Article(id=1209910182600110197, tenantId=1146029695717560320, journalId=1189621681917173762, issueId=1209910182134542453, articleNumber=null, orderNo=null, doi=10.19620/j.cnki.1000-3703.20230877, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=null, receivedDateStr=null, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1766394666575, onlineDateStr=2025-12-22, pubDate=1721750400000, pubDateStr=2024-07-24, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1766394666575, onlineIssueDateStr=2025-12-22, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1766394666575, creator=13701087609, updateTime=1766394666575, updator=13701087609, issue=Issue{id=1209910182134542453, tenantId=1146029695717560320, journalId=1189621681917173762, year='2024', volume='', issue='7', pageStart='1', pageEnd='62', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1766394666465, creator=13701087609, updateTime=1766482240343, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1210277493739753804, tenantId=1146029695717560320, journalId=1189621681917173762, issueId=1209910182134542453, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1210277493739753805, tenantId=1146029695717560320, journalId=1189621681917173762, issueId=1209910182134542453, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=1, endPage=8, ext={EN=ArticleExt(id=1209910182860157048, articleId=1209910182600110197, tenantId=1146029695717560320, journalId=1189621681917173762, language=EN, title=Vehicle Trajectory Prediction Method Based on GRU and Transformer, columnId=1209910182801436791, journalTitle=Automobile Technology, columnName=Feature Topic on Motion Planning and Control Techniques, runingTitle=null, highlight=null, articleAbstract=

In order to enhance the understanding of the dynamic environment of autonomous vehicles and to improve road driving safety, this article proposed a vehicle trajectory prediction STGTF model based on the Gated Recurrent Unit (GRU) and Transformer that used the GRU to extract the historical trajectory features of vehicles, and used a two-layer Multi-Headed Attention (MHA) mechanism to extract the spatio-temporal interaction features of vehicles, generating the predicted trajectories. The experimental results show that the Root-Mean-Square Error (RMSE) of the predicted results decrease by 7.3% on average, STGTF model has different degrees of improvement compared with other existing methods for both short-term prediction and long-term prediction, proving validity of this model.

, correspAuthors=null, 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=Qingrong Wang, Xiaoze Tan, Changfeng Zhu, Yujie Li), CN=ArticleExt(id=1209910186274320590, articleId=1209910182600110197, tenantId=1146029695717560320, journalId=1189621681917173762, language=CN, title=基于门控循环单元和Transformer的车辆轨迹预测方法*, columnId=1209910182935654522, journalTitle=汽车技术, columnName=智能车辆运动规划与控制技术专题, runingTitle=null, highlight=null, articleAbstract=

为增强自动驾驶车辆对动态环境的理解能力及其道路行驶安全性,提出基于门控循环单元(GRU)和Transformer的车辆轨迹预测模型STGTF,使用GRU提取车辆的历史轨迹特征,通过双层多头注意力(MHA)机制提取车辆的时空交互特征,生成预测轨迹。试验结果表明,预测结果的均方根误差(RMSE)平均降低7.3%,STGTF在短期预测和长期预测方面均有不同程度的提升,验证了模型的有效性。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=FCcW12nHifw03wiMtc/m9g==, magXml=AGNePR1CSiatTzGTc7wSjQ==, pdfUrl=null, pdf=6hVPyXmnOlL5ieFRsFGWAQ==, pdfFileSize=4408961, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=j6VgetoqaaElKK77Hh6onQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=J42kpcX/g1ww+UqfsPZSJw==, mapNumber=null, authorCompany=null, fund=null, authors=

修改稿收到日期为2023年11月20日

, authorsList=王庆荣, 谭小泽, 朱昌锋, 李裕杰)}, authors=[Author(id=1210277263183057873, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, orderNo=0, 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=1210277263254361044, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277263183057873, language=EN, stringName=Qingrong Wang, firstName=Qingrong, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1210277263334052825, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277263183057873, language=CN, stringName=王庆荣, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 兰州交通大学,电子与信息工程学院,兰州 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1210277263036257218, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, xref=1, ext=[AuthorCompanyExt(id=1210277263040451523, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070), AuthorCompanyExt(id=1210277263048840133, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 兰州交通大学,电子与信息工程学院,兰州 730070)])]), Author(id=1210277263417938909, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, 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=1210277263485047778, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277263417938909, language=EN, stringName=Xiaoze Tan, firstName=Xiaoze, middleName=null, lastName=Tan, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1210277263564739558, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277263417938909, language=CN, stringName=谭小泽, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 兰州交通大学,电子与信息工程学院,兰州 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1210277263036257218, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, xref=1, ext=[AuthorCompanyExt(id=1210277263040451523, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070), AuthorCompanyExt(id=1210277263048840133, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 