- refactor all EKF backend output predictor pieces into new OutputPredictor class
- output states are now calculated immediately with new high rate IMU rather than after EKF update
- IMU delayed sample is passed as around as control data to avoid storing an extra copy and make the requirement clear
- move EV yaw and EV position to new state machines
- EV yaw and EV pos now configured via EKF2_EV_CTRL (migrated from EKF2_AID_MASK)
- new EV position offset estimator to enable EV position while GPS position is active (no more EV pos delta fusion)
- yaw_align now strictly means north (no more rotate external vision aid mask)
- automatic switching between EV yaw, and yaw align north based on GPS quality
- respect new EKF2_EV_CTRL parameter for VPOS usage
- EV hgt rotate EV position before usage (there's often a small offset in frames)
- EV hgt reset use proper EV velocity body frame
- try to keep EV hgt and EV vel state machines consistent
- small incremental piece of https://github.com/PX4/PX4-Autopilot/pull/19128
- in a lot of cases this won't really matter, although you can see it in logs (especially with things like vehicle_at_rest), but it's something we might as well do properly
Co-authored-by: Mathieu Bresciani <brescianimathieu@gmail.com>
- for accel/gyro/mag estimated bias only consider them stable (valid
for calibration updates) if the value isn't changing (10% of limit) over
the validity period
Also remove the legacy "range aid" than can be achieved by setting the
height reference to range finder and the range finder control parameter
to "conditional".
Conditional range aiding cal also be set when the height reference isn't
the range finder. This prevents the ratchetting effect due to switching
between references.
Instead of having a single height source fused into the EKF and the
other ones "waiting" for a failure or the primary sensor, fuse all
sources in EKF2 at the same time. To prevent the sources from fighting against each
other, the "primary" source is set as reference and the other ones are
running a bias estimator in order to make all the secondary height
sources converge to the primary one.
If the reference isn't available, another one is automatically selected
from a priority list. This secondary reference keeps its current bias
estimate but stops updating it in order to be the new reference as close
as possible to the primary one.
Using the control status flags isn't robust as this part of the code
runs at the EKF update rate while the in_air transition is don at the
prediction rate. It was then likely to miss the transition
- replace float32[21] URT covariances with smaller dedicated position/velocity/orientation variances (the crossterms are unused, awkward, and relatively costly)
- these are easier to casually inspect and more representative of what's actually being used currently and reduces the size of vehicle_odometry_s quite a bit
- ekf2: add new helper to get roll/pitch/yaw covariances
- mavlink: receiver ODOMETRY handle more frame types for both pose (MAV_FRAME_LOCAL_NED, MAV_FRAME_LOCAL_ENU, MAV_FRAME_LOCAL_FRD, MAV_FRAME_LOCAL_FLU) and velocity (MAV_FRAME_LOCAL_NED, MAV_FRAME_LOCAL_ENU, MAV_FRAME_LOCAL_FRD, MAV_FRAME_LOCAL_FLU, MAV_FRAME_BODY_FRD)
- mavlink: delete unused ATT_POS_MOCAP stream (this is just a passthrough)
Co-authored-by: Mathieu Bresciani <brescianimathieu@gmail.com>
