| 3bf4a6d | Mohammed Kabir | 04 September 2018, 01:09:31 UTC | EKF : control : remove state predictor alignment which is not used anywhere else | 13 April 2019, 21:29:08 UTC |
| 886fc83 | Mohammed Kabir | 03 September 2018, 20:42:05 UTC | EKF : control : handle flow data timeouts correctly | 13 April 2019, 21:29:08 UTC |
| b2f9dd8 | Mohammed Kabir | 03 September 2018, 20:38:35 UTC | EKF : control : remove bogus GPS timestamp setting | 13 April 2019, 21:29:08 UTC |
| 4176372 | Mohammed Kabir | 03 September 2018, 20:35:20 UTC | EKF : control : don't check for body tilt angle in flow data update There are additional checks downstream to catch this | 13 April 2019, 21:29:08 UTC |
| 7ecedc1 | Mohammed Kabir | 03 September 2018, 20:34:33 UTC | EKF : estimator_interface : protect against large dts | 13 April 2019, 21:29:08 UTC |
| c4492b1 | Martina | 09 April 2019, 10:27:22 UTC | do not reset yaw if vehicle on ground | 09 April 2019, 10:50:37 UTC |
| 14227ea | Julian Oes | 08 April 2019, 09:17:50 UTC | CMakeLists.txt: fix cmake warning This removes the cmake warning: implicitly converting 'string' to 'STRING' type. | 08 April 2019, 13:54:47 UTC |
| f97f0b6 | Paul Riseborough | 30 March 2019, 02:46:16 UTC | EKF: Remove unnecessary code _flt_mag_align_start_time is set on initial alignment and will always be >0 before this check can run | 07 April 2019, 20:57:17 UTC |
| 43b5e26 | Paul Riseborough | 01 March 2019, 22:03:56 UTC | EKF: Fix bug causing continual yaw reset when EKF2_MAG_TYPE = 2 | 07 April 2019, 20:57:17 UTC |
| f95cd4b | Roman | 22 January 2019, 20:44:57 UTC | ground effect: removed dependency on local position Signed-off-by: Roman <bapstroman@gmail.com> | 19 March 2019, 02:14:57 UTC |
| 7845ff4 | CarlOlsson | 23 January 2019, 13:28:17 UTC | EKF: increase wind process noise scaler to 0.5 Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 19 March 2019, 02:10:12 UTC |
| 32ca6f7 | CarlOlsson | 24 October 2018, 14:36:52 UTC | ekf: scale wind process noise with low pass filtered height rate | 19 March 2019, 02:10:12 UTC |
| 938c8ad | CarlOlsson | 04 February 2019, 08:57:23 UTC | EKF: use hagl estimate if valid for when to trigger yaw reset on takeoff | 19 March 2019, 02:04:54 UTC |
| 8b4ae48 | Hamish Willee | 04 March 2019, 21:18:28 UTC | README: Fix link to EKF/ECL tuning guide | 18 March 2019, 14:21:10 UTC |
| f0889c1 | Carl Olsson | 18 March 2019, 14:20:33 UTC | EKF: fixed some comment typos Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 18 March 2019, 14:20:33 UTC |
| 6e77b19 | Todd Stellanova | 15 March 2019, 04:09:47 UTC | Add DataValidatorGroup tests, add more DataValidator tests (#592) | 15 March 2019, 04:09:47 UTC |
| a892ece | Daniel Agar | 13 March 2019, 03:00:11 UTC | cmake add ECL_TESTS option only enabled in standalone build | 13 March 2019, 03:29:30 UTC |
| fabd216 | Paul Riseborough | 12 March 2019, 21:08:42 UTC | EKF: Documentation edits | 12 March 2019, 21:14:23 UTC |
| 8d6e8ae | CarlOlsson | 12 March 2019, 14:38:30 UTC | EKF: update get_mag_decl_deg() Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 21:14:23 UTC |
| 2b17ced | CarlOlsson | 12 March 2019, 14:08:45 UTC | EKF: Clean up use of magnetometer declination. Before the innovation was not zero in fuseDeclination() Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 21:14:23 UTC |
| 1d91785 | CarlOlsson | 11 March 2019, 14:40:34 UTC | EKF: correct quaternion order | 12 March 2019, 08:07:55 UTC |
