Article(id=1149738773480648933, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149738762382524507, articleNumber=1003-3033(2024)07-0211-08, orderNo=null, doi=10.16265/j.cnki.issn1003-3033.2024.07.0247, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1705161600000, receivedDateStr=2024-01-14, revisedDate=1713369600000, revisedDateStr=2024-04-18, acceptedDate=null, acceptedDateStr=null, onlineDate=1752048684710, onlineDateStr=2025-07-09, pubDate=1722096000000, pubDateStr=2024-07-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752048684710, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752048684710, creator=13701087609, updateTime=1752048684710, updator=13701087609, issue=Issue{id=1149738762382524507, tenantId=1146029695717560320, journalId=1146031787341344770, year='2024', volume='34', issue='7', pageStart='1', pageEnd='252', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752048682065, creator=13701087609, updateTime=1757316437713, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1171833331021824745, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149738762382524507, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1171833331021824746, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1149738762382524507, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=211, endPage=218, ext={EN=ArticleExt(id=1149738773715529958, articleId=1149738773480648933, tenantId=1146029695717560320, journalId=1146031787341344770, language=EN, title=Collision scenario construction and simulation analysis for autonomous driving safety testing, columnId=1149733270084042840, journalTitle=China Safety Science Journal, columnName=Public safety, runingTitle=null, highlight=null, articleAbstract=

To reduce traffic accidents caused by autonomous vehicles and improve the efficiency of vehicle safety testing in simulation environments,an autonomous driving collision test scenario construction method was proposed based on deep reinforcement learning. Firstly,the vehicle's driving process was mapped to a Markov decision process by setting the state,action,and reward functions. Then,the agent was trained to complete the vehicle collision task and generate the collision test scenarios based on the built simulation platform (CARLA-DRL). Finally,500 random collision simulation tests were conducted to analyze the collision success rate,collision time,and collision energy based on the relative distance between the agent and the autonomous vehicle. The results indicated that the agent generated collision trajectories that conformed to vehicle dynamics and could construct refined and multi-type collision test scenarios. The average collision success rate between the agent and the autonomous vehicle was 62.20%,the average collision time was 127.25 s,and the average collision energy value was 175.98 kJ. The proposed method can construct high-frequency,high-efficient,and high-risk autonomous driving vehicle collision test scenarios,increasing the probability of occasional high-risk scenarios in simulation scenarios and enhancing the efficiency of safety testing for autonomous vehicle collision incidents.

, correspAuthors=Liang ZHENG, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Yaohua ZHAO, Yanzhan CHEN, Liang ZHENG, Shukai LI), CN=ArticleExt(id=1149738786004841064, articleId=1149738773480648933, tenantId=1146029695717560320, journalId=1146031787341344770, language=CN, title=面向自动驾驶安全测试的碰撞场景构建与仿真分析, columnId=1149733271510106222, journalTitle=中国安全科学学报, columnName=公共安全, runingTitle=null, highlight=null, articleAbstract=

为减少自动驾驶车辆产生的交通事故,提高车辆在仿真环境中的安全测试效率,提出一种基于深度强化学习的自动驾驶碰撞测试场景构建方法。首先,通过设置状态、动作和奖励函数,将车辆的驾驶过程映射为马尔可夫决策过程;然后,基于搭建的仿真平台(CARLA-DRL)训练智能体完成车辆碰撞任务,生成碰撞测试场景;最后,随机进行500次碰撞仿真试验,根据智能体与自动驾驶车辆的相对距离,分析碰撞成功率、碰撞时间和冲突能量。结果表明:智能体生成符合车辆动力学的碰撞轨迹,能够构建精细化、多类型的碰撞测试场景;智能体与自动驾驶车辆的平均碰撞成功率为62.20%,平均碰撞时间为127.25 s,平均冲突能量值为175.98 kJ。该方法能够构建高频、高效和高危的自动驾驶车辆碰撞测试场景,增加仿真场景中偶发高风险场景的发生概率,提高自动驾驶车辆对于碰撞事故的安全测试效率。

, correspAuthors=郑亮, authorNote=null, correspAuthorsNote=
** 郑亮(1984—),男,湖南衡阳人,博士,教授,主要从事智能交通系统仿真优化、交通系统数字孪生与虚拟仿真、路网交通状态估计与预测等方面的研究。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=mzWy+H/q+ybOiXApy19mGQ==, magXml=Vb57xYK9AFjVq4qRPvtsZw==, pdfUrl=null, pdf=etzTMi/KdohadgV4m4ZNuA==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=fqoXDU47PItviQEVGG5SVQ==, mapNumber=null, authorCompany=null, fund=null, authors=

赵尧华 (1999—),男,重庆人,硕士研究生,主要研究方向为自动驾驶危险测试场景构建与分析。E-mail:

李树凯 教授

, authorsList=赵尧华, 陈延展, 郑亮, 李树凯)}, authors=[Author(id=1171830282446488209, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=yaohuazhao@csu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1171830282580705939, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830282446488209, language=EN, stringName=Yaohua ZHAO, firstName=Yaohua, middleName=null, lastName=ZHAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1171830282672980628, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830282446488209, language=CN, stringName=赵尧华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中南大学 交通运输工程学院,湖南 长沙 410075, bio={"content":"

赵尧华 (1999—),男,重庆人,硕士研究生,主要研究方向为自动驾驶危险测试场景构建与分析。E-mail:

"}, bioImg=null, bioContent=

赵尧华 (1999—),男,重庆人,硕士研究生,主要研究方向为自动驾驶危险测试场景构建与分析。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1171830282240967305, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, xref=1, ext=[AuthorCompanyExt(id=1171830282266133130, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China), AuthorCompanyExt(id=1171830282274521739, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中南大学 交通运输工程学院,湖南 长沙 410075)])]), Author(id=1171830282731700886, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1171830282807198360, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830282731700886, language=EN, stringName=Yanzhan CHEN, firstName=Yanzhan, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1171830282874307225, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830282731700886, language=CN, stringName=陈延展, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中南大学 交通运输工程学院,湖南 长沙 410075, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1171830282240967305, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, xref=1, ext=[AuthorCompanyExt(id=1171830282266133130, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China), AuthorCompanyExt(id=1171830282274521739, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中南大学 交通运输工程学院,湖南 长沙 410075)])]), Author(id=1171830282945610395, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhengliang@csu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1171830283033690781, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830282945610395, language=EN, stringName=Liang ZHENG, firstName=Liang, middleName=null, lastName=ZHENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, **, address=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1171830283096605342, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830282945610395, language=CN, stringName=郑亮, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, **, address=1 中南大学 交通运输工程学院,湖南 长沙 410075, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1171830282240967305, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, xref=1, ext=[AuthorCompanyExt(id=1171830282266133130, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China), AuthorCompanyExt(id=1171830282274521739, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中南大学 交通运输工程学院,湖南 长沙 410075)])]), Author(id=1171830283159519904, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1171830283235017378, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830283159519904, language=EN, stringName=Shukai LI, firstName=Shukai, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1171830283306320547, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, authorId=1171830283159519904, language=CN, stringName=李树凯, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 北京交通大学 轨道交通控制与安全国家重点实验室,北京 100044, bio={"content":"

