Pheromone-based ant colony clustering algorithm in multi-agent cooperation

Cooperation is a key issue of multi-agent system (MAS), pursuit problem as a test platform of multi-agent system and learning algorithm is widely used for testing the performance of cooperation, learning method and communication of MAS. To reduce the large computation of current pursuit methods such as the shortest distance first and contract net protocol, a method of applying pheromone-based ant colony clustering algorithm in multi-agent pursuit problem is proposed. Experiment results show that the proposed method can speed up the pursuit process compared with other algorithms, when increasing the number of predators the proposed method shows its superiority on efficiency and can be applied in multi-agent cooperation effectively.