Research On Coordination Of Multiple Autonomous Underwater Vehicles Based On Swarm Intelligence

The coordination mechanism of multiple autonomous underwater vehicles is an important part of AUV research. Formation control is a typical problem in the coordination of multiple autonomous underwater vehicles, with important research significance and practical value. For the relevant features of underwater acoustic communication, combined with mobile multi-robot coordination methods based on swarm intelligence, the formation control method of multiple autonomous underwater vehicles based on local sensing and limited communication is proposed on the application background of multiple autonomous underwater vehicles formation, in order to solve the underwater acoustic communication delay, communication failure, communication error and other issues.Firstly, the research situation of the coordination of multiple autonomous underwater vehicles and the application of multi-robot coordination based on swarm intelligence are introduced.Secondly, the single AUV architecture based on decision-making and action which can flexibly adapt to dynamically changing environment and the distributed architecture of multiple autonomous underwater vehicles are designed. The single AUV architecture consists of a sensing and communication module, a decision-making module and an action module.Thirdly, the formation control method of multiple autonomous underwater vehicles based on local sensing and limited communication is proposed. AUV achieves local sensing with sensors in order to complete the underlying collaboration such as the formation control between the AUVS. It depends on the limited range of communications between the AUVS for the high-level coordination strategies such as the role assignment, formation transformation, and information exchange which can not be achieved by local sensing. Two modes are executed in parallel, without any centralized controller to coordinate. On this basis, the solutions of communication delay, communication failure and communication error are further proposed, so that the neat formation within the overall scope can be formed by the AUV individual only using local interactions with each other.Finally, the simulation of multi-robot formation based on local sensing and limited communication is introduced, and the corresponding formation performance evaluation as well as the analysis of simulation results is proposed, which verifies the feasibility and effectiveness of the method.