Nonfragile H∞ Filtering for Discrete- time T-S Fuzzy Systems

This paper is concerned with the problem of nonfragile H-infinity filtering for discrete-time nonlinear systems in Takagi-Sugeno's form. Additive interval uncertainty which reflects imprecision in filter implementation due to finite word length effects is considered. Based on vertex theory and probabilistic robust approach, deterministic and randomized filtering algorithms are proposed, respectively. Compared with the deterministic algorithm, advantages of the randomized algorithm are mainly two-fold. First, it has acceptable computational complexity, especially for high-dimensional systems. And second, it reduces the design conservatism. An example about diode-tunnel circuit is given to illustrate the effectiveness of the proposed method.