Non-identifier based adaptive control scheme with guaranteed stability

Abstract

In this paper, we investigate a non-identifier based adaptive control scheme for single-input single-output (SISO) plants. The overall adaptive control system can be viewed as a feedback interconnection of a SISO plant, a set of candidate controllers and a non-identifier based adaptive scheme that supervises the switching process among the candidate controllers. The adaptive scheme is designed without any explicit assumptions on the plant model, and the closed-loop system is guaranteed to be stable as long as at least one of the candidate controllers is stabilizing. The basic idea has been investigated by Safonov et al. in [1, 2], but the candidate controllers are limited to be "stably causally left invertible" (SCLI). In this paper, we show that stronger stability results can be achieved by properly resetting controller states. Furthermore, the requirement for SCLI is eliminated for the case where all the subsystems are linear time invariant (LTI).