摘要
基于对五子棋棋局状态表示、棋局估值函数的设计、搜索算法中搜索深度的选择、后台时间的合理利用和开局残局库的使用等智能五子棋博弈的关键技术的分析,在Windows XP上用Visual C++6.0开发了智能五子棋博弈原型系统:用三维数组表示方法存储棋盘状态,用加权估值法评价状态优劣,用α-β剪枝搜索作为基本的搜索算法,并用启发式规则来选择合适的搜索深度.实现和分析结果表明,该系统与原来的五子棋程序相比,减少了对一些不能取得胜利的空间的搜索,提高了对弈速度.
Based on studying and discussing the key technologies of intelligent Gobang playgame,such as the methods of the game state data structure of Gobang,the design of evaluation function,selecting of search depth,usage of background time and the using methods of opening book and endgame database,the archetype system of intelligent Gobang was developed by Visual C＋ ＋6.0 in Windows XP,which included that the game state data structure of Gobang was showed by three-D array,that the state was evaluated by weighted estimating method,that α-β lopping search was used as basis searching algorithms and hewristic rule was used to select search depth.The implementation and result discussion indicate that compared with the other Gobang programs,the speed of our program is increased by reducing the searching space.
出处
《郑州轻工业学院学报（自然科学版）》
CAS
2010年第6期76-80,共5页
Journal of Zhengzhou University of Light Industry:Natural Science
基金
河南省科技攻关项目(0624220084)
郑州轻工业学院科研基金项目(2007XJJ13)
关键词
智能五子棋
博弈
估值
intelligent Gobang
playgame
evaluation function
