| Bookmark: |
Full-text
| URL: http://arti.vub.ac.be/~edwin/publication...MAcoordination.ps.gz |
| Cached: PDF-186K PS-4540K PS.gz-111K |
| SAVE AS an easy-to-recall long filename: |
| Filename format: author--year--title PDF-186K PS-111K :: About GZip'd PS |
| Filename format: author--year--title--journal|proceedings|...--pages PDF-186K PS-111K |
Related links
| CiteSeer: http://citeseer.ist.psu.edu/37852.html |
| Web search: Google Web Search :: Google Scholar |
| Within this site: Cited by (2) References (15) |
Paper at a Glance
BibTexMultiAgent Coordination by Communication of Evaluations Edwin de Jong Artificial Intelligence Laboratory Vrije Universiteit Brussel Pleinlaan 2, B 1050 Brussels, Belgium email: edwin@arti.vub.ac.be tel: +32 2 629 37 00 fax: +32 2 629 37 29Abstract. A framework for coordination in multiagent systems is intro duced. The main idea of our framework is that an agent with knowledge about the desired behavior in a certain domain will direct other, domain independent agents by means of signals which reflect its evaluation of the coordination between its own actions and their actions. Mechanisms for coordination are required to enable construction of open multiagent sys tems. The goal of this investigation was to test the feasibility of guiding an agent with coordination evaluation signals, and furthermore to gather experience with instantiating the framework on a testbed domain, the Pursuit Problem. In the testbed system, agents have been created which choose their actions by maximizing the coordination evaluation signals they will receive. The performance of these agents turned out to rank among the best results encountered in literature, and behavior guided by coordination evaluation signals can thus be concluded to be useful in this domain.
1 Introduction Without a proper mechanism that guides interactions, the mere result of gath ering many agents into a multiagent system is chaos. Several such mechanisms exist. A short account of these is given in the next section. Most of these mechanisms assume that the agents involved have direct access to the system in which they are incorporated. This implies that only agents that were designed specifically for the application at hand can be put to use. We want to investigate whether this restriction can be overcome. To this end, a coordination mechanism for multiagent systems has been defined that allows domainindependent agents to behave usefully in an unknown environment. To learn how to do this, they are supplied ...
@inproceedings{dejong97multiAgent,
author={E. D. de Jong},
title={Multi-agent coordination by communication of evaluations},
year={1997},
address={Berlin},
editor={Boman, M. and Van de Velde, W.},
publisher={Springer-Verlag},
booktitle={Proceedings of the Eighth European Wrokshop on Modelling Autonomous Agents in a Multi-Agnet World},
url={http://www.isrl.uiuc.edu/~amag/langev/paper/dejong97multiAgent.html}
}
| HOME :: Conference List :: Conference Paper | Comments to: junwang4 you-know-at gmail.com | Last update: 9/15/08 |