Use Tilt Servo as Aileron in Cruise Flight

Hi everybody!

I wonder if the VTOL in the pictures below can be controlled by Ardupilot’s QuadPlane in all flight phases. This aircraft has no dedicated ailerons. It uses the wingtip tilt servo for both tilting the motor and deflecting the wingtip segment (which acts as aileron in cruise flight).


Is there a possibility to use tilt servos as aileron or elevator servos in cruise flight. Or also using differential thrust of the tilted motors for yaw control in cruise?

If it is not directly forseen, do you have ideas on how to implement this? Mixers? External multiplexer device? In the source code?

Thanks for your help and sorry if I might have missed a related thread or ardupilot feature.


1 Like

For differential thrust.

With tailsitters you can use vectored thrust in forward flight, not implemented for tiltrotors yet tho

Looks familiar :wink:
I guess we could also chat about it tomorrow at TUM.

Thanks for your replies so far!

@iampete @vbnhu Is this functionality (tiltwing as aileron) still not implemented? Thanks!

Scripting is the go to solution for this, the mixing can be to subtle to define with a limited number of parameters.

Ok, thanks! I have not previously done any scripting with lua just some reading – so before I delve in, are there any examples of something similar to this type of problem that you know of (don’t seem to be any in /libraries/AP_scripting/examples as far as I can tell)

Yeah, no great examples.

Basically you read in the aileron, rudder and tilt servos do your own mixing and then output to the scripting servos.

Alright, I’ll start giving it a try, thanks. May be seeing me open some topics soon

@iampete Do you know if there is a method to get the current flight type state of the vehicle (e.g., VTOL, transitioning, fixed-wing)?

edit: just saw singleton QuadPlane method in_vtol_mode boolean. Will try that.

edit2: @iampete one issue I have is that I need a good indicator to know when the transition from VTOL to FW is finished. Haven’t come up with solid method yet - some ideas were delays, monitoring PWM outputs – but none really work that well. Any suggestions?