Robot Control For Dynamic Environment Using Vision And Autocalibration

To enhance exibility and extend the area of applications for robotic systems, it is important that the systems are capable of handling uncertainties and respond to (random) human behaviour. A vision system must very often be able to work in a dynam-ical \noisy" world where the placement of objects can vary within certain restrictions. Furthermore it would be useful if the system is able to recover automatically after serious changes have been applied, for instance if the camera has been moved. In this paper an implementation of such a system is described. The system is a robot capable of playing checkers with a human opponent using a camera to determine where the board and pieces are located and when the human has made his move. The paper primarily focuses on the vision aspect of the implemented system.