Adaptive coordination for object-sorting tasks in multi-agent systems

This paper provides a method of adaptive coordination for multi-agent systems to solve object-sorting tasks (OST). It coordinates agents to move objects to destinations efficiently and effectively. Adaptive coordination is achieved by looking up a relationship table from which a relative best coordination algorithm can be found to perform the object-sorting tasks. An adaptive resonance theory network is utilized to cluster the object distribution so as to help construct the relationship table. By using adaptive rules for performing OST, experimental results showed that agents work more efficiently than single coordination solutions. An analysis of object distributions of OST provides a concrete result to make agents perform OST with the best performance