首页|期刊导航|华南理工大学学报(自然科学版)|台风灾害下城市应急无人机三维路径规划与调度

台风灾害下城市应急无人机三维路径规划与调度OA

3D Path Planning and Scheduling for Urban Emergency UAV Operations Under Typhoon Disasters

中文摘要英文摘要

随着低空飞行技术的发展,无人机凭借三维机动性与低基础设施的依赖性,在灾害应急物资配送中成为提升响应效率的工具.以台风灾害背景下城市应急物资配送为背景,构建了综合风场扰动、动态能耗变化与通信约束的三维路径规划模型,建立能耗-时效耦合的多目标优化框架.采用三维立方体障碍建模和AABB相交检测实现路径避障约束,将该约束嵌入NSGA-Ⅱ多目标进化算法中,通过混合编码与惩罚机制对算法进行改进.针对构建的城市建筑群场景和复杂风场环境,通过与MOEA/D方法比较,改进NSGA-Ⅱ的HV均值提高了约67.6%,求解稳定性更高,且求解的解集分布更广,展现更好的算法求解能力与稳定性.开展了风速、通信范围及偏远任务点分布3类敏感性实验,形成自适应应急响应初期、灾后恢复与常态阶段自适应动态调度机制,为极端天气条件下的无人机应急物流调度提供理论支持与技术参考.

With advances of low-altitude flight technology,unmanned aerial vehicles(UAVs)leveraging their three-dimensional maneuverability and low dependence on ground infrastructure,have become an effective means to im-prove response efficiency in disaster emergency material delivery.Focusing on urban emergency material delivery in the context of typhoon disasters,this study develops a three-dimensional path planning model that jointly incorpo-rates wind-field disturbances,dynamical energy consumption variations,and communication constraints,establish-ing a multi-objective optimization framework that couples energy consumption and timeliness.Path obstacle avoi-dance constraints were realized using 3D cuboid obstacle modeling and axis-aligned bounding box(AABB)intersec-tion detection,and these constraints were embedded into the NSGA-Ⅱ multi-objective evolutionary algorithm.The algorithm was further enhanced via hybrid encoding and penalty mechanism.In the constructed urban building complex scenarios and complex wind-field environment,compared with the MOEA/D method,comparative experi-ments against MOEA/D show that the improved NSGA-Ⅱ achieves an average improvement in the Hypervolume(HV)indicator of approximately 67.6%,exhibits higher solution stability and yields a more widely distributed solu-tion set,demonstrating superior algorithmic solving capability and stability.Three types of sensitivity experiments are conducted:wind speed variations,communication range changes,and the distribution of remote mission points.An adaptive dynamic scheduling mechanism is formulated for the initial emergency response phase,post-disaster re-covery phase,and normal operation phase,providing theoretical support and technical reference for UAV emer-gency logistics scheduling under extreme weather conditions.This study offers key decision-making insights for in-tegrating the low-altitude economy with resilient urban emergency management systems.

董智捷;钱媛媛;叶文睿;李旺;顾天奇

东南大学 交通学院,江苏 南京 211189东南大学 交通学院,江苏 南京 211189新加坡国立大学 继续与终身教育学院,新加坡 119077东南大学 交通学院,江苏 南京 211189苏州工业园区蒙纳士科学技术研究院,江苏 苏州 215000||蒙纳士大学 土木环境学院,维多利亚州 墨尔本 3800

交通工程

低空物流无人机集群三维避障控制决策路径规划

low-altitude logisticsUAV swarmsthree-dimensional obstacle avoidancecontrol decisionpath planning

《华南理工大学学报(自然科学版)》 2026 (6)

74-87,14

国家自然科学基金优秀青年科学基金(海外)项目广东省智能交通系统重点实验室开放基金项目(202503001)东南大学博士研究生创新能力提升计划项目(CXJH_SEU 25202)Supported by the National Natural Science Fund for Excellent Young Scientists Fund Program(Overseas)and the Key Laboratory of Intelligent Transportation Systems of Guangdong Provincial(202503001)

10.12141/j.issn.1000-565X.250421

评论