fly316/PX4-Autopilot
7
0
Fork 0
mirror of https://gitee.com/mirrors_PX4/PX4-Autopilot.git synced 2026-10-20 11:23:34 +08:00
Code Issues Packages Projects Releases Wiki Activity
17,946 Commits 852 Branches 161 Tags
c5a6442ce6426bb7351e94de416016400fdd9a06
Commit Graph
96 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Andrew Tridgell 82aca3d65c FMUv4: enable RSSI analog input pin 2016-05-01 14:22:58 +02:00
Andrew Tridgell 6552f925f8 px4fmu: allow control of safety state on FMUv4 
allows for MAVLink control of safety state
 2016-04-25 17:25:47 +02:00
Andrew Tridgell 9daf6c336b px4iofirmware: removed unused PX4IO_P_CONFIG_CONTROL_GROUP_COUNT 2016-04-18 18:44:05 +02:00
Andrew Tridgell 3bad8131cf px4io: don't call io_set_rc_config when rc config disabled 2016-04-18 18:44:05 +02:00
Andrew Tridgell cc0fbdd549 px4fmu: added SET_MODE ioctl 
this gives control over timer capture driver
 2016-04-18 11:45:44 +02:00
Andrew Tridgell e60f2c3f41 drv_pwm: added new PWM_SERVO_SET_MODE ioctl 2016-04-18 11:45:44 +02:00
Andrew Tridgell c66fab01e5 drv_pwm: expand range of PWM ioctl indexes 
we had run out of them. _IOC_MASK is 0xFF, so we can add an extra 16
without an issue
 2016-04-18 11:45:13 +02:00
Andrew Tridgell fde635ef5a px4iofirmware: fixed display of sbus rate in px4io status 2016-04-18 11:43:24 +02:00
Andrew Tridgell c5f5b9f53c drv_pwm_output: added IOCTL for SBUS rate 2016-04-18 11:43:24 +02:00
Andrew Tridgell 30ce7fdd48 px4io: display SBUS rate in px4io status 2016-04-18 11:43:24 +02:00
Andrew Tridgell 6290e621d4 px4io: added control for setting SBUS rate 2016-04-18 11:43:24 +02:00
Andrew Tridgell f19761d278 px4iofirmware: support setting the SBUS output frame rate 
this allows for helicopters and multicopters with SBUS output
 2016-04-18 11:43:24 +02:00
Andrew Tridgell 58122ed33d sbus: added support for changing SBUS output rate 2016-04-18 11:43:24 +02:00
Andrew Tridgell 05790ff540 px4fmu: prevent checking of safety switch parameter in main loop 2016-04-18 11:41:01 +02:00
Andrew Tridgell 3fda934d1b px4io: on config error ask IO to reboot to bootloader 
this allows us to update fw when core config settings have changed,
such as the actuators/servos change that happened recently
 2016-04-18 11:40:07 +02:00
Andrew Tridgell 85e8cd89b6 px4io: set reasonable default voltage scaling 
needed for when px4 param system not available in ardupilot
 2016-04-18 11:39:37 +02:00
Andrew Tridgell 5928569321 px4fmu: fixed signed/unsigned comparison 2016-04-18 11:39:06 +02:00
Andrew Tridgell 9984555faa motor_test: fixed build 
this is C, not C++
 2016-04-18 11:37:43 +02:00
Andrew Tridgell 05c282e761 stm32: fixed build errors with more warnings enabled 
needed for ArduPilot build
 2016-04-18 11:36:28 +02:00
Andrew Tridgell 091515956d dsm: fixed shadowed variables errors 2016-04-18 11:35:21 +02:00
Andrew Tridgell fc6b074d40 irlock: fixed header for other than FMUv2 2016-04-18 11:34:24 +02:00
Andrew Tridgell 943e6ffbc9 px4iofirmware: blink blue LED more rapidly when override is active 
this makes it easier for a user to test that override is working
correctly in pre-flight checks. Otherwise override is hard to
distinguish from normal manual mode
 2016-04-17 19:10:23 +02:00
Andrew Tridgell 43235a8cff perf: fixed mixup of stdout/stdin 2016-04-17 16:07:44 +02:00
Andrew Tridgell efd4552e37 ms5611: reduced OSR to 1024 
this reduces self-heating of the sensor which reduces the amount of
altitude change when warming up. Apparently some individual sensors
are severely affected by this.