| 000aa0d | CarlOlsson | 11 March 2019, 14:17:05 UTC | EKF: do not update the newest attitude estimate in initialiseFilter(). Is is now done in alignOutputFilter() | 12 March 2019, 08:07:55 UTC |
| 96c5c14 | CarlOlsson | 08 March 2019, 13:59:28 UTC | EKF: Change the EKF initialization to 1) init tilt 2) init yaw 3) init tilt uncertainty 4) init yaw uncertainty Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 08:07:55 UTC |
| ea259d3 | CarlOlsson | 08 March 2019, 13:43:02 UTC | EKF: initialize the newest attitude estimate to a valid quaternion Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 08:07:55 UTC |
| 6cf512f | CarlOlsson | 08 March 2019, 13:40:11 UTC | EKF: Also update the newest and oldest quaternion estimate in alignOutputFilter() Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 08:07:55 UTC |
| 78122b2 | CarlOlsson | 08 March 2019, 13:34:17 UTC | EKF: fix the rotation direction of the buffered attitude solutions in alignOutputFilter() Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 08:07:55 UTC |
| 56b6b7a | CarlOlsson | 08 March 2019, 13:29:13 UTC | EKF: cleanup use of _mag_filt_state Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 08:07:55 UTC |
| 37339de | CarlOlsson | 08 March 2019, 13:27:27 UTC | EKF: remove unused variable Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 12 March 2019, 08:07:55 UTC |
| 6c759e0 | Todd Stellanova | 09 March 2019, 19:47:13 UTC | Initial unit tests for DataValidator (#588) | 09 March 2019, 19:47:13 UTC |
| 4e0e68e | Nuno Marques | 09 March 2019, 16:57:35 UTC | Ekf: add more useful methods to interface with the covariances (#543) | 09 March 2019, 16:57:35 UTC |
| a85d3a4 | Daniel Agar | 05 March 2019, 18:22:49 UTC | attitude_fw remove unused | 05 March 2019, 22:26:19 UTC |
| c66f7f4 | Daniel Agar | 05 March 2019, 17:55:26 UTC | EKF drop unused timestamp from collect_gps() and pass data by const reference | 05 March 2019, 22:26:19 UTC |
| 78b899c | Freek van Tienen | 05 March 2019, 17:41:21 UTC | Add Paparazzi compatibility (#580) This adds support for the paparazzi autopilot to use this library as EKF. The header order change has to be done in order to have std.h included to get certain defines at the start of the estimator_interface.h. | 05 March 2019, 17:41:21 UTC |
| 5cd3310 | Daniel Agar | 05 March 2019, 17:27:16 UTC | cmake standalone build pin PX4/Matrix to commit - https://github.com/PX4/Matrix/commit/6b0777d815cd64902eb0575d56ec52f53aebb4a0 | 05 March 2019, 17:35:37 UTC |
| 1378ec1 | Daniel Agar | 08 February 2019, 03:54:45 UTC | EKF mag_fusion avoid unnecessary double promotion | 08 February 2019, 13:24:24 UTC |
| 54ab819 | Carl Olsson | 06 February 2019, 13:49:16 UTC | EKF: add method to get the terrain variance | 06 February 2019, 13:49:16 UTC |
| c5554ca | Daniel Agar | 03 February 2019, 20:56:44 UTC | EKF tests update SWIG usage and use latest containers | 03 February 2019, 23:27:00 UTC |
| dd58e69 | Paul Riseborough | 30 January 2019, 11:26:00 UTC | EKF: Ensure FW yaw alignment method is used on first in-air reset | 31 January 2019, 14:53:57 UTC |
| 3accab1 | Paul Riseborough | 28 January 2019, 05:33:13 UTC | EKF: protect declination fusion from badly conditioned earth field estimates Fusion with large initial magnetometer biases errors can result in the the NE earth field states reducing in magnitude and effectively flipping sign. EKF: Move declination state limiting into a separate function EKF: Limit NE mag states after each 3-axis mag fusion EKF: Fix bug in mag field strength look-up scale factor | 31 January 2019, 14:53:57 UTC |
