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

689 Commits

Author SHA1 Message Date
Daniel Agar 983c44088f fw att fix pitch feedforward constrained roll 2016-11-27 12:34:12 +01:00
Paul Riseborough a776b2c549 EKF: Fix bug preventing reset to GPS position when using optical flow 2016-11-26 10:28:58 +01:00
Paul Riseborough 460c9e5250 EKF: Don't try to fuse air data if not in normal aiding mode 
Taking off before passing GPS checks would cause airspeed or sideslip to be fused when the filter was still using a constant position assumption. This would cause large airspeed innovations, invalid wind estimates and degrade filter performance after GPS was gained and position and velocity was reset.
 2016-11-26 10:28:58 +01:00
Paul Riseborough 5ad329b641 EKF: Fix variable names to match convention 2016-11-07 22:23:12 +11:00
Paul Riseborough afd6e54b2a EKF: Allow for worst case timing jitter when setting observation buffer length 2016-11-07 11:50:16 +11:00
Paul Riseborough 403424d55d EKF: fix cross compiler build errors 2016-11-07 09:38:44 +11:00
Paul Riseborough d5b398ddb1 EKF: Minimise data buffer lengths 
Saves RAM and reduces output predictor errors by using the smallest data buffer length that meets time delay and update rate requirements.
 2016-11-07 09:06:45 +11:00
Paul Riseborough 750a8874c1 EKF: Improve recovery and reporting for badly conditioned sideslip fusion 
Do not reset the entire covariance matrix if we are only using sideslip fusion to update wind states.
Improve reporting.
 2016-11-05 12:19:01 +01:00
Paul Riseborough 6c7703d485 EKF: Update tuning for synthetic sideslip fusion 
Used value that worked well for replay of flight logs
 2016-11-05 12:19:01 +01:00
Paul Riseborough f76a2a5ff7 EKF: fix bug in sideslip fusion and reduce RAM usage 
The bug meant the  Y velocity (state index 5)  covariance was not being updated correctly when sideslip was being used to constrain velocity drift (extended GPS loss).
The rework of the covariance update to reduce RAM follows the same pattern as adopted for other fusion processes
 2016-11-05 12:19:01 +01:00
Paul Riseborough 951c8b77af EKF: Improve wind state reset logic 
Enables wind estimation without an airspeed sensor and enables synthetic sideslip to be used with an airspeed sensor for improved wind state estimation.
Wind states and covariances are reset differently depending on whether airspeed is available.
 2016-11-05 12:19:01 +01:00
Paul Riseborough 2fedbc8a18 EKF: Refresh sideslip fusion auto-code and improve covariance update efficiency 2016-11-05 12:19:01 +01:00
Paul Riseborough 352b4ca177 EKF: Allow sideslip fusion with airspeed fusion 2016-11-05 12:19:01 +01:00
Paul Riseborough fb6f4be453 EKF: don't allow sideslip fusion to affect non-airspeed states if not deae-reckoning 2016-11-05 12:19:01 +01:00
CarlOlsson a9182259a5 fixed bug in beta fusion 2016-11-05 12:19:01 +01:00
CarlOlsson 0e3ae4b323 fixed issue with the covariance update in synthetic sideslip fusion 2016-11-05 12:19:01 +01:00
CarlOlsson e1d9d503e3 updated logic of when to fuse beta. Now synthetic sideslip measurements are fused after 5 seconds of cruise flight with airspeed sensor activated 2016-11-05 12:19:01 +01:00
CarlOlsson dec686afd7 updated parameters 2016-11-05 12:19:01 +01:00
CarlOlsson 5205a8198b added function to control when to fuse beta 2016-11-05 12:19:01 +01:00
CarlOlsson f5d0228cd5 include ecl.h 2016-11-05 12:19:01 +01:00
CarlOlsson beb90f008d adopted code for upstream changes 2016-11-05 12:19:01 +01:00
CarlOlsson b7a4a1358f added sidelip_fusion.cpp 2016-11-05 12:19:01 +01:00
CarlOlsson 12aeaafbd3 adopted estimator_interface.h 2016-11-05 12:19:01 +01:00
CarlOlsson c8e3c24150 adopted ekf_helper.cpp 2016-11-05 12:19:01 +01:00
CarlOlsson 8f6c75bb0b adopted ekf.h 2016-11-05 12:19:01 +01:00
CarlOlsson 15768f6f93 adopted ekf.cpp 2016-11-05 12:19:01 +01:00
CarlOlsson 738414d36e adopted common.h 2016-11-05 12:19:01 +01:00
CarlOlsson 6c9cc5691c adopted EKF/CMakeLists 2016-11-05 12:19:01 +01:00
CarlOlsson daf44dcd67 adopted CMakeLists 2016-11-05 12:19:01 +01:00
Paul Riseborough 283c328c23 Merge pull request #212 from PX4/pr-ekfCriticalBugFix 
EKF: Fix bug causing repeated resets if GPS sensor disconnected
 2016-11-05 09:43:24 +11:00
Paul Riseborough 8aee45dd96 EKF: Fix bug causing repeated resets if GPS sensor disconnected 2016-11-05 09:36:06 +11:00
Paul Riseborough 2bc50489fd Merge pull request #210 from PX4/remove_KH 
EKF: reduce KH to only the elements that are really needed, and merge the loops
 2016-11-05 09:24:13 +11:00
Beat Küng 07c6aabd98 EKF: reduce KH to only the elements that are really needed, and merge the loops 2016-11-02 10:26:40 +01:00
Paul Riseborough 1b59a89a18 EKF: save some stack memory 2016-11-02 09:06:06 +01:00
Beat Küng f472324ae8 DataValidatorGroup: remove the assert 
(avoid crashes in flight)
 2016-11-01 22:49:20 +01:00
Beat Küng 25cb400da9 EKF: remove Ekf::KHP and move KHP to the methods where it's used 
Everywhere where KHP is used, it is first completely reset, thus making
it unnecessary to keep it as a class member.

