Robustness of the characteristic model-based golden-section adaptive control

Abstract

The characteristic model-based golden-section adaptive control law (CM-GSAC) has been developed for over 20 years in China with a broad range of applications in various elds. This paper discusses the optimization of the control coef cients l 1 and l 2 such that the closed-loop system can tolerate the largest parameter estimation error. The Vinnicombe distance is introduced as a metric to evaluate the distance between the true CM and the estimated CM. According to the simulation results, system with CM-GSAC shows better tracking performance than that of the well known model-reference adaptive control and the multiple model adaptive control.