Collaborative Robot System for Playing Chess

Abstract

In recent years, number of collaborative robots industrial applications has made a significant increasment. Implementation of collaborative robots is a safe and effective way for designing robot-human cooperation systems. Combined with constantly developing artificial intelligence, collaborative systems are actually able to solve complex problems that require some sort of intelligence. For humans, board games are a good example of the visualization of robot intelligence. Such systems require estimation and detection of board and pieces in manipulator workspace, some kind of decision-making algorithms and robot control system to move pieces. The flagship of such systems are chess playing robots. The chess game has a defined and easy to understand set of rules which makes it interesting example of intelligent robotics systems application. In this paper, we present an implementation of collaborative robots for chess playing system which was designed to play against human or another robot. The system is able to track state of the game via camera, calculate the optimal move using implemented decision-making algorithm, detect illegal moves and execute pick-and-place task to physically move pieces. We test the developed system in a real-world setup and provide experimental results documenting the performance of proposed approach.