Abstract: The traditional influence map either lacks the representation of dynamic information or is inaccurate in the representation, which will easily lead to the wrong decision of the game artificial intelligence subject. In order to solve the problem that influence map was difficult to describe dynamic information, the propagation mode and attenuation mode of influence map were studied, and a dynamic influence map model based on distance adjustment was proposed. According to the movement trend of the objects affected, the distance to be calculated in the process of impact propagation was adjusted The model can encode dynamic information into the influence map so as to support decision making of the game agent. Experiments show that compared with traditional influence map, this model can effectively improve the accuracy of the dynamic information representation in influence map, thus enhance the performance of the game agent.

Key words: influence map, game artificial intelligence, dynamic information