Unfortunately it raises the noise level, but Paul is confident it
won't be a significant issue
 2016-04-17 16:04:44 +02:00
Andrew Tridgell b8a60f2501 pwm: change default_value for all boards from 1000 to 0 
the value '1000' is not really magic, and it was resulting in sending
a single 1ms pulse on all IO channels for a brief instant on
startup. With some servos this led to the control surfaces being at
extremes (and straining) on startup. It may also contribute to making
ESC calibration harder on multi-rotors

In the worst case it could cause a IC motor with reverse throttle to
go full throttle on startup

A logic analyser shows the problem very clearly. Changing the
default_value fields to 0 from 1000 fixes the issue and no pulses are
generated until an explicit value suitable for the airframe is
provided via one of the many methods PWM output values can be
generated

Conflicts:
	src/drivers/boards/px4fmu-v4/px4fmu_pwm_servo.c
	src/lib/ecl
 2016-01-29 08:57:20 +01:00
Andrew Tridgell d290487382 circuit_breaker: prevent param fetch failure from disabling safety switch 
if the param get failed then an uninitialised stack variable was used
for the safety disable on boot. In ArduPilot that value happened to
equal the correct magic due to stack passing from caller. This forced
safety off on boot
 2015-12-04 21:22:25 +01:00
Andrew Tridgell 46631b5c8a mpu6000: work around apparent ICM20608 bug 
undocumented register 0x11 sometimes starts with value 0, which gives
an offset on the Y accel axis of 2.7m/s/s. It sometimes boots with
0xc9, which gives a zero offset. Force it to 0xc9 to get consistently
correct behaviour
 2015-11-28 09:13:15 +01:00
Andrew Tridgell 4eaa18e6f9 perf_counter: fixed write to correct fd for perf output 
this fixes perf output for nsh over MAVLink
 2015-08-28 21:57:09 +02:00
Andrew Tridgell 41b3452e22 ll40ls: added support for 'blue label' lidars 
the key is splitting up read_reg() into two transfers
 2015-08-17 09:18:38 +02:00
Andrew Tridgell 210bb9e996 px4fmu-v2: added mpu9250 driver to the build 2015-08-17 09:17:14 +02:00
Andrew Tridgell f77620c3f4 mpu6000: check WHOAMI register 
this prevents a mpu9250 being seen as a mpu6000
 2015-08-17 09:17:07 +02:00
Andrew Tridgell 0e6cdecfad drv_sensor: added MPU9250 sensor defines 2015-08-17 09:16:58 +02:00
Andrew Tridgell 39e0a487bb mpu9250: first cut at a MPU9250 driver 2015-08-17 09:16:47 +02:00
Andrew Tridgell 5e27b7bc31 ms5611: fixed the i2c device on NuttX. This was left-over debugging noise from the POSIX bringup. 2015-06-27 15:11:03 +02:00
Andrew Tridgell db8dd000e3 l3gd20: follow same pattern as lsm303d for duplicate rejection 2015-05-07 15:16:34 +10:00
Andrew Tridgell a710159263 mpu6000: sample at 200usec faster rate to avoid aliasing 
this runs the mpu6000 200usec faster than requested then detects and
disccards duplicates by comparing accel values. This avoids a nasty
aliasing issue due to clock drift between the stm32 and mpu6000
 2015-05-07 15:16:32 +10:00
Andrew Tridgell 3ac95fb581 lsm303d: run sampling 200usec faster to avoid aliasing 
this runs the sampling of the accelerometer 200usec faster than
requested and then throw away duplicates using the accelerometer
status register data ready bit. This avoids aliasing due to drift in
the stm32 clock compared to the lsm303d clock
 2015-05-07 15:16:29 +10:00
Andrew Tridgell 219d661888 hmc5883: read the temperature every 10 samples when enabled 2015-04-08 17:45:54 +02:00
Andrew Tridgell 02639411ba hmc5883: added -T option to enable temperature compensation 
this also fixes the behaviour of the -C option
 2015-04-08 17:45:54 +02:00
Andrew Tridgell e17936e237 hmc5883: try to cope with genuine 5883 parts 
if we can't read the temperature registers 10 times then disable the
feature.
 2015-04-08 17:45:54 +02:00
Andrew Tridgell 156a7915ae hmc5883: added support for temperature compensation 
added "hmc5883 tempon" and "hmc5883 tempoff" to enable/disable
 2015-04-08 17:45:53 +02:00
Andrew Tridgell dea97dd277 drv_mag: added ioctl to control temperature compensation 2015-04-08 17:45:53 +02:00
Andrew Tridgell a4bece7595 uavcan-baro: use RingBuffer for read() support 2015-04-04 10:20:51 +02:00
Andrew Tridgell 9442fc13e4 mpu6000: show temperature in "mpu6000 info" 2015-03-17 13:44:53 +11:00
Andrew Tridgell 374c50cab8 lsm303d: support temperature readings in lsm303d 
these are read at the same rate as the mag
 2015-03-17 13:44:51 +11:00
Andrew Tridgell 9d8931328b Tools: added boot_now.py script 2015-03-14 21:12:24 +11:00
Andrew Tridgell 2c2359dcf0 px_uploader.py: added --boot-delay option 
this sets the bootloader delay
 2015-03-14 21:12:21 +11:00
Andrew Tridgell 9efeb4cf0b FMUv2: added bootloader delay signature to text 
this allows for a configurable bootloader delay
 2015-03-14 21:12:19 +11:00
Andrew Tridgell 332d42b105 px_uploader: print chip version 2015-03-07 09:46:27 +01:00
Andrew Tridgell d9b4f5b170 px_loader: added --force option 
this can be used to override the board type check. Useful when
changing bootloaders
 2015-03-07 09:45:51 +01:00
Andrew Tridgell 8240e90d26 bl_update: fixed stat() check 2015-02-26 08:22:23 +01:00
Andrew Tridgell 9a875c53af px4fmu: added "mode_pwm4" startup mode 
this is the default mode ArduPilot uses, and by setting it in the init
script we avoid any pin activity on the two GPIO pins during boot
 2015-02-23 19:52:57 +01:00
Andrew Tridgell 41cc04c064 pwm_input: added PWM input driver 
this allows for input of PWM signals on the FMUv2 aux5 pins
 2015-02-23 19:52:47 +01:00
Andrew Tridgell 322392d853 fmuv2: setup for PWM input on timer4, channel 2 2015-02-23 19:52:38 +01:00
Andrew Tridgell 9d288a9a9a ll40ls: added reset and backoff logic to driver 
this adds automatic resets of the lidar when it becomes unresponsive,
and also tries to cope with NACKs from long acquisition times. It also
adds a "ll40ls regdump" command, and fixes "ll40ls reset" to reset the
sensor.

