Study On Fuzzy Control For Uncertain Systems Via Fuzzy Model

In recent years, fuzzy control for uncertain systems via fuzzy model has been a hot topic in the field of intelligent control. Some correlative problems in this area are studied in this paper.Firstly, a new design scheme of robust fuzzy control for a class of uncertain nonlinear systems based on T-S fuzzy model is proposed. In this scheme the linear matrix inequality (LMI) technique and the concept of the so-called parallel distributed compensation (PDC) is employed to design the state feedback controller. Sufficient conditions with respect to decay rateαare given in the sense of Lyapunov asymptotic stability. By using relaxed stability condition, the scheme has less conservatism.Secondly, a new design scheme of mixed H 2 /H∞fuzzy output feedback tracking control for a class of nonlinear dynamic systems is proposed. By LMI technique, a fuzzy observer-based controller is developed to reduce tracking error as possible and guarantee a desired H∞tracking performance for all bounded reference inputs, and achieve the sub-optimal H 2 control performance with the desired H∞tracking performance. The problem of designing the observer and controller is translated into eigenvalue problem (EVP). The scheme combines the merits of both the H 2 optimal control and the H∞robust control. The state of the systems needs not to be known in the scheme. Furthermore, a novel scheme of H∞fuzzy tracking control for a class of nonlinear dynamic systems is proposed based on the idea of dynamic parallel distributed compensation (DPDC).Thirdly, a new design scheme of direct adaptive fuzzy control for a class of nonlinear discrete-time systems with delay is proposed. The T-S fuzzy model is employed to represent the systems. The concept of the so-called PDC is employed to design the fuzzy controller with unknown parameters and the coefficients of the controller are identified by gradient descent algorithm. By input-to-state stability (ISS) approach, the error between the system output and the reference output is proved to be bounded and to satisfy some average performance. In the scheme, the commonly-used...