This saves 2.3KB RAM.

Stack sizes don't need changing, since there is already a function
Ekf::predictCovariance(), which needs around 3KB of stack and is called
close to where the fuse* functions are called.
 2016-11-01 22:49:20 +01:00
Beat Küng 227938d263 DataValidatorGroup: add add_new_validator() to dynamically add new validators 2016-11-01 22:49:20 +01:00
Beat Küng ded3b1a690 DataValidatorGroup: fix memory leak, delete all DataValidator's in destructor 2016-11-01 22:49:20 +01:00
Beat Küng b870f92e1c DataValidator: use 32bit timeout instead of 64bit, saves memory 2016-11-01 22:49:20 +01:00
Paul Riseborough d07692e87e Merge pull request #204 from PX4/pr-gpsReset 
ekf reset: more granular reset / timeout strategy for gps fusion
 2016-10-26 06:52:10 +11:00
Paul Riseborough 79df7e8889 EKF: Reduce effect of bad airspeed when other aiding data available 2016-10-25 08:34:41 +02:00
Paul Riseborough 968cfae632 EKF: Fix bug preventing fallback from GPS mode if GPS is lost 2016-10-25 08:32:01 +11:00
Roman 63cf3d50be ekf gps reset: do not bother checking for need of reset if gps is not 
available

Signed-off-by: Roman <bapstroman@gmail.com>
 2016-10-24 11:45:36 +02:00
Roman c5f750dcca ekf reset: more granular reset / timeout strategy for gps fusion 
- if both gps position and velocity measurements are rejected for 7 seconds
do a reset
- if only gps position measurements are rejected then wait for 14 seconds
as we still have velocity measurements to constrain the drift in position
- introduced ecl internal parameter for the timeout

Signed-off-by: Roman <bapstroman@gmail.com>
 2016-10-24 11:40:12 +02:00
Roman Bapst 65de4962b5 Merge pull request #203 from PX4/pr-ekfVibeReporting 
EKF: Add IMU vibration calculation and reporting
 2016-10-19 13:32:53 +02:00
Paul Riseborough b10b0184c5 EKF: Add IMU vibration calculation and reporting 
Add calculation and reporting of IMU delta angle and velocity coning and high frequency vibration
 2016-10-19 21:46:55 +11:00
Roman 17c195d1cd ekf2 height reset: do not apply height change to height state twice (bug) 
Signed-off-by: Roman <bapstroman@gmail.com>
 2016-10-18 07:58:19 +02:00
Paul Riseborough 1093e6d789 EKF: fix documentation error 2016-10-13 22:37:14 +02:00
Roman f80b3481a4 reset the newest output sample for all possible resets 
Signed-off-by: Roman <bapstroman@gmail.com>
 2016-10-12 08:54:38 +02:00
Roman 238d947239 proper handling of position and velocity reset 
- position reset method was returning before the actual delta values
were written and applied to the output buffer
- apply reset delta also to the output sample which was already taken out
from the output buffer, otherwise the complementary filter solution is
offset from the ekf solution

Signed-off-by: Roman <bapstroman@gmail.com>
 2016-10-12 08:54:38 +02:00