Article(id=1207343628578562713, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1207343627223802520, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2405917, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1722873600000, receivedDateStr=2024-08-06, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1765782752412, onlineDateStr=2025-12-15, pubDate=1750176000000, pubDateStr=2025-06-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1765782752412, onlineIssueDateStr=2025-12-15, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1765782752412, creator=13701087609, updateTime=1765782752412, updator=13701087609, issue=Issue{id=1207343627223802520, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='17', pageStart='7023', pageEnd='7453', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1765782752085, creator=13701087609, updateTime=1765783816840, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1207348093192872694, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1207343627223802520, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1207348093192872695, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1207343627223802520, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=7373, endPage=7379, ext={EN=ArticleExt(id=1207343630109483680, articleId=1207343628578562713, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Cross-device Tracking of Vehicle Trajectory Based on Radar Data, columnId=1207343630017208991, journalTitle=Science Technology and Engineering, columnName=Papers-Traffics and Transportations, runingTitle=null, highlight=null, articleAbstract=

High-precision vehicle trajectory data is crucial for the realization of intelligent transportation systems. However, existing vehicle trajectory sensing technologies are limited by the range of data collection, making it challenging to obtain full-period and full-area vehicle trajectory data, which cannot meet the demands for trajectory tracking accuracy and real-time performance in practical applications. Considering the characteristics of vehicle trajectory data across radar scenarios, a cross-device vehicle trajectory tracking method was proposed based on radar data. Firstly, trajectory data was filtered based on the lower bound of the confidence interval, and the position and velocity of vehicle trajectories were smoothed and denoised using Kalman filtering. Next, the trajectory timestamp, position coordinates, speed, direction, and lane number from the radar detection area were used as model inputs, while the position information of non-overlapping areas was used as output to construct a non-overlapping area trajectory prediction model based on long short-term memory (LSTM). Subsequently, a fast dynamic time warping (DTW) algorithm based on trajectory condition constraints and search area limitations was proposed to compute trajectory similarity, enabling cross-radar device vehicle trajectory tracking. Finally, verification was conducted using overlapping and non-overlapping scenario data from radar detection on highways. Experimental results show that the trajectory tracking accuracy in overlapping scenarios reaches 92.2%, and the trajectory matching accuracy in non-overlapping areas reaches 90.3%.

, correspAuthors=Jin-yin XUE, 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=Xin YOU, Jin-yin XUE, Bei-hai ZHANG, Yu-hang GAO, Xiang-li TIAN, Jian-dong ZHAO), CN=ArticleExt(id=1207343640146453414, articleId=1207343628578562713, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于跨雷达数据的车辆轨迹跟踪, columnId=1156262730664366426, journalTitle=科学技术与工程, columnName=论文·交通运输, runingTitle=null, highlight=null, articleAbstract=

高精度车辆轨迹数据对于实现智慧交通具有重要意义,然而现有的车辆轨迹感知技术受到采集范围的限制,难以获得全时段全区域的车辆轨迹数据,无法满足实际应用中对车辆轨迹跟踪精度以及实时性的要求。考虑到跨雷达场景下的车辆轨迹数据特征,提出一种基于雷达数据的跨设备车辆轨迹跟踪方法。首先,根据数据特点应用置信区间下限筛选轨迹数据,并通过卡尔曼滤波对车辆轨迹的位置和速度分别进行平滑和降噪处理。其次,将雷达探测区域的车辆轨迹时间戳、位置坐标、行驶速度、行驶方向以及车道编号作为模型输入,非重叠区域的位置信息作为输出,构建了基于长短期记忆(long short term memory, LSTM)的非重叠区域轨迹预测模型。然后,提出基于轨迹条件约束和搜索区域限制的快速动态时间规整算法(dynamic time warping, DTW)用以计算轨迹相似度,从而实现跨雷达设备车辆轨迹跟踪。最后,以高速公路上跨雷达检测的重叠场景和非重叠场景数据为例进行验证,实验结果表明,重叠场景下车辆轨迹跟踪准确度达到92.2%,非重叠区域车辆轨迹匹配正确率达到90.3%。

, correspAuthors=薛金银, authorNote=null, correspAuthorsNote=
*薛金银(1976—),男,汉族,陕西韩城人,博士,副研究员。研究方向:智能交通。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=PD3sc0wFiuKCcCRVfHSmSA==, magXml=d9LW9SE2vhRGjCWHtSRdow==, pdfUrl=null, pdf=xhzzpe1GPJkQS1+LcDL/HQ==, pdfFileSize=4801203, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Ph+AAQ6r/VbS04XkHrBG1A==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=dEtBLGHyRntfF+YFk8i5gQ==, mapNumber=null, authorCompany=null, fund=null, authors=

尤鑫(1986—),男,汉族,辽宁沈阳人,博士,高级工程师。研究方向:智能交通。E-mail:

, authorsList=尤鑫, 薛金银, 张北海, 高宇航, 田向丽, 赵建东)}, authors=[Author(id=1207400964185760290, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=youxin@ktetc.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1207400964319978025, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400964185760290, language=EN, stringName=Xin YOU, firstName=Xin, middleName=null, lastName=YOU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207400964391281195, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400964185760290, language=CN, stringName=尤鑫, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 北京速通科技有限公司, 北京 100161, bio={"content":"

尤鑫(1986—),男,汉族,辽宁沈阳人,博士,高级工程师。研究方向:智能交通。E-mail:

