Optimal control for partially observed nonlinear deterministic systems with fuzzy parameters (Q1863839)

Nonlinear control systems described by nonlinear Takagi-Sugeno fuzzy-models are considered. The membership functions for modeling errors are assumed to be Gaussian with known centers and deviation parameters. This results in nonlinear differential equations for the system state and for the output process. The controller driven by the output process is described by linear fuzzy differential equations with unknown parameters of the Gaussian membership functions and unknown controller parameters. The unknown parameters are to be determined such, that a given quadratic performance criterion is minimized. Necessary conditions of optimality are derived by a variational approach. Based on these conditions an algorithm for calculating the parameters is described. The proposed method is illustrated by two numerical examples.