agar
02ca1a6711
Update submodule mavlink v2.0 to latest Sun Sep 5 12:39:18 UTC 2021 2021-09-05 20:42:54 -04:00
PX4 BuildBot
6278c072c9 Update submodule sitl_gazebo to latest Mon Sep 6 00:40:01 UTC 2021
- sitl_gazebo in PX4/Firmware (51697b0d07665d85ca1ebd9105e3cdc621b41969): 7fda4d311a
    - sitl_gazebo current upstream: 53d606b32f
    - Changes: 7fda4d311a...53d606b32f

    53d606b 2021-09-01 stmoon - fix the bug for the low real time factor in gazebo
2021-09-05 20:42:33 -04:00
Daniel Agar
a4e9444ca4
Makefile: submodulesupdate include fetching all tags recursively 2021-09-05 14:50:56 -04:00
RomanBapst
93075ad5e8 rtl: fixed computation of return altitude if destination is not home
Signed-off-by: RomanBapst <bapstroman@gmail.com>
2021-09-05 10:26:24 +02:00
Jaeyoung-Lim
d32f0d6168 Support headless mode for ignition gazebo
This adds support for `HEADLESS=1` when starting ignition gazebo with headless mode
2021-09-04 14:01:39 +02:00
Jaeyoung-Lim
3aa3f542ae Update simulation-ignition submodule 2021-09-03 15:54:46 +02:00