Trajectory Tracking of Piezoelectric Actuators using State-compensated Iterative Learning Control

This research proposes a novel scheme of state-compensated iterative learning control (SCILC) for trajectory tracking of piezoelectric systems. The new ILC scheme adds a state compensation term to the conventional Arimoto-type ILC formula. The state compensation term is based on the difference in tracking error between current and previous iterations. With the aid of this state compensation, ILC can therefore take effect more precisely as desired. Further, the addition of the current error in the compensation term can also serve the role of feedback control to stabilize the system. The new ILC scheme keeps the simplicity feature of conventional ILC that no model is required for the control purpose. Experimental results demonstrate that a piezoelectric actuator using the proposed learning scheme can achieve error convergence in about three to five iterations, and keep steady-state tracking errors close to the system noise level.