A nonlinear switching control method for a class of non-minimum-phase nonlinear systems

This paper presents a nonlinear control method by combining adaptive-network-based fuzzy inference system (ANFIS) with multiple models for a class of uncertain discrete-time nonlinear systems with unstable zero-dynamics. The proposed method is composed of a linear robust controller, an ANFIS-based nonlinear controller, and a switching mechanism using multiple models technique. The method in this paper has the following three features compared with the results available in the literature. First, this method relaxes the commonly-used global boundedness assumption on the unmodeled dynamics, and thus can cope with a much general class of practical applications. Secondly, ANFIS is used to estimate and compensate for the unmodeled dynamics adaptively in the nonlinear controller design, which improves the relatively low convergence rate of neural networks and reduces the possibility that the networks becomes trapped in local minima. Thirdly, to guarantee the universal approximation property of ANFIS, a “one-to-one mapping” is adapted. A simulation example is exploited to illustrate the effectiveness of the proposed method.