>>2933294
This however might induce a feedback loop, would probably be better if the gyro/accelerometer was on the portion that's moving up & down, then invert the data from it to have the motor/servo negate the motion above.