复杂风场下基于物理感知图神经网络的无人机安全与能效协同路径规划

doi:10.16097/j.cnki.1009-6744.2026.04.018

交通运输系统工程与信息 ›› 2026, Vol. 26 ›› Issue (4): 213-224.DOI: 10.16097/j.cnki.1009-6744.2026.04.018

复杂风场下基于物理感知图神经网络的无人机安全与能效协同路径规划

王闯^{*a ,b, c}，刘佳良^a，许成鹏^a

西安邮电大学，a. 现代邮政学院；b. 现代邮政协同创新中心；c. 计算机学院，西安 710061

收稿日期:2026-02-07 修回日期:2026-06-06 接受日期:2026-06-21 出版日期:2026-08-25 发布日期:2026-08-21
作者简介:王闯（1981— ），男，陕西渭南人，教授，博士
基金资助:
陕西省教育厅重点科学研究计划项目 (23JY072)；陕西省重点研发计划 (2024SF-YBXM- 671)

Synergistic Path Planning of Safety and Energy Efficiency for UAVs in Complex Wind Fields Based on Physics-Informed Graph Neural Networks

WANG Chuang^{*a, b, c}, LIU Jialiang^a, XU Chengpeng^a

a. Industry School of Modern Post; b. Collaborative Innovation Center for Modern Post; c. School of Computer Science & Technology, Xi'an University of Posts and Telecommunications, Xi'an 710061, China

Received:2026-02-07 Revised:2026-06-06 Accepted:2026-06-21 Online:2026-08-25 Published:2026-08-21
Supported by:
Key Scientific Research Project of Shaanxi Provincial Department of Education, China (23JY072); Shaanxi Provincial Key Research and Development Program, China (2024SF-YBXM- 671)

摘要/Abstract

摘要： 针对低空风场中强空间非均匀特征带来的重载无人机能效与安全协同路径规划的非凸优化难题，本文提出一种融合流场动力学特征的物理感知图神经网络引导 A*搜索的协同路径规划框架（PI-GNN-A*）。该框架首先基于流体力学机理构建融合涡度特征的稀疏物理图，赋予模型对流场各向异性的感知能力；其次，设计增强型非对称损失函数，通过惩罚低估误差将安全约束内化为启发式势场的保守偏置，有效平衡搜索效率与风险规避。基于二维稳态兰金涡旋场的理想化原型验证与统计检验表明：本文方法在计算效率、飞行安全和气动能效之间达成高鲁棒性的协同权衡，相比于缺乏显式物理约束的常规神经网络启发算法，本文方法在嵌入涡度特征后有效抑制了空间的盲目试探，搜索效率实现了统计学意义上的显著提升；相比Dijkstra算法，本方法以约8.0%的代价退化为合理让步，换取搜索空间50.8%的缩减；相比传统基于人工势场的风场感知方法，本文方法以适度的搜索开销为合理代价，在保证全局综合代价非劣的同时，将系统的抗风安全裕度从5.08%拓宽至6.17%，相对提升21.46%，为算力受限的机载平台提供更快的路径重规划能力。

关键词: 航空运输, 路径规划, 图神经网络, 无人机, 物理感知, 安全能效协同

Abstract: To address the non-convex optimization challenge of synergizing energy efficiency and safety path planning for heavy-lift Unmanned Aerial Vehicle (UAV) in low-altitude wind fields with strong spatial non-uniformity, this paper proposes a collaborative path planning framework (PI-GNN-A*) using an A* search guided by a Physics-Informed Graph Neural Network integrating flow field dynamic features. First, based on fluid dynamics mechanisms, a sparse physical graph incorporating vorticity features is constructed to empower the model with the capability of perceiving flow field anisotropy. Then, an enhanced asymmetric loss function is designed. By penalizing underestimation errors, this function internalizes safety constraints into a conservative bias within the heuristic potential field, effectively balancing search efficiency with risk avoidance. Idealized prototype validation and statistical tests on two-dimensional steady-state Rankine vortex fields demonstrate that the proposed method achieves a highly robust synergistic trade-off among computational efficiency, flight safety, and aerodynamic energy efficiency. Compared to conventional neural network heuristic algorithms lacking explicit physical constraints, the proposed method, after embedding vorticity features, effectively suppresses blind spatial exploration and achieves a statistically significant improvement in search efficiency. Compared to the Dijkstra algorithm, the proposed method obtains a 50.8% reduction in search space as a reasonable trade-off for an approximate 8.0% degradation in comprehensive cost. Compared to traditional wind-field-aware methods based on artificial potential fields, the proposed method, at the reasonable expense of a moderate search overhead, broadens the wind-resistant safety margin of the system from 5.08% to 6.17% (a relative increase of 21.46%) while ensuring noninferior global comprehensive cost, providing faster path replanning capabilities for airborne platforms with limited computing power.

