A Task Assignment Strategy for Multi-AUV Collaborative Hunting Problem

Abstract

As a typical problem in the research of underwater multi-agent cooperative technology, multi-AUV cooperative hunting is widely used in the military and other fields. In this paper, a hybrid task assignment strategy is proposed to solve the problem of target hunting in complex environments. Firstly, a contract network algorithm based on improved bidder selection and evaluation mechanism (BSECNP) is proposed to improve the efficiency of task assignment and system efficiency. Secondly, a centralized task redistribution mechanism is proposed to solve the problem of individual Autonomous Underwater Vehicles failure that may occur during the collaborative hunting task. Simulation results show that the task assignment strategy composed of the above two can effectively improve the efficiency of the multi-AUV system, and realizes the global coordination of tasks in the case of unforeseen circumstances.