Commit Graph

18717 Commits

Author SHA1 Message Date
Paul Riseborough 60c2e81d19 AP_NavEKF2: Apply optimisations to declination fusion covariance update
These are the same type of optimisations that were successful with the magnetometer fusion
2015-10-20 15:21:41 +11:00
Paul Riseborough 0054291cf4 AP_NavEKF2: Apply optimisations to side-slip fusion covariance update
These are the same optimisations that were successful with the magnetometer fusion
2015-10-20 15:21:40 +11:00
Paul Riseborough 5e4bc4e954 AP_NavEKF2: Apply optimisations to airspeed fusion covariance update
These are the same optimisations that were successful with the magnetometer fusion
2015-10-20 15:21:40 +11:00
Paul Riseborough a29147d6d2 AP_NavEKF2: Apply optimisations to optical flow covariance update
These are the same optimisations that were successful with the magnetometer fusion
2015-10-20 15:21:40 +11:00
Paul Riseborough b5e43288d4 AP_NavEKF2: Down to 12 from 14 usec for perf test[8] 2015-10-20 15:21:40 +11:00
Paul Riseborough 0f530bb5a0 AP_NavEKF2: Correct comments 2015-10-20 15:21:40 +11:00
Andrew Tridgell d1dfd5fd01 AP_NavEKF2: use common header for optimisation level and irq disable 2015-10-20 15:21:40 +11:00
Andrew Tridgell a017ae7e00 AP_NavEKF2: down to 218us for test[9] 2015-10-20 15:21:40 +11:00
Andrew Tridgell fc23be8025 AP_NavEKF2: enable fine grained perf tuning for mag fusion 2015-10-20 15:21:40 +11:00
Andrew Tridgell b6c1352e4c AP_NavEKF2: added 10 test perf counters
used for fine grained performance tuning
2015-10-20 15:21:40 +11:00
Paul Riseborough 824436dfb6 AP_NavEKF2: Explicitly define constants as floats 2015-10-20 15:21:40 +11:00
Paul Riseborough 2fbd050418 AP_NavEKF2: Use blended accelerometer data
If high vibration levels cause offsets between the two, it switches to the accelerometer with lower vibration levels. The default behaviour is to use the average of both accelerometers.
2015-10-20 15:21:40 +11:00
Paul Riseborough 240ea92947 AP_NavEKF: Re-center gyro bias limits on arming
Allows the gyro bias to vary by up to +-10 deg/second before and after arming.
2015-10-20 15:21:39 +11:00
Paul Riseborough 8515dda727 AP_NavEKF: Fix bug in vertical position derivative calculation 2015-10-20 15:21:39 +11:00
Paul Riseborough 2b13020c19 AP_NavEKF2: Fix bug in vertical position derivative calculation 2015-10-20 15:21:39 +11:00
Paul Riseborough f6ad79688e AP_NavEKF2: Update start sequence console messages 2015-10-20 15:21:39 +11:00
Paul Riseborough 6b3e114cd6 AP_NavEKF: Enable EKF1 to be disabled to reduce frame over-runs 2015-10-20 15:21:39 +11:00
Paul Riseborough 1a1236f764 AP_NavEKF2: Update magnetometer fusion documentation 2015-10-20 15:21:39 +11:00
Paul Riseborough 8526a8ba7e AP_NavEKF2: Level processor loading between frames
Don't fuse other measurements on the same frame that magnetometer measurements arrive if running at a high frame rate as there will be insufficient time to complete other operations.
2015-10-20 15:21:39 +11:00
Paul Riseborough 225b81baf4 AP_NavEKF2: Prevent users breaking filter by setting excessive time delays
If time deelays greater than the IMU buffer length are used, then it is possible for measurements to be dropped.
2015-10-20 15:21:39 +11:00
Paul Riseborough a1d6c7ae1e AP_NavEKF2: Improve timing accuracy
This parameter is a compromise between numerical accuracy of the covariance prediction and sensor timing jitter
Further testing has shown that doing covariance prediction and sensor fusion every 10msec has no observable effect on fusion health and reduces timing hitter noise on magnetometer observations during high rate maneovures
2015-10-20 15:21:39 +11:00
Paul Riseborough e0ed2dab63 AP_NavEKF2: Adjust IMU FIFO buffer for different IMU data rates
The values chosen ensure that up to consistent 250 msec of sensor delay compensation is available for different platform types
The revised values also ensure that fusion occurs at different time to when the 10Hz magnetometer measurements are read
2015-10-20 15:21:39 +11:00
Paul Riseborough baa8692960 AP_NavEKF2: Allow use of magnetometer learning during optical flow nav
Adds fusion of the declination when there are no earth relative measurements so that the declination angle and therefore the copters yaw angle have an absolute reference.
This enables the length (but not the declination) of the earth field North/East states to change along with the magnetometer offsets.
