Control of mobile robots using RBF network

Abstract

When designing a control for a nonholonomic mobile robot, the following aspects should be considered: there exists disturbances, nonlinear unknown dynamics of the system, and some parameter uncertainties. Especially, when the mobile robot is used to assist a human in daily life, the mass of the mobile robot changes always. Thus, in that case, there should be a control scheme that can compensate the unknown modeling uncertainty and disturbances. If only conventional kinematic controller is used, then it fails to produce a desired tracking performance. There are several methods to resolve this problem such as sliding mode control and adaptive control. Since it is generally known that neural networks can approximate the given unknown function, it has been widely applied to the control of general robot system. However, since there exist nonholonomic constraint in the mobile robot system, conventional neural network-based controller cannot be applied. In this paper, the RBF network controller is proposed. The proposed RBF network controller learns the unknown dynamics and compensate for it. The stability of the proposed controller is proven by using Lyapunov function. To show the effectiveness of the proposed controller, several simulation results are presented. Through the simulations, it is shown the proposed controller can overcome the modeling uncertainty and the disturbances. Also the proposed RBF controller outperforms the previous works from the viewpoint of computation time, which is a crucial fact for a real-applications.