The Design and Stability Study of Double Inverted Pendulum Controller

Abstract

The double inverted pendulum system is hard to reach steady state, the mathematical model before the simulation and physical control experiment. And the system requires higher performance to the controller. So the improved genetic algorithm is introduced into the wavelet neural network controller. The trained neural network has less number of iterations, smaller error and better global convergence ability. The simulation and actual control experiment results show that, the wavelet neural network controller based on the improved genetic algorithm can achieve stable control of the double inverted pendulum system, It has excellent interference resistance ability and effect meet the requirement of structural performance.