Before the VTOL status would automatically force a commander state update all the time. This saves effort and makes sure the system only updates when it should.
We were running pre-arm checks before when not arming, which led to annoying error messages on vehicles that were on the bench or serviced on the ground. Now we really only run them when trying to arm.
This check is performed for up to 30 seconds after takeoff or until a horizontal speed of 5 m/s has been achieved.
If the vehicle is not landed and the check is active, then the yaw will be declared as failed if the velocity innovations fail for a continuous period of 1 second.
This will cause the local and global position and velocity validity to be latched false to prevent unsafe use of position control modes.
Fix the bug allowing arming without GPS checks passed in the first 20 seconds after gaining GPS lock when COM_ARM_WO_GPS is set to 0
Allow 10 seconds after boot for EKF quality checks to pass before reporting failure to allow EKF to stabilise.
Move GPS quality checking and reporting to after all innovation and bias checks.
Make messages more informative.
Add missing GPS speed accuracy check.
The commander checks use instantaneous values which are vulnerable to false negatives or positives with noisy data or transient faults.
This patch checks the estimators published pre flight check status which is based on persistent checks using filtered data.
This change will force commander into manual reversion mode when an external override device (like PX4IO) overrides the system externally. This is not a functional change on the outputs, as they were in override mode even without this patch. However, this change ensures that the system state is consistent with the output state and also ensures that the pilot and operator has better situational awareness when he / she triggers the manual reversion without realizing it.
arming_state_changed was not set in all places where an arming transition
occurred, for example when calling arm_disarm() from auto-disarm.
We did not notice because the state is published with at least 5 Hz already.
Lorenz Meier