Adaptive fuzzy-neural control with state observer for unknown nonlinear systems via H∞ approaches

In this paper, we developed an observer-based indirect adaptive fuzzy neural control for a certain class of higher order unknown nonlinear dynamic, in which only the system output can be measured. The architecture employs fuzzy-neural network (FNN) system to approximate the unknown system function in designing the FNN controller; a robust control law is designed for compensating the function approximation errors. Moreover, the H^∞ control algorithm obtained by a modified Riccati-like equation can attenuate the effect of the external disturbance on the tracking error to any prescribed level. It is proved that the overall adaptive scheme guarantees the global asymptotic stability in the Lyapunov sense with all signal involved are uniformly bounded. Simulation studies have shown that the proposed controller performs well in superior tracking performance.