Use flow rates instead of integrals in backend. This allows us to delay
the data to the mitpoint integration time and simplifies the code in
general.
Gyro compensation can still be done in EKF2 if needed, but the
flow module normally already appends the correct gyro data to the flow
message.
* created a Performance Model for fixed wing vehicle
- added compensation for maximum climbrate, minimum sinkrate, minimum airspeed and trim airspeed based on weight ratio and air density
- added atmosphere lib to standard atmosphere calculations
---------
Signed-off-by: RomanBapst <bapstroman@gmail.com>
Co-authored-by: Thomas Stastny <thomas.stastny@auterion.com>
- in theory these are helpful to ensure EKF2_HGT_MODE configuration is
consistent with the relevant aid source (GPS, baro, etc), but it can
be a little awkward with users having to fight manual parameter
changes in the right order
Increasing the wind process noise results in a more dynamic
wind estimation, which is capable of catching fast-varying
winds. As wind is used in the lateral guidance it's important
that we don't filter it too much.
Furher the gate of the airspeed fusion is increased, to
reduce the likelihood of airspeed fusion stopping due to
dynamic wind conditions. The airspeed is validated in
the airspeed validator (EKF consumes the validated one).
Signed-off-by: Silvan Fuhrer <silvan@auterion.com>
- if not in air the accel noise is doubled
- if landed don't init unless GPS velocity is non-negligible
- when inactive continue seeding with EKF gyro bias
- reset yaw estimator if GPS fusion is stopped
There is no reason to keep an uncertainty on the origin as it is then
already contained in the local position estimate when GNSS data is fused
in the filter.
* ekf2_derivation: use single source of state definition
The state is defined as an ordered dictionary of group elements and
everything else is generated using that state definition
* ekf2: generated state sample add const reference getter
---------
Co-authored-by: bresch <[brescianimathieu@gmail.com](mailto:brescianimathieu@gmail.com)>
Co-authored-by: Daniel Agar <daniel@agar.ca>
The sparse vector template requires to know which states are non-zero in
the observation jacobian. This complicates the modularity of the code
when the state vector or the derivation is changed.
The computation cost difference is almost negligible for this size.
Even if there is some horizontal motion, a passing check should be
accepted as the terrain can be flat. However, the vehicle must not be
moving horizontally to invalidate the consistency as a change in terrain
can make the kinematic check temporarily fail.
bresch