兰州交通大学,电子与信息工程学院,兰州 730070)])]), Author(id=1210277263703151599, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, 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=1210277263782843379, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277263703151599, language=EN, stringName=Changfeng Zhu, firstName=Changfeng, middleName=null, lastName=Zhu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Institute of Transportation, Lanzhou Jiaotong University, Lanzhou 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1210277263858340857, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277263703151599, language=CN, stringName=朱昌锋, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 兰州交通大学,交通运输学院,兰州 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1210277263107560393, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, xref=2, ext=[AuthorCompanyExt(id=1210277263111754697, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263107560393, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Transportation, Lanzhou Jiaotong University, Lanzhou 730070), AuthorCompanyExt(id=1210277263120143307, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263107560393, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 兰州交通大学,交通运输学院,兰州 730070)])]), Author(id=1210277265078883327, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, 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=1210277265154379778, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277265078883327, language=EN, stringName=Yujie Li, firstName=Yujie, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1210277265221488646, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, authorId=1210277265078883327, language=CN, stringName=李裕杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 兰州交通大学,电子与信息工程学院,兰州 730070, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1210277263036257218, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, xref=1, ext=[AuthorCompanyExt(id=1210277263040451523, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070), AuthorCompanyExt(id=1210277263048840133, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 兰州交通大学,电子与信息工程学院,兰州 730070)])])], keywords=[Keyword(id=1210277265477341202, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, orderNo=1, keyword=Vehicle trajectory prediction), Keyword(id=1210277265569615893, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, orderNo=2, keyword=Gated Recurrent Unit (GRU)), Keyword(id=1210277265636724759, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, orderNo=3, keyword=Transformer), Keyword(id=1210277265716416538, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, orderNo=4, keyword=Vehicle interaction), Keyword(id=1210277265808691230, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, orderNo=5, keyword=Multi-head attention mechanism), Keyword(id=1210277265900965921, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, orderNo=1, keyword=车辆轨迹预测), Keyword(id=1210277266030989349, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, orderNo=2, keyword=门控循环单元), Keyword(id=1210277266131652648, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, orderNo=3, keyword=Transformer), Keyword(id=1210277266232315948, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, orderNo=4, keyword=车辆交互), Keyword(id=1210277266366533681, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, orderNo=5, keyword=多头注意力机制)], refs=[Reference(id=1210277270359511235, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2021, volume=15, issue=9, pageStart=1093, pageEnd=1105, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=HU L, ZHOU X Q, ZHANG X, journalName=IET Intelligent Transport Systems, refType=null, unstructuredReference=HU L, ZHOU X Q, ZHANG X, et al. A Review on Key Challenges in InIntelligent Vehicles: Safety and Driver-Oriented Features[J]. IET Intelligent Transport Systems, 2021, 15(9): 1093-1105., articleTitle=A Review on Key Challenges in InIntelligent Vehicles: Safety and Driver-Oriented Features, refAbstract=null), Reference(id=1210277270468563143, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=978, pageEnd=985, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=LIU J B, MAO X Y, FANG Y Q, journalName=Sanya, refType=null, unstructuredReference=LIU J B, MAO X Y, FANG Y Q, et al. A Survey on Deep-Learning Approaches for Vehicle Trajectory Prediction in Autonomous Driving[C]// 2021 IEEE International Conference on Robotics and Biomimetics (ROBIO). Sanya, China: IEEE, 2021: 978-985., articleTitle=A Survey on Deep-Learning Approaches for Vehicle Trajectory Prediction in Autonomous Driving[C]// 2021 IEEE International Conference on Robotics and Biomimetics (ROBIO), refAbstract=null), Reference(id=1210277270565032137, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2014, volume=1, issue=1, pageStart=1, pageEnd=14, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=LEFÈVRE S, VASQUEZ D, LAUGIER C, journalName=Robomech Journal, refType=null, unstructuredReference=LEFÈVRE S, VASQUEZ D, LAUGIER C. A Survey on Motion Prediction and Risk Assessment for Intelligent Vehicles[J]. Robomech Journal, 2014, 1(1): 1-14., articleTitle=A