"基于深度搜索算法的中国象棋AI引擎设计与实现"

stract Computer games are one of the most challenging research topics in the field of artificial intelligence. Compared to international chess, Chinese chess has a long history, a higher level of game difficulty, a larger board size, more special moves, and more complex variations. Additionally, Chinese chess is a complete knowledge game. A complete Chinese chess engine generally consists of several components: board data structure, position evaluation function, search methods, and opening and endgame libraries. This paper designs and implements a relatively complete Chinese chess game engine, and introduces the basic modules and search methods that a Chinese chess engine should have. The algorithm principles are analyzed in detail, and the improvement points and deficiencies of each module are analyzed through experimental data. The search method in this paper is a combination of various pruning algorithms based on game tree search, with the Alpha-Beta algorithm as the core. It mainly uses the Alpha-Beta algorithm, principal variation search, iterative deepening, and null move pruning to enhance the engine's search capability. According to the experimental results, the engine designed in this paper can solve all first-class endgames and some second-class endgames, and can play against humans and machines. This research mainly focuses on the search method of Chinese chess, while other parts such as evaluation functions are somewhat lacking and need to be improved in the future. Keywords: computer games, Chinese chess, game tree search, Alpha-Beta algorithm, null move pruning