Path planning-by-learning with a Nomad 200 mobile robot

In this article, the authors face the problem of navigating a mobile robot on an indoor environment, where the location and shape of obstacles is assumed to be initially unknown to the robot. They describe an approach for simultaneous learning of a world model, and learning to navigate from a start position to a goal region on the world. These two learning abilities may be seen as cooperating and enhancing each other in order to improve the overall system performance. It is assumed that the robot knows its own current world position. It is only additionally assumed that the mobile robot is able to perform sensor-based obstacle detection (not avoidance), and that it is able to perform straight-line motions. Results of simulation experiments are presented that demonstrate the effectiveness of the approach to navigate a Nomad 200 mobile robot.