Cooperative game method of heterogeneous unmanned surface vehicles based on distributed decision-making framework

Abstract

Due to targeted design, unmanned surface vehicles (USVs) with different functions often show significant heterogeneous characteristics. This heterogeneity allows the USV cluster to perform more complex and diverse tasks. This paper innovatively proposes a cooperative game method for a heterogeneous USV cluster based on a distributed decision-making framework, which aims to significantly improve its decision-making quality. Specifically, we first combine the underactuated characteristics of the USV and the actual mission requirements and use the geometric method to accurately calculate the position of the guidance point. On this basis, we introduce the concept of ’neighbor USV’, and construct a distributed decision-making framework based on neighbor USV. The framework further enhances the coordination ability between USVs through a hierarchical decision-making mechanism. In the hierarchical decision-making system, the upper decision-making is mainly responsible for solving the sub-task allocation problem of each USV. The lower decision-making focuses on the specific action strategy of each USV. In order to verify the feasibility and superiority of the proposed method, we carried out Monte Carlo simulation experiments. The experimental results show that this method significantly improves the coordination efficiency and task execution ability of the USV cluster while ensuring real-time decision-making. Through the research of this paper, we provide a new idea for the cooperative control of heterogeneous USV clusters and provide a reference for the broad application of USV clusters in the future complex task environment.