Throughout the past two decades Iterative Learning Control (ILC) has been analyzed from many different points of view and numerous algorithms for specific classes of systems and applications have been developed. Also, different frameworks have been proposed none of which became dominant for industrial control applications. In the linear case a parametric description has been well established. It forms the basis for the general framework presented here that completely embeds ILC into discrete linear systems theory. In addition to the intuitivity of the approach it is quite general. Parallel and serial combinations of the learning law with conventional feedback control can be considered as well as higher order learning laws. It is valid for arbitrary parametrizations, which allows the development of ILC strategies not only in the time and frequency domain but also in terms of other systems of basis functions.
Print ISSN: 0178-2312
Volume: 55, 03/2007
Pages: 119 - 126