| 7bddbd1 | Paul Riseborough | 28 January 2019, 00:56:39 UTC | EKF: Update cleaned up autocode fragment with sign error fix and missing LD | 31 January 2019, 14:53:57 UTC |
| fe378fd | Paul Riseborough | 21 January 2019, 23:29:42 UTC | EKF: Prevent unwanted declination fusion when re-starting 3-axis fusion | 31 January 2019, 14:53:57 UTC |
| bd1647a | Paul Riseborough | 21 January 2019, 21:58:01 UTC | EKF: Rework use of fuseDeclination() Simplify calling so that it is only called in two ways: 1) Immediately before 3-axis mag fusion if not called since the last earth field covariance reset so that the earth field declination information can be formed. 2) Immediately after 3-axis mag fusion otherwise. | 31 January 2019, 14:53:57 UTC |
| d52f536 | Paul Riseborough | 17 January 2019, 06:36:11 UTC | EKF: Save mag field covariance data before reset | 31 January 2019, 14:53:57 UTC |
| 82ce7a8 | Paul Riseborough | 17 January 2019, 05:52:20 UTC | EKF: Save mag field covariance information on startup | 31 January 2019, 14:53:57 UTC |
| 708c79e | Paul Riseborough | 17 January 2019, 04:51:37 UTC | EKF: Ensure mag field state covariance data is always available for re-use Prevents use of _saved_mag_ef_covmat and _saved_mag_bf_variance before being initialised or when stale. | 31 January 2019, 14:53:57 UTC |
| 8839e4e | Paul Riseborough | 17 January 2019, 03:51:24 UTC | EKF: Don't discard declination certainty information when resuming 3-axis fusion. In the case where the EKF is switching between 3-axis and heading fusion, off-diagonal elements containing the correlation between N,E components of the earth field were being lost on each switch event. These elements contained information about the declination uncertainty and should be preserved. | 31 January 2019, 14:53:57 UTC |
| 25148e1 | Paul Riseborough | 17 January 2019, 02:51:54 UTC | EKF: Prevent rapid changes in declination estimate after a reset Ensures that each time the earth field covariance and variance data is reset, that the off-diagonal elements containing earth field declination angle certainty is restored. | 31 January 2019, 14:53:57 UTC |
| 6e7c119 | CarlOlsson | 16 January 2019, 11:07:52 UTC | EKF: limit yaw variance increase to 0.01 rad^2 to prevent a badly conditioned covariance matrix Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 31 January 2019, 14:53:57 UTC |
| 0896f7b | CarlOlsson | 16 January 2019, 07:52:54 UTC | EKF: Also fill in lower part of covariance matrix in increaseQuatYawErrVariance() Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 31 January 2019, 14:53:57 UTC |
| 4b3140e | Paul Riseborough | 16 January 2019, 06:48:48 UTC | EKF: Fix rebase error | 31 January 2019, 14:53:57 UTC |
| 911d4d8 | Paul Riseborough | 16 January 2019, 06:28:55 UTC | EKF: Fix sign error in increaseQuatYawErrVariance function | 31 January 2019, 14:53:57 UTC |
| a0b9cb0 | Paul Riseborough | 23 December 2018, 22:30:53 UTC | EKF: Use consistent method for recording completion of in-flight yaw alignment | 31 January 2019, 14:53:57 UTC |
| ef5a87c | Paul Riseborough | 21 December 2018, 03:22:54 UTC | EKF: Rework quaternion yaw reset. Use a new method that preserves the roll and pitch information and adds the uncertainty for yaw only. Ensure that correlation information to non-quaternion states is removed when a reset occurs to prevent fusion of subsequent observations (e.g. GPS) causing incorrect yaw. | 31 January 2019, 14:53:57 UTC |
