基于多Agent的交互式动态影响图研究、应用与展望  

Progress for Application and Preparation of Interactive Dynamic Influence Diagrams

作　　者：罗键[1] 李波[1] 潘颖慧[1] 尹华一[1] 吴长庆[1] 

机构地区：[1]厦门大学信息科学与技术学院,福建厦门361005

出　　处：《厦门大学学报（自然科学版）》2011年第2期253-260,共8页Journal of Xiamen University：Natural Science

基　　金：国家自然科学基金项目(60975052)

摘　　要：交互式动态影响图(I-DIDs)是不确定环境下多Agent序贯决策的图模型.该模型突破传统基于公共知识的纳什均衡点的假设,使得该模型具备解决大规模动态决策问题的能力,具有更为广泛的应用前景.在总结、分析动态影响图研究成果的基础上,针对多Agent的I-DIDs状态空间随时间片的增长呈指数级增长问题,提出了基于行为等价和动作等价的两种新型算法并进行算法分析;此外,探讨应用I-DIDs模型解决自动小车存取系统中的路径规划、避免环路死锁问题,扩展了I-DIDs的应用范围.Interactive dynamic influence diagrams（I-DIDs） are graphic models that offer a clear representation for multi-agent decision-making under uncertainty.The solution of I-DIDs is different from the traditional multi-agent solution based on game theory,it breaks through the assumption of traditional Nash equilibrium based on public knowledge,which makes I-DIDs with wide range applications.First of all,this paper discusses the research progress of DID;And then it propose two new algorithms based on behavioral equivalence and actions equivalence to solve the problems of that the state spaces of I-DIDs growth over time exponentially;Moreover,this paper attempts to use the I-DIDs model to solve the problems of path planning and loop-deadlock controlling in automatic vehicle storage and retrieval system,it extends the application scope of I-DIDs.

关 键 词：交互式动态影响图 多AGENT 动态决策 自动小车 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]