基于边聚类的多无人机协同交通巡检优化

Multi-UAV Cooperative Coverage Optimization for Traffic Monitoring Based on Edge Clustering

摘要: 随着城市化进程加快，智能交通系统对高效、灵活的交通监控技术提出了更高要求。无人机凭借其机动性强、成本低等优势，逐渐成为交通监控的重要工具。然而，单架无人机难以覆盖大规模城市路网，多无人机协同巡逻成为研究热点。本文提出一种基于边聚类的多无人机协同覆盖优化方法，通过K-means算法对道路边进行聚类划分，并利用轮廓系数确定最优无人机数量，以实现对路网的高效持续覆盖。实验结果表明，该方法能显著降低平均间隔时间和最差间隔时间，提升覆盖效率，并在不同路网场景中表现出良好的适应性与稳定性。

Abstract: With the acceleration of urbanization, intelligent transportation systems demand more efficient and flexible traffic monitoring technologies. Unmanned Aerial Vehicles (UAVs) have emerged as a promising tool for traffic monitoring due to their high mobility and low cost. However, a single UAV is insufficient to cover large-scale urban road networks, making multi-UAV cooperative patrol a research focus. This paper proposes a multi-UAV cooperative coverage optimization method based on edge clustering. The K-means algorithm is used to cluster road edges, and the silhouette coefficient is employed to determine the optimal number of UAVs for efficient and continuous coverage of the road network. Experimental results demonstrate that the proposed method significantly reduces both the average and worst-case idling times, improves coverage efficiency, and exhibits strong adaptability and stability across different road network scenarios.