Application of Adaline to the Synthesis of Adaptive Controllers for Dynamical Systems

The purpose of this paper is to report on neural network based adaptive controllers and their applications. Specifically, the proposed adaptive controller is applied to the benchmark problem recently proposed by Wie and Bernstein [1]. First, the continuous state space model of the benchmark problem is stabilized. The overall system is then discretized and the proposed adaptive controller is applied. We then analyze the effect of changing the parameters of the update algorithm on the performance of the closed-loop system. Computer simulations are presented to illustrate the discussion.