Iterative learning control for nonlinear nonminimum phase plants

Learning control is a very effective approach for tracking control in proce sses occurring repetitively over a fixed interval of time. In this paper a robust learning algorithm is proposed for a generic family of nonlinear non minimum phase plants with disturbances and initialization error. The "stabl e-inversion" method of Devasia, Chen and Paden is applied to develop a lear ning controller for linear nonminimum phase plants. This is adapted to acco mmodate a more general class of nonlinear plants. The bounds on the asympto tic error for the learned input are exhibited via a concise proof. Simulati on studies demonstrate that in the absence of input disturbances, perfect t racking of the desired trajectory is achieved for nonlinear nonminimum phas e plants. Further, in the presence of random disturbances, the tracking err or converges to a neighborhood of zero. A bound on th tracking error is der ived which is a continous function of the bound on the disturbance. It is a lso observed that perfect tracking of the desired trajectory is achieved if the input disturbance is the same at every iteration.