| fc2a089 | Paul Riseborough | 21 December 2018, 03:18:25 UTC | EKF: Add function to un-correlate quaternion states This is necessary after a quaternion reset to prevent incorrect attitude corrections with subsequent observation fusions. | 31 January 2019, 14:53:57 UTC |
| bce1b96 | Paul Riseborough | 21 December 2018, 00:34:57 UTC | EKF: Add function enabling yaw variance to be increased | 31 January 2019, 14:53:57 UTC |
| bf1f3a2 | Paul Riseborough | 20 December 2018, 23:29:44 UTC | EKF: Derive equations enabling yaw variance to be increased | 31 January 2019, 14:53:57 UTC |
| 81eabc1 | Daniel Agar | 29 January 2019, 21:00:31 UTC | Jenkins update all containers to latest tag 2019-01-28 | 30 January 2019, 21:06:00 UTC |
| a5e6191 | Daniel Agar | 29 January 2019, 21:17:08 UTC | EKF add clarity brackets to avoid potential confusion - fixes https://github.com/PX4/ecl/issues/555 | 30 January 2019, 14:15:53 UTC |
| 721f5e6 | Anna Dai | 18 December 2018, 13:26:38 UTC | increase optical flow buffer to imu buffer length | 16 January 2019, 23:23:15 UTC |
| 44200e9 | Anna Dai | 18 December 2018, 13:24:39 UTC | add GPS drop out case to GPS fusion logic EKF waits 10s after GPS signal is lost before setting GPS control status flag to false. As the position information given by the alternative position sources drifts from the last GPS position, the controller over corrects. With this update, the time horizon until GPS control flag set to false is reduced and only alternative position source is used for estimation. tested with optical flow as position souce log from as is https://review.px4.io/plot_app?log=d624af5e-dde4-40ab-ba5b-a693a49f5a36 log with update https://review.px4.io/plot_app?log=13ed6dc3-22dd-43f8-b898-4add41d60439 | 16 January 2019, 23:23:15 UTC |
| 8a813c5 | CarlOlsson | 10 January 2019, 13:44:27 UTC | EKF: fix comment typo | 13 January 2019, 22:57:44 UTC |
| 472f228 | CarlOlsson | 10 January 2019, 12:32:28 UTC | EKF: initialize covariances before we reset the heading in order to preserve the yaw uncertainty Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 13 January 2019, 22:57:44 UTC |
| 4908882 | CarlOlsson | 08 January 2019, 09:53:33 UTC | EKF: Remove flag prefix Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 13 January 2019, 22:48:46 UTC |
| d223680 | CarlOlsson | 04 January 2019, 13:15:32 UTC | EKF: add flt_mag_align_complete to control_status flags Signed-off-by: CarlOlsson <carlolsson.co@gmail.com> | 13 January 2019, 22:48:46 UTC |
| b920910 | Daniel Agar | 06 January 2019, 16:37:44 UTC | replace <cfloat> with <float.h> - <cfloat> isn't available in the NuttX c++ library | 06 January 2019, 17:09:05 UTC |
| c032126 | Paul Riseborough | 04 January 2019, 09:30:51 UTC | EKF: Fix bug causing incorrect yaw variance value to be used This bug causes the variance of the Z axis rotation vector uncertainty to not be reset to the correct value. | 04 January 2019, 10:01:06 UTC |
| 48d83f3 | Paul Riseborough | 03 January 2019, 14:41:55 UTC | EKF: Fix vulnerability to ground level mag anomally when using heading fusion (#544) Previously, the reset of the yaw when climbing above 1.5m was not performed until 3-axis fusion was enabled. This could result in loss of navigation depending on the value of EKF2_MAG_TYPE and the flight profile. | 03 January 2019, 14:41:55 UTC |
| 4822bf8 | Hamish Willee | 07 December 2018, 02:53:40 UTC | Fix links to moved files in readme | 10 December 2018, 02:34:53 UTC |
| 1a969ae | CarlOlsson | 21 November 2018, 14:02:42 UTC | EKF: use low pass filtered mag measurements for heading initialization | 05 December 2018, 09:08:11 UTC |