"}, bioImg=null, bioContent=

尤鑫(1986—),男,汉族,辽宁沈阳人,博士,高级工程师。研究方向:智能交通。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207400963892158984, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=1, ext=[AuthorCompanyExt(id=1207400963900547593, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China), AuthorCompanyExt(id=1207400963992822288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 北京速通科技有限公司, 北京 100161)])]), Author(id=1207400964508721716, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=xuejinyin@ktetc.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1207400964600996413, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400964508721716, language=EN, stringName=Jin-yin XUE, firstName=Jin-yin, middleName=null, lastName=XUE, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207400964701659716, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400964508721716, language=CN, stringName=薛金银, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 北京速通科技有限公司, 北京 100161, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207400963892158984, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=1, ext=[AuthorCompanyExt(id=1207400963900547593, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China), AuthorCompanyExt(id=1207400963992822288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 北京速通科技有限公司, 北京 100161)])]), Author(id=1207400964798128712, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, orderNo=2, 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=1207400964911374927, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400964798128712, language=EN, stringName=Bei-hai ZHANG, firstName=Bei-hai, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207400964999455315, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400964798128712, language=CN, stringName=张北海, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 北京速通科技有限公司, 北京 100161, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207400963892158984, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=1, ext=[AuthorCompanyExt(id=1207400963900547593, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China), AuthorCompanyExt(id=1207400963992822288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 北京速通科技有限公司, 北京 100161)])]), Author(id=1207400965091730007, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, 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=1207400965213364830, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400965091730007, language=EN, stringName=Yu-hang GAO, firstName=Yu-hang, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207400965297250912, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400965091730007, language=CN, stringName=高宇航, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 北京交通大学交通运输学院, 北京 100044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207400964080902680, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=2, ext=[AuthorCompanyExt(id=1207400964089291288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China), AuthorCompanyExt(id=1207400964097679898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学交通运输学院, 北京 100044)])]), Author(id=1207400965376942695, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, orderNo=4, 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=1207400965506966127, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400965376942695, language=EN, stringName=Xiang-li TIAN, firstName=Xiang-li, middleName=null, lastName=TIAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207400965616018037, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400965376942695, language=CN, stringName=田向丽, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 北京交通大学交通运输学院, 北京 100044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207400964080902680, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=2, ext=[AuthorCompanyExt(id=1207400964089291288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China), AuthorCompanyExt(id=1207400964097679898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学交通运输学院, 北京 100044)])]), Author(id=1207400966903669377, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, orderNo=5, 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=1207400967033692807, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400966903669377, language=EN, stringName=Jian-dong ZHAO, firstName=Jian-dong, middleName=null, lastName=ZHAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207400967142744715, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, authorId=1207400966903669377, language=CN, stringName=赵建东, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 北京交通大学交通运输学院, 北京 100044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207400964080902680, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=2, ext=[AuthorCompanyExt(id=1207400964089291288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China), AuthorCompanyExt(id=1207400964097679898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学交通运输学院, 北京 100044)])])], keywords=[Keyword(id=1207400967327294104, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, orderNo=1, keyword=highway), Keyword(id=1207400967440540319, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, orderNo=2, keyword=millimeter-wave radar), Keyword(id=1207400967507649188, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, orderNo=3, keyword=dynamic time warping), Keyword(id=1207400967591535273, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, orderNo=4, keyword=trajectory tracking), Keyword(id=1207400967700587183, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, orderNo=5, keyword=LSTM), Keyword(id=1207400967792861876, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, orderNo=1, keyword=高速公路), Keyword(id=1207400967880942264, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, orderNo=2, keyword=毫米波雷达), Keyword(id=1207400968002577085, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, orderNo=3, keyword=动态时间规整算法), Keyword(id=1207400968107434692, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, orderNo=4, keyword=轨迹跟踪), Keyword(id=1207400968199709384, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, orderNo=5, keyword=LSTM)], refs=[Reference(id=1207400972633088838, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2021, volume=3, issue=2, pageStart=141, pageEnd=157, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Dhatbale R, Chilukuri B R, journalName=Journal of Big Data Analytics in Transportation, refType=null, unstructuredReference=Dhatbale R, Chilukuri B R. Deep learning techniques for vehicle trajectory extraction in mixed traffic[J]. Journal of Big Data Analytics in Transportation, 2021, 3(2): 141-157., articleTitle=Deep learning techniques for vehicle trajectory extraction in mixed traffic, refAbstract=null), Reference(id=1207400972838609737, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2019, volume=129, issue=null, pageStart=156, pageEnd=169, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=Chen T, Shi X, Wong Y D, journalName=Accident Analysis & Prevention, refType=null, unstructuredReference=Chen T, Shi X, Wong Y D. Key feature selection and risk prediction for lane-changing behaviors based on vehicles’ trajectory data[J]. Accident Analysis & Prevention, 2019, 129: 156-169., articleTitle=Key feature selection and risk prediction for lane-changing behaviors based on vehicles’ trajectory data, refAbstract=null), Reference(id=1207400972943467342, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=2118, pageEnd=2125, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=Krajewski R, Bock J, Kloeker L, journalName=null, refType=null, unstructuredReference=Krajewski R, Bock J, Kloeker L, et al. The highd dataset:a drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems[C]//2018 21st International Conference on Intelligent Transportation Systems (ITSC). New York: IEEE, 2018: 2118-2125., articleTitle=The highd dataset:a drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems[C], refAbstract=null), Reference(id=1207400973069296464, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2021, volume=39, issue=2, pageStart=61, pageEnd=69, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=冯汝怡, 李志斌, 吴启范, journalName=交通信息与安全, refType=null, unstructuredReference=冯汝怡, 李志斌, 吴启范, 等. 航拍视频车辆检测目标关联与时空轨迹匹配[J]. 交通信息与安全, 2021, 39(2): 61-69, 77., articleTitle=航拍视频车辆检测目标关联与时空轨迹匹配, refAbstract=null), Reference(id=1207400973174154069, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2021, volume=39, issue=2, pageStart=61, pageEnd=69, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=Feng Ruyi, Li Zhibin, Wu Qifan, journalName=Journal of Transport Information and Safety, refType=null, unstructuredReference=Feng Ruyi, Li Zhibin, Wu Qifan, et al. Association of vehicle object detection and the time-space trajectory matching from aerial videos[J]. Journal of Transport Information and Safety, 2021, 39(2): 61-69, 77., articleTitle=Association of vehicle object detection and the time-space trajectory matching from aerial videos, refAbstract=null), Reference(id=1207400973266428761, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=12, pageStart=181, pageEnd=192, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=王俊骅, 宋昊, 景强, journalName=中国公路学报, refType=null, unstructuredReference=王俊骅, 宋昊, 景强, 等. 基于毫米波雷达组群的全域车辆轨迹检测技术方法[J]. 中国公路学报, 2022, 35(12): 181-192., articleTitle=基于毫米波雷达组群的全域车辆轨迹检测技术方法, refAbstract=null), Reference(id=1207400973341926237, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2022, volume=35, issue=12, pageStart=181, pageEnd=192, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=Wang Junhua, Song Hao, Jing Qiang, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=Wang Junhua, Song Hao, Jing Qiang, et al. Road-range tracking of vehicle trajectories based on millimeter-wave radar[J]. China Journal of Highway and Transport, 2022, 35(12): 181-192., articleTitle=Road-range tracking of vehicle trajectories based on millimeter-wave radar, refAbstract=null), Reference(id=1207400973430006625, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=11, pageStart=4874, pageEnd=4880, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=顾思思, 常世新, 韩明敏, journalName=科学技术与工程, refType=null, unstructuredReference=顾思思, 常世新, 韩明敏, 等. 毫米波雷达数据驱动的桥梁挠度预测[J]. 