Detune to stop ground oscillation

[lthall]

This pr provides the ability to reduce the gains of roll, pitch, and yaw to remove ground oscillation while landed. Each axis has a gain reduction variable that is set when the aircraft is landed. The gain returns to normal in 2 x ATC_INPUT_TC.

This is a common problem on aircraft with flexible landing gear. It is often seen as a yaw oscillation during spool up on more solidly built aircraft also.

[rmackay9]

@lthall I'm sure this will be popular. Do we need to have both Roll and Pitch parameters? I wonder if we can imagine a frame where we'd want them to be different.

[bnsgeyer]

@lthall I'm sure this will be popular. Do we need to have both Roll and Pitch parameters? I wonder if we can imagine a frame where we'd want them to be different.

I think for heli there could be significantly different gains between pitch and roll. But I don't know if the same backoff factor could work for both axes.

[lthall]

I considered this but I believe it would be a mistake to not separate them. We don't want to compromise control authority unnecessarily. There are numerous aircraft that have significantly different ground interaction modes between all three axis.

[IamPete1]

Looks good. Just to confirm, it does should up in the ATSC log message?

[lthall]

@IamPete1 Sorry mate, I didn't understand what you were asking.

[IamPete1]

@IamPete1 Sorry mate, I didn't understand what you were asking.

The ATSC message contains the used values of the angle P scale and the rate PD scale if not all ones. The logging should "just work" with your changes, but it would be worth double checking just so we can track what is going on. You should see them slew up and down, Once they get to all ones the log should stop.

[lthall]

[rmackay9]

LGTM, thanks for the changes