复杂城市低空无人机安全风险评估与三维路径规划  

Safety risk assessment and three-dimensional path planning for UAV in complex urban low-altitude airspace

作　　者：谢华[1,2] 韩斯特 尹嘉男[1,3] 纪晓辉 杨逸晨 XIE Hua;HAN Site;YIN Jianan;JI Xiaohui;YANG Yichen(State Key Laboratory of Air Traffic Management System,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China;College of General Aviation and Flight,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China;College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学空中交通管理系统全国重点实验室,南京210016 [2]南京航空航天大学通用航空与飞行学院,南京210016 [3]南京航空航天大学民航学院,南京210016

出　　处：《安全与环境学报》2024年第7期2490-2507,共18页Journal of Safety and Environment

基　　金：国家重点研发计划项目(2022YFB4300905);国家自然科学基金项目(52002178);江苏省自然科学基金项目(BK20222013,BK20190416)。

摘　　要：针对复杂城市环境内低空无人机飞行安全与效率亟待提升的问题,提出了复杂城市低空无人机安全风险评估与三维路径规划方法。首先,设计了无人机越界冲突率、缓冲空域占比指标,建立了无人机地理围栏安全缓冲间距优化模型,对最佳缓冲间距和栅格粒度进行了标定;然后,构建了由人口密度层、遮蔽层和障碍层构成的无人机风险地图,建立了弹道下降和失控滑行两种模式下的无人机对地风险评估模型,生成了精细化、组合化的城市低空概率风险地图;最后,综合利用地理围栏、概率风险地图和跳点搜索算法,对无人机三维路径进行了初始规划和优化重构。结果表明:弹道下降模式的伤亡风险是失控滑行下降模式的5~75倍;与A*算法相比,跳点搜索算法有效减少了飞行路径的转弯数量,缩短了求解时长,更适合规划无人机飞行路径;与不采用风险地图的方法相比,基于风险地图的无人机路径规划减少了50%的较高风险节点,相应的路径长度仅增加了7.2%和11.4%,整体路径节点的伤亡风险明显降低。研究成果可为复杂城市低空无人机飞行计划制定及安全运行监管提供理论依据和方法支撑。To meet the pressing need for improved safety and efficiency of low-altitude Unmanned Aerial Vehicle(UAV)flights in complex urban environments,a method for complex urban low-altitude UAV safety risk assessment and three-dimensional(3D)path planning is proposed.Firstly,based on a hierarchical structure of origin/destination/warning boundaries,UAV out-of-bounds conflict rate and buffer airspace occupancy ratio indicators are designed.An optimization model for UAV geofence safety buffer distance is established,and the optimal buffer distance of 28.90 m is scientifically calibrated,serving as the basis for setting the optimal grid granularity of 30 m for the low-altitude airspace.Then,a UAV risk map consisting of a population density layer,sheltered layer,and obstacle layer is constructed.A risk-to-ground assessment model for UAVs is established for two descent modes:ballistic descent and uncontrolled glide,generating a fine-grained and composite urban low-altitude probability risk map.Finally,initial planning,optimization reconstruction,and simulation experiments are conducted for UAV 3D path utilizing geofences,probability risk map,and Jump Point Search algorithm.The effectiveness of the 3D path planning method is verified,and the parameter sensitivity of the path planning model is analyzed.Experimental results demonstrate that the casualty risk of fixed-wing UAVs is 3 to 50 times higher than that of rotary-wing UAVs,and the casualty risk of ballistic descent mode is 5 to 75 times higher than that of uncontrolled glide de-scent mode.Compared to the A*algorithm,the Jump Point Search algorithm effectively reduces the number of turns and solution time for flight paths,making it more suitable for planning UAV flight paths.Compared to the method without a risk map,UAV path planning based on a risk map reduces 50%of high-risk nodes,with an increase in path length of only 7.2%and 11.4%.The overall casualty risk of path nodes is significantly reduced,effectively ensuring the safety of UAV operations and demonstrating signi

关 键 词：安全系统学 城市低空 无人机(UAV) 地理围栏 安全评估 路径规划 跳点搜索算法 

分 类 号：X951[环境科学与工程—安全科学]