High-order parameter-optimization iterative learning control algorithm

A high-order parameter-optimization iterative learning control algorithm is presented for solving the tracking problems of a class of linear time-invariant discrete system. The proposed algorithm is based on a quadratic performance objective function with the tracking errors from earlier trials. By solving this function we obtain the optimal time-varying parameters as the learning gain of the iterative update law. It is proved theoretically that when applied to the relaxed linear discrete system, the proposed algorithm guarantees the tracking error to converge to zero monotonically even the original system is nonpositive. Moreover, since more information of previous iterations is considered in the proposed algorithm, the robustness and convergence performance of the algorithm are improved accordingly. Finally, a case study is carried out to illustrate the performance of this new algorithm. ©, 2015, South China University of Technology. All right reserved.