The Design And Implement Of Chinese Chess Game

Despite the Gobang, chess, chess game program has made great success, but Chinese chess because of their late start, high complexity, there are still many technical problems which are not resolved. Based on the analysis and research of the existing Chinese chess computer game software, this paper tries to make some improvements. Because the game algorithms are generally based on α-β pruning algorithm, so we can use the same framework. However, we have increased the number of layers of search, and the use of internal iteration and optimization. Our method does not affect the search results, but more pruning, so the efficiency is obviously improved. In addition, this paper uses a variety of heuristic algorithms, such as replacement table heuristic, static evaluation heuristic, killer inspired, inspired by history, and so on. We define different priorities according to their importance. The aim is to further optimize the law in order to better optimize the α-β pruning strategy. Most of the existing game algorithms are used to adjust the evaluation function of neural network, this paper also uses the same, but joined the normalization, constraint method, hand manipulation and other methods to select the appropriate evaluation function. This is our technical point. We conducted experiments with the software of man-machine chess. Experimental results show that, but in many cases, the computer can overcome the high level chess player, but does not defeat the apex master and master chess player.