Stabilization of inverted pendulum by the genetic algorithm

Considers the stabilization of an inverted pendulum which can be controlled by moving a cart in an intelligent way. We adopt a PID (proportional + integral + derivative) control method to stabilize the pendulum. This controller requires the determination of PID control gains, but it is difficult to select the best gains theoretically. Thus, there have been many approaches to determine them empirically; most of these are based on the experience of operators and knowledge. We propose a method to use neural networks to tune the PID gains in the same way that human operators tune the gains adaptively according to the environmental condition and systems specification. The tuning method is based on the error backpropagation method, and hence it may be trapped in a local minimum. In order to avoid the local minimum problem, we use a genetic algorithm to find the initial values of the connection weights of the neural network and the initial values of the PID gains. Experimental results show the effectiveness of the approach.