The invention discloses an intersection safety evaluation method based on unmanned aerial vehicle detection. Comprising the following steps: 1, flying an unmanned aerial vehicle to a proper height above an intersection, and carrying out video shooting on the intersection; 2, constructing an intersection area rectangular coordinate system, and tracking and recording vehicle tracks in the video by using Tracker software; and 3, counting three index data including post-intrusion time PET, parking distance proportion PSD and maximum deceleration maxDRAC for avoiding collision. And 4, performing clustering analysis on the sample data of the three indexes by using a grey clustering method, and dividing four grades to judge the operation safety condition of the target intersection. The conventional method has the defects of inaccurate data acquisition, macroscopic and single evaluation index and the like. Accurate vehicle space-time trajectory data are obtained through unmanned aerial vehicle detection, safety evaluation is carried out on a plurality of microscopic operation indexes of the intersection by adopting a gray clustering method, and comprehensive traffic safety evaluation considering intersection time and space element indexes can be realized.
本发明公开了一种基于无人机检测的交叉口安全评价方法。包括以下步骤:一、将无人机飞至交叉口上方合适高度,对交叉口进行视频拍摄。二、构建交叉口区域直角坐标系,使用Tracker软件对视频中车辆的轨迹进行追踪与记录。三、统计出后侵入时间PET、停车距离比例PSD、避免碰撞的最大减速度maxDRAC三个指标数据。四、利用灰色聚类方法对三个指标的样本数据进行聚类分析,并划分4个等级评判目标交叉口的运行安全情况。以往的方法有数据获取不准确,评价指标宏观,单一等缺陷。本发明通过无人机检测获取精准的车辆时空轨迹数据,采用灰色聚类方法对交叉口多个微观运行指标进行安全评价,能够实现考虑交叉口时间、空间要素指标的综合化交通安全评价。
Intersection safety evaluation method based on unmanned aerial vehicle detection
一种基于无人机检测的交叉口安全评价方法
2024-04-12
Patent
Elektronische Ressource
Chinesisch
Vehicle intersection collision avoidance system and method based on unmanned aerial vehicle
Europäisches Patentamt | 2021
|Europäisches Patentamt | 2021
|Europäisches Patentamt | 2020
|