Key words: air transportation, path planning, graph neural network, unmanned aerial vehicle (UAV), physics-informed, safetyenergy synergy

中图分类号:

王闯, 刘佳良, 许成鹏. 复杂风场下基于物理感知图神经网络的无人机安全与能效协同路径规划[J]. 交通运输系统工程与信息, 2026, 26(4): 213-224.

WANG Chuang, LIU Jialiang, XU Chengpeng. Synergistic Path Planning of Safety and Energy Efficiency for UAVs in Complex Wind Fields Based on Physics-Informed Graph Neural Networks[J]. Journal of Transportation Systems Engineering and Information Technology, 2026, 26(4): 213-224.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2026.04.018

http://www.tseit.org.cn/CN/Y2026/V26/I4/213

参考文献

[1] 伍景琼, 奠然, 字太升, 等. 无人机配送研究: 关于技术、效益及应用的系统综述[J]. 交通运输系统工程与信息, 2025, 25 (6): 34-49. [WU J Q, DIAN R, ZI T S, et al. Drone delivery: A systematic review on technology, efficiency, and applications[J]. Journal of Transportation Systems Engineering and Information Technology, 2025, 25(6): 34-49.]

[2] 张洪海, 张连东, 刘皞, 等. 城市低空物流无人机航迹规划模型研究[J]. 交通运输系统工程与信息, 2022, 22 (1): 256-264. [ZHANG H H, ZHANG L D, LIU H, et al. Track planning model for logistics unmanned aerial vehicle in urban low-altitude airspace[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(1): 256-264.]

[3] 胡小兵, 杨常澍, 周隽. 复杂城市环境下无人机路网模型研究[J]. 交通运输系统工程与信息, 2023, 23(4): 251-261. [HU X B, YANG C S, ZHOU J. Route network modeling for unmanned aerial vehicle in complex urban environment[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(4): 251-261.]

[4] 赵嶷飞, 谷瑞嘉, 任新惠. 考虑城市低空风场的小型无人机路径规划方法[J]. 北京航空航天大学学报, 2026, 52(6): 1777-1788. [ZHAO Y F, GU R J, REN X H. Method for small unmanned aerial vehicles path planning considering urban low-altitude wind fields[J]. Journal of Beijing University of Aeronautics and Astronautics, 2026, 52(6): 1777-1788.]

[5] 程明, 黄泓鸣. 天气对低空航线网络抗毁性影响分析[J]. 交通运输系统工程与信息, 2025, 25(6): 101-108, 275. [CHENG M, HUANG H M. Impact of weather on invulnerability of low-altitude route network[J]. Journal of Transportation Systems Engineering and Information Technology, 2025, 25(6): 101-108, 275.]

[6] ZHANG Y, LYU J, FU L. Energy-efficient trajectory design for UAV-aided maritime data collection in wind [J]. IEEE Transactions on Wireless Communications, 2022, 21(12): 10871-10886.

[7] AGGARWAL S, KUMAR N. Path planning techniques for unmanned aerial vehicles: A review, solutions, and challenges[J]. Computer Communications, 2020, 149: 270-299.

[8] JAYAWEERA H M P C, HANOUN S. Path planning of unmanned aerial vehicles (UAVs) in windy environments [J]. Drones, 2022, 6(5): 101.

[9] 陈明强, 周子杨, 张勇,等. 基于改进RRT*算法的无人机三维航迹规划[J]. 兵器装备工程学报, 2025, 46(4): 192-199, 234. [CHEN M Q, ZHOU Z Y, ZHANG Y, et al. Three-dimensional path planning of UAV based on improved RRT* algorithm[J]. Journal of Ordnance Equipment Engineering, 2025, 46(4): 192-199, 234.]

[10] RUBI B, MORCEGO B, PEREZ R. Deep reinforcement learning for quadrotor path following with adaptive velocity[J]. Autonomous Robots, 2021, 45(1): 119-134.

[11] 李凯文, 张涛, 王锐, 等. 基于深度强化学习的组合优化研究进展[J]. 自动化学报, 2021, 47(11): 2521-2537. [LI K W, ZHANG T, WANG R, et al. Research reviews of combinatorial optimization methods based on deep reinforcement learning[J]. Acta Automatica Sinica, 2021, 47(11): 2521-2537.]

[12] DU P, SHI Y, CAO H, et al. AI-enabled trajectory optimization of logistics UAVs with wind impacts in smart cities[J]. IEEE Transactions on Consumer Electronics, 2024, 70(1): 3885-3897.