李树凯 教授

"}, bioImg=null, bioContent=

李树凯 教授

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1171830282354213516, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, xref=2, ext=[AuthorCompanyExt(id=1171830282362602125, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282354213516, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China), AuthorCompanyExt(id=1171830282370990734, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282354213516, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学 轨道交通控制与安全国家重点实验室,北京 100044)])])], keywords=[Keyword(id=1171830283465704100, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, orderNo=1, keyword=autonomous driving), Keyword(id=1171830283524424357, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, orderNo=2, keyword=safety testing), Keyword(id=1171830283578950310, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, orderNo=3, keyword=collision test scenarios), Keyword(id=1171830283646059175, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, orderNo=4, keyword=simulation experiment), Keyword(id=1171830283713168040, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, orderNo=5, keyword=deep reinforcement learning), Keyword(id=1171830283767693993, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, orderNo=1, keyword=自动驾驶), Keyword(id=1171830283826414250, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, orderNo=2, keyword=安全测试), Keyword(id=1171830283889328811, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, orderNo=3, keyword=深度强化学习), Keyword(id=1171830283943854764, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, orderNo=4, keyword=碰撞测试场景), Keyword(id=1171830284006769325, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, orderNo=5, keyword=仿真试验)], refs=[Reference(id=1171830285508330179, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=8248, pageEnd=8254, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=KENDALL A, HAWKE J, JANZ D, journalName=2019 International Conference on Robotics and Automation (ICRA). IEEE, refType=null, unstructuredReference=KENDALL A, HAWKE J, JANZ D, et al. Learning to drive in a day[C]. 2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019: 8248-8254., articleTitle=Learning to drive in a day, refAbstract=null), Reference(id=1171830285562856132, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=3, pageStart=41, pageEnd=54, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=李瑞敏, 戴晶辰, journalName=交通运输工程学报, refType=null, unstructuredReference=李瑞敏, 戴晶辰. 自动驾驶影响下的出行行为研究综述[J]. 交通运输工程学报, 2022, 22(3): 41-54., articleTitle=自动驾驶影响下的出行行为研究综述, refAbstract=null), Reference(id=1171830285621576389, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=3, pageStart=41, pageEnd=54, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=LI Ruimin, DAI Jingchen, journalName=Journal of Traffic and Transportation Engineering, refType=null, unstructuredReference=LI Ruimin, DAI Jingchen. Review on impact of autonomous driving on travel behaviors[J]. Journal of Traffic and Transportation Engineering, 2022, 22(3): 41-54., articleTitle=Review on impact of autonomous driving on travel behaviors, refAbstract=null), Reference(id=1171830285680296646, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2023, volume=615, issue=7953, pageStart=620, pageEnd=627, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=FENG Shuo, SUN Haowei, YAN Xintao, journalName=Nature, refType=null, unstructuredReference=FENG Shuo, SUN Haowei, YAN Xintao, et al. Dense reinforcement learning for safety validation of autonomous vehicles[J]. Nature, 2023, 615(7953): 620-627., articleTitle=Dense reinforcement learning for safety validation of autonomous vehicles, refAbstract=null), Reference(id=1171830285734822599, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2016, volume=94, issue=null, pageStart=182, pageEnd=193, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=NIDHI K, SUSAN M P, journalName=Transportation Research Part A: Policy and Practice, refType=null, unstructuredReference=NIDHI K, SUSAN M P. Driving to safety: how many miles of driving would it take to demonstrate autonomous vehicle reliability[J]. Transportation Research Part A: Policy and Practice, 2016, 94: 182-193., articleTitle=Driving to safety: how many miles of driving would it take to demonstrate autonomous vehicle reliability, refAbstract=null), Reference(id=1171830285789348552, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=1, pageStart=316, pageEnd=333, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=邓伟文, 李江坤, 任秉韬, journalName=中国公路学报, refType=null, unstructuredReference=邓伟文, 李江坤, 任秉韬, 等. 面向自动驾驶的仿真场景自动生成方法综述[J]. 中国公路学报, 2022, 35(1): 316-333., articleTitle=面向自动驾驶的仿真场景自动生成方法综述, refAbstract=null), Reference(id=1171830285848068809, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=1, pageStart=316, pageEnd=333, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=DENG Weiwen, LI Jiangkun, REN Bingtao, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=DENG Weiwen, LI Jiangkun, REN Bingtao, et al. A survey on automatic simulation scenario generation methods for autonomous driving[J]. China Journal of Highway and Transport, 2022, 35(1): 316-333., articleTitle=A survey on automatic simulation scenario generation methods for autonomous driving, refAbstract=null), Reference(id=1171830285910983370, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2020, volume=33, issue=11, pageStart=125, pageEnd=138, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=余荣杰, 田野, 孙剑, journalName=中国公路学报, refType=null, unstructuredReference=余荣杰, 田野, 孙剑. 高等级自动驾驶汽车虚拟测试:研究进展与前沿[J]. 中国公路学报, 2020, 33(11): 125-138., articleTitle=高等级自动驾驶汽车虚拟测试:研究进展与前沿, refAbstract=null), Reference(id=1171830285982286539, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2020, volume=33, issue=11, pageStart=125, pageEnd=138, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=YU Rongjie, TIAN Ye, SUN Jian, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=YU Rongjie, TIAN Ye, SUN Jian. Highly automated vehicle virtual testing: a review of recent developments and research frontiers[J]. China Journal of Highway and Transport, 2020, 33(11): 125-138., articleTitle=Highly automated vehicle virtual testing: a review of recent developments and research frontiers, refAbstract=null), Reference(id=1171830286049395404, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2021, volume=12, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=FENG Shuo, YAN Xintao, SUN Haowei, journalName=Nature Communications, refType=null, unstructuredReference=FENG Shuo, YAN Xintao, SUN Haowei, et al. Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment[J]. Nature Communications, 2021, 12(1): DOI:10.1038/s41467-021-21007-8., articleTitle=Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment, refAbstract=null), Reference(id=1171830286112309965, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=null, volume=23, issue=6, pageStart=10, pageEnd=77, url=null, language=null, rfNumber=[8], rfOrder=10, authorNames=null, journalName=交通运输工程学报, refType=null, unstructuredReference=赵祥模国家重点研发计划项目(2021YFB2501200)团队. 自动驾驶测试与评价技术研究进展[J]. 交通运输工程学报, 2023, 23(6): 10-77., articleTitle=自动驾驶测试与评价技术研究进展, refAbstract=null), Reference(id=1171830286162641614, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=null, volume=23, issue=6, pageStart=10, pageEnd=77, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=null, journalName=Journal of Traffic and Transportation Engineering, refType=null, unstructuredReference=ZHAO Xiangmo's team supported by the National Key Research and Development Program of China (2021YFB2501200). Research progress in testing and evaluation technologies for autonomous driving[J]. Journal of Traffic and Transportation Engineering, 2023, 23(6): 10-77., articleTitle=Research progress in testing and evaluation technologies for autonomous driving, refAbstract=null), Reference(id=1171830286221361871, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=孔祥时, journalName=自动驾驶道路测试行政许可法律问题研究, refType=null, unstructuredReference=孔祥时. 自动驾驶道路测试行政许可法律问题研究[D]. 北京: 中国人民公安大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1171830286275887824, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=KONG Xiangshi, journalName=Research on legal issues of administrative licensing for autonomous driving road testing, refType=null, unstructuredReference=KONG Xiangshi. Research on legal issues of administrative licensing for autonomous driving road testing[D]. Beijing: People's Public Security University of China, 2023., articleTitle=null, refAbstract=null), Reference(id=1171830286347190993, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2019, volume=32, issue=6, pageStart=1, pageEnd=19, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=朱冰, 张培兴, 赵健, journalName=中国公路学报, refType=null, unstructuredReference=朱冰, 张培兴, 赵健, 等. 基于场景的自动驾驶汽车虚拟测试研究进展[J]. 中国公路学报, 2019, 32(6): 1-19., articleTitle=基于场景的自动驾驶汽车虚拟测试研究进展, refAbstract=null), Reference(id=1171830286405911250, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2019, volume=32, issue=6, pageStart=1, pageEnd=19, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=ZHU Bing, ZHANG Peixing, ZHAO Jian, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=ZHU Bing, ZHANG Peixing, ZHAO Jian, et al. Review of scenario-based virtual validation methods for automated vehicles[J]. China Journal of Highway and Transport, 2019, 32(6): 1-19., articleTitle=Review of scenario-based virtual validation methods for automated vehicles, refAbstract=null), Reference(id=1171830286473020115, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2024, volume=22, issue=1, pageStart=1, pageEnd=24, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=张坤鹏, 常成, 王世璞, journalName=交通运输工程与信息学报, refType=null, unstructuredReference=张坤鹏, 常成, 王世璞, 等. 自动驾驶汽车仿真器综述:能力、挑战和发展方向[J]. 交通运输工程与信息学报, 2024, 22(1): 1-24., articleTitle=自动驾驶汽车仿真器综述:能力、挑战和发展方向, refAbstract=null), Reference(id=1171830286531740372, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2024, volume=22, issue=1, pageStart=1, pageEnd=24, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=ZHANG Kunpeng, CHANG Cheng, WANG Shipu, journalName=Journal of Transportation Systems Engineering and Information Technology, refType=null, unstructuredReference=ZHANG Kunpeng, CHANG Cheng, WANG Shipu, et al. Review of autonomous vehicle simulators: capabilities, challenges, and development directions[J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 22(1): 1-24., articleTitle=Review of autonomous vehicle simulators: capabilities, challenges, and development directions, refAbstract=null), Reference(id=1171830286586266325, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=491, pageEnd=496, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=JUNIETZ P, WACHENFELD W, KLONECKI K, journalName=2018 21st International Conference on Intelligent Transportation Systems (ITSC). IEEE, refType=null, unstructuredReference=JUNIETZ P, WACHENFELD W, KLONECKI K, et al. Evaluation of different approaches to address safety validation of automated driving[C]. 