Stability analysis, synthesis and optimization of radial-basis-function neural-network based controller for nonlinear systems

This paper presents the stability analysis, synthesis, and performance optimization of a radial-basis-function neural-network based control system. Global stability conditions will be derived in terms of matrix measure. Based on the derived stability conditions, connection weights of the radial-basis-function neural-network based controller can be optimized by genetic algorithm (GA) subject to the system stability. Furthermore, the system performance will also be optimized by the GA. An application example on stabilizing an inverted pendulum will be given to illustrate the design procedure and merits of the proposed approach.