Survey on Motion Prediction and Risk Assessment for Intelligent Vehicles, refAbstract=null), Reference(id=1210277270665695435, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=454, pageEnd=469, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=CHEN W H, WANG F F, SUN H B, journalName=Asian Conference on Machine Learning, refType=null, unstructuredReference=CHEN W H, WANG F F, SUN H B. S2TNet: Spatio-Temporal Transformer Networks for Trajectory Prediction in Autonomous Driving[C]// Asian Conference on Machine Learning. Online: PMLR, 2021: 454-469., articleTitle=S2TNet: Spatio-Temporal Transformer Networks for Trajectory Prediction in Autonomous Driving, refAbstract=null), Reference(id=1210277270766358731, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=11, pageStart=1, pageEnd=9, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=高振海, 鲍明喜, 高菲, journalName=汽车技术, refType=null, unstructuredReference=高振海, 鲍明喜, 高菲, 等. 智能驾驶系统对周边交通车辆预期轨迹的单模态网络预测方法[J]. 汽车技术, 2022(11): 1-9., articleTitle=智能驾驶系统对周边交通车辆预期轨迹的单模态网络预测方法, refAbstract=null), Reference(id=1210277270850244812, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=11, pageStart=1, pageEnd=9, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=GAO Z H, BAO M X, GAO F, journalName=Automobile Technology, refType=null, unstructuredReference=GAO Z H, BAO M X, GAO F, et al. A Uni-Modal Network Prediction Method for Surrounding Vehicle Expected Trajectory in Intelligent Driving System[J]. Automobile Technology, 2022(11): 1-9., articleTitle=A Uni-Modal Network Prediction Method for Surrounding Vehicle Expected Trajectory in Intelligent Driving System, refAbstract=null), Reference(id=1210277270976073935, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=1468, pageEnd=1476, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=DEO N, TRIVEDI M M, journalName=Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. Salt Lake City, UT, refType=null, unstructuredReference=DEO N, TRIVEDI M M. Convolutional Social Pooling for Vehicle Trajectory Prediction[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. Salt Lake City, UT, USA: IEEE, 2018: 1468-1476., articleTitle=Convolutional Social Pooling for Vehicle Trajectory Prediction, refAbstract=null), Reference(id=1210277271055765713, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=961, pageEnd=971, url=null, language=null, rfNumber=[7], rfOrder=7, authorNames=ALAHI A, GOEL K, RAMANATHAN V, journalName=Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, refType=null, unstructuredReference=ALAHI A, GOEL K, RAMANATHAN V, et al. Social LSTM: Human Trajectory Prediction in Crowded Spaces[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 961-971., articleTitle=Social LSTM: Human Trajectory Prediction in Crowded Spaces, refAbstract=null), Reference(id=1210277271148040403, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2023, volume=36, issue=1, pageStart=226, pageEnd=239, url=null, language=null, rfNumber=[8], rfOrder=8, authorNames=李文礼, 韩迪, 石晓辉, journalName=中国公路学报, refType=null, unstructuredReference=李文礼, 韩迪, 石晓辉, 等. 基于时-空注意力机制的车辆轨迹预测[J]. 中国公路学报, 2023, 36(1): 226-239., articleTitle=基于时-空注意力机制的车辆轨迹预测, refAbstract=null), Reference(id=1210277271219343574, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2023, volume=36, issue=1, pageStart=226, pageEnd=239, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=LI W L, HAN D, SHI X H, journalName=China Journal of Highway and Transport, refType=null, unstructuredReference=LI W L, HAN D, SHI X H, et al. Vehicle Trajectory Prediction Based on Spatial-Temporal Attention Mechanism[J]. China Journal of Highway and Transport, 2023, 36(1): 226-239., articleTitle=Vehicle Trajectory Prediction Based on Spatial-Temporal Attention Mechanism, refAbstract=null), Reference(id=1210277271299035351, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=2255, pageEnd=2264, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=AGRIM G, JUSTIN J, LI F F, journalName=Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, refType=null, unstructuredReference=AGRIM G, JUSTIN J, LI F F, et al. Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 2255-2264., articleTitle=Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks, refAbstract=null), Reference(id=1210277271378727129, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2022, volume=27, issue=6, pageStart=5690, pageEnd=5700, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=LI L, ZHAO W Z, WANG C Y, journalName=IEEE/ASME Transactions on Mechatronics, refType=null, unstructuredReference=LI L, ZHAO W Z, WANG C Y, et al. BRAM-ED: Vehicle Trajectory Prediction Considering the Change of Driving Behavior[J]. IEEE/ASME Transactions on Mechatronics, 2022, 27(6): 5690-5700., articleTitle=BRAM-ED: Vehicle Trajectory Prediction Considering the Change of Driving Behavior, refAbstract=null), Reference(id=1210277271450030297, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2021, volume=6, issue=1, pageStart=175, pageEnd=185, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=MESSAOUD K, YAHIAOUI A, VERROUST-BLONDET A, journalName=IEEE Transactions on Intelligent Vehicles, refType=null, unstructuredReference=MESSAOUD K, YAHIAOUI A, VERROUST-BLONDET A, et al. Attention Based Vehicle Trajectory