科学技术与工程, 2023, 23(11): 4874-4880., articleTitle=毫米波雷达数据驱动的桥梁挠度预测, refAbstract=null), Reference(id=1207400973518087012, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=11, pageStart=4874, pageEnd=4880, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=Gu Sisi, Chang Shixin, Han Mingmin, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Gu Sisi, Chang Shixin, Han Mingmin, et al. Bridge deflection prediction driven by millimeter wave radar data[J]. Science Technology and Engineering, 2023, 23(11): 4874-4880., articleTitle=Bridge deflection prediction driven by millimeter wave radar data, refAbstract=null), Reference(id=1207400973639721832, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2020, volume=62, issue=4, pageStart=1279, pageEnd=1312, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=Lu Q, Wang R, Yang B, journalName=Knowledge and Information Systems, refType=null, unstructuredReference=Lu Q, Wang R, Yang B, et al. Trajectory splicing[J]. Knowledge and Information Systems, 2020, 62(4): 1279-1312., articleTitle=Trajectory splicing, refAbstract=null), Reference(id=1207400973731996523, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2020, volume=29, issue=null, pageStart=5191, pageEnd=5205, url=null, language=null, rfNumber=[8], rfOrder=10, authorNames=He Y, Wei X, Hong X, journalName=IEEE Transactions on Image Processing, refType=null, unstructuredReference=He Y, Wei X, Hong X, et al. Multi-target multi-camera tracking by tracklet-to-target assignment[J]. IEEE Transactions on Image Processing, 2020, 29: 5191-5205., articleTitle=Multi-target multi-camera tracking by tracklet-to-target assignment, refAbstract=null), Reference(id=1207400973849437039, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=24, pageStart=88, pageEnd=97, url=null, language=null, rfNumber=[9], rfOrder=11, authorNames=苟铃滔, 宋焕生, 张朝阳, journalName=计算机工程与应用, refType=null, unstructuredReference=苟铃滔, 宋焕生, 张朝阳, 等. 融合时空特征的隧道场景跨相机车辆实时跟踪方法[J]. 计算机工程与应用, 2023, 59(24): 88-97., articleTitle=融合时空特征的隧道场景跨相机车辆实时跟踪方法, refAbstract=null), Reference(id=1207400973929128817, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2023, volume=59, issue=24, pageStart=88, pageEnd=97, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=Gou Lingtao, Song Huansheng, Zhang Chaoyang, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=Gou Lingtao, Song Huansheng, Zhang Chaoyang, et al. Real-time cross-camera vehicle tracking method for tunnel scenes by fusing spatiotemporal features[J]. Computer Engineering and Applications, 2023, 59(24): 88-97., articleTitle=Real-time cross-camera vehicle tracking method for tunnel scenes by fusing spatiotemporal features, refAbstract=null), Reference(id=1207400974122066804, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2012, volume=61, issue=3, pageStart=959, pageEnd=970, url=null, language=null, rfNumber=[10], rfOrder=13, authorNames=Felguera-Martín D, González-Partida J T, Almorox-González P, journalName=IEEE Transactions on Vehicular Technology, refType=null, unstructuredReference=Felguera-Martín D, González-Partida J T, Almorox-González P, et al. Vehicular traffic surveillance and road lane detection using radar interferometry[J]. IEEE Transactions on Vehicular Technology, 2012, 61(3): 959-970., articleTitle=Vehicular traffic surveillance and road lane detection using radar interferometry, refAbstract=null), Reference(id=1207400974231118713, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=15, pageStart=6654, pageEnd=6660, url=null, language=null, rfNumber=[11], rfOrder=14, authorNames=牛大伟, 李一贤, 崔玮, journalName=科学技术与工程, refType=null, unstructuredReference=牛大伟, 李一贤, 崔玮, 等. 基于超距雷达数据的车辆跟驰行为分析及建模[J]. 科学技术与工程, 2023, 23(15): 6654-6660., articleTitle=基于超距雷达数据的车辆跟驰行为分析及建模, refAbstract=null), Reference(id=1207400974306616189, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=15, pageStart=6654, pageEnd=6660, url=null, language=null, rfNumber=[11], rfOrder=15, authorNames=Niu Dawei, Li Yixian, Cui Wei, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Niu Dawei, Li Yixian, Cui Wei, et al. Analysis and modeling of high speed car-following behavior based on ultra-range radar data[J]. Science Technology and Engineering, 2023, 23(15): 6654-6660., articleTitle=Analysis and modeling of high speed car-following behavior based on ultra-range radar data, refAbstract=null), Reference(id=1207400974382113662, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2020, volume=33, issue=2, pageStart=146, pageEnd=157, url=null, language=null, rfNumber=[12], rfOrder=16, authorNames=李颖, 赵莉, 赵祥模, journalName=中国公路学报, refType=null, unstructuredReference=李颖, 赵莉, 赵祥模, 等. 基于大货车GPS数据的轨迹相似性度量有效性研究[J]. 中国公路学报, 2020, 33(2): 146-157., articleTitle=基于大货车GPS数据的轨迹相似性度量有效性研究, refAbstract=null), Reference(id=1207400974495359872, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2020, volume=33, issue=2, pageStart=146, pageEnd=157, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=Li Ying, Zhao Li, Zhao Xiangmo, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=Li Ying, Zhao Li, Zhao Xiangmo, et al. Effectiveness of trajectory similarity measures based on truck GPS data[J]. China Journal of Highway and Transport, 2020, 33(2): 146-157., articleTitle=Effectiveness of trajectory similarity measures based on truck GPS data, refAbstract=null), Reference(id=1207400974583440258, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=353, pageEnd=359, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=Altché F, de La Fortelle A, journalName=2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), refType=null, unstructuredReference=Altché F, de La Fortelle A. An LSTM network for highway trajectory prediction[C]// 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC). New York: IEEE, 2017: 353-359., articleTitle=An LSTM network for highway trajectory prediction, refAbstract=null)], funds=[Fund(id=1207400972310127423, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, awardId=2022ZD0115605, language=CN, fundingSource=国家重点研发计划(2022ZD0115605), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1207400963892158984, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=1, ext=[AuthorCompanyExt(id=1207400963900547593, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China), AuthorCompanyExt(id=1207400963992822288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400963892158984, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 北京速通科技有限公司, 北京 100161)]), AuthorCompany(id=1207400964080902680, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, xref=2, ext=[AuthorCompanyExt(id=1207400964089291288, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China), AuthorCompanyExt(id=1207400964097679898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, companyId=1207400964080902680, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学交通运输学院, 北京 100044)])], figs=[ArticleFig(id=1207400968417813203, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.1, caption=Comparison of confidence filtering for radar1, figureFileSmall=VvX8ZzgPvzUI2w9axV1i0Q==, figureFileBig=o3HugOuW0deHKI/lHjg7sQ==, tableContent=null), ArticleFig(id=1207400968543642328, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图1, caption=radar1数据置信筛选对比, figureFileSmall=VvX8ZzgPvzUI2w9axV1i0Q==, figureFileBig=o3HugOuW0deHKI/lHjg7sQ==, tableContent=null), ArticleFig(id=1207400968656888539, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.2, caption=Position smoothing effect, figureFileSmall=WTejUiE1V56ks6dcDGDYZQ==, figureFileBig=uxsgb8Tp/NN86gs9RZL3DA==, tableContent=null), ArticleFig(id=1207400968770134752, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图2, caption=位置平滑效果, figureFileSmall=WTejUiE1V56ks6dcDGDYZQ==, figureFileBig=uxsgb8Tp/NN86gs9RZL3DA==, tableContent=null), ArticleFig(id=1207400968870798054, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.3, caption=Velocity smoothing effect, figureFileSmall=Or/DncL8Vyf2z37E+/DCVQ==, figureFileBig=UhQIN49Ni3mcYHBBdKrt+g==, tableContent=null), ArticleFig(id=1207400969000821483, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图3, caption=速度平滑效果, figureFileSmall=Or/DncL8Vyf2z37E+/DCVQ==, figureFileBig=UhQIN49Ni3mcYHBBdKrt+g==, tableContent=null), ArticleFig(id=1207400969097290480, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.4, caption=Optimal alignment path of different radar trajectory sequences, figureFileSmall=H0Mr6X0drO3kEaVvlTzh2g==, figureFileBig=lOX7mm2QV/fKT2mnrBXZuA==, tableContent=null), ArticleFig(id=1207400969206342389, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图4, caption=不同雷达轨迹序列的最优对齐路径, figureFileSmall=H0Mr6X0drO3kEaVvlTzh2g==, figureFileBig=lOX7mm2QV/fKT2mnrBXZuA==, tableContent=null), ArticleFig(id=1207400969315394294, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.5, caption=RR- DTW path search strategy, figureFileSmall=pNvny91mWNHRI7CVws70iw==, figureFileBig=9u4E3CQ1BGkxzMSH28gBKw==, tableContent=null), ArticleFig(id=1207400969382503162, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图5, caption=RR-DTW路径搜索策略, figureFileSmall=pNvny91mWNHRI7CVws70iw==, figureFileBig=9u4E3CQ1BGkxzMSH28gBKw==, tableContent=null), ArticleFig(id=1207400969453806334, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.6, caption=Vehicle trajectory prediction model based on LSTM, figureFileSmall=KEgQM8HZltve1AuvckhRDw==, figureFileBig=Ch6TGFvjm99IETIM3QSiow==, tableContent=null), ArticleFig(id=1207400969546081027, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图6, caption=基于LSTM的车辆轨迹预测模型, figureFileSmall=KEgQM8HZltve1AuvckhRDw==, figureFileBig=Ch6TGFvjm99IETIM3QSiow==, tableContent=null), ArticleFig(id=1207400969684493062, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.7, caption=Results of overlapping scene trajectory tracking, figureFileSmall=SY01XjeXs8JGSZ6TdBoUEQ==, figureFileBig=bmkHvW6IGwHrCPcuPraQAQ==, tableContent=null), ArticleFig(id=1207400969852265229, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图7, caption=重叠场景轨迹跟踪结果, figureFileSmall=SY01XjeXs8JGSZ6TdBoUEQ==, figureFileBig=bmkHvW6IGwHrCPcuPraQAQ==, tableContent=null), ArticleFig(id=1207400969961317137, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Fig.8, caption=Non-overlapping scenario trajectory tracking results (radar1,radar2), figureFileSmall=ObMkn9OXJWIMm4DzilvwQg==, figureFileBig=mr3CVrb7i/AoKFHi37Ow3w==, tableContent=null), ArticleFig(id=1207400970049397524, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=图8, caption=非重叠场景轨迹跟踪结果(radar1,radar2), figureFileSmall=ObMkn9OXJWIMm4DzilvwQg==, figureFileBig=mr3CVrb7i/AoKFHi37Ow3w==, tableContent=null), ArticleFig(id=1207400970204586782, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Table 1, caption=

