1st IEEE Int. Conf. Machine Learning and Cybernetics,
Volume 3,
pgs. 1263--1266,
2002
In this paper, iterative learning cnntrol prnhlem is investigated for a class of time-varying discrete nonlinear systems in the presence of different disturbances. Conditions for guaranieeing the robustness of open-dosed-lwp iterative learning control system are presented. It is shown that the tracking errors of ihe cnutrnl system are n n i f o d y hounded in the presence of hounded state uncertainty, output dbturhance and initial state error. The system can track desired trajectory completely if all the uncertainties are g r a d d y repeated. The conditionsshow that mast coeftlcients of the iterative learning eontroller have no effects on the system robustness.