Model Identification of Cart-plus-Pendulum System Using Genetic Algorithm

This paper proposes an identification of a cart-plus-pendulum (CPP) system model via the genetic algorithm (GA), one of the most efficient AI searching techniques to compare with the Box-Jenkins (BJ) model obtained from the conventional identification method based on the regression analysis. Under testing, the system was excited by an uniformly distributed random input. The pendulum angles, output of the system, were monitored by the encoder. From the results of model identification and model validation, it was found that the GA gives the model representing system dynamics superior to the BJ model. Details and results of identification and validation are discussed and shown in the paper