CTSTC: An Energy-Efficient Coverage Path Planning Algorithm for Unmanned Surface Vehicles
Xiaowei Li, Junhan Huang, Yangmin Xie, Dong Yue Qu, Yuxuan Zhong · 2023
Marine environment investigation and patrol operations typically encompass a range of tasks involving Unmanned Surface Vehicles (USVs) and Coverage Path Planning (CPP). The assessment of CPP performance generally revolves around three key metrics: coverage time, extent of overlapping areas, and coverage rate. However, the consideration of energy consumption metrics is often overlooked. Especially for prolonged patrol missions or extensive coverage activities in larger regions, the minimization of energy consumption plays a pivotal role in enhancing the operational endurance of the USV while sustaining its hardware efficiency. We introduce a methodology aimed at minimizing the energy consumption associated with coverage tasks performed by USVs. Our approach draws on the principles of the Scanning Tree Coverage (STC) and the Dynamic Minimum Spanning Tree (DMST) algorithms. Adapting the STC model to the distinctive marine environment, we quantify the energy consumption resulting from ocean currents and the turns of the USV as path length. We present a method for creating a set of longest edges as a crucial part of the spanning tree, which involves dividing the map into multiple elongated regions. This approach serves to simplify the complexity of the problem. Subsequently, we employ the partial DMST algorithm to further generate tree branches. This dynamic approach aims to interconnect the longest edges based on the quantified energy consumption, ultimately striving to minimize energy consumption during coverage. By employing the method, we achieve a dual advantage: a reduction in the number of required turns, coupled with a comprehensive consideration of the influence of ocean currents. This combination empowers the USV to efficiently accomplish its tasks while consuming less energy.