Real-time lane-changing crash prediction model at the individual vehicle level using real-world trajectories prior to crashes

IF 7.6 1区工程技术 Q1 TRANSPORTATION SCIENCE & TECHNOLOGY

Transportation Research Part C-Emerging Technologies Pub Date : 2025-05-14 DOI:10.1016/j.trc.2025.105171

Kequan Chen , Zhibin Li , Pan Liu , Chengcheng Xu , Yuxuan Wang

{"title":"Real-time lane-changing crash prediction model at the individual vehicle level using real-world trajectories prior to crashes","authors":"Kequan Chen , Zhibin Li , Pan Liu , Chengcheng Xu , Yuxuan Wang","doi":"10.1016/j.trc.2025.105171","DOIUrl":null,"url":null,"abstract":"<div><div>This study aims to develop a real-time crash prediction model for individual lane-changing (LC) maneuvers by considering interactions between the LC vehicle and surrounding vehicles. Vehicle trajectories prior to real-world LC crashes are extracted for modeling. Risky events are identified based on the remaining distance between vehicles to develop Generalized Extreme Value (GEV) distributions. Driving-related factors, such as the relative distance, speed, and acceleration between vehicles during the LC maneuver, are considered to address the non-stationary issue. A real-time LC crash prediction model is established by quantifying the differences between non-stationary GEV distributions under LC crash and non-crash conditions. The results show that incorporating driving-related factors significantly improves the goodness-of-fit of GEV distribution. Our model shows satisfactory LC crash prediction performance, with the Area Under the Curve (AUC) values ranging from 0.92 to 0.98. The proposed model improves by an average of 75% over traditional Time-to-Collision (TTC), and 49% over Two-Dimensional TTC.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"176 ","pages":"Article 105171"},"PeriodicalIF":7.6000,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transportation Research Part C-Emerging Technologies","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0968090X25001755","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"TRANSPORTATION SCIENCE & TECHNOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

This study aims to develop a real-time crash prediction model for individual lane-changing (LC) maneuvers by considering interactions between the LC vehicle and surrounding vehicles. Vehicle trajectories prior to real-world LC crashes are extracted for modeling. Risky events are identified based on the remaining distance between vehicles to develop Generalized Extreme Value (GEV) distributions. Driving-related factors, such as the relative distance, speed, and acceleration between vehicles during the LC maneuver, are considered to address the non-stationary issue. A real-time LC crash prediction model is established by quantifying the differences between non-stationary GEV distributions under LC crash and non-crash conditions. The results show that incorporating driving-related factors significantly improves the goodness-of-fit of GEV distribution. Our model shows satisfactory LC crash prediction performance, with the Area Under the Curve (AUC) values ranging from 0.92 to 0.98. The proposed model improves by an average of 75% over traditional Time-to-Collision (TTC), and 49% over Two-Dimensional TTC.

查看原文本刊更多论文

实时变道碰撞预测模型在个别车辆水平使用真实世界的轨迹之前的碰撞

本研究旨在建立考虑变道车辆与周围车辆相互作用的实时变道碰撞预测模型。在实际LC碰撞之前的车辆轨迹被提取用于建模。基于车辆之间的剩余距离对危险事件进行识别，得到广义极值（GEV）分布。在LC机动过程中，考虑了车辆之间的相对距离、速度和加速度等与驾驶相关的因素来解决非静止问题。通过量化LC碰撞与非碰撞条件下非平稳GEV分布的差异，建立了LC碰撞实时预测模型。结果表明，纳入驾驶相关因素显著提高了GEV分布的拟合优度。我们的模型显示了令人满意的LC崩溃预测性能，曲线下面积（AUC）值在0.92到0.98之间。该模型比传统的碰撞时间（TTC）平均提高75%，比二维TTC平均提高49%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Transportation Research Part C-Emerging Technologies 工程技术-运输科技

CiteScore

15.80

自引率

12.00%

发文量

332

审稿时长

64 days

期刊介绍： Transportation Research: Part C (TR_C) is dedicated to showcasing high-quality, scholarly research that delves into the development, applications, and implications of transportation systems and emerging technologies. Our focus lies not solely on individual technologies, but rather on their broader implications for the planning, design, operation, control, maintenance, and rehabilitation of transportation systems, services, and components. In essence, the intellectual core of the journal revolves around the transportation aspect rather than the technology itself. We actively encourage the integration of quantitative methods from diverse fields such as operations research, control systems, complex networks, computer science, and artificial intelligence. Join us in exploring the intersection of transportation systems and emerging technologies to drive innovation and progress in the field.