Information of road-side detection equipment

, figureFileSmall=null, figureFileBig=null, tableContent=
立杆桩号 设备
类型
检测
范围/m
备注
K227+100 radar1 533 与radar4有 215 m检测区域重叠
radar2 438 与radar1检测区域间隔46 m
camera1 用于图像捕捉
K227+600 radar3 506 与radar4检测区域间隔25 m
radar4 571
camera10 用于图像捕捉
), ArticleFig(id=1207400971446100771, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=表1, caption=

路段检测设备基本情况

, figureFileSmall=null, figureFileBig=null, tableContent=
立杆桩号 设备
类型
检测
范围/m
备注
K227+100 radar1 533 与radar4有 215 m检测区域重叠
radar2 438 与radar1检测区域间隔46 m
camera1 用于图像捕捉
K227+600 radar3 506 与radar4检测区域间隔25 m
radar4 571
camera10 用于图像捕捉
), ArticleFig(id=1207400971576124198, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Table 2, caption=

Format of vehicle trajectory data

, figureFileSmall=null, figureFileBig=null, tableContent=
字段 字段说明 格式/单位
TIMESTAMP 时间戳 YYYY-MM-DD:
hh-mm-ss: ms
DEVICEID 设备编号
PTCTYPE 车辆类型 0-大货车,1-小客车
PTCID 车辆识别号
WIDTH 车辆宽度 m
HEIGHT 车辆长度 m
HEADING 车辆朝向
LONGITUDE 经度 °E
LATITUDE 纬度 °N
VELOCITYX 纵向车速 m/s
VELOCITYY 横向车速 m/s
LANEID 车道编号
), ArticleFig(id=1207400971664204586, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=表2, caption=

车辆轨迹数据输出格式

, figureFileSmall=null, figureFileBig=null, tableContent=
字段 字段说明 格式/单位
TIMESTAMP 时间戳 YYYY-MM-DD:
hh-mm-ss: ms
DEVICEID 设备编号
PTCTYPE 车辆类型 0-大货车,1-小客车
PTCID 车辆识别号
WIDTH 车辆宽度 m
HEIGHT 车辆长度 m
HEADING 车辆朝向
LONGITUDE 经度 °E
LATITUDE 纬度 °N
VELOCITYX 纵向车速 m/s
VELOCITYY 横向车速 m/s
LANEID 车道编号
), ArticleFig(id=1207400971773256494, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Table 3, caption=

Sample results of overlapping scene matching

, figureFileSmall=null, figureFileBig=null, tableContent=
radar1车辆ID radar4车辆ID DTW距离
254 29 0.023 876
28 63 0.002 198
15 45 0.015 617
), ArticleFig(id=1207400971873919793, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=表3, caption=

重叠场景匹配结果样本

, figureFileSmall=null, figureFileBig=null, tableContent=
radar1车辆ID radar4车辆ID DTW距离
254 29 0.023 876
28 63 0.002 198
15 45 0.015 617
), ArticleFig(id=1207400972003943220, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=EN, label=Table 4, caption=

Sample results of non-overlapping scenario matching (radar1,radar2)

, figureFileSmall=null, figureFileBig=null, tableContent=
radar1车辆ID radar2车辆ID DTW距离
244 32 0.068 231
232 52 0.078 432
199 51 0.091 341
), ArticleFig(id=1207400972092023610, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1207343628578562713, language=CN, label=表4, caption=

非重叠场景匹配结果样本(radar1,radar2)

, figureFileSmall=null, figureFileBig=null, tableContent=
radar1车辆ID radar2车辆ID DTW距离
244 32 0.068 231
232 52 0.078 432
199 51 0.091 341
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, 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=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, 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.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, 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.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, 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/kxjsygc/CN/10.12404/j.issn.1671-1815.2405917, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2405917, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2405917, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2405917, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于跨雷达数据的车辆轨迹跟踪
收藏切换
PDF下载
尤鑫 1 , 薛金银 1, * , 张北海 1 , 高宇航 2 , 田向丽 2 , 赵建东 2
科学技术与工程 | 论文·交通运输 2025,25(17): 7373-7379
收起
收藏切换
科学技术与工程 | 论文·交通运输 2025, 25(17): 7373-7379
基于跨雷达数据的车辆轨迹跟踪
全屏
尤鑫1 , 薛金银1, * , 张北海1, 高宇航2, 田向丽2, 赵建东2
作者信息
  • 1 北京速通科技有限公司, 北京 100161
  • 2 北京交通大学交通运输学院, 北京 100044
  • 尤鑫(1986—),男,汉族,辽宁沈阳人,博士,高级工程师。研究方向:智能交通。E-mail:

通讯作者:

*薛金银(1976—),男,汉族,陕西韩城人,博士,副研究员。研究方向:智能交通。E-mail:
Cross-device Tracking of Vehicle Trajectory Based on Radar Data
Xin YOU1 , Jin-yin XUE1, * , Bei-hai ZHANG1, Yu-hang GAO2, Xiang-li TIAN2, Jian-dong ZHAO2
Affiliations
  • 1 Beijing Sutong Technology Co., Ltd., Beijing 100161, China
  • 2 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China
出版时间: 2025-06-18 doi: 10.12404/j.issn.1671-1815.2405917
文章导航
收藏切换

高精度车辆轨迹数据对于实现智慧交通具有重要意义,然而现有的车辆轨迹感知技术受到采集范围的限制,难以获得全时段全区域的车辆轨迹数据,无法满足实际应用中对车辆轨迹跟踪精度以及实时性的要求。考虑到跨雷达场景下的车辆轨迹数据特征,提出一种基于雷达数据的跨设备车辆轨迹跟踪方法。首先,根据数据特点应用置信区间下限筛选轨迹数据,并通过卡尔曼滤波对车辆轨迹的位置和速度分别进行平滑和降噪处理。其次,将雷达探测区域的车辆轨迹时间戳、位置坐标、行驶速度、行驶方向以及车道编号作为模型输入,非重叠区域的位置信息作为输出,构建了基于长短期记忆(long short term memory, LSTM)的非重叠区域轨迹预测模型。然后,提出基于轨迹条件约束和搜索区域限制的快速动态时间规整算法(dynamic time warping, DTW)用以计算轨迹相似度,从而实现跨雷达设备车辆轨迹跟踪。最后,以高速公路上跨雷达检测的重叠场景和非重叠场景数据为例进行验证,实验结果表明,重叠场景下车辆轨迹跟踪准确度达到92.2%,非重叠区域车辆轨迹匹配正确率达到90.3%。

高速公路  /  毫米波雷达  /  动态时间规整算法  /  轨迹跟踪  /  LSTM

High-precision vehicle trajectory data is crucial for the realization of intelligent transportation systems. However, existing vehicle trajectory sensing technologies are limited by the range of data collection, making it challenging to obtain full-period and full-area vehicle trajectory data, which cannot meet the demands for trajectory tracking accuracy and real-time performance in practical applications. Considering the characteristics of vehicle trajectory data across radar scenarios, a cross-device vehicle trajectory tracking method was proposed based on radar data. Firstly, trajectory data was filtered based on the lower bound of the confidence interval, and the position and velocity of vehicle trajectories were smoothed and denoised using Kalman filtering. Next, the trajectory timestamp, position coordinates, speed, direction, and lane number from the radar detection area were used as model inputs, while the position information of non-overlapping areas was used as output to construct a non-overlapping area trajectory prediction model based on long short-term memory (LSTM). Subsequently, a fast dynamic time warping (DTW) algorithm based on trajectory condition constraints and search area limitations was proposed to compute trajectory similarity, enabling cross-radar device vehicle trajectory tracking. Finally, verification was conducted using overlapping and non-overlapping scenario data from radar detection on highways. Experimental results show that the trajectory tracking accuracy in overlapping scenarios reaches 92.2%, and the trajectory matching accuracy in non-overlapping areas reaches 90.3%.