[13] ZENG Y, XU J, ZHANG R. Energy minimization for wireless communication with rotary-wing UAV[J]. IEEE Transactions on Wireless Communications, 2019, 18(4): 2329-2345.

[14] STOLAROFF J K, SAMARAS C, O'NEILL E R, et al. Energy use and life cycle greenhouse gas emissions of drones for commercial package delivery[J]. Nature Communications, 2018, 9(1): 409.

[15] 李安醍, 武丁杰, 李诚龙. 低空无人机自主避障算法综述[J]. 电光与控制, 2021, 28(8): 59-64. [LI A D, WU D J, LI C L. A survey on autonomous collision avoidance algorithms for UAVs at low altitude electro-optics and control[J] Electronics Optics & Control, 2021, 28(8): 59- 64.]

[16] MO X, HUANG Z, XING Y, et al. Multi-agent trajectory prediction with heterogeneous edge-enhanced graph attention network[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(7): 9554-9567.

[17] REKABI- BANA F, HU J, KRAJNIK T, et al. Unified robust path planning and optimal trajectory generation for efficient 3D area coverage of quadrotor UAVs[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 25(3): 2492-2507.

复杂风场下基于物理感知图神经网络的无人机安全与能效协同路径规划

Synergistic Path Planning of Safety and Energy Efficiency for UAVs in Complex Wind Fields Based on Physics-Informed Graph Neural Networks

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	高远, 黄慧, 孙赫迎, 张文会. 面向城市低空物流的分层航路网构建与韧性恢复策略[J]. 交通运输系统工程与信息, 2026, 26(4): 52-64.
[2]	赵小奇, 唐铁桥, 李斌, 闫娜. 基于大语言模型端到端的机场离港旅客流量预测方法[J]. 交通运输系统工程与信息, 2026, 26(4): 309-320.
[3]	王涛, 魏婷婷, 王越超, 李艳华. 考虑旅客取消行为与航班结构性调整的机场离港中断恢复方法[J]. 交通运输系统工程与信息, 2026, 26(4): 332-343.
[4]	赵瀚玮, 陈杰, 孟宪锋, 高学奎, 江子超, 钟秋玥. 飞行监控数据驱动的飞机着陆荷载分析与建模研究[J]. 交通运输系统工程与信息, 2026, 26(4): 344-354.
[5]	张荣, 张茜, 史文萱. 不同空间频率对比敏感度与飞行疲劳的关系模型研究[J]. 交通运输系统工程与信息, 2026, 26(4): 368-379.
[6]	程华, 张扬, 向飞, 马吉震, 唐明杰. 基于深度强化学习的机场地面保障调度方法[J]. 交通运输系统工程与信息, 2026, 26(4): 380-390.
[7]	任广建, 李艳华, 王越超. 机场群空间异质性的多尺度实证分析[J]. 交通运输系统工程与信息, 2026, 26(3): 103-113.
[8]	万莉莉, 徐舒梦, 黄嘉慧, 张庆阳, 袁振宇. 基于模糊最大覆盖模型的无人机应急配送中心选址研究[J]. 交通运输系统工程与信息, 2026, 26(3): 166-175.
[9]	郑展骥, 廖方正, 李燊, 冯昌奎, 凃强, 张河山, 徐进. 无人机夜间航拍视角下小目标车辆精确检测方法[J]. 交通运输系统工程与信息, 2026, 26(3): 247-258.
[10]	唐立, 何彪, 唐昕琛, 王焜. 融合速度障碍与人工势场法的电动垂直起降飞行器路径规划[J]. 交通运输系统工程与信息, 2026, 26(3): 348-359.
[11]	胡小兵, 那容菲, 李航. 多机协同定检智能排班优化方法研究[J]. 交通运输系统工程与信息, 2026, 26(2): 148-158.
[12]	李艳华, 王越超, 任广建, 王涛, 祁嘉仪. 可持续航空燃料与机型和航线三维耦合优化方法[J]. 交通运输系统工程与信息, 2026, 26(2): 169-179.
[13]	张宝成, 谢亚辰. 考虑飞行机组编制规模的机组配对优化方法[J]. 交通运输系统工程与信息, 2026, 26(2): 289-299.
[14]	耿劭卿, 翟一冰, 曹允春. 异构无人机两级协同配送网络选址-路径联合优化[J]. 交通运输系统工程与信息, 2026, 26(1): 34-44.
[15]	马健霄, 王雨, 陆涛, 白莹佳, 王羽尘, 赵顗. 融合多源势场的公路作业区智能驾驶动态路径规划[J]. 交通运输系统工程与信息, 2026, 26(1): 115-124.