Experimental implementation and validation of dual adaptive control for mobile robots

Abstract

This paper presents experimental results which validate the use of a novel dual adaptive controller for mobile robots operating in the presence of dynamic uncertainty. The control scheme, recently proposed by the same authors, has so far been tested by simulations only. The presented results show, for the first time, the successful application of neural network dual adaptive control in a practical mobile robot scenario. In contrast to other adaptive controllers hitherto proposed for mobile robots, the dual adaptive approach employed in this scheme does not treat estimation and control as two separate tasks, but aims to strike a balance between the two at all times. This improves the overall performance. The implementation details of the robot designed for the purpose of this research are also presented in this paper.