| 140 | 6 | 26 |
| 下载次数 | 被引频次 | 阅读次数 |
对抗性多机器人系统是 1个典型的多智能体系统 (MAS)。在 MAS中 ,建立对手模型 ,并进行意图和规划识别是对抗活动的基本前提。BDI是描述思维的有效手段 ,可用来建立通用的对手思维状态模型。本文在 BDI模型的基础上结合换位原理 ,提出了 1种对抗性多机器人系统中的对手建模模型 ,并将此方法应用到足球机器人系统 ,获得了较好的效果。
Abstract：An adversarial multi-robot system is a typical multi-agent system (MAS). Developing an opponent model and recognizing opponent's intentions and plans are prerequisite to an MAS. BDI (Belief, Desire, Intention) is an effective means to describe a mental state, and it can be used to develop a generic opponent mental state model. In this paper, based on the model of BDI and combined with the transposition principle, a new method of modeling opponents in an adversarial multi-robot system is proposed. The method has been applied to a soccer robot system and the experimental results show that it can well enhance the system's performance and intelligence degree.
[1] Findler,Elder.Multi- agent coordination and cooperation in a dynamic environm ent with limited resources[J].ArtificialIntelligence inEngineering,1995,9:229-238.
[2] 马光伟,徐晋晖.Agent思维状态模型[J].软件学报,1999,10(4):342-347.
[3] MilindTambe.AgentArchitectures forFlexible,PracticalTeamwork[C].Proceedings of theFourteenthNa-tionalConference onArtificialIntelligence,Providence:AAAIPress,1997.22-28.
[4] 李 毅,石纯一.对群体Agent的意图跟踪[J].软件学报,2002,13(7):1298-1302.
[5] 李 毅,石纯一.基于BDI的足球 agent和TH- soccer平台[J].计算机研究与发展,2001.38(8):911-915.
[6] 周昌乐.认知逻辑导论[M].北京:清华大学出版社,2001.144-152.
[7] ZhuangXiaodong,MengQingchun.MobileRobotControl inMulti-ObstacleEnvironmentBased onReinforce-ment antLearning[C].Proceedings of the2002InternationalConference onControl andAutomation.Xiamen:IEEE Press,2002.856-860.
[8] ZhuangXiaodong,MengQingchun,WeiTianbin.Robot path planning in dynam ic environment based on rein-forcementlearning[J].Journal ofHarbinInstitute ofTechnology(Chinese),2001,8(3):253-255.
基本信息:
DOI：10.16441/j.cnki.hdxb.2004.01.019
中图分类号:TP242
引用信息:
[1]顿文力,孟庆春,宋长虹,张艳.基于换位原理的对手建模模型[J].中国海洋大学学报(自然科学版),2004(01):109-114.DOI:10.16441/j.cnki.hdxb.2004.01.019.
基金信息:
山东省自然科学基金 ( Y2 0 0 2 G18)资助
2004-01-30
2004-01-30