Design of a fuzzy control system for two wheeled inverted

In this paper, a control system for two wheeled inverted pendulum robot is designed to have more stable balancing capability than conventional control system. Fuzzy control system is selected for the two wheeled inverted pendulum mobile robot, and fuzzy membership function factors for the control system are obtained for specified 3 users' weights using trial-and-error method. Next a neural network is employed to generate fuzzy membership function factors for more stable control performance when the user's weight is arbitrarily selected. Through the simulation study, we find that the designed fuzzy controller using the neural network is better than the conventional fuzzy controller.