Prediction[J]. IEEE Transactions on Intelligent Vehicles, 2021, 6(1): 175-185., articleTitle=Attention Based Vehicle Trajectory Prediction, refAbstract=null), Reference(id=1210277271542304987, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=7573, pageEnd=7582, url=null, language=null, rfNumber=[12], rfOrder=13, authorNames=LIU Y C, ZHANG J H, FANG L J, journalName=Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville, TN, refType=null, unstructuredReference=LIU Y C, ZHANG J H, FANG L J, et al. Multimodal Motion Prediction with Stacked Transformers[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville, TN, USA: IEEE, 2021: 7573-7582., articleTitle=Multimodal Motion Prediction with Stacked Transformers, refAbstract=null), Reference(id=1210277271621996766, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=7, pageStart=24, pageEnd=30, url=null, language=null, rfNumber=[13], rfOrder=14, authorNames=赵树恩, 苏天彬, 赵东宇, journalName=汽车技术, refType=null, unstructuredReference=赵树恩, 苏天彬, 赵东宇. 基于图神经网络的交互车辆驾驶意图识别及轨迹预测[J]. 汽车技术, 2023(7): 24-30., articleTitle=基于图神经网络的交互车辆驾驶意图识别及轨迹预测, refAbstract=null), Reference(id=1210277271680717025, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=7, pageStart=24, pageEnd=30, url=null, language=null, rfNumber=[13], rfOrder=15, authorNames=ZHAO S E, SU T B, ZHAO D Y, journalName=Automobile Technology, refType=null, unstructuredReference=ZHAO S E, SU T B, ZHAO D Y. Interactive Vehicle Driving Intention Recognition and Trajectory Prediction Based on Graph Neural Network[J]. Automobile Technology, 2023(7): 24-30., articleTitle=Interactive Vehicle Driving Intention Recognition and Trajectory Prediction Based on Graph Neural Network, refAbstract=null), Reference(id=1210277271743631587, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2019, volume=151, issue=null, pageStart=327, pageEnd=334, url=null, language=null, rfNumber=[14], rfOrder=16, authorNames=CHOI S, KIM J, YEO H, journalName=Procedia Computer Science, refType=null, unstructuredReference=CHOI S, KIM J, YEO H. Attention-Based Recurrent Neural Network for Urban Vehicle Trajectory Prediction[J]. Procedia Computer Science, 2019, 151: 327-334., articleTitle=Attention-Based Recurrent Neural Network for Urban Vehicle Trajectory Prediction, refAbstract=null), Reference(id=1210277271827517671, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=1, pageEnd=2, url=null, language=null, rfNumber=[15], rfOrder=17, authorNames=ZHANG Z J, journalName=Banff, AB, refType=null, unstructuredReference=ZHANG Z J. Improved Adam Optimizer for Deep Neural Networks[C]// 2018 IEEE/ACM 26th International Symposium on Quality of Service (IWQoS). Banff, AB, Canada: IEEE, 2018: 1-2., articleTitle=Improved Adam Optimizer for Deep Neural Networks[C]// 2018 IEEE/ACM 26th International Symposium on Quality of Service (IWQoS), refAbstract=null), Reference(id=1210277271919792359, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=598, pageEnd=614, url=null, language=null, rfNumber=[16], rfOrder=18, authorNames=SONG H R, DING W C, CHEN Y X, journalName=European Conference on Computer Vision. Glasgow, refType=null, unstructuredReference=SONG H R, DING W C, CHEN Y X, et al. PiP: Planning-Informed Trajectory Prediction for Autonomous Driving[C]// European Conference on Computer Vision. Glasgow, United Kingdom: Springer, Cham, 2020: 598-614., articleTitle=PiP: Planning-Informed Trajectory Prediction for Autonomous Driving, refAbstract=null)], funds=[Fund(id=1210277270028161202, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, awardId=71961016, language=CN, fundingSource=*国家自然科学基金项目(71961016), fundOrder=null, country=null), Fund(id=1210277270099464375, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, awardId=72161024, language=CN, fundingSource=国家自然科学基金项目(72161024), fundOrder=null, country=null), Fund(id=1210277270179156154, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, awardId=GSSYLXM-04, language=CN, fundingSource=甘肃省教育厅“双一流”重大研究项目(GSSYLXM-04), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1210277263036257218, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, xref=1, ext=[AuthorCompanyExt(id=1210277263040451523, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070), AuthorCompanyExt(id=1210277263048840133, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263036257218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 兰州交通大学,电子与信息工程学院,兰州 730070)]), AuthorCompany(id=1210277263107560393, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, xref=2, ext=[AuthorCompanyExt(id=1210277263111754697, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263107560393, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Institute of Transportation, Lanzhou Jiaotong University, Lanzhou 730070), AuthorCompanyExt(id=1210277263120143307, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, companyId=1210277263107560393, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 兰州交通大学,交通运输学院,兰州 