随着无人机技术的成熟以及相关政策的完善，对无人机的应用模式、航线网络等进行管理以保证运输过程的快捷流畅和飞行安全是低空经济蓬勃发展的基础。通过物流无人机需求生成、起降点选址和分区以及配送路径规划可以构建完整的技术体系：首先，使用兴趣点（point of interest, POI）数据按照指定的规则生成物流速递、餐饮外卖、同城跑腿3类无人机需求；其次，出于无人机在空域安全飞行的需要，以固定路网为基础，采用带权重的K-means聚类分析法进行选址以及Voronoi图对无人机负责区域进行划分；最后，结合固定路网模式，在其中某个区域构建路径规划模型，并使用快速非支配排序遗传算法进行求解。以唐山市为实例进行验证分析，生成了80个无人机需求，选出21个起降点，并对区域3中的12个需求进行路径规划，求解结果为由两架无人机进行配送，飞行总距离为12.26 km，任务均衡度为0.275 8。为城市场景下的物流无人机起降点选址以及路径规划提供了体系化的技术方法。

With the maturity of drone technology and the improvement of related policies, managing the application mode and route network of drones to ensure fast and smooth transportation and flight safety is the foundation for the vigorous development of low altitude economy. A complete technical system can be constructed through the generation of logistics drone demand, location and zoning of take-off and landing points, and distribution path planning. Firstly, using point of interest(POI) data, three types of drone demand can be generated according to specified rules: logistics express delivery, catering takeout, and local errand running. Secondly, in order to ensure the safe flight of drones in airspace, based on a fixed road network, the weighted K-means clustering analysis method was used for site selection, and the Voronoi diagram was used to divide the drone's responsible area. Finally, combined with a fixed road network model, a path planning model was constructed in a certain area and solved using a fast non dominated sorting genetic algorithm. Taking Tangshan City as an example for verification and analysis, 80 drone demands were generated, 21 takeoff and landing points were selected, and path planning was carried out for 12 demands in region 3. The solution result show that two drones are used for delivery, with a total flight distance of 12.26 km and a task balance of 0.275 8. A systematic technical method for the location selection and path planning of logistics unmanned aerial vehicle take-off and landing points in urban scenarios is provided.