A Variable Structure Observer Based Control Design for a Class of Large Scale MIMO Nonlinear Systems

This paper fully discusses how to design an observer based decentralized fuzzy adaptive controller for a class of large scale multivariable non-canonical nonlinear systems with unknown functions of subsystems’ states. On-line tuning mechanisms, which adjust both the parameters of the direct adaptive controller and the observer that guarantee the ultimately boundedness of the tracking error as well as that of the observer error, are derived through Lyapunov stability analysis. The most important merits of both the proposed controller and the observer are their robustness against external disturbances. The observer proposed in the paper is designed based on a reconfiguration of the system in which the dynamics of the model reference is taken into account. It should be emphasized that compared to the other methods recently cited in the literature, this paper designs proper controllers for a class of nonlinear large scale MIMO systems in which merely the structure of the systems is known. The simulation results easily approve the remarkable capability of the proposed method.