| 27f7846 | Roman | 02 December 2018, 15:34:28 UTC | optical flow: fixed sign in calculation of optical flow sensor velocity - the velocity of the optical flow sensor relative to the IMU expressed in body frame is the cross product of the angular velocity with the vector from IMU to the sensor. If we use the angular velocity stored in the flow sample struct we need to use a negative sign as that angular velocity follows the opposite sign convention. Signed-off-by: Roman <bapstroman@gmail.com> | 03 December 2018, 14:26:40 UTC |
| c49ab76 | Paul Riseborough | 20 June 2018, 04:06:41 UTC | EKF: Allow EKF to start without using mag data when not using earth frame data | 14 November 2018, 22:03:03 UTC |
| 572710e | Daniel Agar | 14 November 2018, 19:46:04 UTC | EKF add typed Quatf getter | 14 November 2018, 21:50:39 UTC |
| 8a2e512 | CarlOlsson | 28 June 2018, 12:50:42 UTC | terrain_estimator: add vehicle_variance_scaler | 14 November 2018, 21:38:48 UTC |
| acde4eb | Paul Riseborough | 14 November 2018, 10:41:11 UTC | EKF: fix comment | 14 November 2018, 20:16:31 UTC |
| df57120 | Paul Riseborough | 13 November 2018, 20:19:23 UTC | EKF: Fix bug preventing saved declination being used for first alignment | 14 November 2018, 20:16:31 UTC |
| 4657a9c | Paul Riseborough | 13 November 2018, 05:09:37 UTC | EKF: Ensure yaw gets reset when declination is set | 14 November 2018, 20:16:31 UTC |
| 68a3cbc | Roman | 08 November 2018, 15:33:47 UTC | addressed review comment Signed-off-by: Roman <bapstroman@gmail.com> | 12 November 2018, 21:42:28 UTC |
| a74b7ef | Roman | 07 November 2018, 10:55:45 UTC | 3D only mag fusion: make sure _flt_mag_align_complete is set - _flt_mag_align_complete was not set when choosing pure 3D mag fusion. one effect of this was that the declination used in the filter was not the one calculated from the magnetic field states. Signed-off-by: Roman <bapstroman@gmail.com> | 12 November 2018, 21:42:28 UTC |
| 10a0fef | jie.zhang | 17 October 2018, 19:50:00 UTC | EKF/control: remove unnecessary yaw_align update | 17 October 2018, 19:50:00 UTC |
| 8bbaf89 | Daniel Agar | 17 October 2018, 19:30:25 UTC | EKF add new simple setIMUData() interface (#512) | 17 October 2018, 19:30:25 UTC |
| e6cd6ef | Paul Riseborough | 02 September 2018, 22:17:39 UTC | EKF: Fix bug preventing use of flow sensors without gyros The handling of invalid flow data when on ground is performed in controlOpticalFlowFusion() where it is able to handle flow sensors that don't publish gyro data. | 10 October 2018, 20:12:54 UTC |
| 2c3a064 | Daniel Agar | 19 September 2018, 00:16:29 UTC | README.md update Jenkins build status icon and url | 19 September 2018, 00:16:29 UTC |
| f240eaa | Paul Riseborough | 18 September 2018, 10:25:10 UTC | EKF: Fix test script | 18 September 2018, 20:32:43 UTC |
| 5cb228e | Paul Riseborough | 18 September 2018, 10:00:31 UTC | EKF: Remove duplicate line | 18 September 2018, 20:32:43 UTC |
| 0220f46 | Paul Riseborough | 18 September 2018, 09:59:02 UTC | EKF: Use more generic variable name for bad yaw fusion flag This flag now reports on fusion of data that is not from a magnetometer. | 18 September 2018, 20:32:43 UTC |
| eb52f7c | Paul Riseborough | 02 September 2018, 23:25:19 UTC | EKF: Remove implicit conversion and use single precision trig operator Double precision accuracy is not required for this operation. | 18 September 2018, 20:32:43 UTC |
