倒立振子のモデル化と安定化のためのニューロPD制御

抄録

In Japan, about 84 percent of real plants adopt PID controllers. However, when we use them, we are required much effort and time for tuning the PID gains. In this paper, we propose a method using a neural network stabilization of the inverted pendulum which is the single input and two outputs system. Here, we propose a neuro-PD control system to stabilize the inverted pendulum, which consists of two PD controllers and a neural network to tune the PD gains. Finally, the experimental results to stabilize the inverted pendulum are illustrated to show effectiveness of the present method.