Abstract: With the rise of drones in urban distribution, under the joint distribution mode, the location planning of distribution centre and the path planning between distribution centre and automatic airport have become one of the urgent problems to be solved under the joint distribution. Firstly, a location-path model aiming at the best comprehensive operating cost and minimum energy consumption is constructed according to the UAV operating cost, battery energy consumption and distribution centre management cost. Secondly, the optimal location of distribution centre is found based on K-means algorithm, and the genetic algorithm is used to solve the distribution route. Finally, the main urban area of Tianjin is taken as an example to apply the model and analyze the results. The results show that a total of 40 distribution centres are needed in the main urban area of Tianjin. As demand points increase, the number and cost of drones increase, and total energy consumption increases in a fluctuating manner. It is found that the distribution effect is better with the use of drones with 3.5~5.0?kg payload and distribution centres with coverage within 3~4?km. The research results provide constructive suggestions for urban UAV distribution centre location, path planning, aircraft type selection, service range and so on.

Key words: location-path planning, distribution of logistic drones, dynamic energy consumption, unmanned automatic vertiport, distribution centre

摘要： 随着无人机在城市配送的兴起，在联合配送模式下，配送中心的选址规划及与全自动机场之间的路径规划成为亟待解决的问题之一。根据无人机运营成本、电池能耗和配送中心管理成本构建以最佳综合运营成本和最小能耗为目标的选址-路径模型；基于K-means算法寻找最佳配送中心选址，运用遗传算法对配送路径进行求解；以天津市主城区为例进行模型应用及结果分析。结果表明天津市主城区共需40个配送中心，随着需求点增多，无人机数量和成本增加，总能耗呈波动式上升；分析发现使用载荷3.5~5.0?kg的无人机机型，覆盖范围在3~4?km内的配送中心使得配送效果更优。研究结果为城市无人机配送中心选址、路径规划以及机型选择、服务范围等提供了建设性建议。

关键词: 选址-路径规划, 物流无人机配送, 动态能耗, 无人机全自动机场, 配送中心