Takeover performance according to the level of disengagement during automated driving

Evan Gallouin, Xuguang Wang, P. Beillas, T. Bellet
{"title":"Takeover performance according to the level of disengagement during automated driving","authors":"Evan Gallouin, Xuguang Wang, P. Beillas, T. Bellet","doi":"10.17077/dhm.31754","DOIUrl":null,"url":null,"abstract":"Taking over the manual control of a car after Automated Driving (AD) is a key issue for future road safety. However, performance to resume this manual control may be dependant of the driver’s level of engagement in driving during AD. Indeed, according to the level of automation (from L2 to L3 of the SAE), drivers will be in charge of monitoring the driving situation, or will be allowed to perform non-driving related tasks (NDRT) and thus, to be fully disengaged of the driving task. In this context, the present study aims to investigate the influence of the driver’s level of engagement/disengagement during AD on takeover performance using a driving simulator. Four levels of engagement/disengagement were studied: (C1) being engaged in driving situation monitoring without TakeOver Request (TOR) to resume the manual control, (C2) being engaged in driving situation monitoring with a TOR to resume the manual control, (C3) being disengaged of the driving monitoring by performing a cognitively demanding secondary task with a TOR to resume the manual control, and (C4) being disengaged of the driving monitoring in a relaxed position situation with eyes closed and with a TOR to resume the manual control. Forty participants were performed sixteen critical takeover scenarios involving different critical takeover situations. Drivers reaction times and collision risks were measured to assess their takeover performances and to investigate the safety of automation levels 2 and 3. Driving situation monitoring with a TOR (C2) induce shortest reaction times and a lower number of collisions. For the relaxed posture (C4), drivers took longer time to react than the other three conditions. Driving situation monitoring without TOR (C1), had the highest number of collisions. This suggests that the engagement in driving is not always effective and efficient without TOR. Moreover, being in a relaxed position during automated driving decreases takeover performance.","PeriodicalId":111717,"journal":{"name":"Proceedings of the 7th International Digital Human Modeling Symposium (DHM 2022) and Iowa Virtual Human Summit 2022 -","volume":"64 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 7th International Digital Human Modeling Symposium (DHM 2022) and Iowa Virtual Human Summit 2022 -","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.17077/dhm.31754","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Taking over the manual control of a car after Automated Driving (AD) is a key issue for future road safety. However, performance to resume this manual control may be dependant of the driver’s level of engagement in driving during AD. Indeed, according to the level of automation (from L2 to L3 of the SAE), drivers will be in charge of monitoring the driving situation, or will be allowed to perform non-driving related tasks (NDRT) and thus, to be fully disengaged of the driving task. In this context, the present study aims to investigate the influence of the driver’s level of engagement/disengagement during AD on takeover performance using a driving simulator. Four levels of engagement/disengagement were studied: (C1) being engaged in driving situation monitoring without TakeOver Request (TOR) to resume the manual control, (C2) being engaged in driving situation monitoring with a TOR to resume the manual control, (C3) being disengaged of the driving monitoring by performing a cognitively demanding secondary task with a TOR to resume the manual control, and (C4) being disengaged of the driving monitoring in a relaxed position situation with eyes closed and with a TOR to resume the manual control. Forty participants were performed sixteen critical takeover scenarios involving different critical takeover situations. Drivers reaction times and collision risks were measured to assess their takeover performances and to investigate the safety of automation levels 2 and 3. Driving situation monitoring with a TOR (C2) induce shortest reaction times and a lower number of collisions. For the relaxed posture (C4), drivers took longer time to react than the other three conditions. Driving situation monitoring without TOR (C1), had the highest number of collisions. This suggests that the engagement in driving is not always effective and efficient without TOR. Moreover, being in a relaxed position during automated driving decreases takeover performance.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
根据自动驾驶过程中的脱离程度接管性能
在自动驾驶(AD)之后,如何取代人工控制汽车是未来道路安全的关键问题。然而,恢复这种手动控制的性能可能取决于驾驶员在AD期间驾驶的参与程度。事实上,根据自动驾驶级别(SAE从L2到L3),驾驶员将负责监控驾驶情况,或者允许驾驶员执行与驾驶无关的任务(NDRT),从而完全脱离驾驶任务。在此背景下,本研究旨在利用驾驶模拟器研究AD期间驾驶员的投入/脱离水平对接管绩效的影响。研究了四个层次的投入/脱离:(C1)在没有接管请求(TOR)的情况下进行驾驶状态监控以恢复手动控制,(C2)在有接管请求(TOR)的情况下进行驾驶状态监控以恢复手动控制,(C3)在有接管请求(TOR)的情况下执行认知要求较高的次要任务以恢复手动控制,(C4)在闭上眼睛的放松位置情况下脱离驾驶状态监控并有接管请求(TOR)以恢复手动控制。40名参与者进行了16个关键接管场景,涉及不同的关键接管情况。测试了驾驶员的反应时间和碰撞风险,以评估他们的接管性能,并调查自动化级别2和3的安全性。使用TOR (C2)进行驾驶状态监控可以缩短反应时间,减少碰撞次数。对于放松姿势(C4),驾驶员的反应时间比其他三种情况要长。无TOR (C1)的驾驶状态监测,碰撞次数最多。这表明,如果没有TOR,参与驾驶并不总是有效和高效的。此外,在自动驾驶过程中处于放松状态会降低接管性能。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Design concept evaluation in digital human modeling tools Prediction of walking kinematics and muscle activities under idealized lower limb exoskeleton assistances Forward and Backwards Reaching Inverse Kinematics (FABRIK) solver for DHM: A pilot study Methods for including human variability in system performance models Identifying the best objective function weightings to predict comfortable motorcycle riding postures
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1