Decentralized decision making approaches for Dynamic Coalition of Resource-Bounded Agents
Résumé
Purpose - The purpose of this paper is to extend the existing approaches of coalition formation to how to adapt dynamically the size of the coalition according to the complexity of the task to be accomplished. Design/methodology/approach - Aconsiderable amount of attention has been paid to the coalition formation problem to deal efficiently with tasks needing more than one agent (i.e. robot). However, little attention has been paid to the problem of monitoring a coalition during the execution by modifying it according to the progress of the accomplishment of the task. In this paper, the authors consider a coalition of resource-bounded autonomous agents with anytime behavior solving a common complex task. There is no central control component. Agents can observe the effect of the other agents' actions. They can decide whether they should continue to contribute in solving the common task or to stop their contribution and to leave the coalition. This decision is made in a distributed way. The objective is to avoid the waste of resources and time by using the same coalition along the task accomplishment while some agents become unnecessary to pursue the accomplishment of the task. The authors formalize this decentralized decision-making problem as a decentralized Markov decision process (DEC-MDP). Findings - The paper results in a framework leading to Coal-DEC-MDP, which allows each agent to decide whether to stay in the coalition or leave it by estimating the progress on the task accomplishment. Research limitations/implications - The approach could be extended to deal with more than one coalition. Practical implications - Decentralized control of a fleet of robots accomplishing a mission. Originality/value - The paper deals with a new problem of adapting dynamically the coalition to the target task and the use of DEC-MDPs.
Origine : Fichiers produits par l'(les) auteur(s)