keeping orientation with gyro

[sparrowkc]

I am using Kevin Watson's sensor code, And I want to keep the robot in a constant orientation (We have mecannums).

My current method tends to over correct while strafing, and starts to oscilate. Has anybody done this in the past? Tomorrow I am going to do some testing to get some values for an equation that will take into account rotational velocity and determine a coast down time so that the robot doesn't overshoot so much.

Currently I find a turn value with
turn = .14111*tempgyro+127;
and then put turn through my limit function.

[JohnC]

The limit function should be a closed-loop feedback function, it sounds like. And if it's oscillating, it sounds like you're using just Integrated control (from the PID term you keep hearing). Try a good whitepaper on that subject...

[sparrowkc]

I got it, and it is beautiful. This was a good day.