Suppr超能文献

基于驾驶员视觉感知的城市主次道路右转车辆安全可靠性评价

Safety reliability evaluation when vehicles turn right from urban major roads onto minor ones based on driver's visual perception.

机构信息

Key Laboratory of Road and Traffic Engineering of the Ministry of Education, Traffic Engineering School of Tongji University, 4800 Cao an Highway, Shanghai 201804, China.

出版信息

Accid Anal Prev. 2016 Oct;95(Pt B):487-494. doi: 10.1016/j.aap.2015.08.014. Epub 2015 Sep 1.

Abstract

Turning right has a significant impact on urban road traffic safety. Driving into the curve inappropriately or with improper turning speed often leads to a series of potential accidents and hidden dangers. For a long time, the design speed at intersections has been used to determine the physical radius of curbs and channelization, and drivers are expected to drive in accordance with the design speed. However, a large number of real vehicle tests show that for the road without an exclusive right-turn lane, there is not a good correlation between the physical radius of curbs and the turning right speeds. In this paper, shape parameters of the driver's visual lane model are put forward and they have relatively high correlations with right-turn speeds. Hence, an evaluation method about safety reliability of turning right from urban major roads onto minor ones based on driver's visual perception is proposed. For existing roads, the evaluation object could be real driving videos; for those under construction roads, the evaluation object could be visual scenes obtained from a driving simulation device. Findings in this research will make a contribution to the optimization of right-turn design at intersections and lead to the development of auxiliary driving technology.

摘要

右转对城市道路交通安全有重大影响。驶入弯道不当或转弯速度不当,往往会导致一系列潜在的事故和隐患。长期以来,交叉口的设计速度一直被用来确定路缘石和渠化的物理半径,并且期望驾驶员按照设计速度行驶。然而,大量的实际车辆测试表明,对于没有专用右转车道的道路,路缘石的物理半径与右转速度之间没有很好的相关性。在本文中,提出了驾驶员视觉车道模型的形状参数,它们与右转速度有较高的相关性。因此,提出了一种基于驾驶员视觉感知的城市主干道右转安全可靠性评价方法。对于现有道路,评价对象可以是实际驾驶视频;对于在建道路,可以是从驾驶模拟装置获得的视觉场景。本研究的结果将有助于优化交叉口的右转设计,并为辅助驾驶技术的发展做出贡献。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验