Searching Pareto-optimal solutions for the problem of forming and restructuring coalitions in multi-agents systems

Coordination is one of the fundamental research issues in distributed artificial intelligence and multi-agent systems. Current multi-agent coalition formation methods present two limits: First, computation must be completely restarted when a change occurs. Second, utility functions of the agents are either global or aggregated. We present a new algorithm to cope with these limits. The first part of this paper presents a coalition formation method for multi-agent systems which finds a Pareto optimal solution without aggregating the preferences of the agents. This protocol is adapted to problems requiring coordination by coalition formation, where it is undesirable, or not possible, to aggregate the preferences of the agents. The second part of this paper proposes an extension of this method enabling dynamic restructuring of coalitions when changes occur in the system.