2018 21st International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2018: 491-496., articleTitle=Evaluation of different approaches to address safety validation of automated driving, refAbstract=null), Reference(id=1171830286640792278, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2017, volume=2626, issue=1, pageStart=25, pageEnd=33, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=KHAN M A, ECTORS W, BELLEMANS T, journalName=Transportation Research Record, refType=null, unstructuredReference=KHAN M A, ECTORS W, BELLEMANS T, et al. Unmanned aerial vehicle-based traffic analysis: Methodological framework for automated multivehicle trajectory extraction[J]. Transportation Research Record, 2017, 2626(1): 25-33., articleTitle=Unmanned aerial vehicle-based traffic analysis: Methodological framework for automated multivehicle trajectory extraction, refAbstract=null), Reference(id=1171830286720484055, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2020, volume=22, issue=5, pageStart=3190, pageEnd=3202, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=CHEN Xinqiang, LI Zhibin, YANG Yongsheng, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=CHEN Xinqiang, LI Zhibin, YANG Yongsheng, et al. High-resolution vehicle trajectory extraction and denoising from aerial videos[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(5): 3190-3202., articleTitle=High-resolution vehicle trajectory extraction and denoising from aerial videos, refAbstract=null), Reference(id=1171830286791787224, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2023, volume=14, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=YAN Xintao, ZOU Zhengxia, FENG Shuo, journalName=Nature Communications, refType=null, unstructuredReference=YAN Xintao, ZOU Zhengxia, FENG Shuo, et al. Learning naturalistic driving environment with statistical realism[J]. Nature Communications, 2023, 14(1): DOI:10.1038/s41467-023-37677-5., articleTitle=Learning naturalistic driving environment with statistical realism, refAbstract=null), Reference(id=1171830286846313177, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=1714, pageEnd=1721, url=null, language=null, rfNumber=[16], rfOrder=22, authorNames=WILLIAMS G, WAGENER N, GOLDFAIN B, journalName=2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, refType=null, unstructuredReference=WILLIAMS G, WAGENER N, GOLDFAIN B, et al. Information theoretic MPC for model-based reinforcement learning[C]. 2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2017: 1714-1721., articleTitle=Information theoretic MPC for model-based reinforcement learning, refAbstract=null), Reference(id=1171830286900839130, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2021, volume=23, issue=6, pageStart=4909, pageEnd=4926, url=null, language=null, rfNumber=[17], rfOrder=23, authorNames=KIRAN B R, SOBH I, TALPAERT V, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=KIRAN B R, SOBH I, TALPAERT V, et al. Deep reinforcement learning for autonomous driving: a survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(6): 4909-4926., articleTitle=Deep reinforcement learning for autonomous driving: a survey, refAbstract=null), Reference(id=1171830286951170779, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=[18], rfOrder=24, authorNames=DOSOVITSKIY A, ROS G, CODEVILLA F, journalName=Conference on Robot Learning. PMLR, refType=null, unstructuredReference=DOSOVITSKIY A, ROS G, CODEVILLA F, et al. CARLA: an open urban driving simulator[C]. Conference on Robot Learning. PMLR, 2017: 1-16., articleTitle=CARLA: an open urban driving simulator, refAbstract=null), Reference(id=1171830287014085340, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=25, authorNames=LI D, OKHRIN O, journalName=Transportation Research Part C: Emerging Technologies, refType=null, unstructuredReference=LI D, OKHRIN O. Modified DDPG car-following model with a real-world human driving experience with CARLA simulator[J]. Transportation Research Part C: Emerging Technologies, 2023, 147: DOI:10.1016/j.trc.2022.103987., articleTitle=Modified DDPG car-following model with a real-world human driving experience with CARLA simulator, refAbstract=null), Reference(id=1171830287068611293, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=1996, volume=4, issue=null, pageStart=237, pageEnd=285, url=null, language=null, rfNumber=[20], rfOrder=26, authorNames=KAELBLING L P, LITTMAN M L, MOORE A W, journalName=Journal of Artificial Intelligence Research, refType=null, unstructuredReference=KAELBLING L P, LITTMAN M L, MOORE A W. Reinforcement learning: a survey[J]. Journal of Artificial Intelligence Research, 1996, 4: 237-285., articleTitle=Reinforcement learning: a survey, refAbstract=null), Reference(id=1171830287123137246, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2015, volume=518, issue=7540, pageStart=529, pageEnd=533, url=null, language=null, rfNumber=[21], rfOrder=27, authorNames=MNIH V, KAVUKCUOGLU K, SILVER D, journalName=Nature, refType=null, unstructuredReference=MNIH V, KAVUKCUOGLU K, SILVER D, et al. Human-level control through deep reinforcement learning[J]. Nature, 2015, 518(7540): 529-533., articleTitle=Human-level control through deep reinforcement learning, refAbstract=null), Reference(id=1171830287177663199, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, doi=null, pmid=null, pmcid=null, year=2011, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=28, authorNames=DIJKSTRA A, journalName=En route to safer roads: how road structure and road classification can affect road safety, refType=null, unstructuredReference=DIJKSTRA A. En route to safer roads: how road structure and road classification can affect road safety[D]. Enschede: University of Twente, 2011., articleTitle=null, refAbstract=null)], funds=[Fund(id=1171830285244089024, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, awardId=72371251, language=CN, fundingSource=国家自然科学基金面上项目资助(72371251), fundOrder=null, country=null), Fund(id=1171830285327975105, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, awardId=2024JJ2080, language=CN, fundingSource=湖南省自然科学基金杰出青年项目资助(2024JJ2080), fundOrder=null, country=null), Fund(id=1171830285378306754, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, awardId=RCS2022K004, language=CN, fundingSource=轨道交通控制与安全国家重点实验室(北京交通大学)开放课题基金资助(RCS2022K004), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1171830282240967305, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, xref=1, ext=[AuthorCompanyExt(id=1171830282266133130, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China), AuthorCompanyExt(id=1171830282274521739, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282240967305, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中南大学 交通运输工程学院,湖南 长沙 410075)]), AuthorCompany(id=1171830282354213516, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, xref=2, ext=[AuthorCompanyExt(id=1171830282362602125, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282354213516, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China), AuthorCompanyExt(id=1171830282370990734, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, companyId=1171830282354213516, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学 轨道交通控制与安全国家重点实验室,北京 100044)])], figs=[ArticleFig(id=1171830284153569966, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.1, caption=Simulation platform framework, figureFileSmall=4Dg+NiyCp/i2eHE50ObZDA==, figureFileBig=wR1AqxvhYZnN55OzrI/vhg==, tableContent=null), ArticleFig(id=1171830284199707311, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图1, caption=仿真平台搭建框架, figureFileSmall=4Dg+NiyCp/i2eHE50ObZDA==, figureFileBig=wR1AqxvhYZnN55OzrI/vhg==, tableContent=null), ArticleFig(id=1171830284254233264, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.2, caption=State space, figureFileSmall=IEw1Vm7RfBe7c5gdX+2Ajg==, figureFileBig=gxjY6iaJcsXSSuvstdxatg==, tableContent=null), ArticleFig(id=1171830284300370609, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图2, caption=状态空间, figureFileSmall=IEw1Vm7RfBe7c5gdX+2Ajg==, figureFileBig=gxjY6iaJcsXSSuvstdxatg==, tableContent=null), ArticleFig(id=1171830284350702258, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.3, caption=Action space, figureFileSmall=bqBiVzAt3V2ZTAVl908zew==, figureFileBig=MD+08xnCHcMoQf2qMDIITw==, tableContent=null), ArticleFig(id=1171830284409422515, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图3, caption=动作空间, figureFileSmall=bqBiVzAt3V2ZTAVl908zew==, figureFileBig=MD+08xnCHcMoQf2qMDIITw==, tableContent=null), ArticleFig(id=1171830284455559860, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.4, caption=Training process of DQN, figureFileSmall=Xe06XYl6toYu+4G6qhtdrA==, figureFileBig=0j8a7IgXl4OH6LMlo+hYLw==, tableContent=null), ArticleFig(id=1171830284514280117, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图4, caption=DQN训练流程, figureFileSmall=Xe06XYl6toYu+4G6qhtdrA==, figureFileBig=0j8a7IgXl4OH6LMlo+hYLw==, tableContent=null), ArticleFig(id=1171830284568806070, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.5, caption=Cumulative reward during training, figureFileSmall=FxkCNaNJjFLdS5bv01GkHQ==, figureFileBig=dVljS9qwFcTPyGfKdYyQEQ==, tableContent=null), ArticleFig(id=1171830284627526327, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图5, caption=训练过程中的累积奖励, figureFileSmall=FxkCNaNJjFLdS5bv01GkHQ==, figureFileBig=dVljS9qwFcTPyGfKdYyQEQ==, tableContent=null), ArticleFig(id=1171830284686246584, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.6, caption=Vehicle collision process, figureFileSmall=59P1Yz85p86Y9VHLo7E5JA==, figureFileBig=Ohvsolvfr/K/ZkjD42UVRA==, tableContent=null), ArticleFig(id=1171830284732383929, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图6, caption=车辆碰撞过程, figureFileSmall=59P1Yz85p86Y9VHLo7E5JA==, figureFileBig=Ohvsolvfr/K/ZkjD42UVRA==, tableContent=null), ArticleFig(id=1171830284782715578, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Fig.7, caption=Vehicle initial position and grid number, figureFileSmall=o/TxtPMQIYsgvM4W6s8iXg==, figureFileBig=osWUvtn2wvZynIAv6sd6iA==, tableContent=null), ArticleFig(id=1171830284833047227, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=图7, caption=车辆起始位置与网格编号, figureFileSmall=o/TxtPMQIYsgvM4W6s8iXg==, figureFileBig=osWUvtn2wvZynIAv6sd6iA==, tableContent=null), ArticleFig(id=1171830284891767484, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Table 1, caption=