highway  /  millimeter-wave radar  /  dynamic time warping  /  trajectory tracking  /  LSTM
尤鑫, 薛金银, 张北海, 高宇航, 田向丽, 赵建东. 基于跨雷达数据的车辆轨迹跟踪. 科学技术与工程, 2025 , 25 (17) : 7373 -7379 . DOI: 10.12404/j.issn.1671-1815.2405917
Xin YOU, Jin-yin XUE, Bei-hai ZHANG, Yu-hang GAO, Xiang-li TIAN, Jian-dong ZHAO. Cross-device Tracking of Vehicle Trajectory Based on Radar Data[J]. Science Technology and Engineering, 2025 , 25 (17) : 7373 -7379 . DOI: 10.12404/j.issn.1671-1815.2405917
目前,雷达检测设备已对高速公路实现了全面覆盖,为获取车辆轨迹数据提供了技术保障。与传统交通流传感器如地磁线圈和红外线传感器相比,轨迹数据记录了覆盖区域内道路使用者的连续动作,包含丰富的时空信息,对分析车辆行驶行为和运行状态具有重要意义,为研究道路使用者的行为模式提供了更多可能性[1-2]。因此,轨迹数据感知和挖掘逐渐成为交通领域的重要研究方向。
由于视频数据采集方式简单且蕴含丰富信息,加之计算机视觉技术的快速发展,基于视觉技术的轨迹数据采集方法已被广泛应用。德国亚琛工业大学汽车工程研究所利用无人机航拍技术创建了HighD数据集[3],其中包含高速公路车辆的轨迹数据,为研究高速公路车辆驾驶行为提供了重要的数据资源。冯汝怡等[4]利用卷积神经网络YOLOv5构建了一种车辆检测目标跨帧关联与轨迹匹配融合方法,并通过无人机在南京市快速路拍摄了两组航拍视频,对算法效果进行测试。
然而,由于无人机或摄像机等视觉传感器在轨迹数据采集方面的限制,所获得的数据集普遍存在覆盖时间短、覆盖距离有限以及覆盖场景受限的问题[5]。毫米波雷达系统主要由天线、射频模块、数据采样及处理模块和频率综合器构成[6]。由于其操作频段高,使得信号衰减速度较慢,因此具有更广泛的探测距离和探测范围。相比其他技术,毫米波雷达突破了连续观测的时长与距离限制,支持实时采集大范围的轨迹数据并提供高精度监测。这种特性为从微观层面持续研究车辆交通行为提供了重要支持与创新优势。
此外,由于单个路侧设备的感知范围有限,常需多设备之间的轨迹拼接[7]来实现车辆轨迹的连续跟踪,增加了轨迹匹配和跟踪的复杂性。目前大部分轨迹拼接研究都是针对视觉传感器开展的,He等[8]为解决跨相机轨迹匹配问题,提出轨迹片段到目标分配方法,并采用受限非负矩阵分解算法计算满足约束的最优分配方案。苟铃滔等[9]提出一种融合时空特征的跨相机多目标跟踪方法,通过改进YOLOv7模型并引入归一化注意力模块,并在隧道场景中测试跨相机目标匹配正确率。国内外学者对基于毫米波雷达的车辆轨迹采集方法的研究较少,其中Felguera-Martín等[10]基于干涉线性调频连续波雷达改进了道路超速监控技术,实现了对目标区域内多辆车的速度、距离和车道的同时监测,从而准确识别超速车辆。牛大伟等[11]基于超距雷达数据分析高速公路上的车辆跟驰特性,并基于长短期记忆(long short term memory, LSTM)模型进行车辆跟驰速度预测,结果显示LSTM模型性能优于传统支持向量回归(support vector regression, SVR)模型。但目前基于雷达的车辆轨迹跟踪方法仍然存在准确性低和误报率高的问题,且在不能实现跨设备连续跟踪车辆。
综上所述,随着深度学习技术和雷达检测设备的快速发展,雷达轨迹数据的挖掘能力得到了显著提升。特别是长短期记忆神经网络的应用使得基于历史信息的轨迹预测和趋势分析成为可能。因此,现提出基于长短期记忆神经网络的非重叠区域轨迹预测模型,以及基于改进动态时间规整(dynamic time warping,DTW)算法的车辆轨迹匹配方法,旨在克服雷达设备感知范围的局限性,实现跨雷达设备的高效、准确的车辆连续轨迹跟踪,并通过高速公路数据验证和评估模型的表现。
研究所采用的车辆轨迹数据来自京哈高速的特定路段,该路段不在桥梁或隧道中,为普通路段,检测器安装在道路两侧的立杆或桩号位置。这些位置用于支撑监控和检测设备,确保设备能够有效覆盖车道和交通流量。研究路段K227+000位置,安装了radar1、radar2、camera1和边缘计算(mobile edge computing,MEC)设备,其中radar1和radar2的检测范围分别为533 m和438 m,并在46 m范围内重叠,用于检测车辆的速度、位置和加速度等,camera1用于捕捉实时图像,MEC设备负责现场数据处理,减少数据传输延迟。研究路段K227+600位置,安装了radar3、radar4和camera10设备,radar3和radar4的检测范围分别为506 m和571 m,并在25 m范围内重叠,camera10用于图像捕捉和视觉数据支持。研究路段的检测设备的基本情况说明如表1所示。
选取京哈高速公路上的4个相邻的毫米波雷达采集的车辆轨迹数据集,雷达检测器采集约2 km单向直线路段的交通数据,包含了采集设备编号以及单向四车道上检测目标的时间戳、车辆类型、车辆识别号、车辆朝向、车辆宽度、车辆长度、位置信息、纵向车速、横向车速、行驶方向、车道编号等信息。数据输出格式如表2所示。
以radar1检测数据为例,由于时间段的选择是随机的,既有车辆完整行驶整个区域,也有车辆刚刚驶入或即将驶出。因此,为了更精确地进行匹配,需要筛选出轨迹点密集的车辆轨迹。根据radar1检测数据集,可以统计出所有检测到的车辆轨迹的平均轨迹点数,然后计算95%的置信区间,小于此区间的都剔除。这种方法称为置信区间下限,是统计学中用于估计总体参数范围的一种方法,置信区间下限L可表示为
$L=\overline{x}-{Z}_{\alpha /2}\frac{\sigma }{\sqrt{n}}$
式(1)中:$\overline{x}$为样本均值;${Z}_{\alpha /2}$为标准正态分布的临界值,对于95%置信水平,${Z}_{0.025}\approx 1.96$;$\sigma $为总体标准差;n为样本量。
在进行数据清洗前,车辆的平均轨迹点数为169.81,表示所有经过雷达检测范围内的车辆平均被记录了约170次。标准差为54.94,表示车辆轨迹点数的分布有一定的波动性。具体来说,有些车辆的轨迹点数会显著高于或低于 平均值。由此,95%的置信区间为(159.86, 179.77),表示有95%的信心,车辆的真实平均轨迹点数落在这个区间内。因此,置信区间下限为159.86。使用这个值作为过滤标准,剔除了轨迹点数低于159.86的车辆。在清洗后的数据集中,所有车辆轨迹包含的轨迹点数至少为160。原始数据与筛选后数据的轨迹点数分布如图1所示。
此外,由于原始轨迹数据由于存在数据丢失、定位故障、传输错误等原因存在噪声,通过卡尔曼滤波的方法对轨迹进行平滑处理。卡尔曼滤波是一种线性二次估计算法,用于线性动态系统中的状态估计。它将先前的状态估计(即当前轨迹点的预测位置)与当前观测数据(当前轨迹点的记录位置)相结合,以获得当前状态的最优估计。
卡尔曼滤波的实现涉及使用先前的最佳结果预测当前值,然后使用观测值校正当前值以获得最佳结果。这种方法有效地减少了噪声的影响,可以更准确地估计车辆的实际状态。以radar1数据集中车辆编号为232的轨迹为例,位置及速度的平滑效果如图2图3所示。
DTW算法由日本学者Itakura提出,起初用于处理相似度计算中的扭曲时间问题,后来被广泛应用于轨迹相似性计算[12]。DTW算法根据时间序列的数值相似性对时间维度进行规整,并基于此寻找两个时间序列的最优对齐路径,使得两个序列之间的累积距离最小化,从而衡量序列的相似度。
假设两条车辆轨迹的时间序列分别是$X=({x}_{1},{x}_{2},\dots,{x}_{m})$$Y=({y}_{1},{y}_{2},\dots,{y}_{n})$。其中,XY的长度分别为mn。首先,构建一个m×n的矩阵网格D,其中矩阵元素D(i,j)表示xiyj两点之间的欧式距离$d({x}_{i},{y}_{j})=({x}_{i}-{y}_{j}{)}^{2}$,也被称作两点之间的局部距离。规整路径W
$W=\left[{w}_{x}\right(k),{w}_{y}(k\left)\right],k=\mathrm{1,2},\dots,l$
式(2)中:wx(k)为轨迹序列X中的点xi的索引;wy(k)为轨迹序列Y中的点yi的索引;l为规整路径W的长度,并且$l\in \left[\mathrm{m}\mathrm{a}\mathrm{x}\right(m,n),m+n]$
最优规整路径W表示为
$\mathrm{D}\mathrm{T}\mathrm{W}(X,Y)=\mathrm{m}\mathrm{i}\mathrm{n}[\sqrt{\stackrel{k}{\sum _{k=1}}w\left(k\right)}/k]$
利用最优规整路径W,能够将两个轨迹序列点xiyj更新为两个新的轨迹序列点$\stackrel{-}{x}\left(k\right)$$\stackrel{-}{y}\left(k\right)$,表示为
$\left\{\begin{array}{l}\stackrel{-}{x}\left(k\right)=x\left[{w}_{x}\right(k\left)\right]\\ \stackrel{-}{y}\left(k\right)=y\left[{w}_{y}\right(k\left)\right]\end{array},\right.\mathrm{k}=\mathrm{1,2},\dots,\mathrm{l}$
