Ferrianto Gozali, Daniel Adrian


This research aims to design and implement methods of color tracking on a chess robot to recognize chess pieces. Color tracking is used to obtain the position of chess pieces on a chess board placed distinguished by different colors depending on chess. The usual method for chess robot is to detect changes in the frame on the image and differentiate the type of chess based feature. In the method of detecting changes in the frame, the robot needs to know the previous frame so that the position of the pieces can not be in real-time. In the method to distinguish the type of pawn based feature errors may occur due to image features that are not detected. By using color tracking, the detection becomes more simple chess robot, real-time and a small error rate. There are three components of the system are the webcam as input, the computer as a process and a robot controller as output. Webcam function to get the image that will be used to process color tracking. This process is done through a program on a computer. Results of color tracking form 8x8 matrix that represents the position of the pieces on the chessboard. This matrix is then translated into FEN is one of the chess notation used to describe the chess board on the program. Program connected to a chess engine. Chess engine aims to calculate bestmove for computer. Communication between the chess engine and a program using the UCI protocol. After getting the results from the chess engine, program send command to the robot controller to move the chess pieces. Robot controller and program communicates via Bluetooth. The capability of color tracking method in recognizing the chess pieces without any error eventhough the lumen of the environments varies is also useful in other object recognition processes, especially in industrial application.

Keywords: camera, chess,chess board, color tracking, robot


  • There are currently no refbacks.