Abstract: Aiming at the problem of urban regional distribution path planning of logistics UAV, the improved grid method is used to model the environment, and the cost function of path security is designed to improve the path security. In order to solve the problem that Theta* algorithm changes the altitude of the path greatly in urban regional planning, and to reduce the number of path nodes and improve the smoothness of the path, the altitude change cost and attitude adjustment cost are introduced into the actual cost function of the algorithm. In the process of searching path nodes, a preset strategy of line of sight is proposed to reduce the redundant checking process and improve the search efficiency of the algorithm. Compared with A* algorithm and Theta* algorithm, the applicability of the improved Theta* algorithm is verified. The optimal values of objective function weight and evaluation function weight are determined by design of control test. The simulation results in different random environments and real environments show that compared with the traditional algorithm, the altitude change, path number, planning time and total path cost of planned path of the improved Theta* algorithm are significantly reduced, which proves the effectiveness of the improved strategy and has certain application value in the UAV path planning in metropolitan.

Key words: 3D path planning, logistics unmanned aerial vehicle（UAV）, improved Theta* algorithm, urban area distribution

摘要： 针对物流无人机城域配送路径规划问题，使用改进栅格法进行环境建模，设计路径安全代价函数提高路径安全性。为了解决Theta*算法在城市区域规划时路径海拔变化较大的问题，同时为减少路径节点数，提高路径平滑性，在算法实际代价函数中引入海拔变化代价及姿态调整代价；在算法搜索路径节点过程中，通过提出视线存在预设策略减少多余检查过程，提高算法搜索效率。与A*算法及Theta*算法对比验证了改进Theta*算法适用性。设计对照实验确定了目标函数权重及估价函数权重最优取值。在随机不同环境及真实环境下进行仿真，结果表明：相比于传统算法，改进Theta*算法规划路径的海拔变化、路径点数、规划时间及路径总代价均有较明显减少，证明了改进策略的有效性，在城域无人机路径规划方面有一定应用价值。

关键词: 三维路径规划, 物流无人机, 改进Theta*算法, 城市区域配送