A novel predictive and iterative learning control algorithm

Abstract

In this article, an iterative learning control algorithm enhanced with predictive features is developed for time-varying linear systems. An error model is introduced that can represent the transition of tracking error in two successive trials. Based on this model, a predictive and iterative learning control algorithm is derived that is based only on the trial number (or repetition index). A rigourous analysis of the convergence of this hybrid algorithm is provided. In addition, the robustness of the algorithm against modelling errors, initial errors, and the presence of disturbances is discussed. To highlight the key features of the algorithm, an example is provided to demonstrate the performance achievable in an injection-molding application.