this zeros the delta angle and delta velocity accumulators when a
sensor is unavailable for a period of 0.1s. This prevents garbage
values being passed into the EKF when a sensor dies and then becomes
available again some time later
this we ensures we get new data for all active IMUs on each loop,
rather than sometimes returning with some IMUs not having data.
This matters as not having a sample on an IMU for a single loop can
cause an EKF IMU failover, which will degrade the learned bias
variances
The issue is usually only seen under high load, such as requesting a
loop rate beyond what the hardware is capable of
this allows us to learn the gyro biases each lane would need if it had
to switch to another gyro due to a sensor failure. This prevents a
sudden change in gyro bias on IMU failure
this allows Linux boards to trigger an external script on GPIO
writes. This can be hooked up to relays with either RC options or
mavlink or mission items to trigger actions that can be scripted
this ensures we consistently fly with EKF lane1 if it is healthy at
the point we arm. Otherwise the choice of lane will be a lottery.
This is important as many systems have quite different filtering and
vibration characteristics on their different IMUs. We by default
enable fast sampling only on the first IMU for example, which means
the 2nd and 3rd IMUs are more vulnerable to high freq causing
aliasing.
this ensures we consistently fly with EKF lane1 if it is healthy at
the point we arm. Otherwise the choice of lane will be a lottery.
This is important as many systems have quite different filtering and
vibration characteristics on their different IMUs. We by default
enable fast sampling only on the first IMU for example, which means
the 2nd and 3rd IMUs are more vulnerable to high freq causing
aliasing.
Andrew Tridgell
Randy Mackay
Michael du Breuil
Peter Barker
Rajat Singhal
Atxheli
Hwurzburg
Tom Pittenger
Alexey Kozin