关键词: 交通运输工程；无人机医药物流；任务分配；航迹规划；k-means聚类分析；A*算法

Abstract: Aiming at the problem of task assignment and trajectory planning for using UAVs for pharmaceutical transportation in urban environments, a UAV task assignment model under multiple constraints based on the coordinates of demand points and daily demand volume was established, with the objective function of minimizing cost. K-means clustering analysis was embedded into the genetic algorithm for solving UAV takeoff and landing site selection and task allocation. A task line was selected on the basis of task assignment, and the grid method was used to perform spatial 3D modeling of the terrain and obstacle environment around the task line. And then, the A* algorithm was introduced to search for the optimal trajectory for this task. The simulation of trajectory planning based on A* algorithm, ACO algorithm and RRT algorithm were carried out respectively on 3D grid map. The simulation results show that in the process of trajectory planning, the A* algorithm has shorter trajectories, higher search return rate and smoother trajectories compared to the other two algorithms, which verifies the feasibility and effectiveness of the A* algorithm.

Key words: traffic and transportation engineering; drones for pharmaceutical logistics; task assignment; trajectory planning; K-means clustering analysis; the A* algorithm