I tried to plot power levels vs angular speed (in rad/s) and discovered goooood linearity ( remember how non-linear was RCX PWM)? The PID loop is a great idea, and I suppose that PID parameters should be tweakable. In SDK or HDK ( don't remember (...) (18 years ago, 3-Aug-06, to lugnet.robotics, lugnet.robotics.nxt)
(...) I've done the same measurements/plots and found same linearity when 'brake' on inactive PWM cycles is used. Not so good when 'float' is used. This leads to a conundrum that I simply can't explain. If you use 'inactive brake' mode the motors (...) (18 years ago, 3-Aug-06, to lugnet.robotics, lugnet.robotics.nxt)
