基于完善交通信息收集的UAV路径规划

针对部分路段不能获取完整交通信息的问题，提出使用无人机对未布设固定型交通信息检测器路段进行交通巡视，完善交通信息。通过时空网络建立了一个总飞行时间最短、最大单机飞行时间最短的多目标模型，确定最佳的无人机数量和交通信息收集路径。新模型不仅利用时空网络技术细致刻画了无人机在巡视过程中的飞行轨迹，而且加入了对未布设固定型交通检测器路段的巡视次数以及巡视时间间隔约束问题，使巡视路径更加合理。情景分析表明，使用两架无人机进行巡视时，总飞行时间最短为37min，在23min 内完成巡视任务；随着最大单机飞行时间权重的增大，无人机的总飞行时间增加9.76%，最大单机飞行时间减少8.70%。算例分析表明，所建模型和方法能够解决大规模路网的多无人机调度问题，能够根据实际需求得到满意的巡视路径。

UAV Path Planning Based on Perfect Traffic Information Collection

In order to solve the problem that some road segments cannot obtain complete traffic information, and to perfect traffic information, UAV(Unmanned Aerial Vehicle) was proposed to conduct traffic inspections on unfixed traffic information detector road sections. Based on space-time network, a multi-objective model with the minimum total flight time and the minimum single flight time was established, and the optimal number of UAVs and traffic information collection path were determined. The new model not only made use of the space-time network technology to meticulously depict the flight trajectory of the UAV in the process of inspection, but also added the times of patrol and restrictions on the patrol time interval for the unfixed traffic detector sections, which made the patrol route more reasonable. Scenario analysis showed that when using two UAVs for patrol, the total flight time was as short as 37 minutes, and the patrol mission was completed within 23 minutes. As the maximum single UAV flight time weight increasing, the total flight time of the UAV increased by 9.76%, and the maximum single flight time reduced by 8.70%. The example shows that the proposed model and method can solve the multi-UAV scheduling problem of large-scale road network, and can obtain a satisfactory patrol path according to actual needs.