Observer-based adaptive control using multiple-models switching and tuning

Abstract

Despite the remarkable theoretical accomplishments and successful applications of adaptive control, this field is not mature enough to solve challenging problems where strict performance and robustness guarantees are required. The needs of an approach that explicitly accounts for robust performance and stability specifications is a critical to the design of practical adaptive control systems. Towards this goal, this study extends the robust adaptive controller using multiple models, switching and tuning to multiple input multiple output and non-linear systems. The use of `extended superstability', instead of superstability, allows us to establish overall performance guarantees and reduce the conservativeness of the resulting closed-loop system. The authors show that under the proposed framework, the output and states remain bounded for bounded disturbances, as a direct consequence of the passivation properties of superstability. The effectiveness of the proposed algorithm is demonstrated in numerical simulations of a non-linear continuous stirred tank reactor.