One-step-ahead optimal adaptive control based on ELS algorithm

An optimal adaptive controller for multidimensional disturbed systems is concerned in this paper. The system parameters, especially the first parameters of the controller, are unknown a prior. The one-step-ahead adaptive controller is designed based on the input matching technique and the extended least-squares (ELS) algorithm. It is shown that the system identification is consistent and the adaptive system is closed-stable globally. The adaptive controller converges to the one-step-ahead adaptive controller. Finally, some simulation examples are included to illustrate the effectiveness and efficiency of the optimal controller.