| 6847acb | Paul Riseborough | 02 September 2018, 11:22:58 UTC | EKF: Add protection for undefined GPS antenna array heading offset | 18 September 2018, 20:32:43 UTC |
| d1f3f4c | Paul Riseborough | 01 September 2018, 23:32:01 UTC | EKF: Use #define function for isfinite check | 18 September 2018, 20:32:43 UTC |
| 91f886c | Paul Riseborough | 27 August 2018, 21:22:48 UTC | EKF: Add support for use of GPS heading data. Heading data is assumed to be from a dual antenna array at a specified yaw angle offset in body frame, but with the heading data already corrected for antenna offset. The offset is required to apply the correct compensation for combined rotations and to determine when the yaw observation has become badly conditioned. | 18 September 2018, 20:32:43 UTC |
| af60e6b | Paul Riseborough | 27 August 2018, 21:17:31 UTC | EKF: Add derivation for for dual antenna yaw observations | 18 September 2018, 20:32:43 UTC |
| 342c3ab | Paul Riseborough | 27 August 2018, 21:00:53 UTC | EKF: Fix timeout parameter documentation and name The parameter used to control the maximum dead reckoning time had 'gps' in the parameter name which was confusing because it was used for all measurement types capable of constraining horizontal velocity error growth. The parameter variable has been renamed and the documentation for it improved. The parameter used to control the maximum time since fusing a measurement before the measurement is considered to be not contributing to aiding had misleading documentation which has been updated. | 18 September 2018, 20:32:43 UTC |
| 2354c30 | Dion Gonano | 10 September 2018, 20:36:13 UTC | Add vision height covariance passthrough (#507) * Add vision height covariance passthrough * Fix hgtErr layout | 10 September 2018, 20:36:13 UTC |
| a724522 | Kabir Mohammed | 03 September 2018, 04:26:43 UTC | EKF : Fix divergence when optical flow is not fused for a long time (#503) * terrain_estimator : guard against case where latest range sample is newer than IMU sample * EKF : control : correct detection of no optical flow fusion over a time period | 03 September 2018, 04:26:43 UTC |
| b861594 | Paul Riseborough | 01 September 2018, 11:10:57 UTC | EKF: enable optical flow use to bootstrap | 01 September 2018, 12:49:45 UTC |
| 3c74dc2 | Daniel Agar | 30 August 2018, 15:48:55 UTC | ECL_Controller base needs virtual destructor | 30 August 2018, 16:07:25 UTC |
| 0d39072 | Daniel Agar | 30 August 2018, 15:48:38 UTC | EKF add missing header guards | 30 August 2018, 16:07:25 UTC |
| a6a1edb | Kabir Mohammed | 30 August 2018, 08:26:17 UTC | EKF : Fix type of flow data navigation variable (#499) | 30 August 2018, 08:26:17 UTC |
| a53ad9c | Paul Riseborough | 24 August 2018, 22:25:03 UTC | EKF: Add missing optical flow ground motion protection Motion compensated optical flow rates are supposed to be zeroed if reported flow quality is below the minimum threshold value when on ground. The comments and logic have been amended to be consistent and make the design intent clearer. | 28 August 2018, 19:31:50 UTC |
| 56b8bb0 | dakejahl | 23 August 2018, 20:42:58 UTC | added a checking for range_aid to determine if we are relying on the range finder as our primary source of height observation (#491) | 23 August 2018, 20:42:58 UTC |
| aa134d5 | Roman | 20 August 2018, 19:29:42 UTC | l1 position controller: make sure that a NAN roll setpoint does not keep the roll setpoint state at NAN forever Signed-off-by: Roman <bapstroman@gmail.com> | 21 August 2018, 14:16:33 UTC |
