System Design And Research On Key Algorithms Of The Chess Game Robot

Computer game has gradually become a very popular field direction.Many research institutions at home and abroad have developed and manufactured game robots with human-machine game functions,but there are still some deficiencies: Most robotic motion systems are industrial robot arms,it is difficult to popularize due to size,cost,power consumption,etc.Special chessboards or high-definition cameras are used in vision system.Recognition algorithm has high requirements on the environment and is not universal.Strategy system is mostly designed for one kind of chess.The single goal is to pursue the winning rate.And the man-machine game experience is poor.The traditional game algorithms are difficult to meet the requirements of multiple chess games.This article focuses on the perception system,decision system and control system of the game robot,designs a chess game system with multi-game capability,and develops the robot comprehensive experiment platform.Main tasks and research contents of this thesis are as follows:(1)The robot game system is studied and designed.The common methods in the field of computer games are studied.The overall structure of the robot game system is designed.An intelligent algorithm is designed for the strategy generation function,focusing on the design of strategy-value network,loss function and input feature representation.The neural network is trained from scratch through self-playing and finally learned the rules and skills of the game.The limitations of the algorithm and room for improvement are analyzed.(2)The robot vision system is designed.The overall structure of the game robot vision system is designed.According to the different functions of each module in the chessboard information acquisition process,the camera collection module,the information processing module and the classification recognition module are designed.For the recognition tasks of different chess piece types,two recognition methods have been proposed.Finally,the effectiveness of proposed method is verified by experiments.(3)The chess gaming robot platform is developed.After analyzing the functional requirements of each module of the chess robot system,the overall design structure of the game robot is proposed.By selection of key hardware such as the core computing module,power module,camera and motion motor of the robot,an overall hardware structure platform is built.The comprehensive development of robot hardware and software is completed.The development of chess game robot system platform was completed.And human-machine game experiments were conducted to verify the feasibility of the system.This system realizes the function of playing games with human players,and has practical significance for improving the intelligence of chess game robots and improving the experience of man-machine games.