Using Meta-Agents for Multi-Agents in Networks
2007 (English)In: The 2007 International Conference on Artificial Intelligence: The 2007 World Congress in Computer Science, Artificial Intelligence / [ed] H. Arabnia et al, 2007, 561-567 p.Conference paper (Refereed)
In this paper, we propose an approach usingmeta-agents for monitoring multi-agents and controllingtheir behaviour while moving between states in a network.The meta-agents are built on the agents in the system andthen used to inspect the agents’ behaviour when reachinga result as well as perceiving the reason for that result.Each meta-agent can comprise everything from one agentto dozen or more agents depending on the task assigned tothe system. The benefit of using meta-agents for a multiagentsystem is the ability to provide the fastest waybetween nodes under given circumstances and to handle avast number of nodes in graphs and networks such astravelling salesman problem (TSP).
Place, publisher, year, edition, pages
2007. 561-567 p.
Intelligent Agents, Multi-agent Systems, Meta-agents, Graphs, Networks
IdentifiersURN: urn:nbn:se:kth:diva-44435OAI: oai:DiVA.org:kth-44435DiVA: diva2:450460
QC 201608312011-10-202011-10-202016-08-31Bibliographically approved