Mobile robot control using XCS

In the field of robot control and navigation, alternative methods, such as various techniques from artificial intelligence are becoming widely adopted in these days. In this paper we present a middle-size omni-directional robotic vehicle with Mecanum wheels and its simulated counterpart trained to move along the walls and avoid obstacles on the basis of learning classifier systems, namely using the accuracy-based classifier system (XCS). The XCS operates in real time and produces highly compact solutions in very short time, therefore we regard it as a good solution for building a control architecture for a mobile robot.