dark0707
/
px4-firmware
forked from Archive/PX4-Autopilot
1
0
Fork 0
Code Pull Requests Activity
1,184 Commits 514 Branches 136 Tags 234 MiB
Commit Graph

1184 Commits

Author SHA1 Message Date
Paul Riseborough 310bd97080 EKF: Improve protection against bad pos vel fusion 2016-05-07 14:22:59 +10:00
Paul Riseborough da9d894441 EKF: Improvements to covariance reset 2016-05-07 12:57:41 +10:00
Lorenz Meier 199c423f1f Attitude fw: Remove builtin perf counters 2016-05-05 20:06:01 +02:00
Paul Riseborough d9aeace6ac EKF: tuning update 2016-05-05 21:23:03 +10:00
Paul Riseborough f1b7e7714e EKF: Make average update rate of EKF closer to target 
With the EKF, the average update rate is more important than the instantaneous value as it affects tuning. This patch ensures that the EKF prediction cycle will be performed early if the previous one was late in an attempt to maintain the target update rate.
 2016-05-05 21:23:03 +10:00
Paul Riseborough 1daa6c7866 EKF: update tuning defaults 2016-05-05 21:23:03 +10:00
Paul Riseborough fc6c0c7ebf EKF: Update covariance prediction 
Refresh auto generated code
Make 3-axis accel fusion the default and no accel fusion the option
 2016-05-05 21:23:03 +10:00
Paul Riseborough ecaffe63af EKF: update tuning defaults 2016-05-05 21:23:03 +10:00
Paul Riseborough 7a74d58591 EKF: reduce initial accel bias uncertainty 2016-05-05 21:23:03 +10:00
Paul Riseborough a7d08d08fc EKF: Relax tilt alignment check 2016-05-05 21:23:03 +10:00
Paul Riseborough 5523a4f225 EKF: Fix IMU bias compensation scale error in output filter 
Delta angle and velocities are calculated assuming data is at the filter update rate, not the IMU update rate
 2016-05-05 21:23:03 +10:00
Paul Riseborough fd109b00ab EKF: replace in-line code with existing function 2016-05-05 21:23:03 +10:00
Paul Riseborough 727a43764f EKF: update initial angle alignment check 2016-05-05 21:23:03 +10:00
Paul Riseborough bc34b14779 EKF: Initialise height state variance to a value compatible with the measurement 
A large height state variance and small measurement variance can destabilise the filter in the first few seconds after alignment
 2016-05-05 21:23:03 +10:00
Paul Riseborough 29a361d3a3 EKF: Use intuitive units for gyro and accel bias process noise 
This also makes the tuning less sensitive to changes in the EKF update rate.
 2016-05-05 21:23:03 +10:00
Paul Riseborough 799865c934 EKF: update default tuning parameters 2016-05-05 21:23:03 +10:00
Paul Riseborough f01f4ae4b1 EKF: tidy up mag declination fusion 2016-05-05 21:23:03 +10:00
Paul Riseborough 0d31aad33a EKF: Fix bad limit on magnetometer noise parameter 2016-05-05 21:23:03 +10:00
Paul Riseborough 7069fb7a04 EKF: refresh auto code for mag fusion 2016-05-05 21:23:03 +10:00
Paul Riseborough ffea65e1a4 EKF: fix bug in state prediction 2016-05-05 21:23:03 +10:00
Paul Riseborough 63f093a10a EKF: Update upper limits for state variances 
This limiting is a  last resort measure and the limiting for normal operation should be achieved by modification in the covariance prediction
 2016-05-05 21:23:03 +10:00
Paul Riseborough 26a51c667d EKF: Improve robustness to initial gyro bias errors 
Allows for ~6 deg/sec of startup bias
 2016-05-05 21:23:03 +10:00
Paul Riseborough 7530e30012 EKF: Improve IMU data collection 
Use IMU driver published delta angle integration time to determine when enough data has been accumulated.
Perform basic first order sculling corrections on delta velocity data
Comment steps to make make method clearer
 2016-05-05 21:23:03 +10:00
Paul Riseborough f23f0b1035 EKF: fix bugs in stationary process model covariance prediction 
Noise variance for stationary states was  being overwritten by the covariance prediction operations for the kinematic states
Unwanted states could end up with non-zero covariance values
The forced symmetry operation was being applied too late to be effective
 2016-05-05 21:23:03 +10:00
Paul Riseborough a373ada858 EKF: explicitly define floating point type in hardcoded constants 2016-05-05 21:23:03 +10:00
Paul Riseborough a189c60d33 EKF: update code fragment text files 2016-05-05 21:23:03 +10:00
Paul Riseborough f4e84a2234 EKF: fix axis label error in comments 2016-05-05 21:23:03 +10:00
Paul Riseborough 57d3036ee7 EKF: update derivation files 2016-05-05 21:23:03 +10:00
Paul Riseborough 82da832816 EKF: tidy up Kalman gain calculations 
Inhibiting of states is controlled via zeroing rows and columns in the covariance prediction so conditional logic in gain calculations is unnecessary.
 2016-05-05 21:23:03 +10:00
Paul Riseborough 311d046a26 EKF: Enable optional activation of XY delta velocity bias estimation 2016-05-05 21:23:03 +10:00
Paul Riseborough 572c2280bd EKF: tidy up covariance prediction 2016-05-05 21:23:03 +10:00
Paul Riseborough 54e713969d EKF: Fix rebase error 2016-05-05 21:23:03 +10:00
Paul Riseborough 67d23948b5 EKF: preliminary re-tune 2016-05-05 21:23:03 +10:00
Paul Riseborough 03c35266ef EKF: remove unused parameter 2016-05-05 21:23:03 +10:00
Paul Riseborough fe9f88a8b4 EKF: test new derivation 
Use direct attitude parameterisation
Discard scale factors
Add accel bias
 2016-05-05 21:23:03 +10:00
Daniel Agar 22d18d638c enable Wshadow 2016-05-04 19:45:40 -04:00
Paul Riseborough c6775b5bf2 Merge pull request #118 from PX4/fix_member_init 
ekf: initialize _terrain_var
 2016-05-03 10:26:48 +10:00
Beat Küng f2dda3c183 ekf: initialize _terrain_var, fixes access to uninitialized value 
in PX4 Firmware, Ekf::get_terrain_vert_pos depends on it. This is the
valgrind output:
==15439== Thread 14 ekf2:
==15439== Conditional jump or move depends on uninitialised value(s)
==15439==    at 0x5610087: sqrtf (in /usr/lib64/libm-2.22.so)
==15439==    by 0x4D1AF0: Ekf::get_terrain_vert_pos(float*) (terrain_estimator.cpp:135)
==15439==    by 0x46641F: Ekf2::task_main() (ekf2_main.cpp:655)
==15439==    by 0x42BF3D: entry_adapter(void*) (px4_posix_tasks.cpp:103)
==15439==    by 0x4E3C609: start_thread (in /usr/lib64/libpthread-2.22.so)
==15439==    by 0x5BF7A4C: clone (in /usr/lib64/libc-2.22.so)
 2016-05-02 17:35:42 +02:00
Paul Riseborough 0a6d95c6c2 Merge pull request #116 from PX4/pr-covReset 
after a height reset zero the cross covariances for height and vertical
 2016-05-01 21:44:38 +10:00
Roman 0c77ebf6bd EKF: Reduce likelihood of bad covariance values after height resets 
The cross terms (covariances) are now reset before the variance variance is set.