2015-10-20 15:21:38 +11:00
Paul Riseborough 1b8a93ef0c AP_NavEKF2: Update function header comments 2015-10-20 15:21:38 +11:00
Paul Riseborough 6899767d28 AP_NavEKF2: Disable magnetic field learning if we have no absolute position reference 2015-10-20 15:21:38 +11:00
Paul Riseborough 75a61df627 Copter: Enable access to EKF2 height tuning 2015-10-20 15:21:38 +11:00
Paul Riseborough d0aba09503 Copter: Access EKF healthy through AHRS object
Enables EKF2 use
2015-10-20 15:21:38 +11:00
Paul Riseborough 19785c1033 Copter: Access EKF origin through AHRS object
Enables EKF2 use
2015-10-20 15:21:38 +11:00
Paul Riseborough 7c6b31b585 Copter: Access EKF variance checks through AHRS object
Supports flight using EKF2
2015-10-20 15:21:38 +11:00
Paul Riseborough 0722ebe8a0 AP_AHRS: Add EKF variance checks 2015-10-20 15:21:38 +11:00
Paul Riseborough 242b9641c8 AP_NavEKF2: Fix in-air logic bug 2015-10-20 15:21:38 +11:00
Paul Riseborough e7de2d3ea3 AP_NavEKF: Update magnetic field learning options
Provide an option to always do learning
Make field learning decision logic clearer
Change defaults so that plane learns when airborne
Change defaults so that Rover does not learn (large external magnetic interference)
2015-10-20 15:21:38 +11:00
Paul Riseborough 495809f211 AP_NavEKF2: Allow magnetic field learning to be faster 2015-10-20 15:21:38 +11:00
Paul Riseborough 84a02efd52 AP_NavEKF2: Improve efficiency of Z magnetometer fusion 2015-10-20 15:21:37 +11:00
Paul Riseborough 35b08849f7 AP_NavEKF2: Reduce effect of rounding errors on covariance prediction 2015-10-20 15:21:37 +11:00
Paul Riseborough 4e928bf294 AP_NavEKF2: Fix bug preventing pre-takeoff reference measurements 2015-10-20 15:21:37 +11:00
Paul Riseborough 71c399674a AP_NavEKF2: Delay use of magnetic field states until off-ground
Magnetic interference whilst on the ground can adversely affect filter states. This patch ensures that the simpler and more robust magnetic heading observation method is used until the vehicle has cleared the ground.
2015-10-20 15:21:37 +11:00
Paul Riseborough 99c2dc41e0 AP_NavEKF2: Miscellaneous tuning changes 2015-10-20 15:21:37 +11:00
Paul Riseborough 2d44441d6b DataFlash: Make EKF2 PosDownDerivative interface follow coding conventions
Updates arising from peer review.
2015-10-20 15:21:37 +11:00
Paul Riseborough ae830b44c4 AP_AHRS: Make EKF2 PosDownDerivative interface follow coding conventions
Updates arising from peer review.
2015-10-20 15:21:37 +11:00
Paul Riseborough 840f307d58 AP_NavEKF: Make EKF2 PosDownDerivative interface follow coding conventions
Updates arising from peer review.
2015-10-20 15:21:37 +11:00
Paul Riseborough 1323db10e8 DataFlash: Make EKF2 PosDownDerivative interface follow coding conventions
Updates arising from peer review.
2015-10-20 15:21:37 +11:00
Paul Riseborough e8037f94ad AP_AHRS: Make EKF2 PosDownDerivative interface follow coding conventions
Updates arising from peer review
2015-10-20 15:21:37 +11:00
Paul Riseborough 47ae0f35f6 AP_NavEKF2: Make EKF2 PosDownDerivative interface follow coding conventions
Updates arising from peer review.
2015-10-20 15:21:37 +11:00
Paul Riseborough f63c32531c AP_InertialNav: Address peer review comments
The decision to set the origin to zero has been moved out of the AHRS library and into the inertial nav library as this is consumer specific.
2015-10-20 15:21:36 +11:00
Paul Riseborough 58d57994f4 AP_AHRS: Rectify coding errors found during peer review
Fixes code that didn't take into account fall-through behaviour of C++ switch statements
Makes get_rigin furnction more generic allowing the consumer to decide what to do with an invalid origin
2015-10-20 15:21:36 +11:00
Paul Riseborough 27393855f1 AP_NavEKF2: Change assumed gyro calibration accuracy
If an external gyro calibration has been performed, we should assume that it has been done under static conditions
Otherwise it is pointless and we should allow the EKF to find its own gyro bias offsets.
2015-10-20 15:21:36 +11:00
Paul Riseborough 67a669fcdc AP_NavEKF2: Correct comments for height measurement 2015-10-20 15:21:36 +11:00
Paul Riseborough 0dc570b5a5 AP_NavEKF2: Improve optical flow takeoff detection logic
Ensure takeoff detect status goes to false when on ground
2015-10-20 15:21:36 +11:00
Paul Riseborough d3a6690e4f AP_NavEKF2: Fix bugs preventing use of optical flow navigation 2015-10-20 15:21:36 +11:00