The default acquisition period is changed to 50ms, with backoff to
100ms on a NACK

Note that there are still times when the sensor can get into an
unrecoverable state.

Thanks to Austin, Dennis and Bob from PulseLight for assistance in
tracking down the problems.
 2015-02-20 09:10:24 +01:00
Andrew Tridgell 91aab00fa0 build: avoid wiping an existing PYTHONPATH variable 2015-02-16 22:54:32 +01:00
Andrew Tridgell 618d51dab1 hmc5883: fixed build errors with gcc 2.7.4 and -O3 2015-02-16 06:54:05 +01:00
Andrew Tridgell 4c8ac04877 PX4: better error msg in python uorb generation error 2015-02-16 06:52:21 +01:00
Andrew Tridgell 36870cf44b Tools: improve python package error message 2015-02-16 06:52:00 +01:00
Andrew Tridgell 1b8a830a38 i2c: prevent double free of _dev pointer 
this caused heap corruption
 2015-02-14 11:16:02 +01:00
Andrew Tridgell 0801dbda38 l3gd20: checking status only makes sense if we have DRDY 
it makes no sense on the external SPI bus
 2015-02-11 12:48:59 +01:00
Andrew Tridgell e0ac0c4a4b l3gd20: fixed "l3gd20 test" 
don't reset after test, and leave us in correct polling mode
 2015-02-11 12:48:52 +01:00
Andrew Tridgell fc40a5c641 systemcmds: added usb_connected command 
this is used in APM startup to see if we should start nsh on usb when
no microSD is inserted
 2015-02-11 17:36:07 +11:00
Andrew Tridgell 52bff67076 ms5611: removed debug code 2015-02-01 22:13:09 +01:00
Andrew Tridgell 2f25469bbe px4fmu-v1: removed baro I2C address in board_config.h 
the ms5611 can be on two addresses. The driver handles this, not the
board config
 2015-02-01 13:09:55 +01:00
Andrew Tridgell 04a9050410 ms5611: allow for all 4 bus combinations in ms5611 driver 
this uses the same table driven approach as the hmc5883 driver, and
allows for a ms5611 baro on any of the 4 bus combinations. A simple
"ms5611 start" will start all baros that are found.
 2015-02-01 13:09:44 +01:00
Andrew Tridgell 9b10395e94 hmc5883: fixed DEVIOCGDEVICEID ioctl 
we need to pass the ioctl through to the bus interface

