Ar_Di,
Did you come to closure on Problem #1?
Did you mean that both roll and yaw were inverted (both individually negative), or did you mean that the roll and yaw axes were swapped?
My issue is that, if calibrated to have a standard (mode 2) control layout in multicopter mode (with pitch and roll on the right stick, and thrust and yaw on the left stick), then the roll and yaw axes are swapped after transition. That is, in fixed wing flight, a right roll is executed with a left yaw stick input, and a right yaw is executed with a right roll stick input. If you came to closure on problem 1, it could help solve this.
I would like to know what the intended behavior is. Currently, the control axes are fixed to the body the same way regardless of mode. Should the control axes be switching to the correct roll and yaw sense after transition is accomplished?
