Robust H8 and Guaranteed Cost Filtering for T-S Fuzzy Systems with Multipath Quantizations

This paper investigates the problems of robust H-infinity and guaranteed cost filtering for discrete-time uncertain Takagi-Sugeno (T-S) fuzzy systems with multipath quantizations. The "multipath quantizations " mean that both the measurement output and estimated output of the uncertain T-S fuzzy systems are quantized by two different dynamic quantizers before they are transmitted. The unknown uncertain parameters are assumed to be norm bounded. Through applying the S-procedure and introducing some slack matrix variables, new sufficient conditions about the robust asymptotical stability with specific performance measures for quantized filtering error system have been developed via the fuzzy-basis-dependent Lyapunov function approach. The desired robust H-infinity filter, robust guaranteed cost filter and dynamic quantizer parameters can be easily obtained by means of linear matrix inequalities (LMIs). Finally, a practical example about the mass-spring-damper mechanical system is given.