thanks to Jon Challinger for noticing this bug
 2015-01-20 15:08:22 +11:00
Andrew Tridgell 0af6303d0b hmc5883: fixed mixup of internal and external hmc5883 I2C bus options 2015-01-20 14:30:46 +11:00
Andrew Tridgell 54a22aed94 hmc5883: fixed handling of 3 bus options 
use a table of possible bus options. This prevents us starting two
drivers on the same bus
 2015-01-09 19:42:27 +11:00
Andrew Tridgell 500ac1443b lsm303d: detect large fixed values and report an error 
we have logs where the lsm303d gets large fixed values for long
periods. This will detect that error case and raise error_count so the
higher level sensor integration code can choose another sensor
 2015-01-02 16:47:48 +11:00
Andrew Tridgell 7b6620b319 ll40ls: fixed exit code on external sensor startup failure 2015-01-01 13:06:46 +01:00
Andrew Tridgell 8470105f9e lsm303d: fixed build warning 2015-01-01 10:04:09 +11:00
Andrew Tridgell 7864958f1c l3gd20: fixed build warning 2015-01-01 10:04:07 +11:00
Andrew Tridgell 82b9e08e4c mpu6000: removed unsafe printf in interrupt context 
instead delay 3ms between register writes. This seems to give a quite
high probability of correctly resetting the sensor, and does still
reliably detect the sensor going bad, which is the most important
thing in this code
 2015-01-01 08:51:57 +11:00
Andrew Tridgell b5f6fedfa8 lsm303d: show all perf counters in "info" 2015-01-01 08:51:54 +11:00
Andrew Tridgell 083d9e3e69 lsm303d: check DRDY after check_registers() 
this allows recovery from a state where DRDY is not set
 2015-01-01 08:51:52 +11:00
Andrew Tridgell 1de7fab974 lsm303d: removed unsafe printf in interrupt context 2015-01-01 08:51:49 +11:00
Andrew Tridgell e4318345f3 lsm303d: added two more checked registers 
these are key for DRDY behaviour
 2015-01-01 08:51:47 +11:00
Andrew Tridgell 2ac3a6fcf4 l3m303d: added testerror command 
useful for testing error handling
 2015-01-01 08:51:44 +11:00
Andrew Tridgell 5c1da70d0c l3gd20: fixed reporting of error count 2015-01-01 08:51:42 +11:00
Andrew Tridgell 1bb3974c84 l3gd20: check DRDY after check_registers() is called 
this allows us to recover from an error that disables data ready
 2015-01-01 08:51:40 +11:00
Andrew Tridgell 8db2066151 l3gd20: removed printf in interrupt context 
this is not safe
 2015-01-01 08:51:37 +11:00
Andrew Tridgell 353ea9beac l3gd20: added testerror command 
useful for testing error handling
 2015-01-01 08:51:34 +11:00
Andrew Tridgell a8cea3a4da l3gd20: use the I2C disable bit on l3gd20H 
this seems to prevent a mpu6000 reset from causing an issue on the
l3gd20H
 2014-12-30 20:55:57 +01:00
Andrew Tridgell b455d647b2 l3gd20: added "l3gd20 regdump" command 2014-12-30 20:55:57 +01:00
Andrew Tridgell 4a81384b2d mpu6000: make register fixup much closer to a reset() 
this may help automatic reset on the faulty boards
 2014-12-30 20:55:57 +01:00
Andrew Tridgell e537de20e3 mpu6000: wait for 10ms after a full reset 
this prevents the mpu6000 getting in a really weird state!
 2014-12-30 20:55:57 +01:00
Andrew Tridgell 333039d3db mpu6000: added "mpu6000 testerror" command 
used to generate a error case for reset testing
 2014-12-30 20:55:57 +01:00
Andrew Tridgell b157466608 mpu6000: monitor some more registers 2014-12-30 20:55:57 +01:00
Andrew Tridgell e5d3c80686 mpu6000: added factory self-test function 
this follows the factory calibration self-test method in the datasheet
to see if the sensor still has the same calibration it had in the factory
 2014-12-30 20:55:57 +01:00
Andrew Tridgell ca47952281 l3gd20: added register checking 
this checks at runtime that key registers have correct values
 2014-12-30 20:55:57 +01:00
Andrew Tridgell ae3a92569d mpu6000: try resetting the mpu6000 up to 5 times 
this mirrors the ardupilot driver. We have seen situations where the
mpu6000 on the Pixhawk comes up in SLEEP mode, despite a reset
 2014-12-30 20:55:57 +01:00
Andrew Tridgell 03c40d5d91 lsm303d: replace old register checking with new check_registers() method 
this uses the same method as is now used in the MPU6000 to check that
the sensor retains its correct values

Conflicts:
	src/drivers/lsm303d/lsm303d.cpp
 2014-12-30 20:55:57 +01:00
Andrew Tridgell 3e06a65516 mpu6000: monitor some key registers for correct values 
this will catch both bad SPI bus comms and a sensor that has been
reset causing incorrect configuration.
 2014-12-30 20:55:57 +01:00
Andrew Tridgell 2bcf4385f6 build: use unqualified com port names on windows 
the qualified names were not working with current versions of python
 2013-08-21 15:00:19 +02:00
Andrew Tridgell 43eca3eb0e build: use unqualified com port names on windows 
the qualified names were not working with current versions of python
 2013-07-10 15:36:58 +10:00