摘要
森林火灾的动态蔓延可视为一种动态障碍系统,其非线性扩散行为对应急救援路径的安全性及时效性构成挑战。现有路径规划方法难以实现火场演化进程与路径决策的时序匹配,导致避障失效。因此针对森林火灾动态蔓延环境下应急救援路径的时空避障难题,首先通过森林火灾蔓延模型构建动态障碍场;然后提出一种基于地图代数动态欧氏障碍空间的最短路径算法生成动态绕障距离场;最后在绕障距离场基础上实施路径抽取。最终实现了规划路径与火场扩散的时序匹配,在动态火场障碍场景下计算最优安全路径,同时划定极端风险区域。以“3·30”西昌森林火灾为原型构建火场动态障碍,对不同行进速度、不同安全阈值、人工动态障碍进行多场景对比实验,实验结果均表明,所提方法所规划最短路径结果可有效规避火场范围,划分极端风险区域,兼顾安全性与时效性。该研究为动态障碍环境下的应急救援提供了可解释性强、鲁棒性高的解决方案,在森林消防指挥系统中具有显著应用价值。
Objectives:The dynamic spread of forest fires can be regarded as a dynamic obstacle system,and its nonlinear diffusion behavior poses a challenge to the safety and timeliness of emergency rescue paths.The existing path planning methods are difficult to achieve temporal matching between the evolution process of the fire scene and path decision-making,resulting in obstacle avoidance failure or suboptimal route costs.Methods:Focused on the spatiotemporal obstacle avoidance problem of emergency rescue paths in the dynamic spread environment of forest fires,a dynamic obstacle field is constructed firstly through a forest fire spread prediction model.Then,based on this field and a static obstacle field,a map algebra dynamic euclidean space shortest path with obstacles algorithm is proposed to generate a dynamic obstacle avoidance Euclidean distance field and its corresponding source-point field.Finally,path extraction is implemented based on the obstacle avoidance distance field and source-point field.The timing matching between the planned path and the fire spread was achieved,and the optimal safe path was calculated in dynamic fire obstacle scenarios,while extreme-risk areas were delineated.Results:Using the“3·30”Xichang forest fire as a prototype,dynamic obstacles were constructed in the fire scene.Multi-scenario comparative experiments evaluated varied travel speeds and time safety thresholds,including cases with complex multi-ignition scenarios featuring artificially superimposed dynamic obstacles.The results show that the planned shortest path effectively avoided the fire-affected zones,with extreme-risk areas delineated through the dynamic obstacle avoidance distance field,while maintaining safety-timeliness trade-off balance.Critically,setting context-appropriate time safety thresholds can significantly mitigate fire exposure risks without compromising rescue path costs.Conclusions:This proposed method provides a highly interpretable and robust solution for emergency rescue in dynamic obstacle environments,achieving fire path timing matching through distance field with optimal safety threshold calibration,which has significant application value in forest fire command systems.
作者
魏自立
胡海
李思宇
韩超然
任福
张浡聪
WEI Zili;HU Hai;LI Siyu;HAN Chaoran;REN Fu;ZHANG Bocong(School of Resource and Environmental Sciences,Wuhan University,Wuhan 430079,China;Technology Innovation Center for Spatiotemporal Information and Equipment of Intelligent City,Ministry of Natural Resources,Chongqing 401120,China)
出处
《武汉大学学报(信息科学版)》
北大核心
2025年第11期2325-2338,共14页
Geomatics and Information Science of Wuhan University
基金
国家重点研发计划(2022YFC3005704)
自然资源部智能城市时空信息与装备工程技术创新中心开放基金(STIEICKF202309)。
关键词
路径规划
动态障碍
距离变换
森林火灾
path planning
dynamic obstacles
distance transform
forest fire