730070)])], figs=[ArticleFig(id=1210277266605609020, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=dIYpbRhsOnr5lx68pCtCJw==, figureFileBig=j+NaQFHkkTdMlBOUfzGiOg==, tableContent=null), ArticleFig(id=1210277266693689412, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图1, caption=坐标系旋转, figureFileSmall=dIYpbRhsOnr5lx68pCtCJw==, figureFileBig=j+NaQFHkkTdMlBOUfzGiOg==, tableContent=null), ArticleFig(id=1210277266794352714, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=gjpw81yC8QzCpv0VhudmNQ==, figureFileBig=kNqAYvdih19r7p2yaZoD/A==, tableContent=null), ArticleFig(id=1210277266886627404, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图2, caption=STGTF模型结构, figureFileSmall=gjpw81yC8QzCpv0VhudmNQ==, figureFileBig=kNqAYvdih19r7p2yaZoD/A==, tableContent=null), ArticleFig(id=1210277267016650833, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=sJpC4+u7cb5jhMWXLp0wjg==, figureFileBig=vQZifQ59QPNGXL7xc3lP9w==, tableContent=null), ArticleFig(id=1210277267113119830, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图3, caption=GRU单元结构, figureFileSmall=sJpC4+u7cb5jhMWXLp0wjg==, figureFileBig=vQZifQ59QPNGXL7xc3lP9w==, tableContent=null), ArticleFig(id=1210277267230560347, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=8ry12YbQktKcyC/c4WnVNQ==, figureFileBig=4sR9hd2PiInulvwSPfCi0w==, tableContent=null), ArticleFig(id=1210277267306057822, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图4, caption=Transformer编码器结构, figureFileSmall=8ry12YbQktKcyC/c4WnVNQ==, figureFileBig=4sR9hd2PiInulvwSPfCi0w==, tableContent=null), ArticleFig(id=1210277267394138210, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=JtoPljo61NtIxvn6V7da8g==, figureFileBig=FKGE1jFHc01GjHxuMoppFQ==, tableContent=null), ArticleFig(id=1210277267482218601, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图5, caption=注意力机制, figureFileSmall=JtoPljo61NtIxvn6V7da8g==, figureFileBig=FKGE1jFHc01GjHxuMoppFQ==, tableContent=null), ArticleFig(id=1210277267549327469, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=TvP+8V76nevXGXd26QCK/A==, figureFileBig=NcwndSm7LzygmunqhYFCHw==, tableContent=null), ArticleFig(id=1210277267612242033, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图6, caption=研究区域示意及其覆盖范围, figureFileSmall=TvP+8V76nevXGXd26QCK/A==, figureFileBig=NcwndSm7LzygmunqhYFCHw==, tableContent=null), ArticleFig(id=1210277267691933813, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=MD51tmegXhoxGW4ZxZdeLQ==, figureFileBig=xHo7kshXwjpDu95xvLceBA==, tableContent=null), ArticleFig(id=1210277267788402807, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图7, caption=交互车辆选取范围, figureFileSmall=MD51tmegXhoxGW4ZxZdeLQ==, figureFileBig=xHo7kshXwjpDu95xvLceBA==, tableContent=null), ArticleFig(id=1210277267884871802, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=cRSWyurq6Yv79OYNVNTaFQ==, figureFileBig=57Zojja6DBLLuGCmULxnbw==, tableContent=null), ArticleFig(id=1210277267968757889, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图8, caption=预测结果对比, figureFileSmall=cRSWyurq6Yv79OYNVNTaFQ==, figureFileBig=57Zojja6DBLLuGCmULxnbw==, tableContent=null), ArticleFig(id=1210277268056838276, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=05ICqLw9ed7YA6zWwLqCHw==, figureFileBig=OdWOTB51AVfUgb+Fo8CmXA==, tableContent=null), ArticleFig(id=1210277268140724362, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图9, caption=消融实验结果对比, figureFileSmall=05ICqLw9ed7YA6zWwLqCHw==, figureFileBig=OdWOTB51AVfUgb+Fo8CmXA==, tableContent=null), ArticleFig(id=1210277268228804752, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=Cm/QX2wPEBLDJshEK0q4XA==, figureFileBig=IO14FoQA7m6dPeVY4/qSzg==, tableContent=null), ArticleFig(id=1210277268295913618, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图10, caption=不同场景预测轨迹, figureFileSmall=Cm/QX2wPEBLDJshEK0q4XA==, figureFileBig=IO14FoQA7m6dPeVY4/qSzg==, tableContent=null), ArticleFig(id=1210277268379799701, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=Gmk5DqMdyI+MarkYd/ffNQ==, figureFileBig=dGBksG4kDtjfVnxQKsARIQ==, tableContent=null), ArticleFig(id=1210277269550010523, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=图11, caption=换道轨迹对比, figureFileSmall=Gmk5DqMdyI+MarkYd/ffNQ==, figureFileBig=dGBksG4kDtjfVnxQKsARIQ==, tableContent=null), ArticleFig(id=1210277269621313696, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
方法 RMSE/m
tp=1 s tp=2 s tp=3 s tp=4 s tp=5 s 平均值
S-LSTM 0.67 1.34 2.21 3.33 4.62 2.43
S-GAN 0.60 1.35 2.26 3.31 4.52 2.40
CS-LSTM 0.64 1.31 2.15 3.16 4.47 2.35
PiP 0.58 1.23 2.01 2.95 4.14 2.18
STGTF 0.51 1.12 1.83 2.74 3.89 2.02
), ArticleFig(id=1210277269696811171, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=表1, caption=

各模型预测结果

, figureFileSmall=null, figureFileBig=null, tableContent=
方法 RMSE/m
tp=1 s tp=2 s tp=3 s tp=4 s tp=5 s 平均值
S-LSTM 0.67 1.34 2.21 3.33 4.62 2.43
S-GAN 0.60 1.35 2.26 3.31 4.52 2.40
CS-LSTM 0.64 1.31 2.15 3.16 4.47 2.35
PiP 0.58 1.23 2.01 2.95 4.14 2.18
STGTF 0.51 1.12 1.83 2.74 3.89 2.02
), ArticleFig(id=1210277269768114345, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
方法 RMSE/m
tp=1 s tp=2 s tp=3 s tp=4 s tp=5 s 平均值
STTF 0.53 1.20 1.92 2.94 4.07 2.13
TGTF 0.55 1.24 2.06 3.12 4.26 2.25
SGTF 0.52 1.16 1.89 2.81 3.95 2.07
STGTF 0.51 1.12 1.83 2.74 3.89 2.02
), ArticleFig(id=1210277269856194731, tenantId=1146029695717560320, journalId=1189621681917173762, articleId=1209910182600110197, language=CN, label=表2, caption=

消融实验结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
方法 RMSE/m
tp=1 s tp=2 s tp=3 s tp=4 s tp=5 s 平均值
STTF 0.53 1.20 1.92 2.94 4.07 2.13
TGTF 0.55 1.24 2.06 3.12 4.26 2.25
