Motion Control of an Inverted Pendulum Using Neural Networks based on optimal Tracking Control

This paper deals with motion control of an inverted pendulum using neural networks (NN)based on optimal tracking control. First, we examined the control performance in the nonlinear region, applying the optimal tracking control theory to the inverted pendulum. Second, the control system in which the optical tracking control system was replaced with two NNs and a feedback (FB)was designed. In the nonlinear region, the NN control performance was compared with optimal tracking control from the points of view of simulation and experiment. As a result, it was shown that NN control is valid for nonlinear problems in the case where the pendulum angle is large and the motor is saturated.