Predictor-Based Model Reference Adaptive Control

Abstract

This paper is devoted to robust, Predictor-based Model Reference Adaptive Control (PMRAC) design. The proposed adaptive system is compared with the now-classical Model Reference Adaptive Control (MRAC) architecture. Simulation examples are presented. Numerical evidence indicates that the proposed PMRAC tracking architecture has better than MRAC transient characteristics. In this paper, we presented a state-predictor based direct adaptive tracking design methodology for multi-input dynamical systems, with partially known dynamics. Efficiency of the design was demonstrated using short period dynamics of an aircraft. Formal proof of the reported PMRAC benefits constitute future research and will be reported elsewhere.