SGTF 0.52 1.16 1.89 2.81 3.95 2.07
STGTF 0.51 1.12 1.83 2.74 3.89 2.02
)], attaches=null, journal=Journal(id=1149693407745847311, delFlag=0, nameCn=汽车技术, nameEn=Automobile Technology, nameHistory1=null, nameHistory2=null, issn=1000-3703, eissn=null, cn=22-1113/U, coden=null, periodic=0, language=CN, oaType=null, 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=rYFtDx/CU9+iX8QTM0ckbw==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1752037868679, updatedTime=1761735668047, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=A, firstLetterEn=A, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=rYFtDx/CU9+iX8QTM0ckbw==, picEn=oFT2NmUwKPUjZ27C1+d9pw==, jcr=null, cjcr=null, exts=[JournalExt(id=1190368926564450443, language=CN, name=汽车技术, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735668078, updatedTime=1761735668078, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://qcjs.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://qcjs.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://qcjs.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190368926618976396, language=EN, name=Automobile Technology, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735668091, updatedTime=1761735668091, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://qcjs.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://qcjs.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://qcjs.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1189621681917173762, websiteList=[Website(id=1189624193747526544, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189621681917173762, 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/qcjs/CN, language=CN, createTime=1761558109939, createBy=18614031015, updateTime=1761558140534, updateBy=18614031015, name=汽车技术-中, tplId=1146099689490845704, title=汽车技术, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189625424704451180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=articleTextType, value=kx, createTime=1761558403421, updateTime=1761558403421, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424675091049, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=banner, value=null, createTime=1761558403414, updateTime=1761558403414, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424733811311, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=grayFlag, value=0, createTime=1761558403428, updateTime=1761558403428, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424658313832, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=logo, value=https://castjournals.cast.org.cn/joweb/qcjs/CN/file/pic?fileId=7En9rzX2QCa/1J8NnKt/Fg==, createTime=1761558403410, updateTime=1761558403410, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424746394225, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=minRunFlag, value=0, createTime=1761558403431, updateTime=1761558403431, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424691868267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcjs/CN/file/pic, createTime=1761558403418, updateTime=1761558403418, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424742199920, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=silenceFlag, value=0, createTime=1761558403430, updateTime=1761558403430, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424683479658, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761558403416, updateTime=1761558403416, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424712839789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=themeColor, value=null, createTime=1761558403423, updateTime=1761558403423, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625424725422702, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193747526544, code=themeStyle, value=null, createTime=1761558403426, updateTime=1761558403426, creator=18614031015, updator=18614031015)]), Website(id=1189624193869161363, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189621681917173762, 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/qcjs/EN, language=EN, createTime=1761558109967, createBy=18614031015, updateTime=1761558340679, updateBy=18614031015, name=汽车技术-英文, tplId=1146101810881728533, title=Automobile Technology, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1189625550722311064, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=articleTextType, value=kx, createTime=1761558433466, updateTime=1761558433466, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550688756629, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=banner, value=null, createTime=1761558433458, updateTime=1761558433458, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550739088283, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=grayFlag, value=0, createTime=1761558433470, updateTime=1761558433470, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550676173716, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=logo, value=https://castjournals.cast.org.cn/joweb/qcjs/EN/file/pic?fileId=7En9rzX2QCa/1J8NnKt/Fg==, createTime=1761558433455, updateTime=1761558433455, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550751671197, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=minRunFlag, value=0, createTime=1761558433473, updateTime=1761558433473, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