Design of Parallel Fuzzy Neural Network Controller for a Invert-Pendulum

This paper reviews the current fuzzy control technology from the engineering point of view, and presents a new method for using parallel fuzzy neural network for invert-pendulum. The neural network is designed to be a treble forward direction network structure with two inputs and three outputs. Considering the reiteration of neural network training, a fuzzy controller is added which is used to compensate the output of the neural network controller. It stabilizes the output variable at some minimum value. Then the weights of neural network are kept at a stable value and the whole controller retains in stabilization. Simulation results show that the fuzzy neural network controller with a simple and intuitive structure can keep an invert-pendulum system completely in stabilization.