【作者】 李春光；

【导师】 刘国栋；

【作者基本信息】 江南大学 ， 控制理论与控制工程， 2004， 硕士

【摘要】 随着控制论及计算机技术的发展，分布式人工智能中多智能体系统的理论及相关的应用研究已成人工智能和智能控制研究的热点。RoboCup（机器人世界杯足球赛）是一项旨在提高相关领域的教育和研究水平而举行的国际性的大型比赛和学术活动。通过提供一个标准任务来促进分布式人工智能、智能控制及机器人技术等相关领域的研究与发展。RoboCup的研究目标是：计划经过五十年左右的研究，使机器人足球队能战胜人类足球冠军队。本文作者借鉴近年来国内外RoboCup仿真球队的经验和教训，采用层结构、WIN2000平台及VC6.0开发环境，设计了江南大学RoboCup AFU2003仿真机器人足球队（阿福2003），并在以下几个方面进行了较为深入的研究工作：1.深入研究了RoboCup仿真比赛的SoccerServer模型，这是设计RoboCup仿真球队的基础；2.完整设计了带球模块、传球模块、点球模块、动态阵形模块、防守模块、在线教练及球员异构模块等；3.设计了基于Levenberg-Marquardt BP算法的神经网络训练器；4.通过研究Markov决策过程与再励学习算法，设计了基于Q学习方法的射门模块。更多还原

【Abstract】 With the development of cybernetics and computer technology , the theories of multi-agent system and relevant application study in distributed artificial intelligence already become the study focus of artificial intelligence and intellectual control.The Robot World Cup (RoboCup) Initiative is an attempt to foster artificial intelligence (AI) and intelligent robotics research by providing a standard problem where a wide range of technologies can be integrated and examined. RoboCup’s ultimate long-term goal is stated as follows:By mid-21st century, a team of fully autonomous humanoid robot soccer players shall win a soccer game, complying with the official rules of the FIFA, against the winner of the most recent world cup for human players.By learning the experience and lessons of other robotic soccer simulation teams, adopting a layer structure, Windows 2000 platform, and VC 6.0 development environment, we have designed AFU2003 robotic soccer simulation team of Southern Yangtze University (SYTU), and carried on relatively deep research work in following: 1.Further study SoccerServer model of the robotic soccer simulation tournament, this is the foundation of designing a robotic soccer simulation team； 2.Have designed dribbling module, pass module, penalty kick module, defence module,flexible team strategy module,online coach and heterogeneous players module,etc； 3.Have designed the neural network training tool based on Levenberg-Marquardt BP algorithm； 4.Through studying Markov decision-making process and reinforcement learning algorithm, have designed the shoot module based on Q learning method.更多还原