Research And Implementation Of The System Of Connect6 Game

Computer game is an important research topic in Artificial Intelligence, it plays a significant role in the development of Artificial Intelligence. Chess Computer Game, chess, Chinese chess, backgammon have better solutions. Connect6 invented by Professor Wu in Taiwan and improved from the backgammon, are simple game. There are two plays, we called Black and White. Black plays first and puts only one black stone on one unoccupied intersection. Subsequently, Black and White alternately put two of their own stones on two unoccupied grids. Complexity has been assessed as second only to Go and Shogi, is much higher than backgammon and chess, and chess comparable or slightly higher. Fairness, because the parties after each hand is finished, one more disk than the other , therefore, the game can reach a balanced state of nature, which makes equity greatly improve.Connect6 appears, caused widespread concern of researchers, is a good test environment for the development of Artificial Intelligence, how to improve the process of thinking depth is a major problem in Artificial Intelligence. At the same time, developed a considerable level of human players in Connect6 will also help the understanding of human cognitive abilities. Therefore, a computer chess game six sub-study has important theoretical and practical value.This study has a brief introduction to current research on the status of the computer game and Connect6, describes the improvement of evaluation function and parameters in detail. The main method of improvements in evaluation function is space valuation method and chess-shaped method of valuation. Space valuation method is assessed using the current vacancy, chess-shaped method of valuation is assessed using chess shaped vector method. In assessing the function of the parameters of the major improvements is the application of particle swarm optimization algorithm, using the method of the system to improve chess-type parameters, and with the improvement of the system before the game. Specifically, we have completed the major work and innovation, including the following:First, use the revaluation method in the system, which introduces the situation factor, as the situation factor changes, can reflect the different needs of the situation on the chessboard.Second, use particle swarm optimization algorithm to optimize parameter.Third, simulation of human thinking proposed method of space valuation. Human chess players in games, will think every empty position on the possible impact of the current situation. Inhibition of the other premise situation, the player will choose the most advantageous position expanded store-opening. In this paper, the method of valuation is to achieve this idea, through the experience of a number of chess masters, set the valuation of five vacancies, the method comparison experiment has been the ideal result.Fourth, the valuation functions are generally used on isolated pieces of assessment, but the pieces and pieces in different locations will be different impacts, the proposed method of valuation chess form. The method used the chess-type and the positional relationship between the pieces. Chess shape and space vector combination of valuation methods, applied to the system.