A hybrid heuristic particle swarm optimization for coordinated multi-target assignment

Target assignment of coordinated distributed multi-agent system is an important yet difficult task. Previous methods (e.g., neural network, genetic algorithm, ant colony algorithm, particle swarm optimization and auction algorithm) used to address this problem have proved to be either too slow or not stable as far as converging to the global optimum is concerned. To address this problem, a new algorithm is proposed which combines heuristic particle swarm optimization and decentralized cooperative auction. Based on the particle swarm optimization, the decentralized cooperative auction is used to construct particles' original solutions which replaced previous random generation solutions, and then the original solutions are improved by the heuristic approach to increase the stability of system. Simulation experiment results show our method can converge to the global optimum more stably and faster by comparing with the original methods.