Simulation parameters

, figureFileSmall=null, figureFileBig=null, tableContent=
参数名称 数值
经验池容量 100 000
批量大小 64
目标值更新权重 0.001
学习率 0.000 5
网络更新时间步 4
奖励折扣因子 0.99
), ArticleFig(id=1171830284954682045, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=表1, caption=

仿真试验参数

, figureFileSmall=null, figureFileBig=null, tableContent=
参数名称 数值
经验池容量 100 000
批量大小 64
目标值更新权重 0.001
学习率 0.000 5
网络更新时间步 4
奖励折扣因子 0.99
), ArticleFig(id=1171830285021790910, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=EN, label=Table 2, caption=

Statistics of 500 random simulation results

, figureFileSmall=null, figureFileBig=null, tableContent=
网格
编号
网格试验
次数
碰撞成功
次数
碰撞成功
率/%
最长碰撞
时间/s
平均碰撞
时间/s
经向冲突
次数
平均经向冲
突能量/kJ
合流冲突
次数
平均合流冲
突能量/kJ
A1 34 26 76.47 192 137 18 223.08 8 59.77
A2 31 29 93.55 177 117 17 176.47 12 80.36
A3 34 28 82.35 170 117 23 193.93 5 76.29
A4 17 14 82.35 177 138 11 182.46 3 138.46
B1 12 11 91.67 114 137 11 248.74
B2 25 21 84.00 116 89 18 218.94 3 171.47
B3 61 48 78.69 201 133 34 202.54 14 123.51
B4 31 20 64.52 184 123 16 247.08 4 139.19
C1 30 26 86.67 201 139 25 274.67 1 204.57
C2 20 19 95.00 148 129 15 262.19 4 152.86
C3 58 25 43.10 162 131 19 191.85 6 101.97
C4 25 5 25.00 150 146 5 151.91
D1 21 12 57.14 190 123 10 301.79 2 172.65
D2 18 11 61.11 191 121 11 274.79
D3 59 11 22.92 159 121 6 217.23 5 75.77
D4 24 5 20.83 135 135 3 228.94 2 63.59
平均值 31.25 19.44 62.20 166.69 127.25 15.80 229.65 5.29 122.31
), ArticleFig(id=1171830285109871295, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1149738773480648933, language=CN, label=表2, caption=

