Q学习角色值法在机器人足球比赛中的应用.PDFVIP

第 36 卷 第 4 期 电 子 科 技 大 学 学 报 Vol.36 No.4 2007 年 8 月 Journal of University of Electronic Science and Technology of China Aug. 2007 Q学习角色值法在机器人足球比赛中的应用 向中凡 (西华大学机械工程与自动化学院 成都 610039) 【摘要】提出了基于Q学习的角色值方法，避免了在比赛中由于机器人之间的频繁角色转换而造成的系统效率损失及系 统不稳定。该方法完善了多智能体系统的整体调整方法，有效地解决了在实际系统设计和实现过程中遇到的问题。经FIRA仿 真比赛检验，该方法是有效的，降低了机器人丢球、漏球、不作为的可能性，弥补了按区域分配固定角色的不足，有较好的 实用性。 关 键 词 多智能体系统; 强化学习; 机器人; 角色值 中图分类号 TP24 文献标识码 B Application of Role Value to Robot Soccer Based on Q-Learning XIANG Zhong-fan (School of Mechanical Engineering and Automation, Xihua University Chengdu 610039) Abstract Multi-Agent System (MAS) designing has faced some challenging work such as cooperation among agents which are vital to the performance of this system. A much advanced agent role-value method based upon Q-learning is proposed in this paper to avoid the unstabilizing factors and the loss of efficiency caused by possibility of too frequent role switching between robots. Other new methods based on this role model are suggested to solve the problems associated with system designing and implementation. Application to Federation of International Robot-Soccer Association (FIRA) simulation system proves that this method is effective, and reduces the possibility that the robots loss ball, fumble ball and nonfeasance, and remedies the shortage that roles are assigned according to fixed regions. Key words multi-agent system; reinforcement learning; robot; role value Multi-Agent System(MAS) 是由自主的Agent 通 避障；而且各个机器人的“体力”消耗不同( 如 过协作完成某些任务或达到某些目标的计算系统，