Iterative Learning Control of Discrete Systems With Actuator Backlash Using a Weighted Sum of Previous Trial Control Signals.

This letter considers iterative learning control design for discrete dynamics in the presence of backlash in the actuators. A new control design for this problem is developed based on the stability theory for nonlinear repetitive processes. An example demonstrates the effectiveness of the new design where the system model is constructed from data collected from frequency response tests on a physical system.