Ant Colony Parallel Algorithm Based Multi-Robots Cooperation And Coordination Research

Multi-robot cooperation and coordination is of great academic and applied significance. With the development of Distributed Artificial Intelligence theories and applications, Multi-agent system theory has been used in the area of multi-robot system cooperation and coordination for its flexibility and applicability, and this has become the robots research focus. While Multi-agent cooperation and coordination technology is not mature, and the research work is not enough far away in the aspects of mechanism, such as the time, the course, the mechanism and the agreement of cooperation and coordination. So we need to thoroughly study about multi-agent cooperation and coordination formers and methods.Multi-robot system coalition composing is the importance way. In recent years coalition mechanism has been the important aspect,and it is interested in more and more. In MAS that agents compose coalition not only can solve the works that the single robot can not finish but also can adjust the goals, resolve the conflict, share the resources and construct optimizing means together. So all the MAS system can solve some task and get the most profit in the optimal deploying and the tiptop efficiency. Agent coalition composing is to compose in the optimal combination and getting the most profit targeting to the work. For every Agent can coordinate with others possibly in MAS, we must take into account most coalition combination in order to get the satisfactory result. We can find that multi-agent coalition is a complicated combination optimizing problem.Ant Colony Algorithms (ACO) is a kind of intelligent optimizing algorithms .It has achieved a very good effect to take for the combination optimizing problem, and it has showed the advantage to solve the complicated combination optimizing problem. ACO has latent parallelism. It shows off every intelligent ant's self-organization.This paper adopts Ant Colony Algorithm and its parallel running method and self-organization combining multi-robots coordination mechanism, to generate multi-robot dynamic optima coalitions. This method is designed for avoiding the coalition locking and resource wasting, in order to work's efficacies and to reduce the searching time and completing works effectively. The result and analysis of simulation show this way is effective and feasibility.