A market based distributed cooperation mechanism in a multi-robot transportation problem

Abstract

An artificial capital market is employed to solve a multi-robot transportation problem. In this artificial market, each robot is considered as an investor agent where contributing in carrying the object is considered as asset. It should be decided which agents are going to do investment. A negotiation protocol is proposed to support the society to make decision by a distributed approach. The agents are self-interested trying to increase their own profit but acting in the framework of the negotiation protocol yields a decision that is the best for whole society. The best decision for the society corresponds to minimum fuel consumption. Simulation results showed that the convergence of the algorithm is very fast, the results are not dependent on starting point and the solution is close to optimal solution.