Dynamic Area Coverage with Multi-USV in Fully Connected State using Graph Theory

Exploration of marine resources is becoming increasingly important in both civil and military domains as marine research and technology advance. However, ocean exploration is often accompanied by high-intensity and high-risk missions, so governments of various countries vigorously develop unmanned systems for ocean missions. With the advancement of artificial intelligence technology, the robustness and scalability of multi-unmanned systems have been significantly improved. In recent years, the control technology of Unmanned Surface Vehicle (USV) has become increasingly mature. Based on the monitoring of the area, this article analyzes the multi-USV system to perform coverage control tasks. This article proposes an algorithm based on the idea of virtual potential field, which enables three USVs to achieve 100% coverage of the area of interest (AOI) and maintain full connectivity as much as possible throughout the process, in order to increase the safety and reliability of the system. The algorithm uses the adjacency matrix to limit the path of the USVs from the perspective of graph theory. Once the multi-USV system cannot maintain full connectivity, measures are taken to bring the USVs close to each other to reach full connectivity again, and finally achieve 100% coverage of AOI, with the multi-USV system remaining fully connected to wait for the next instruction.