可用更新轨迹序列点$\stackrel{-}{x}\left(k\right)$$\stackrel{-}{y}\left(k\right)$的欧氏距离表示原轨迹序列点xiyi之间的规整距离,表达式为
$\mathrm{D}\mathrm{T}\mathrm{W}(x,y)=\stackrel{l}{\sum _{k=1}}[\stackrel{-}{x}(k)-\stackrel{-}{y}{(k)]}^{2}$
累积距离矩阵D的起点和终点分别为(1,1)和(m,n)。每个元素D(i,j)表示从起点到位置(i,j)的累积最小距离,表达式为
$\begin{array}{l}D(i,j)=D({X}_{i},{Y}_{j})+\mathrm{m}\mathrm{i}\mathrm{n}\left[D\right(i-1,j),\\ D(i,j-1),D(i-1,j-1)]\end{array}$
D(i,j)即为从(1,1)到(i,j)的累积最小距离,通过回溯法可找到最优路径,从(m,n)开始,沿着累积最小距离的方向移动,直至到达(1,1)。最优对齐路径如图4所示。
DTW算法需要进行m×n次计算,计算量大,所需时间较长,在处理大规模雷达数据时,其计算成本高,执行速度慢。为了提高DTW算法处理雷达数据的效率,通过缩小搜索区域以及增加轨迹特征约束条件,提出区域限制动态时间规整(reduced range dynamic time wrapping, RR-DTW)算法。首先,以损失一定精度为代价,缩小搜索范围,提高路径搜索效率,优化搜索策略如图5所示。具体而言,在计算两个轨迹序列的最优局部距离时,将较短轨迹放置于i方向,较长轨迹放置于j方向,此后对阴影部分进行搜索以获得最优局部距离。按照此策略进行搜索,仅需计算m×(2n-m+1)/2次,较原始算法减少了m×(m-1)/2次计算。
此外,为减少计算成本,通过重叠时长约束和车道限制约束来对轨迹进行过滤,从而减少需要进行相似度矩阵计算的轨迹对。定义ts(X)ts(Y)分别为轨迹X和轨迹Y的起始时间,te(Y)te(Y)为轨迹X和轨迹Y的结束时间,则轨迹X和轨迹Y的重叠区域的起始时间tos(X,Y)与结束时间toe(X,Y)可表示为
${t}_{\mathrm{o}\mathrm{s}}(X,Y)=\mathrm{m}\mathrm{a}\mathrm{x}\left[{t}_{\mathrm{s}}\right(X),{t}_{\mathrm{s}}(Y\left)\right]$
${t}_{\mathrm{o}\mathrm{e}}(X,Y)=\mathrm{m}\mathrm{i}\mathrm{n}\left[{t}_{\mathrm{e}}\right(X),{t}_{\mathrm{e}}(Y\left)\right]$
轨迹X和轨迹Y重叠区域时长To(X,Y)和总时长Tt(X,Y)可表示为
${T}_{\mathrm{o}}(X,Y)={T}_{\mathrm{o}\mathrm{e}}(X,Y)-{T}_{\mathrm{o}\mathrm{s}}(X,Y)$
$\begin{array}{l}{T}_{\mathrm{t}}(\mathrm{X},\mathrm{Y})=max\left[{t}_{\mathrm{e}}\right(\mathrm{X}),{t}_{\mathrm{e}}(\mathrm{Y}\left)\right]-\\    min\left[{t}_{\mathrm{s}}\right(\mathrm{X}),{t}_{\mathrm{s}}(\mathrm{Y}\left)\right]\end{array}$
因此,轨迹X和轨迹Y的重叠比例R可表示为
$R(X,Y)={T}_{\mathrm{o}}(X,Y)/{T}_{\mathrm{t}}(X,Y)$
设定重合阈值为${R}_{\mathrm{m}\mathrm{i}\mathrm{n}}=0.4$,仅当轨迹X和轨迹Y的重合比例$R(X,Y)\ge {R}_{\mathrm{m}\mathrm{i}\mathrm{n}}$时对其进行相似度计算。
长短期记忆网络(LSTM)作为循环神经网络(recurrent neural network,RNN)的一种高级形式,专门针对RNN在处理长序列数据时遇到的梯度消失和梯度爆炸问题进行了优化,它通过引入独特的门控机制,显著提升了模型在捕捉长序列中长期依赖关系的能力,提高了时间序列预测的精度。雷达数据中的车辆位置坐标、速度、加速度等参数在时间维度上具有明显相关性,因此应用LSTM网络对两个雷达之间的非重叠区域轨迹进行预测。
LSTM模块的输入包括当前时间点的输入xt,上一时间点传递给当前时间点的单元状态ct-1以及隐藏状态ht-1;输出包括当前时间点的单元状态ct以及隐藏状态ht。前向计算过程如下。
${f}_{t}=\sigma ({W}_{\mathrm{f}}\left|{h}_{\mathrm{t}-1},{x}_{\mathrm{t}}\right|+{b}_{\mathrm{f}})$
${i}_{t}=\sigma ({W}_{\mathrm{i}}\left|{h}_{\mathrm{t}-1},{x}_{t}\right|+{b}_{\mathrm{i}})$
${\stackrel{~}{c}}_{t}=\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h}({W}_{\mathrm{c}}\left|{h}_{\mathrm{t}-1},{x}_{t}\right|+{b}_{\mathrm{c}})$
${c}_{t}={f}_{t}\mathrm{*}{c}_{\mathrm{t}-1}+{i}_{t}\mathrm{*}{\stackrel{~}{c}}_{t}$
${o}_{t}=\sigma ({W}_{\mathrm{o}}\left|{h}_{\mathrm{t}-1},{x}_{t}\right|+{b}_{\mathrm{o}})$
${h}_{t}={o}_{t}\mathrm{*}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h}\left({c}_{t}\right)$
式中:ft为遗忘门的输出值;it为输入门的输出值;${\stackrel{~}{c}}_{t}$为输入的单元状态;ct为当前时刻的单元状态;ht为当前时刻的输出;WfWiWcWo为权重;bfbibcbo为偏置项;*表示卷积操作。
LSTM神经网络由输入层、隐藏层、全连接层以及输出层构成。选取雷达检测区域轨迹的时间戳、位置坐标、行驶速度、行驶方向以及车道编号作为模型的输入参数;选取非重叠时间窗口内的位置坐标作为输出参数;构建一个由128个LSTM单元构成的单层隐藏层;构建两个分别由256和128个神经元构成的全连接层[13];设置模型的损失函数为均方误差(mean squared error, MSE);优化器为Adam算法作为优化器;dropout为0.1;设置epochs为300次;选用sigmoid和tanh函数作为门控单元内部激活函数;选取tanh作为输出层的激活函数。
根据以上网络参数设置,构建模型结构如图6所示。
在高速公路上,当相邻雷达检测设备之间的空间距离较小时,检测记录区域会出现部分重叠现象。在这种情况下,车辆的轨迹数据信息,如车道、时间窗口和经纬度信息等,也会产生一定程度的重叠。在本文的研究区域内,雷达设备radar1和radar4的检测区域存在部分重叠。其中,radar1的检测区域长度为533 m,radar4的检测区域长度为571 m,二者的重叠区域长度为215 m。
通过RR-DTW算法对radar1和radar4采集的所有轨迹数据进行相似度计算,获得了跨设备重叠区域内的车辆轨迹匹配结果。随机抽取了6对数据,记录了这些数据中的车辆ID相关信息和DTW距离,并通过绘制不同颜色的轨迹曲线展示了轨迹跟踪结果。具体的匹配结果如表3所示。轨迹跟踪结果如图7所示。
表3展示了部分模型输出结果,即车辆匹配结果与轨迹跟踪结果。第一列是来自radar1数据集的车辆ID,第二列是来自radar4数据集的车辆ID,第三列是这两个雷达中不同ID的车辆在重叠区域的轨迹相似度度量结果。例如,来自radar1的车辆ID为244,与radar4记录的车辆ID为22的车辆,其轨迹相似度度量为0.000 896,表明这两段轨迹来自同一辆车。图7展示了4组车辆轨迹跟踪的结果,其中蓝色曲线代表radar1的车辆轨迹,红色曲线代表radar4的车辆轨迹。可以看到,两段曲线在重叠区域的轨迹高度吻合,表明这些轨迹来自同一车辆。
当相邻雷达检测设备之间的空间距离较大时,检测记录区域可能存在间隙,这种间隙会影响数据的准确性和连续性。研究区域内,雷达设备radar1与radar2之间的检测区域存在46 m的间隔,而雷达设备radar3与radar4之间的检测区域则存在25 m的间隔。
采用基于LSTM的非重叠区域轨迹预测模型,对上游雷达检测到的车辆轨迹进行预测。预测得到的轨迹与下游雷达检测到的轨迹在重叠区域被定义为虚拟重合区域。利用RR-DTW算法计算虚拟重合区域内轨迹的相似度,并基于此相似度进行轨迹匹配与跟踪。表4展示了从radar1与radar2抽取的3组匹配结果。图8为radar1与radar2中3组车辆轨迹跟踪的结果。
图8展示了radar1-radar2数据集的轨迹匹配结果。可以看出,黄色曲线与绿色曲线部分重叠,表明存在重叠的检测场景,并且重叠部分的吻合度较高。
