Research On Robot Soccer Strategy Based On Fuzzy Technology

Robot soccer is robotics research and competition project which developed in recent years. It is no doubt that soccer robot is a new interdisciplinary research area, referring to diverse fields like robotics, artificial intelligence, intelligent control, real-time vision, wireless communication, integration of machinery and so on. Robot soccer system is a typical multi-agent system and multi-robot cooperative autonomous system. Furthermore, it has provided a standard experimental platform for multi-agent system fundamental research and model test. Robot system can be divided into four subsystems: car subsystem, vision subsystem, strategy subsystem and communication subsystem. Strategy subsystem is equal to the trainer of modern real soccer competition, which affects success or failure of competition directly. It can manifest intelligence of robot soccer system. It has significant theory significance and reality application value to study strategy system. This paper analyzed the existing strategy system to use hands over layered lamination model framework, then pointed out that strategy system established a decision-making model with difficulty because of many influence factors, big decision-making information, and complex decision-making quantities and so on. The method to describe the soccer robot position and posture as well as field situation with fuzzy set, to use fuzzy logic and fuzzy decision-making method to carry on modeling was proposed. The paper used four layered decision-making inference model to take strategy framework, and used fuzzy technology to analyze strategy system situation, formation choice, role assignment, action choice. The paper constructed robot soccer strategy system based on classical fuzzy set, had realized the following separately. Situation analysis: using fuzzy set to structure reflection attack and defense situation, then using fuzzy logic to chose attack or defense strategy; Formation choice: using the result of situation analysis to chose formation; Role assignment: because of importance of role assigns in entire strategy system, proposing separately two fuzzy decision-makings methods of fuzzy consistent relation (FCR) and fuzzy analytic hierarchy process (FAHP) to carry on role assignment. Finally the paper used simulation platform to prove algorithm and strategy. Robot soccer strategy system using vague set and its logic was proposed with analyzing fuzzy set and fuzzy logic in carrying on situation analysis. According to result of situation analysis, fuzzy decision-making algorithm for formation choice and role assignment was proposed.