摘要
服务区入口处车辆频繁变道、合流与分流,导致车流冲突频发,严重影响道路通行效率和行车安全。为此,本文提出基于车辆轨迹大数据分析的服务区入口车流冲突风险实时评估方法。本文利用Tracker软件提取服务区入口车辆的轨迹数据,并对原始数据进行平滑处理。本文基于平滑处理后的车辆轨迹数据,计算TTC与ETTC,作为服务区入口车辆冲突风险的量化指标。本文引入事故树分析法,结合TTC与ETTC指标数据计算服务区入口车流冲突概率值,进而确定风险等级,实现实时评估。实例分析结果表明,该方法下服务区入口车流冲突风险实时评估结果的均方根误差为0.38%,R^(2)为0.9417,实时评估效果良好。
Vehicles at the entrance of the service area frequently change lanes,merge and divert,resulting in frequent traffic conflicts and seriously affecting road traffic efficiency and driving safety.Therefore,this article proposes a real-time assessment method for traffic conflict risk at service area entrances based on vehicle trajectory big data analysis.We extract trajectory data of vehicles entering the service area using Tracker software and smooth the raw data.Based on the smoothed vehicle trajectory data,we calculate TTC and ETTC as quantitative indicators of vehicle conflict risk at the entrance of the service area.We introduce fault tree analysis,and combine TTC and ETTC data to calculate the probability of traffic conflict at the entrance of the service area,in order to determine the risk level and achieve real-time evaluation.The case analysis results show that the root mean square error of the real-time assessment of traffic conflict risk at the entrance of the service area under the research method is 0.38%,and R^(2) is 0.9417,indicating a good real-time assessment effect.
作者
蔡利国
汪俊彬
Cai Liguo;Wang Junbin(Guangdong Road&Bridge Construction Development Co.,Ltd.,Guangzhou,Guangdong 510000,China;Guangdong Communications Planning&Design Institute Group Co.,Ltd.)
出处
《计算机时代》
2026年第3期55-60,共6页
Computer Era
关键词
车辆轨迹大数据
大数据分析
服务区入口
车流冲突
冲突风险
实时评估
Vehicle Trajectory Big Data
Big Data Analysis
Service Area Entrance
Traffic Conflict
Conflict Risk
Real-Time Evaluation