500次随机仿真试验结果统计

, figureFileSmall=null, figureFileBig=null, tableContent=
网格
编号
网格试验
次数
碰撞成功
次数
碰撞成功
率/%
最长碰撞
时间/s
平均碰撞
时间/s
经向冲突
次数
平均经向冲
突能量/kJ
合流冲突
次数
平均合流冲
突能量/kJ
A1 34 26 76.47 192 137 18 223.08 8 59.77
A2 31 29 93.55 177 117 17 176.47 12 80.36
A3 34 28 82.35 170 117 23 193.93 5 76.29
A4 17 14 82.35 177 138 11 182.46 3 138.46
B1 12 11 91.67 114 137 11 248.74
B2 25 21 84.00 116 89 18 218.94 3 171.47
B3 61 48 78.69 201 133 34 202.54 14 123.51
B4 31 20 64.52 184 123 16 247.08 4 139.19
C1 30 26 86.67 201 139 25 274.67 1 204.57
C2 20 19 95.00 148 129 15 262.19 4 152.86
C3 58 25 43.10 162 131 19 191.85 6 101.97
C4 25 5 25.00 150 146 5 151.91
D1 21 12 57.14 190 123 10 301.79 2 172.65
D2 18 11 61.11 191 121 11 274.79
D3 59 11 22.92 159 121 6 217.23 5 75.77
D4 24 5 20.83 135 135 3 228.94 2 63.59
平均值 31.25 19.44 62.20 166.69 127.25 15.80 229.65 5.29 122.31
)], attaches=null, journal=Journal(id=1123942128916217864, delFlag=0, nameCn=中国安全科学学报, nameEn=China Safety Science Journal, nameHistory1=null, nameHistory2=null, issn=1003-3033, eissn=, cn=11-2865/X, coden=null, periodic=0, language=CN, oaType=0, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=fkqsFM6VKlHC4gCtS5XqTw==, journalPrice=null, startedYear=null, abbrevIsoEn=Chin Saf Sci J, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754269350027, createdBy=null, updatedBy=13701087609, firstLetterCn=C, firstLetterEn=C, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=fkqsFM6VKlHC4gCtS5XqTw==, picEn=SHn9HgqSxtJrOcAxqD++4Q==, jcr=null, cjcr=null, exts=[JournalExt(id=1159052918994595848, language=CN, name=中国安全科学学报, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.cssjj.com.cn/, createdTime=1754269350050, updatedTime=1754269350050, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.cssjj.com.cn/CN/column/item15.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""}), JournalExt(id=1159052919040733193, language=EN, name=China Safety Science Journal, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.cssjj.com.cn/EN/1003-3033/home.shtml, createdTime=1754269350061, updatedTime=1754269350061, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=https://synbioj.cip.com.cn/EN/column/column3.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031787341344770, websiteList=[Website(id=1148243202345263519, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN, language=CN, createTime=1751692112766, createBy=18614031015, updateTime=1753502583634, updateBy=18614031015, name=《中国安全科学学报》中文站点, tplId=1146099689490845704, title=中国安全科学学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148618794941046792, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=articleTextType, value=kx, createTime=1751781661020, updateTime=1751781661020, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794911686661, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=banner, value=null, createTime=1751781661012, updateTime=1751781661012, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794894909444, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1751781661008, updateTime=1751781661008, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794932658183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751781661017, updateTime=1751781661017, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794924269574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751781661015, updateTime=1751781661015, creator=18614031015, updator=18614031015)]), Website(id=1155836763751993353, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN, language=EN, createTime=1753502558893, createBy=18614031015, updateTime=1753524450387, updateBy=18614031015, name=《中国安全科学学报》英文站点, tplId=1146101810881728533, title=China Safety Science Journal, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155895925743669425, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=articleTextType, value=kx, createTime=1753516664205, updateTime=1753516664205, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925722697902, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=banner, value=null, createTime=1753516664200, updateTime=1753516664200, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925714309293, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1753516664198, updateTime=1753516664198, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925735280816, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753516664203, updateTime=1753516664203, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925731086511, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753516664202, updateTime=1753516664202, creator=18614031015, updator=18614031015)])], journalTitle=中国安全科学学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=China Safety Science Journal, journalPhotoCn=fkqsFM6VKlHC4gCtS5XqTw==, journalPhotoEn=SHn9HgqSxtJrOcAxqD++4Q==, journalFirstLetter=C, journalRecommend=null, journalNew=null, journalCollection=1, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2024.07.0247, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2024.07.0247, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/PDF/10.16265/j.cnki.issn1003-3033.2024.07.0247, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/PDF/10.16265/j.cnki.issn1003-3033.2024.07.0247, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
面向自动驾驶安全测试的碰撞场景构建与仿真分析
收藏切换
PDF下载
赵尧华 1 , 陈延展 1 , 郑亮 1, ** , 李树凯 2
中国安全科学学报 | 公共安全 2024,34(7): 211-218
收起
收藏切换
中国安全科学学报 | 公共安全 2024, 34(7): 211-218
面向自动驾驶安全测试的碰撞场景构建与仿真分析
全屏
赵尧华1 , 陈延展1, 郑亮1, ** , 李树凯2
作者信息
  • 1 中南大学 交通运输工程学院,湖南 长沙 410075
  • 2 北京交通大学 轨道交通控制与安全国家重点实验室,北京 100044
  • 赵尧华 (1999—),男,重庆人,硕士研究生,主要研究方向为自动驾驶危险测试场景构建与分析。E-mail:

    李树凯 教授

通讯作者:

** 郑亮(1984—),男,湖南衡阳人,博士,教授,主要从事智能交通系统仿真优化、交通系统数字孪生与虚拟仿真、路网交通状态估计与预测等方面的研究。E-mail:
Collision scenario construction and simulation analysis for autonomous driving safety testing
Yaohua ZHAO1 , Yanzhan CHEN1, Liang ZHENG1, ** , Shukai LI2
Affiliations
  • 1 School of Traffic and Transportation Engineering,Central South University,Changsha Hunan 410075,China
  • 2 State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China
出版时间: 2024-07-28 doi: 10.16265/j.cnki.issn1003-3033.2024.07.0247
文章导航
收藏切换

为减少自动驾驶车辆产生的交通事故,提高车辆在仿真环境中的安全测试效率,提出一种基于深度强化学习的自动驾驶碰撞测试场景构建方法。首先,通过设置状态、动作和奖励函数,将车辆的驾驶过程映射为马尔可夫决策过程;然后,基于搭建的仿真平台(CARLA-DRL)训练智能体完成车辆碰撞任务,生成碰撞测试场景;最后,随机进行500次碰撞仿真试验,根据智能体与自动驾驶车辆的相对距离,分析碰撞成功率、碰撞时间和冲突能量。结果表明:智能体生成符合车辆动力学的碰撞轨迹,能够构建精细化、多类型的碰撞测试场景;智能体与自动驾驶车辆的平均碰撞成功率为62.20%,平均碰撞时间为127.25 s,平均冲突能量值为175.98 kJ。该方法能够构建高频、高效和高危的自动驾驶车辆碰撞测试场景,增加仿真场景中偶发高风险场景的发生概率,提高自动驾驶车辆对于碰撞事故的安全测试效率。

自动驾驶  /  安全测试  /  深度强化学习  /  碰撞测试场景  /  仿真试验

To reduce traffic accidents caused by autonomous vehicles and improve the efficiency of vehicle safety testing in simulation environments,an autonomous driving collision test scenario construction method was proposed based on deep reinforcement learning. Firstly,the vehicle's driving process was mapped to a Markov decision process by setting the state,action,and reward functions. Then,the agent was trained to complete the vehicle collision task and generate the collision test scenarios based on the built simulation platform (CARLA-DRL). Finally,500 random collision simulation tests were conducted to analyze the collision success rate,collision time,and collision energy based on the relative distance between the agent and the autonomous vehicle. The results indicated that the agent generated collision trajectories that conformed to vehicle dynamics and could construct refined and multi-type collision test scenarios. The average collision success rate between the agent and the autonomous vehicle was 62.20%,the average collision time was 127.25 s,and the average collision energy value was 175.98 kJ. The proposed method can construct high-frequency,high-efficient,and high-risk autonomous driving vehicle collision test scenarios,increasing the probability of occasional high-risk scenarios in simulation scenarios and enhancing the efficiency of safety testing for autonomous vehicle collision incidents.

