Iterative learning control using information database

Abstract

In this paper, the problem of convergence rate in the initial iterations for iterative learning control schemes is addressed. Iterative learning control using information database (ILCID) is proposed for improving the convergence rate in the initial iterations. It is assumed that proper and efficient selection of initial control input can improve the convergence rate in the initial iterations. The initial control input for a new desired trajectory is constructed by using the information database based on a similarity index defined in this paper. This method is very general and can be applicable to all iterative learning algorithms.