Adaptive Control using Multiple Models Switching

Abstract

The purpose of this paper is to marry the two concepts of multiple model adaptive control and safe adaptive control. In its simplest form, multiple model adaptive control involves a supervisory switching among one of a finite number of controllers as more is learnt about the plant, until one of the controllers is finally selected and remains unchanged. Safe adapative control is concerned with ensuring that when the controller is changed the closed-loop is ever unstable. This paper introduces a receding horizon multiple model, switching and tuning control scheme based on an on-line redesign of the controller.. This control scheme has a natural two-stage adaptive control algorithm: identification of the closest model and design of the control law. The computational complexity aspects of this approach to adaptive control are discussed briefly. A nonlinear system is used to illustrate the ideas.