autonomous driving  /  safety testing  /  collision test scenarios  /  simulation experiment  /  deep reinforcement learning
赵尧华, 陈延展, 郑亮, 李树凯. 面向自动驾驶安全测试的碰撞场景构建与仿真分析. 中国安全科学学报, 2024 , 34 (7) : 211 -218 . DOI: 10.16265/j.cnki.issn1003-3033.2024.07.0247
Yaohua ZHAO, Yanzhan CHEN, Liang ZHENG, Shukai LI. Collision scenario construction and simulation analysis for autonomous driving safety testing[J]. China Safety Science Journal, 2024 , 34 (7) : 211 -218 . DOI: 10.16265/j.cnki.issn1003-3033.2024.07.0247
自动驾驶车辆的出现改变了传统的驾驶模式,其主要优点是能够满足驾驶安全性和出行的舒适性,自动驾驶具有从根本上改变人类交通方式的潜力,其巨大的发展前景引起国内外汽车制造商和科技公司的重视[1-3]。目前,高等级自动驾驶技术的研发瓶颈主要是自动驾驶车辆在生产前必须对车辆的驾驶安全性能进行测试和评估[3]。但是,驾驶环境的高维性和关键测试场景的稀疏性导致自动驾驶安全测试需要数亿英里[4],因此,自动驾驶安全测试场景的构建对加速自动驾驶测试效率和推动自动驾驶发展具有重要意义。
诸多学者研究了自动驾驶车辆的安全测试方法[5-7]。许多车辆安全标准为车辆的生产材料、结构设计和操作系统等规定了最低安全性能标准,但设计规范只针对车辆生产前各个部件的安全性能,缺少整车级的安全性能测试。自动驾驶实车测试是最直接的安全性能验证方法,众多汽车公司已经完成了实车道路测试。赵祥模等[8]对比了不同测试方法的优缺点,指出道路的测试结果虽然有较高参考意义,但测试结果和场景难以复现,且将具有潜在危险的车辆直接放入道路中,存在不可控的交通风险;孔祥时[9]分析了国内外自动驾驶测试场地,提出道路测试不仅面临着交通法规的约束,还需要昂贵的设备成本和时间成本。交通仿真软件的出现不仅提升了测试效率,而且降低了测试成本,但仿真软件中所包含的场景有限,难以表征出一些现实中极端但存在的关键测试场景,朱冰等[10]总结了测试场景的要素种类、数据来源和2种典型的场景构建方法,用以解决虚拟仿真场景不足的问题;张坤鹏等[11]系统分析了自动驾驶仿真软件的发展过程和未来挑战,认为将新兴技术与仿真软件结合,可解决仿真软件的场景局限性。
车辆碰撞是常见的交通安全事件,车辆的碰撞轨迹可作为车辆行驶过程中偶发高风险场景的有效表征[12]。精细化的车辆碰撞轨迹有助于识别事故的关键因素和还原事故过程,其特征参数可用于构建车辆碰撞的仿真场景。KHAN等[13]提出一个基于无人机视频自动处理的方法,用于提取特定路段多辆汽车的行驶轨迹,并将轨迹参数用于交通安全分析;CHEN Xinqiang等[14]通过目标检测技术和滤波器技术快速定位目标车辆,并利用映射算法完成对车辆轨迹的实时提取;YAN Xintao等[15]基于车辆事故数据和深度学习框架,学习车辆碰撞轨迹的特征,在CARLA软件平台中控制车辆生成类似的碰撞轨迹。虽然目标检测和目标跟踪算法可完成车辆轨迹提取,但天气情况、摄像头角度等会直接影响轨迹的精确度;而基于数据驱动的车辆碰撞轨迹生成方法需要大量的事故数据作为输入,但事故随机性和隐私保护导致被完整记录的车辆碰撞过程极少。
鉴于此,笔者拟通过强化学习的自主探索特性,提出一种无需数据驱动的车辆碰撞轨迹生成方法;并设置合理的深度强化学习训练模型,构建适用于自动驾驶汽车安全测试的仿真场景;根据500次碰撞仿真试验的结果,分析车辆间的碰撞成功率、碰撞时间和冲突能量,以期提高自动驾驶车辆在仿真环境中的安全测试效率。
深度强化学习不需要大量的标签数据来训练模型,其核心是训练智能体在一个复杂且不确定的环境中找到最优策略。适用于一些高维度、大规模和非线性的任务,如机器人控制和车辆控制。因此,深度强化学习已经被广泛用于自动驾驶车辆的研究领域[16-17]。交通仿真软件能够降低自动驾驶车辆的研究成本,而CARLA软件是目前自动驾驶研究中的主流仿真工具[18-19]。通过CARLA中逼真的车辆动力学模型和3D场景模拟现实驾驶环境,采用Python语言对CARLA进行二次开发,搭建基于CARLA仿真软件的深度强化学习训练平台(CARLA-DRL)。仿真平台搭建思路如图1所示。
仿真平台以CARLA为仿真环境,将用于碰撞测试场景生成的背景车辆作为智能体,学习任务为智能体与自动驾驶车辆产生碰撞。模型训练过程为:智能体通过各种传感器与仿真环境进行交互,从而获取当前 t时刻的观测状态 s t,再根据驾驶策略 π选择动作 a t,并在仿真环境中执行。当执行完一个动作时,智能体会从环境中得到新的观测状态 s t + 1和奖励 r t,其中, s t + 1用于支持下一个时间步的控制策略,奖励 r t作为当前动作的价值反馈,并以此更新网络参数。当进入终止状态时,环境给出当前获得的累计奖励,智能体进行下一轮训练。此外,为加速智能体的训练效率,将训练得到的信息集合 D ( t ) = ( s t a t r t s t + 1 )存储在经验回放池。最终,基于训练后的智能体构建出用于自动驾驶车辆安全测试的碰撞场景。
深度强化学习由智能体、环境、状态、动作和奖励组成[20]。马尔可夫决策过程(Markov Decision Processes,MDP)是深度强化学习的模型基础,其核心是状态空间 S、动作空间 A和奖励函数 R
状态空间是智能体通过传感器对周围环境的观测状态集合。在CARLA-DRL仿真平台中,使用其自带的定位系统,能够获得不同车辆的位置、速度、加速度等,状态空间如图2所示。因此,状态空间 S表示为:
S = { l r l l x r y r v x b v y b θ b v x a v y a θ a }
式中: l r l l为智能体与道路边缘的左、右距离,m; ( x r y r )为智能体和自动驾驶车辆的相对距离,m;( v x b v y b )为智能体的横、纵向速度,m/s; ( v x a v y a )为自动驾驶车辆的横、纵向速度,m/s; θ为车辆行驶方向与路沿的夹角,(°)。
车辆的操作主要由方向盘、油门和刹车决定,而方向盘、油门和刹车在不同状态下对应不同的控制动作,由此定义出控制车辆的离散动作,如图3所示。动作空间 A表示为:
A = { S l S r D 0 D 1 A 0 A 1 }
式中: S l S r分别为车辆左转或右转; D 0 D 1分别为车辆不制动或制动; A 0 A 1分别为车辆不加速或加速。
奖励函数定义了智能体的行为目标,是评估和调整智能体策略的关键。奖励函数的设置主要由训练任务决定,不同的奖励方式会影响智能体的探索策略和学习效果。训练任务是让智能体与自动驾驶车辆产生碰撞,所以将智能体的碰撞作为结束状态,并根据碰撞的目标给予奖励值 r c,同时对智能体第 i个时间步的速度大小给予奖励值 r i,奖励函数 r t表示如下:
r t = r c + r i
r c = + 12   000 - 10   000
r i = - 5 ( v x b ) 2 + ( v y b ) 2 < 20 k m / h - 1 ( v x b ) 2 + ( v y b ) 2 > 20 k m / h
式中: r c为智能体的碰撞奖励,当与自动驾驶车辆碰撞时,给予较大奖励值(+12 000),当与其他物体碰撞时,给予较大惩罚(-10 000),目的是引导智能体完成车辆碰撞任务,且符合正常驾驶规则; r i为智能体速度的奖励值; ( v x b ) 2 + ( v y b ) 2为智能体的速度。限制智能体速度的目的是防止训练出惰性智能体,避免驾驶策略陷入局部最优,该方法如果不约束速度,智能体最终会原地绕圈,以此逃避错误碰撞引起的较大惩罚。而 r c远大于 r i的目的是引导智能体将车辆碰撞作为首要学习任务。
智能体在和仿真环境交互时,由于是以离散动作来控制车辆,且CARLA-DRL具有高维度的驾驶环境,因此,采用DQN(Deep Q-learning)算法[21]。DQN算法通过构造Q-神经网络 Q ( s a θ )提升训练效率,让智能体在与环境不断交互的过程中学习到最优策略,并通过定时更新参数提升学习稳定性。完整的训练流程如图4所示。
基于CARLA-DRL平台进行智能体的训练和仿真分析试验,选择一条4车道的高速公路作为试验路段,利用CLRLA中提供的车辆传感器获取状态信息。将智能体和自动驾驶车辆的速度初始值设为0,行驶方向固定为同向行驶,并随机设置2类车的初始车道。首先,对智能体进行1 500次任务训练,分析累计奖励的变化情况、车辆碰撞测试场景类型和精细化程度;然后,在CARLA-DRL平台中进行500次随机仿真试验,根据智能体和自动驾驶车辆的相对距离分析碰撞成功率、碰撞时间和冲突能量。仿真试验参数见表1