550713922455, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/qcjs/EN/file/pic, createTime=1761558433464, updateTime=1761558433464, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550743282588, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=silenceFlag, value=0, createTime=1761558433471, updateTime=1761558433471, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550705533846, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761558433462, updateTime=1761558433462, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550726505369, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=themeColor, value=null, createTime=1761558433467, updateTime=1761558433467, creator=18614031015, updator=18614031015), WebsiteProps(id=1189625550734893978, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189624193869161363, code=themeStyle, value=null, createTime=1761558433469, updateTime=1761558433469, creator=18614031015, updator=18614031015)])], journalTitle=汽车技术, weixinUrl=null, journalUrl=null, iacademicId=null, status=1, seqNo=null, journalTitleEn=Automobile Technology, journalPhotoCn=rYFtDx/CU9+iX8QTM0ckbw==, journalPhotoEn=oFT2NmUwKPUjZ27C1+d9pw==, journalFirstLetter=A, 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=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/qcjs/CN/10.19620/j.cnki.1000-3703.20230877, detailUrlEn=https://castjournals.cast.org.cn/joweb/qcjs/EN/10.19620/j.cnki.1000-3703.20230877, pdfUrlCn=https://castjournals.cast.org.cn/joweb/qcjs/CN/PDF/10.19620/j.cnki.1000-3703.20230877, pdfUrlEn=https://castjournals.cast.org.cn/joweb/qcjs/EN/PDF/10.19620/j.cnki.1000-3703.20230877, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于门控循环单元和Transformer的车辆轨迹预测方法*
收藏切换
PDF下载
王庆荣 1 , 谭小泽 1 , 朱昌锋 2 , 李裕杰 1
汽车技术 | 智能车辆运动规划与控制技术专题 2024,(7): 1-8
收起
收藏切换
汽车技术 | 智能车辆运动规划与控制技术专题 2024, (7): 1-8
基于门控循环单元和Transformer的车辆轨迹预测方法*
全屏
王庆荣1, 谭小泽1, 朱昌锋2, 李裕杰1
作者信息
  • 1 兰州交通大学,电子与信息工程学院,兰州 730070
  • 2 兰州交通大学,交通运输学院,兰州 730070
Vehicle Trajectory Prediction Method Based on GRU and Transformer
Qingrong Wang1, Xiaoze Tan1, Changfeng Zhu2, Yujie Li1
Affiliations
  • 1 Institute of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070
  • 2 Institute of Transportation, Lanzhou Jiaotong University, Lanzhou 730070
出版时间: 2024-07-24 doi: 10.19620/j.cnki.1000-3703.20230877
文章导航
收藏切换

为增强自动驾驶车辆对动态环境的理解能力及其道路行驶安全性,提出基于门控循环单元(GRU)和Transformer的车辆轨迹预测模型STGTF,使用GRU提取车辆的历史轨迹特征,通过双层多头注意力(MHA)机制提取车辆的时空交互特征,生成预测轨迹。试验结果表明,预测结果的均方根误差(RMSE)平均降低7.3%,STGTF在短期预测和长期预测方面均有不同程度的提升,验证了模型的有效性。

车辆轨迹预测  /  门控循环单元  /  Transformer  /  车辆交互  /  多头注意力机制

In order to enhance the understanding of the dynamic environment of autonomous vehicles and to improve road driving safety, this article proposed a vehicle trajectory prediction STGTF model based on the Gated Recurrent Unit (GRU) and Transformer that used the GRU to extract the historical trajectory features of vehicles, and used a two-layer Multi-Headed Attention (MHA) mechanism to extract the spatio-temporal interaction features of vehicles, generating the predicted trajectories. The experimental results show that the Root-Mean-Square Error (RMSE) of the predicted results decrease by 7.3% on average, STGTF model has different degrees of improvement compared with other existing methods for both short-term prediction and long-term prediction, proving validity of this model.

Vehicle trajectory prediction  /  Gated Recurrent Unit (GRU)  /  Transformer  /  Vehicle interaction  /  Multi-head attention mechanism
王庆荣, 谭小泽, 朱昌锋, 李裕杰. 基于门控循环单元和Transformer的车辆轨迹预测方法*. 汽车技术, 2024 , (7) : 1 -8 . DOI: 10.19620/j.cnki.1000-3703.20230877
Qingrong Wang, Xiaoze Tan, Changfeng Zhu, Yujie Li. Vehicle Trajectory Prediction Method Based on GRU and Transformer[J]. Automobile Technology, 2024 , (7) : 1 -8 . DOI: 10.19620/j.cnki.1000-3703.20230877
自动驾驶系统中,车辆轨迹预测直接关联车辆的决策与控制,是实现智能交通的核心技术之一。车辆行驶轨迹通常受周围车辆影响[1],为确保自动驾驶车辆能够安全、有效地适应环境变化,需实时了解临近车辆的驾驶状态和意图[2]。因此,车辆轨迹预测对防止交通事故的发生、提升道路行驶的安全性具有重要意义。
传统车辆轨迹预测方法主要依靠物理模型或机器学习方法,如运动学和动力学模型、隐马尔可夫模型和卡尔曼滤波等[3]。此类方法适用于驾驶意图识别和短期的轨迹预测,无法准确理解复杂场景,导致长期预测效果不佳[4]。随着深度学习的发展,人工神经网络逐渐应用于轨迹预测问题[5]。Deo等[6]提出车辆轨迹预测模型卷积社交长短时记忆(Convolutional Social Long Short Term Memory,CS-LSTM)网络,使用卷积池化层代替社交长短时记忆(Social Long Short Term Memory,S-LSTM)网络[7]的池化层,以提取周围车辆的交互信息进行轨迹预测。李文礼等[8]在社交生成对抗网络(Social Generative Adversarial Network,S-GAN)模型[9]的基础上考虑车辆的行驶状态参数和形状尺寸,并建立车辆间交互影响力场模型,体现了车辆间的复杂空间交互影响特性。为加强车辆间交互关系,可通过引入注意力机制聚焦历史轨迹数据中的关键信息,从而提取车辆间交互特征[10]。Messaoud等[11]提出基于长短时记忆(Long Short Term Memory,LSTM)网络的编、解码器框架,并采用多头注意力(Multi-Head Attention,MHA)机制强调相邻车辆间的交互作用以提高轨迹预测的精度。Liu等[12]提出mmTransformer(Multimodal Motion Transformer)模型,使用多头自注意机制捕捉相邻车辆的交互特征,并采用基于区域的训练策略实现了多模态轨迹预测,但此类方法仅关注交互车辆的空间相关性,忽略了交互的时间依赖性。
在动态密集的驾驶环境中,车辆运动受到周围车辆的交互作用通常随着时间动态变化。鉴于此,本文提出基于门控循环单元(Gated Recurrent Unit,GRU)和Transformer的车辆轨迹预测模型STGTF(Spatio-Temporal GRU-Transformer),利用GRU编码历史轨迹数据提取车辆的历史轨迹特征,设计具有双层多头注意力机制的Transformer编码器,同时捕获车辆间的时空交互特征,使用Transformer解码器生成预测轨迹。
根据目标车辆的历史轨迹以及交通环境信息准确预测目标车辆未来一段时间的行驶轨迹[13]。以场景内所有车辆在历史时段Tp内的轨迹序列P作为输入:
P = p T - T p , , p t , , p T p t = p t 1 , , p t i , , p t n p t i = x t i , y t i , v t i , u t i , c t i
式中:T为当前时刻;Tp为历史时间;n为车辆总数量;ptt时刻车辆的历史特征; p t it时刻车辆i的历史特征,包括车辆it时刻的横坐标 x t i、纵坐标 y t i、瞬时速度 v t i、瞬时加速度 u t i和车辆类型 c t i
模型的输出为场景内所有车辆在未来时段Tf内的轨迹序列F
