COORDINATION AND REORGANIZATION IN MULTI-AGENTS SYSTEMS, I
Alina Georgiana Baciu, Adina Nagy · 2003
Abstract. A method of considering coordination and reorganization as keys in achieving (organizational) multi-agent system adaptation in unknown situations is proposed. Within a not totally predictable environment multi-agent systems are prone to failures. In such unpredicted situations the system must be able to adapt in order to accomplish its purpose. The proposed system architecture is a combination of MOISE+ and MOCA concepts. The main inconvenient of MOCA platform is that the mechanism of dynamic role allocation is entirely left to the designer. The inconvenient of MOISE+ platform is that agents ’ behavior is not considered. MOCA gives a structural vision on multi-agents systems based on individual and collective patterns of behavior. MOISE+ model describes how agents endorse roles in order to achieve their individual and collective goals. In Part I the main concepts of MOCA and MOISE+ models are presented.