Adaptive Fuzzy Control For A Class Of Non - Strictly Feedback Nonlinear Systems

So far, fuzzy control has been considered as an effective approach to solve the problem of modeling and control for complex nonlinear systems. In particular, due to its universal approximation property, fuzzy logic system is usually used to approximate the unknown nonlinear functions in controlled systems, and adaptive fuzzy control method is derived. In recent years, adaptive fuzzy control has becomes an active research direction in control theory field. The main advantage of adaptive fuzzy control lies in that it can be used to deal with the problem of stability analysis and control synthesis of strict feedback nonlinear systems. At present, all of backstepping-based adaptive fuzzy control strategies are proposed for the strict feedback nonlinear systems. However, in practice there exist some controlled systems in non-strict feedback form. How to contain such a class of non-strict feedback systems is still a difficult and opened problem. In this thesis, we presents a variable separation method, by which a system function with all the state variables can be decomposed into a sum of some continuous single-variable functions. Furthermore, an adaptive fuzzy controller can be constructed with adaptive fuzzy control method and backstepping technique. The developed control strategy ensures the stability of the closed-loop system.The following is the detail: Firstly, adaptive fuzzy control has been regarded for a class of single input single output non-strict feedback nonlinear systems. To solve the control design problem for non-strict feedback systems, it is assumed that the system functions can be bounded by monotonously increasing functions. Based on such an assumption, a variable separation method is proposed. In the procedure of controller design, fuzzy logic systems are used to approximate the unknown nonlinear functions, and combined with backstepping adaptive technique is used to construct adaptive fuzzy controller.Secondly, based on the above work control systems with disturbance has been further addressed. The proposed adaptive controller ensures that all the signals of the closed-loop system is semi-globally bounded, and all state variables will converge to a sufficiently small neighborhood of the origin.Thirdly, A practical example has been used to illustrate effectiveness of the proposed method.