The vertical velocity state variance following a reset without GPS has been reduced.
 2016-05-01 21:39:56 +10:00
Paul Riseborough 6c57bea1c3 EKF: fix bug in IMU error application timing patch 2016-04-30 15:11:12 +10:00
Paul Riseborough 6ecbefd7e4 EKF: add matlab script to test complementary filter gains 2016-04-30 11:05:03 +10:00
Paul Riseborough 15df20831a EKF: improve output complementary filter tracking 
Optimise for height tracking and adjust gains automatically with changes in time delay.
 2016-04-30 11:03:45 +10:00
Paul Riseborough 4fa1e9c651 EKF: Fix timing errors in state prediction 
IMU corrections were being applied at the wrong time horizon to the EKF states
IMU downsampling rotates the delta velocities into the orientation produced by the last delta angle, but this wasn't consistent with the way the state prediction was using the delta velocity data.
 2016-04-30 11:02:19 +10:00
Paul Riseborough 4c9c102597 Merge pull request #115 from PX4/fix_isfinite_again 
ekf: another stab at fixing isfinite for all
 2016-04-29 09:12:59 +10:00
Julian Oes f8a48f9cc5 ekf: now it should even compile for NuttX 2016-04-28 16:30:49 +02:00
Julian Oes 12f4d6f703 ekf: another stab at fixing isfinite for all 
The previous solution did not work for Snapdragon, so I needed to copy
what is used in px4_defines.h.
 2016-04-28 16:09:12 +02:00
Paul Riseborough 22ad87599c Merge pull request #112 from PX4/fix_ubuntu1604_2 
ekf: compile fix for Ubuntu 16.04
 2016-04-27 21:21:04 +10:00
Julian Oes 46e207106b ekf: use `std::isfinite` instead of `isnan` 2016-04-27 12:58:22 +02:00
Lorenz Meier 7b702ee37e Merge pull request #105 from georgehines/matlab-log-import 
fix a field name that I missed last time
 2016-04-26 10:03:37 +02:00