Empirical study of coordination strategies in cooperative multi-agentsystems

• Main Author: Arias Fernandez, jose
• Format: Others
• Language: English
• Published: Uppsala universitet, Institutionen för informationsteknologi 2014
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-232492

The growing importance of the Internet Of Things in our society leads to the necessity of developing systems where each thing reacts to the environment, taking its own decisions. Agent based systems fit perfectly in resolving those challenges. The aim of this thesis is to study the communication in multi-agent systems through coordination and negotiation techniques, applying them in a practical case and analyzing the results given after the execution. The implementation of the study-case has been developed using the Gaia methodology  since it is easy, but complex enough to develop a small multi-agent system from an initial set of requirements. The strategies analyzed are the Global Planning, Structural Organizations, Contractual Networks, Social Election, Coalitions and Negotiation based on Game Theory. Analysis of the results leads to the conclusion that no strategy is the best in all cases. Each strategy has different strengths, such as efficiency, scalability and robustness. Besides, negotiation techniques yield to poor results when all the agents try to achieve a common goal.