Experimental implementation of iterative learning control for processes with stochastic disturbances

A number of iterative learning control algorithms have been developed in a stochastic setting in recent years. The results currently available are in the form of fundamental systems theoretical properties and associated algorithm development. This paper reports results from the application of a stochastic algorithm on a gantry robot system that has been used in the benchmarking a range of deterministic algorithms. These results confirm that this algorithm is capable of delivering good performance in the experimental domain, including comparison against an alternative.