Application Of Ffnn For Path-finding In Game-map

Today, the game industry brought about the rapid development of artificial intelligence, game development, more and more of the game using artificial intelligence technology to improve game playing. This paper introduces the current situation and some important technical index of path-finding in game map, and also describes the advantage and shortage of the traditional path-finding algorithm; then introduces the application of neural network model for real-time path-finding in computer games. Finally, a hope is proposed about the future of path-finding in computer games.In this paper, the characteristics of the search path in the Common heuristic search algorithm and technology on the basis of a detailed analysis and study, the combination of practical application in the flash game development, for a number of A*algorithm optimization.This Paper put-forward neural network,it influxes the advantage of neural network,gather learning association and self-adaptation together. Tradition learning method has disadvantages such as long learning and slow convergences speed. This Paper uses improved genetic algorithm adjusting the weights of neural network and using these optimized parameters as the initial weight values of neural network. Aiming at the structure characters of neural controller,using genetic algorithm training the weights of neural network,the compound controller has good character. So the methods remains the global stochastically searching ability of genetic algorithm the robustness and self-leaning ability of neural network and the extensive mapping ability of neural network and rapid global convergences of genetic algorithm.The Environment Game system is made by author with Visual Basic 6.0.This system has good user interfaces, and especially the player can be Neural Network Model in this system.