Trajectory tracking control of a wheeled mobile robot using an ADALINE neural network

Abstract

This paper presents the design of an adaptive neural kinematic controller, to guide a nonholonomic wheeled mobile robot throughout a predefined path. The ADAptive LINEar (ADALINE) neural network has been used to implement the controller. The Widrow-Hoff algorithm has been used to train the adaptive neural network, first offline and then online, in order to calculate the required control signals to guide the mobile robot over the predefined path. The simulation and experimental results show that the presented controller can be used effectively to guide the robot along the desired path.