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.