Cooperation of autonomous underwater vehicle fleet based on MOOS-IvP

This paper describes an investigation into the cooperation of multiple autonomous underwater vehicle system( MAUVS). A distributed control architecture MOOS-IvP (which was proposed by Newman at Oxford and Benjamin at MIT) is applied. In this approach, fully distributed control of underwater vehicles is realized and behavior-based control strategy is used to ensure a timely reaction of the vehicles in complex and dynamic environment. In order to tackle the problem of task level cooperation of MAUVS in task execution, a market-based approach to coordination of MAUVS is proposed and software modules compatible with the MOOS-IvP architecture are devised. The modules, working in accordance with other public modules of MOOS-IvP, supports a variety of fleet behaviors useful for ocean exploration. It is shown in this paper that the integration of market based coordination strategy and MOOS-IvP architecture can enable the MAUVS to work efficiently in tasks require both loosely-coupled and tightly-coupled cooperations.