Studying the predictability of crash risk caused by manual takeover of autonomous vehicles in mixed traffic flow

IF 3.3 3区工程技术 Q2 TRANSPORTATION

Transportation Letters-The International Journal of Transportation Research Pub Date : 2024-11-25 DOI:10.1080/19427867.2023.2279807

Qingchao Liu , Ruohan Yu , Yingfeng Cai , Long Chen

{"title":"Studying the predictability of crash risk caused by manual takeover of autonomous vehicles in mixed traffic flow","authors":"Qingchao Liu , Ruohan Yu , Yingfeng Cai , Long Chen","doi":"10.1080/19427867.2023.2279807","DOIUrl":null,"url":null,"abstract":"<div><div>This study explores how to reduce the cost of prediction as much as possible while ensuring the prediction accuracy of a real-time crash risk model. The extreme gradient enhancement (XGBoost) algorithm was used to predict the crash risk of autonomous vehicles in different sections of highway. The results show that the prediction performance of the model is the best when the threshold value is 0.05. Choosing two variables to predict can ensure high accuracy and simultaneously reduce the cost of prediction when the accuracy of crash risk prediction of the three sections can reach 73%, 62%, and 70%. However, when only one variable can be selected due to sensor or system failure, the speed difference between the takeover car and the front car can be chosen to achieve the greatest benefit. These findings could provide a reference for technicians to design safer and more economical autonomous vehicles.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"16 10","pages":"Pages 1205-1223"},"PeriodicalIF":3.3000,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transportation Letters-The International Journal of Transportation Research","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/org/science/article/pii/S1942786723002515","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"TRANSPORTATION","Score":null,"Total":0}

引用次数: 0

Abstract

This study explores how to reduce the cost of prediction as much as possible while ensuring the prediction accuracy of a real-time crash risk model. The extreme gradient enhancement (XGBoost) algorithm was used to predict the crash risk of autonomous vehicles in different sections of highway. The results show that the prediction performance of the model is the best when the threshold value is 0.05. Choosing two variables to predict can ensure high accuracy and simultaneously reduce the cost of prediction when the accuracy of crash risk prediction of the three sections can reach 73%, 62%, and 70%. However, when only one variable can be selected due to sensor or system failure, the speed difference between the takeover car and the front car can be chosen to achieve the greatest benefit. These findings could provide a reference for technicians to design safer and more economical autonomous vehicles.

查看原文本刊更多论文

研究混合交通流中人工接管自动驾驶汽车碰撞风险的可预测性

本研究探讨如何在保证实时碰撞风险模型预测精度的同时，尽可能降低预测成本。采用极限梯度增强（XGBoost）算法对自动驾驶汽车在不同路段的碰撞风险进行预测。结果表明，当阈值为0.05时，模型的预测性能最好。当三个路段的碰撞风险预测精度分别达到73%、62%和70%时，选择两个变量进行预测可以保证较高的预测精度，同时降低预测成本。但是，当由于传感器或系统故障只能选择一个变量时，可以选择接管车与前车的速度差，以实现最大的效益。这些发现可以为技术人员设计更安全、更经济的自动驾驶汽车提供参考。

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 Letters-The International Journal of Transportation Research TRANSPORTATION SCIENCE & TECHNOLOGY-

CiteScore

6.40

自引率

14.30%

发文量

审稿时长

>12 weeks

期刊介绍： Transportation Letters: The International Journal of Transportation Research is a quarterly journal that publishes high-quality peer-reviewed and mini-review papers as well as technical notes and book reviews on the state-of-the-art in transportation research. The focus of Transportation Letters is on analytical and empirical findings, methodological papers, and theoretical and conceptual insights across all areas of research. Review resource papers that merge descriptions of the state-of-the-art with innovative and new methodological, theoretical, and conceptual insights spanning all areas of transportation research are invited and of particular interest.