A novel Lyapunov-stability-based recurrent-fuzzy system for the Identification and adaptive control of nonlinear systems

The requirement to manage complicated nonlinear systems with high uncertainty is one of the main drivers of progress in the identification and control discipline. Since most of the systems are nonlinear, exact identification and control of them using traditional techniques is a challenging task. This paper proposes a novel Takagi–Sugeno (T-S) type Recurrent Fuzzy System (RFS) for nonlinear system identification and adaptive control. To ensure the system’s overall stability, a novel Lyapunov-stability-based learning algorithm is developed to obtain the parameter update equations for the proposed model. The performance of the proposed model is also evaluated with the well-known gradient-descent-based Back Propagation (BP) learning method and compared with the Lyapunov-stability method. To judge the efficacy of the proposed model and the learning algorithm we have performed an extensive comparative study by considering other popular soft-computing models such as Jordan Recurrent Neural Network (JRNN), Feed-Forward Neural Network (FFNN), Radial Basis Function Network (RBFN) and a conventional recurrent fuzzy system. The results obtained from the proposed method (RFS ＋ Lyapunov-stability-based learning algorithm) are compared with those obtained from other models and are found to be superior.