多agent系统的联盟框架及形成机制  

Architecture and Mechanism of Coalition Formation in Multi-Agent Systems

作　　者：叶斌[1] 马忠贵[1] 曾广平[2] 涂序彦[2] 

机构地区：[1]北京理工大学信息科学技术学院自动控制系 [2]北京科技大学信息工程学院,北京100083

出　　处：《北京理工大学学报》2005年第10期864-867,共4页Transactions of Beijing Institute of Technology

基　　金：国家自然科学基金资助项目(60375038)

摘　　要：基于集合分割、R ough集理论、大系统控制论和人工社会理论对多agen t系统协作的联盟框架及形成机制进行了研究,提出了2类6种联盟框架.在知识层次上定义了联盟精度等概念,并给出了联盟形成的算法描述,算法易于编程实现,并给出一个应用实例,有效解决了agen t的能力和任务所形成的知识集合边界不分明的问题.The architecture and mechanism of coalition formation that cooperate in multi-agent systems, based on set partitioning, rough sets, large systems cybernetics and artificial society theory are studied. Two kinds of coalition architecture and six kinds of coalition model are presented. Several concepts, such as the concept of degree of coalition accuracy, are presented from the point of view of the knowledge capability. The algorithm of coalition formation is described and analyzed. The algorithm is easy to implement. An application example is given. These work presented an effective approach to solve the vague boundary region problem of knowledge sets made up of the capability and task of agent.

关 键 词：分布式人工智能 MULTI—AGENT 联盟框架 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]