Nonlinear control system for a nonlinear time-varying plant

The output of a nonlinear plant varies according to the initial state of the plant. Therefore, if one tries to make the plant output track a reference output exactly, one must identify initial states and parameters of the plant, and synthesize control signals by taking account of the effect of the initial states of the plant. In this paper the authors propose a quick identification method which can identify initial states of the nonlinear plant as well as the time varying parameters, and propose a nonlinear robust controller, consisting of a simulator, that is, a physical model of the plant with the same structure as the object plant, with a computing predictor network which predicts dynamics of the plant from the information of internal states of the simulator. Validity of the proposed control system was verified by a simulation study in which an arbitrary control performance of the system for a nonlinear time-varying plant was verified.<<ETX>>