Adaptive Smart Neural Network Tracking Control of Wheeled Mobile Robots

Abstract

Adaptive smart neural network controller design is presented in this paper for wheeled mobile robots with unknown dynamics. The controller is constructed at the dynamical level. The smart neural control scheme is designed such that the current control action not only can utilize the knowledge that neural networks learned from the past experience, but also keep the learning ability in the operational phase and finish the same control task in a 'smarter' way. The proposed neural control scheme can act smartly in the operational phase after the networks have been well trained in the training phase, in a way similar to the control process of human in learning to accomplish some complicated control tasks. All the system states are shown to be able to track the desired trajectory. Numerical simulation is conducted to verify the effectiveness of the proposed method