An iterative learning controller with initial state learning

In iterative learning control (ILC), a common assumption is that the initia l states in each repetitive operation should be inside a given ball centere d at the desired initial states which may be unknown. This assumption is cr itical to the stability analysis, and the size of the hall will directly af fect the final output trajectory tracking errors. In this paper, this assum ption is removed by using an initial state learning scheme together with th e traditional D-type ILC updating law. Both linear and nonlinear time-varyi ng uncertain systems are investigated. Uniform bounds for the final trackin g errors are obtained and these bounds are only dependent on the system unc ertainties and disturbances, yet independent of the initial errors. Further more, the desired initial states can be identified through learning iterati ons.