智能体通过1 500轮的训练后,所获奖励值呈现为收敛状态,训练过程中的累计奖励收敛曲线如图5所示。滑动平均奖励值反映了智能体每50轮训练所获的平均奖励值;每回合累计奖励反映了智能体在单轮训练中的表现,可精确反映智能体的学习效果。在前400轮训练中,智能体获得的奖励值集中分布在 ( - 9   000 - 12   000 ),在400~1 000轮训练中,其奖励值集中分布在0左右,在1 200~1 500轮训练中,其奖励值集中分布在(9000,10 000)。这表明智能体能够不断根据每回合的奖励去调整驾驶策略,最终完成车辆碰撞任务。
车辆轨迹能完整记录车辆的行驶参数,同时,也反映出车辆的驾驶策略。训练后的智能体在CARLA-DRL平台中生成2种车辆碰撞轨迹,分别为侧碰和追尾,如图6所示。
图6可知:车辆碰撞轨迹具有以下特点:①有效性,智能体在驾驶过程中能够避免与其他物体碰撞,可完成与自动驾驶车辆的碰撞任务。②多样性,图6中车辆的碰撞轨迹各不相同,说明智能体的驾驶策略具有多样性,能多角度测试自动驾驶车辆的安全性能,而不是单一的固定模式。③精确性,生成的轨迹是连续变化的曲线,可精细化地记录不同时刻下车辆的行驶参数,精确的轨迹数据有利于还原事故场景。试验结果说明智能体能够生成多类型、精细化和符合车辆动力学特征的碰撞过程。多类型、精细化的碰撞过程有助于构建出自动驾驶碰撞测试场景,将高维复杂的驾驶环境转换成低维的碰撞测试场景,提高自动驾驶车辆的安全测试效率。
智能体与自动驾驶车辆的相对位置会影响碰撞成功率和碰撞时间,将智能体和测试车辆的位置限定在一定范围内,随机进行500次仿真试验。车辆的起始位置分布如图7a所示,智能体和测试车辆的起始位置基本覆盖了全部仿真道路。计算仿真试验开始时智能体与自动驾驶车辆的横向相对距离 x r与纵向相对距离 y r,设置一个4×4的网格,如图7b所示,仿真试验统计结果见表2。结果表明:单个网格最多仿真次数59(D3),最少仿真次数12(B1)。仿真试验累计碰撞成功311次,碰撞失败189次,平均碰撞成功率62.20%。
相对距离 ( x r y r )的增大会影响智能体的碰撞成功率。其中,最高碰撞成功率95.00%(C2),最低碰撞成功率20.83%(D4)。当 0 m < x r < 11.0   m时,yr的变化对碰撞成功率的影响较小;当 11.0 m < x r < 22.0   m时, y r的变化对碰撞成功率的影响较大;当 19.0 m < y r < 27.0   m时, x r的变化对碰撞成功率的影响较小;当 27.0 m < y r < 35.0   m时, x r的变化对碰撞成功率的影响较大。因此,最具鲁棒性的碰撞距离为 0 < x r < 11.0   m 19.0   m < y r < 31.0   m
安全事故发生的时间长短决定自动驾驶车辆安全测试的效率,由表2可知:成功碰撞的311次仿真试验中最长碰撞时间201s,平均碰撞时间为127.25s。16个网格区域的碰撞时间长短与相对距离 ( x r y r )之间的关系较弱,主要原因是智能体每一次仿真试验中都会采用不同的驾驶策略,使得车辆的行驶参数不断变化,削弱了碰撞时间与相对距离 ( x r y r )的相关性。试验结果表明:车辆碰撞事故发生时间相较于传统仿真软件急剧缩短,该场景能够增加偶发高风险场景的发生概率,将自动驾驶车辆的平均安全测试时间控制在3min以内。
交通冲突技术与代理安全评价方法是评估交通安全的主要方法,以潜在冲突能量(Potential Conflict Eenergy,PCE)[22]为评价指标。根据车辆的速度与质量,计算车辆碰撞前的冲突能量,以此刻画碰撞事故的严重性。试验中智能体生成的碰撞事故有追尾和侧碰,因此,冲突类型只考虑经向冲突与合流冲突。车辆的冲突能量计算式如下:
1) 经向冲突:
P C E = 1 2 ( m a v a 2 - m b v b 2 )
2) 合流冲突:
P C E = 1 4 ( m a v a 2 + m b v b 2 )
式中: m a m b分别为智能体和自动驾驶车辆的质量,kg; v a v b分别为智能体和自动驾驶车辆在碰撞前的瞬时速度,m/s。文中假设ma=mb=1 500 kg。
冲突能量的数值主要与智能体和自动驾驶车辆的速度和碰撞类型有关,由表2可知:全部仿真试验中累计经向冲突237次,累计合流冲突74次。其中,平均经向冲突能量值为229.65 kJ,平均经向冲突能量值为122.31 kJ,平均冲突能量值为175.98 kJ,证明智能体生成的车辆碰撞事故是一个高危的仿真测试场景。而相同网格内经向冲突能量的平均值更大,不同网格内的合流冲突能量波动较大,说明相对距离对车辆的合流冲突能量影响较大。同时,C1网格的平均冲突能量数值最大,可用于自动驾驶车辆的防撞性能测试,C4网格没有产生经向冲突,即C4网格内只会发生车辆侧碰事故;B1和D2没有产生合流冲突,即B1和D2网格内只会发生车辆追尾事故。因此,C4网格可用于自动驾驶车辆防侧撞性能测试;B1和D2网格可用于自动驾驶车辆防追尾性能测试。
1) 通过设置合理的状态空间、动作空间和奖励函数,将复杂的驾驶过程转变为可学习过程。同时,该方法不需要大量的车辆碰撞事故数据集作为输入,智能体训练过程中的累计奖励收敛且达到理论最大值,其生成的碰撞场景有侧碰和追尾2种类型,能够精细化的还原碰撞过程,车辆运动过程符合车辆动力学;训练后的智能体能够产生一个高效、高频和高危的碰撞测试场景,该场景可提高自动驾驶车辆对于碰撞事故的安全测试效率。
2) 搭建基于CARLA仿真软件的深度强化学习仿真平台(CARLA-DRL),该平台集成3D试验场景、逼真的车辆动力学模型和深度强化学习算法,可为智能体的训练和仿真试验分析提供高保真的试验环境。该平台也可用于其他仿真场景构建的研究,以及部署训练后的仿真测试场景。
3) 以智能体和自动驾驶车辆的初始位置为影响因素,得出车辆碰撞成功率最具鲁棒性的距离,该结果可最大化车辆碰撞事故的发生率,加速自动驾驶车辆对于碰撞事故的仿真过程。
  • 国家自然科学基金面上项目资助(72371251)
  • 湖南省自然科学基金杰出青年项目资助(2024JJ2080)
  • 轨道交通控制与安全国家重点实验室(北京交通大学)开放课题基金资助(RCS2022K004)
参考文献 引证文献
排序方式:
[1]
KENDALL A, HAWKE J, JANZ D, et al. Learning to drive in a day[C]. 2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019: 8248-8254.
[2]
李瑞敏, 戴晶辰. 自动驾驶影响下的出行行为研究综述[J]. 交通运输工程学报, 2022, 22(3): 41-54.
LI Ruimin, DAI Jingchen. Review on impact of autonomous driving on travel behaviors[J]. Journal of Traffic and Transportation Engineering, 2022, 22(3): 41-54.
[3]
FENG Shuo, SUN Haowei, YAN Xintao, et al. Dense reinforcement learning for safety validation of autonomous vehicles[J]. Nature, 2023, 615(7953): 620-627.
[4]
NIDHI K, SUSAN M P. Driving to safety: how many miles of driving would it take to demonstrate autonomous vehicle reliability[J]. Transportation Research Part A: Policy and Practice, 2016, 94: 182-193.
[5]
邓伟文, 李江坤, 任秉韬, 等. 面向自动驾驶的仿真场景自动生成方法综述[J]. 中国公路学报, 2022, 35(1): 316-333.
DENG Weiwen, LI Jiangkun, REN Bingtao, et al. A survey on automatic simulation scenario generation methods for autonomous driving[J]. China Journal of Highway and Transport, 2022, 35(1): 316-333.
[6]
余荣杰, 田野, 孙剑. 高等级自动驾驶汽车虚拟测试:研究进展与前沿[J]. 中国公路学报, 2020, 33(11): 125-138.
YU Rongjie, TIAN Ye, SUN Jian. Highly automated vehicle virtual testing: a review of recent developments and research frontiers[J]. China Journal of Highway and Transport, 2020, 33(11): 125-138.
