Integrated fuzzy modeling and adaptive control for nonlinear systems

A systematic design methodology for integrating fuzzy modeling and adaptive control is proposed and developed in this paper. This design procedure provides a real-time system identification scheme using less fuzzy rules than that of the other existing methods due to a new sliding-mode learning mechanism embedded in the identified model, which has robust stability not only for stabilization of the identified system but also for trajectory tracking control. The integration of the identification and the adaptive control schemes ensures the suggested methodology overall advantageous and more attractive as compared to the other existing, usually separated, design approaches. Two typical complex systems are simulated, showing some convincing stabilization and tracking performance of the proposed integrated fuzzy system. © 2003 Elsevier Science Inc. All rights reserved.