Incorporation of experience in iterative learning controllers using locally weighted learning

A method of incorporating experience in iterative learning controllers is p roposed in this paper. Importance of the selection of initial control input in the convergence of error is highlighted. It is proposed that if previou s experience of the controller can be incorporated in the selection of the initial control input for a new desired trajectory tracking task, the conve rgence of error can be improved without modifying the structure of the cont roller. Therefore, the proposed method is very general and is applicable to most of the iterative learning control algorithms. (C) 2001 Elsevier Scien ce Ltd. All rights reserved.