Research And Development On Control System And Decision-making System Of Robot Soccer

Robot soccer system is kind of typical multiple mobile robots rivalry system, with the characteristic of high real time request, high dynamic and existing unexpected states. It connects research of high-tech and soccer game, tests many kinds of AI theories, methods and technologies via soccer robot games. The robot soccer game has got attentions of researchers in AI and related fields and is becoming a standard problem.This paper pays more attention to the problems of research and developments of Control System and Decision-making System on Robot Soccer, including interrelated AI studies.First, this paper analyses and sums up the system and structure of present robot soccer, pointed out their strongpoint and insufficiency. We explore the resolution to this question.Second, the thesis studies the characteristics of centralizing control system and distributing control system. It pointed out the common insufficiency of some typical decision-making systems, then we brought out our-own centralizing control system which is introduced local control of distributing control.Third, how to enhance the real time ability of Multi-agent system, our advice is to simplify the motion function of soccer robot, we had made the feasibility analysis.Forth, the thesis pays attention to two kinds of coordination of multiple soccer robots. Stratagem coordination happens in the large area, with low real time request. We have researched the application of Q-study in MiroSot robot, it aims at the dynamic adjustment on the roles of all robots.Fifth, the thesis studies the sampling cycle of digital simulation system's influence to the collision models of soccer robots, analyzes the states in the collision, and brings out a new collision model; considering the regular figures of objects and real time request of the simulation system, the thesis designed a fast collision detecting method based on collision chain;This dissertation focus on MiroSot game organized by FIRA, presents the system architecture, introduces the research of control system and decision-making system. Studies the mathematic model and collision tests .Some results and conclusions of the paper have some significance to the design of control and strategy system of other intelligent robot system.