[7]
FENG Shuo, YAN Xintao, SUN Haowei, et al. Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environment[J]. Nature Communications, 2021, 12(1): DOI:10.1038/s41467-021-21007-8.
[8]
赵祥模国家重点研发计划项目(2021YFB2501200)团队. 自动驾驶测试与评价技术研究进展[J]. 交通运输工程学报, 2023, 23(6): 10-77.
ZHAO Xiangmo's team supported by the National Key Research and Development Program of China (2021YFB2501200). Research progress in testing and evaluation technologies for autonomous driving[J]. Journal of Traffic and Transportation Engineering, 2023, 23(6): 10-77.
[9]
孔祥时. 自动驾驶道路测试行政许可法律问题研究[D]. 北京: 中国人民公安大学, 2023.
KONG Xiangshi. Research on legal issues of administrative licensing for autonomous driving road testing[D]. Beijing: People's Public Security University of China, 2023.
[10]
朱冰, 张培兴, 赵健, 等. 基于场景的自动驾驶汽车虚拟测试研究进展[J]. 中国公路学报, 2019, 32(6): 1-19.
ZHU Bing, ZHANG Peixing, ZHAO Jian, et al. Review of scenario-based virtual validation methods for automated vehicles[J]. China Journal of Highway and Transport, 2019, 32(6): 1-19.
[11]
张坤鹏, 常成, 王世璞, 等. 自动驾驶汽车仿真器综述:能力、挑战和发展方向[J]. 交通运输工程与信息学报, 2024, 22(1): 1-24.
ZHANG Kunpeng, CHANG Cheng, WANG Shipu, et al. Review of autonomous vehicle simulators: capabilities, challenges, and development directions[J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 22(1): 1-24.
[12]
JUNIETZ P, WACHENFELD W, KLONECKI K, et al. Evaluation of different approaches to address safety validation of automated driving[C]. 2018 21st International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2018: 491-496.
[13]
KHAN M A, ECTORS W, BELLEMANS T, et al. Unmanned aerial vehicle-based traffic analysis: Methodological framework for automated multivehicle trajectory extraction[J]. Transportation Research Record, 2017, 2626(1): 25-33.
[14]
CHEN Xinqiang, LI Zhibin, YANG Yongsheng, et al. High-resolution vehicle trajectory extraction and denoising from aerial videos[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(5): 3190-3202.
[15]
YAN Xintao, ZOU Zhengxia, FENG Shuo, et al. Learning naturalistic driving environment with statistical realism[J]. Nature Communications, 2023, 14(1): DOI:10.1038/s41467-023-37677-5.
[16]
WILLIAMS G, WAGENER N, GOLDFAIN B, et al. Information theoretic MPC for model-based reinforcement learning[C]. 2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2017: 1714-1721.
[17]
KIRAN B R, SOBH I, TALPAERT V, et al. Deep reinforcement learning for autonomous driving: a survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(6): 4909-4926.
[18]
DOSOVITSKIY A, ROS G, CODEVILLA F, et al. CARLA: an open urban driving simulator[C]. Conference on Robot Learning. PMLR, 2017: 1-16.
[19]
LI D, OKHRIN O. Modified DDPG car-following model with a real-world human driving experience with CARLA simulator[J]. Transportation Research Part C: Emerging Technologies, 2023, 147: DOI:10.1016/j.trc.2022.103987.
[20]
KAELBLING L P, LITTMAN M L, MOORE A W. Reinforcement learning: a survey[J]. Journal of Artificial Intelligence Research, 1996, 4: 237-285.
[21]
MNIH V, KAVUKCUOGLU K, SILVER D, et al. Human-level control through deep reinforcement learning[J]. Nature, 2015, 518(7540): 529-533.
[22]
DIJKSTRA A. En route to safer roads: how road structure and road classification can affect road safety[D]. Enschede: University of Twente, 2011.
2024年第34卷第7期
PDF下载
355
140
引用本文
BibTeX
文章信息
doi: 10.16265/j.cnki.issn1003-3033.2024.07.0247
  • 接收时间:2024-01-14
  • 首发时间:2025-07-09
  • 出版时间:2024-07-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-01-14
  • 修回日期:2024-04-18
基金
国家自然科学基金面上项目资助(72371251)
湖南省自然科学基金杰出青年项目资助(2024JJ2080)
轨道交通控制与安全国家重点实验室(北京交通大学)开放课题基金资助(RCS2022K004)
作者信息
    1 中南大学 交通运输工程学院,湖南 长沙 410075
    2 北京交通大学 轨道交通控制与安全国家重点实验室,北京 100044

通讯作者:

** 郑亮(1984—),男,湖南衡阳人,博士,教授,主要从事智能交通系统仿真优化、交通系统数字孪生与虚拟仿真、路网交通状态估计与预测等方面的研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2024.07.0247
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