Article Details

Abstract

To address the problem that the coexistence of straight aisles, intersections, and narrow passages in warehouse aisles causes indoor UAV paths to hug shelves and make sharp turns, while fixed weights generalize poorly across scenarios, this paper proposes a structure-aware adaptive cost scheduling method based on a 2.5D layered occupancy grid. The method uses the Euclidean Signed Distance Field (ESDF) and free-space skeletons for structural partitioning, and dynamically adjusts shelf-proximity and turning penalties in Weighted A* according to region type and local corridor width. Experimental results in the S1 straight-aisle and S2 intersection scenarios show that, compared with WA*-Fix, the proposed method increases the minimum obstacle distance from 0.57 m and 0.52 m to 0.72 m and 0.69 m, reduces the turning severity from 3.62 rad and 5.95 rad to 3.10 rad and 3.40 rad, and increases the planning time only from 36 ms and 45 ms to 38 ms and 47 ms. These results demonstrate that the proposed method improves both safety margin and path smoothness while maintaining real-time performance, making it suitable for global path planning of indoor UAVs in warehouse aisle environments.

Keywords

path planning; indoor UAV; warehouse aisles; 2.5D layered occupancy grid; adaptive cost scheduling

References

1. [1] Hart P E，Nilsson N J，Raphael B. A Formal Basis for the Heuristic Determination of Minimum Cost Paths [J].IEEE Trans on Systems Science and Cybernetics，1968，4（2）：100-107.
2. [2] Pohl I. Heuristic Search Viewed as Path Finding in a Graph [J].Artificial Intelligence，1970，1（3/4）：193-204.
3. [3] Wang Ershen, Zhang Shenglong, Guo Jing, etc A review of unmanned aerial vehicle path planning technology for low altitude economic application scenarios [J]. Telecommunications Science, 2025, 41 (12): 27-43
4. [4] Meysami A，Kelouwani S，Cuillière J C，et al. An Efficient Indoor Large Map Global Path Planning for Robot Navigation [J/OL].Expert Systems With Applications，2024，248：123388.https://doi.org/10.1016/j.eswa.2024.123388.
5. [5] Yang Tongyao Research on UAV trajectory hierarchical planning for path search and task allocation [D]. Taiyuan: North China University, 2024
6. [6] Mellinger D，Kumar V. Minimum Snap Trajectory Generation and Control for Quadrotors [C]//Proceedings of the IEEE International Conference on Robotics and Automation (ICRA)，2011：2520-2525.
7. [7] Yu Y，Nagpal K，Mceowen S，et al. Real-time Quadrotor Trajectory Optimization with Time-triggered Corridor Constraints [J].Journal of Guidance, Control, and Dynamics，2023，46（6）：1197-1205.
8. [8] Noël T，Lehuger A，Marchand É，et al. Skeleton Disk-graph Roadmap: A Sparse Deterministic Roadmap for Safe 2D Navigation and Exploration [J].IEEE Robotics and Automation Letters，2024，9（1）：555-562.
9. [9] Zhu J，Wong M O，Nisbet N，et al. Semantics-based Connectivity Graph for Indoor Pathfinding Powered by IFC-Graph [J/OL].Automation in Construction，2025，171：106019.https://doi.org/10.1016/j.autcon.2025.106019.
10. [10] Alqobali R，Alnasser R，Rashidi A，et al. A Real-Time Semantic Map Production System for Indoor Robot Navigation [J/OL].Sensors，2024，24（20）：6691.https://doi.org/10.3390/s24206691.
11. [11] Malang C，Charoenkwan P，Wudhikarn R. Implementation and Critical Factors of Unmanned Aerial Vehicle (UAV) in Warehouse Management: A Systematic Literature Review [J/OL].Drones，2023，7（2）：80.https://doi.org/10.3390/drones7020080.
12. [12] Lin H Y，Chang K L，Huang H Y. Development of Unmanned Aerial Vehicle Navigation and warehouse inventory system based on Reinforcement Learning [J/OL].Drones，2024，8（6）：220.https://doi.org/10.3390/drones8060220.
13. [13] Guzman C J P D，Chua A Y，Chu T S C. Path planning of Multiple Quadrotors with Ultrawideband Localization for Warehouse Inventory Management [J/OL].Journal of Robotics，2024，2024：2566619.https://doi.org/10.1155/2024/2566619.
14. [14] Panupong V，Simon E，Ma Kuseemasae，et al. Locating Charging Stations and Routing Drones for Efficient Automated Stocktaking [J].European Journal of Operational Research，2024，316（3）：1129-1145.

Funding

Research on Key Technologies and Applications of Unmanned Aerial Vehicle Intelligent Operation in Warehouse Management for China Southern Power Grid Co., Ltd. (ZBKJXM20240284)

About the authors

Li Bo (September 1990-), male, Han ethnicity, from Neijiang, Sichuan, engineer, deputy manager, master's student, research interests include material supply management; Corresponding author: Xu Wanwei (November 2000-), female, Han ethnicity, from Foshan, Guangdong, specialist, master's student, research interests include machine learning, edge computing and federated learning; Ding Ming (December 1998-), male, Han ethnicity, from Chaozhou, Guangdong, specialist, master's student, research interests include logistics enterprises, multi-center demand and vehicle routing; Zhang Endong (April 2003-), male, Tujia ethnicity, from Zhangjiajie, Hunan, specialist, bachelor's degree, research interests include optimal design of power electronic converters; Yu Wuyi (May 2003-), male, Han ethnicity, from Yunmeng, Hubei, specialist, bachelor's degree, research interests include smart grids.