Introduction to ILC: Concepts, Schematics, and Implementation

Abstract

In this chapter we review several important concepts, basic schematics, and implementation issues associated with real-time ILC. We first introduce five basic configurations of ILC for linear processes, including the previous cycle learning, current cycle learning, previous and current cycle learning, cascade learning, and incremental cascade learning. Next, we focus on ILC for non-linear processes, make clear two conditions intrinsic to ILC applications – the global Lipschitz continuity condition and identical initialization condition, and explore possible extensions. Finally we discuss three practical issues encountered in real-time ILC implementation – repetitive control tasks, robustness and filter design, as well as the sampling effect.