针对基于雷达的车辆轨迹跟踪方法准确性低且难以实现跨设备连续跟踪的问题,提出一种基于LSTM模型以及改进DTW算法的跨设备轨迹跟踪模型。主要研究结论如下。
(1)在雷达监测区域存在重叠的场景中,由于检测设备型号和数据上传延迟等因素导致的不同轨迹数据之间长度不相等,无法直接进行轨迹匹配。为了解决这一问题,采用了动态时间规整算法,并针对其计算量大的缺点进行了优化,设计了一种快速搜索策略。此外,还引入了重叠时长约束和车道限制约束,对轨迹数据进行二次筛选,仅对符合条件的轨迹进行相似度计算,从而提高了DTW算法的执行效率,实现了轨迹数据的快速匹配。
(2)在雷达监测区域非重叠的场景中,即两雷达设备间存在的空白区域,无法直接进行轨迹相似度计算。针对此问题,建立了基于LSTM模型的非重叠区域轨迹预测模型。该模型将检测区域轨迹的时间戳、位置坐标和行驶速度等作为输入参数,并将非重叠时间窗口内的位置坐标作为输出参数。通过测试均方误差的大小,确定了模型结构、训练次数、激活函数和Dropout率等参数,从而有效提高了轨迹预测的准确性。
(3)通过高速公路雷达检测数据集对模型性能进行验证,radar1和radar4的检测范围存在重叠区域,经过数据预处理后,其中radar1有71条轨迹,radar4有67条轨迹,通过改进DTW算法进行轨迹相似度计算,根据计算结果进行轨迹匹配,结果显示radar1和radar4的轨迹匹配准确率为92.2%。radar1和radar2的检测范围不存在重叠区域,通过基于LSTM模型,根据radar1的车辆轨迹数据对非重叠区域的轨迹进行预测,使得radar1的轨迹在时间上和radar2的轨迹存在重合,根据预测结果的重叠区域进行轨迹相似度匹配,结果显示radar1和radar2的轨迹匹配准确率为90.3%。
  • 国家重点研发计划(2022ZD0115605)
参考文献 引证文献
排序方式:
[1]
Dhatbale R, Chilukuri B R. Deep learning techniques for vehicle trajectory extraction in mixed traffic[J]. Journal of Big Data Analytics in Transportation, 2021, 3(2): 141-157.
[2]
Chen T, Shi X, Wong Y D. Key feature selection and risk prediction for lane-changing behaviors based on vehicles’ trajectory data[J]. Accident Analysis & Prevention, 2019, 129: 156-169.
[3]
Krajewski R, Bock J, Kloeker L, et al. The highd dataset:a drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems[C]//2018 21st International Conference on Intelligent Transportation Systems (ITSC). New York: IEEE, 2018: 2118-2125.
[4]
冯汝怡, 李志斌, 吴启范, 等. 航拍视频车辆检测目标关联与时空轨迹匹配[J]. 交通信息与安全, 2021, 39(2): 61-69, 77.
Feng Ruyi, Li Zhibin, Wu Qifan, et al. Association of vehicle object detection and the time-space trajectory matching from aerial videos[J]. Journal of Transport Information and Safety, 2021, 39(2): 61-69, 77.
[5]
王俊骅, 宋昊, 景强, 等. 基于毫米波雷达组群的全域车辆轨迹检测技术方法[J]. 中国公路学报, 2022, 35(12): 181-192.
Wang Junhua, Song Hao, Jing Qiang, et al. Road-range tracking of vehicle trajectories based on millimeter-wave radar[J]. China Journal of Highway and Transport, 2022, 35(12): 181-192.
[6]
顾思思, 常世新, 韩明敏, 等. 毫米波雷达数据驱动的桥梁挠度预测[J]. 科学技术与工程, 2023, 23(11): 4874-4880.
Gu Sisi, Chang Shixin, Han Mingmin, et al. Bridge deflection prediction driven by millimeter wave radar data[J]. Science Technology and Engineering, 2023, 23(11): 4874-4880.
[7]
Lu Q, Wang R, Yang B, et al. Trajectory splicing[J]. Knowledge and Information Systems, 2020, 62(4): 1279-1312.
[8]
He Y, Wei X, Hong X, et al. Multi-target multi-camera tracking by tracklet-to-target assignment[J]. IEEE Transactions on Image Processing, 2020, 29: 5191-5205.
[9]
苟铃滔, 宋焕生, 张朝阳, 等. 融合时空特征的隧道场景跨相机车辆实时跟踪方法[J]. 计算机工程与应用, 2023, 59(24): 88-97.
Gou Lingtao, Song Huansheng, Zhang Chaoyang, et al. Real-time cross-camera vehicle tracking method for tunnel scenes by fusing spatiotemporal features[J]. Computer Engineering and Applications, 2023, 59(24): 88-97.
[10]
Felguera-Martín D, González-Partida J T, Almorox-González P, et al. Vehicular traffic surveillance and road lane detection using radar interferometry[J]. IEEE Transactions on Vehicular Technology, 2012, 61(3): 959-970.
[11]
牛大伟, 李一贤, 崔玮, 等. 基于超距雷达数据的车辆跟驰行为分析及建模[J]. 科学技术与工程, 2023, 23(15): 6654-6660.
Niu Dawei, Li Yixian, Cui Wei, et al. Analysis and modeling of high speed car-following behavior based on ultra-range radar data[J]. Science Technology and Engineering, 2023, 23(15): 6654-6660.
[12]
李颖, 赵莉, 赵祥模, 等. 基于大货车GPS数据的轨迹相似性度量有效性研究[J]. 中国公路学报, 2020, 33(2): 146-157.
Li Ying, Zhao Li, Zhao Xiangmo, et al. Effectiveness of trajectory similarity measures based on truck GPS data[J]. China Journal of Highway and Transport, 2020, 33(2): 146-157.
[13]
Altché F, de La Fortelle A. An LSTM network for highway trajectory prediction[C]// 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC). New York: IEEE, 2017: 353-359.
2025年第25卷第17期
PDF下载
181
92
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2405917
  • 接收时间:2024-08-06
  • 首发时间:2025-12-15
  • 出版时间:2025-06-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-08-06
基金
国家重点研发计划(2022ZD0115605)
作者信息
    1 北京速通科技有限公司, 北京 100161
    2 北京交通大学交通运输学院, 北京 100044

通讯作者:

*薛金银(1976—),男,汉族,陕西韩城人,博士,副研究员。研究方向:智能交通。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2405917
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