F = f T + 1 , ,   f t , ,   f T + T f f t = f t 1 , ,   f t i , ,   f t n f t i = x t i , y t i
式中:ftt时刻车辆的预测特征; f t it时刻车辆i的预测特征,包括目标车辆的横坐标 x t i和纵坐标 y t i
t时刻,以采集区域内目标车辆的前轴中心坐标(x0,y0)为原点,以行驶方向为x轴、垂直方向为y轴建立坐标系,如图1所示。假设周围车辆使用局部坐标(xn,yn),所在坐标系的相对坐标为 x n ' , y n ',则有:
x n ' = x n - x 0 y n ' = y n - y 0
当目标车辆行驶方向的偏转角为θ时,周围车辆的行驶方向不变,此时周围车辆的相对坐标为 x n θ , y n θ
θ = a r c t a n x 0 t - x 0 t - 1 y 0 t - y 0 t - 1
x n θ = x n ' c o s θ + y n ' s i n θ y n θ = - x n ' c o s θ + y n ' s i n θ
式中: x 0 t , y 0 tt时刻车辆的前轴中心坐标。
为实现车辆在复杂交通场景中轨迹的准确预测,本文提出基于GRU和改进Transformer的时空交互注意网络STGTF,如图2所示。STGTF分为以下3个部分:
a. 轨迹特征提取模块,通过GRU对输入的车辆历史轨迹序列P进行编码,提取历史轨迹的时间序列特征。
b. 交互特征提取模块,使用双层多头注意力机制的Transformer编码器对编码向量进行注意力加权,先获取车辆之间的空间交互特征,再获取空间交互特征序列的时间依赖关系,最终得到车辆的时空交互特征。
c. 轨迹解码模块,通过Transformer解码器实现对时空交互特征的解码,生成最终预测轨迹序列F
轨迹特征提取模块首先通过多层感知机(Multi-Layer Perceptron,MLP)将车辆各时刻运动状态转换成嵌入表示:

et=fMLP(pt,We)

式中:We为嵌入权重矩阵,fMLP为MLP层。
利用GRU对嵌入编码并提取历史轨迹特征,如图3所示。其中:ztt时刻更新门(Update Gate)的输出,控制要被继续传递到未来的信息量;rtt时刻重置门(Reset Gate)的输出,控制将要遗忘的信息量。
通过重置门对t时刻车辆的运动状态et和上一时刻传递的隐藏状态ht-1计算出需要被遗忘的信息量。同时,使用更新门计算需要被传递的信息量,将重置门的输出rt与上一时刻的隐藏状态ht-1相乘并与et进行映射,计算候选隐藏状态。最后,将上述记忆内容与当前记忆内容叠加进行更新记忆,得到t时刻的隐藏状态:
r t = σ W r [ h t - 1 , e t ] + b r z t = σ W z [ h t - 1 , e t ] + b z h ˜ t = t a n h W h [ r t , h t - 1 , e t ] + b h
h t = z t × h t + ( 1 - z t ) × h t - 1
式中:σ为S型函数(Sigmoid函数),htt时刻的隐藏状态, h ˜ tt时刻候选隐藏状态,tanh为激活函数,WrWzWh分别为重置门、更新门和候选隐藏单元的权重矩阵,brbzbh分别为重置门、更新门和候选隐藏单元的偏置向量。
对所有嵌入进行处理,得到所有车辆的运动隐藏状态后进行信息编码,作为车辆的历史轨迹特征H={ht1,…,hti,…,htn}。
在历史轨迹特征提取模块中,各车辆均以独立的GRU单元编码,不考虑各车辆间交互关系。实际上,车辆的行驶轨迹受自身驾驶状态影响的同时,还会受到其他车辆驾驶行为交互作用的影响。由于不同车辆对目标车辆的影响程度存在差异,为了考虑车辆交互作用的影响程度,使用Transformer编码器的多头注意力机制对车辆历史轨迹的编码信息建立各车辆间的联系。
Transformer编码器结构如图4所示,包含位置编码(Positional Encoding,PE)、多头注意力机制层、前馈网络(Feed Forward Network,FFN)层、残差连接和层归一化(Residual Connection and Layer Normalization,Add&Norm)。车辆间交互作用存在空间性及时间依赖性,因此,本文将Transformer编码器中的多头注意力机制层设计成双层多头注意力机制。
首层MHA中,为获取交互作用的空间性,在每个时间戳内计算车辆之间交互作用程度,对各车辆与周围车辆的历史轨迹特征进行相似度计算,通过加权求和得到周围车辆的注意力权重。第二层MHA中,计算不同时间戳之间的注意力权重,使模型更多地关注上层中有强交互作用的时间戳内的信息。
由于Transformer中不存在处理时序数据的循环神经网络(Recurrent Neural Network,RNN)[14]结构,所以在输入嵌入(Input Embedding)后增加位置编码(Positional Encoding),以提供输入的时序信息:
P E ( l , 2 i ) = s i n ( l / 10   000 2 i / d m o d e l ) P E ( l , 2 i + 1 ) = c o s ( l / 10   000 2 i / d m o d e l )
式中:PE(l,2i)PE(l,2i+1)分别为偶数位置和奇数位置的编码,l为轨迹点在整条轨迹中的位置,i为对应轨迹点向量的维度,dmodel为嵌入的总维度。
计算首层MHA中各车辆与其他车辆间在空间中交互作用的程度,注意力机制的原理如图5所示。
首先,将车辆的历史特征通过3个权重矩阵 W q W k W v分别映射得到对应的查询(Query)矩阵Q、键(Key)矩阵K和值(Value)矩阵V
Q = q t 1 , , q t i , , q t n K = k t 1 , , k t i , , k t n V = v t 1 , , v t i , , v t n
q t i = W q h t i k t i = W k h t i v t i = W v h t i
式中: q t i k t i v t i分别为t时刻车辆i的查询向量、键向量和值向量。
通过计算车辆i q t i和车辆j k t i的点积得到车辆j对车辆i的注意力得分为 s t ( i , j ) = q t i k t i,并通过Softmax函数进行归一化处理,再将其与车辆j v t i相乘,对所有加权值求和,得到对应的注意力权重:
a t , i = i = 1 n j = 1 n S o f t m a x s t ( i , j ) d k e y v t j
式中:dkey为值向量的维度。
将多个原始特征向量WaWkWv映射到不同空间,计算相似度。通过拼接(Concat)操作拼接各注意力权重,从而得到包含所有注意力头的权重:
A t = f C o n c a t ( a t 1 , , a t h ) W o
式中: f C o n c a t为拼接(Concat)函数,Ath个注意力头的权重, a t h为第h个头的注意力权重,Wo为线性转换的权重矩阵,h为映射的空间的数量,即头数。
车辆在各时间戳内均需进行计算,因此,车辆的空间交互特征序列 H - = A T - T p , , A T。将空间交互特征序列 H -作为第二层MHA的输入,同理,车辆的时空交互特征序列 H ˜为:
H ˜ = { A 1 , , A i , , A n } A i = f C o n c a t ( a t 1 , , a t h ) W o a i , t = t = T - T p T t ' = T - T p T S o f t m a x q t k t ' d k e y v t '
使用Transformer解码器进行解码,包括掩码多头注意力(Masked Multi-Head Attention,MMHA)层、编码器-解码器注意力(Encoder-Decoder Attention,EDA)层、前馈网络层、残差连接和层归一化。
MMHA是解码器特有结构,为保证未来的信息不被提前解码,需要对当前时刻之后的信息进行掩码处理。EDA通过计算编码器的输出Key、Value和解码器上一时刻的输出Query之间的权重,使解码器能够将注意力集中于对当前车辆交互影响更大的车辆。此外,编码器或解码器均由数量为N的相同的编码层或解码层堆叠而成。
解码后,需对解码器的输出进行全连接线性映射,并通过Softmax函数归一化预测值,输出轨迹的预测位置,获得最终预测轨迹序列。
本文使用美国联邦公路管理局采集的公开轨迹数据集NGSIM(Next Generation Simulation),由多个高架摄像机以10 Hz的采样频率记录的真实高速公路驾驶场景组成。本文使用US 101和I-80两个场景下的数据,如图6所示。每个数据集有45 min数据,其中,稀疏、中等和稠密3种交通状况各15 min,数据内容包括车辆的坐标、速度、加速度等。
本文提出的模型在NVIDIA GeForce RTX 3090图形处理器的试验环境下进行,在PyTorch框架下使用Python语言完成。其中,GRU隐藏单元数为64,并使用tanh函数将输出约束在[-1,1]范围内。MHA头数为8,Transformer编解码器设置为6层。采用自适应矩估计(Adaptive Moment Estimation,Adam)优化器[15],初始学习率设置为0.000 5。每次训练的样本批量大小为32,模型迭代次数为400次。每个场景共有8 s的观测数据,以前3 s的数据作为历史轨迹数据,后5 s的数据作为需要预测的未来轨迹数据,并以7∶1∶2的比例划分为训练集、验证集与测试集。
在训练过程中,将最小平方误差损失作为模型的损失函数,即 l L o s s = t = T T + T f f t - f ^ t 2。其中,ft为模型预测的轨迹, f ^ t为真实轨迹。
模型评估采用的评价指标为均方根误差(Root Mean Square Error,RMSE),即在预测时间内预测轨迹位置和真实轨迹位置的平均欧几里得距离(Mean Euclidean Distance,MED)之间的差异,反映模型的平均预测性能。RMSE的值越小,表示模型的预测轨迹与真实轨迹之间的偏移越小,预测效果越好:
e R M S E = 1 n i = 1 n x t i - x ^ t i 2 + y t i - y ^ t i 2
式中: x t i y t i分别为t时刻轨迹预测位置的横、纵坐标, x ^ t i y ^ t i分别为t时刻轨迹真实位置的横、纵坐标。
为了验证临近车辆与待预测车辆的交互作用范围,选择纵向距离前、后分别为0 m、50 m、100 m、150 m和横向距离左、右分别为0 m、4 m、8 m进行分析,试验结果如图7所示。
图7a可知,当周围车辆与待预测目标车辆的纵向距离为前、后100 m时,模型可以取得优于其他距离的结果。由图7b可知,当周围车辆与待预测目标车辆的横向距离为左、右4 m时,模型可以取得优于其他距离的结果。以上结果说明,不考虑周围车辆的交互影响或考虑过多周围车辆的交互影响均会降低预测精度。
为了评估模型性能,将本文模型STGTF与基准模型进行比较,不同预测时域tp条件下的预测结果如表1图8所示。对比模型包括S-LSTM模型、S-GAN模型、CS-LSTM模型和规划轨迹预测(Planning-informed trajectory Prediction,PiP)[16]模型。
由预测结果可知,本文模型STGTF的性能均优于其他基准模型,且在3~5 s内的长期预测中优势明显。在1~2 s内各模型预测性能相近源于短期预测更多受到车辆历史轨迹和即时状态的影响,比长期预测更容易取得较好性能,交互作用的影响在长期预测上表现更为明显。S-LSTM虽然考虑了空间交互作用的影响,但与其他模型相比范围较小,故预测结果较差。S-GAN使用生成式对抗网络,交互作用的范围更大,使得预测效果有所提升。CS-LSTM使用了卷积池化层,相较于S-LSTM的最大池化层能够更好地捕捉到车辆之间的交互,但预测精度仍略低于考虑自车运动趋势的PiP模型。本文提出的STGTF同时考虑了交互的空间相关性和时间依赖性,使模型能够更稳健地对车辆之间的交互作用进行建模,从而预测结果更佳。
为进一步验证本文模型各模块的有效性,对STGTF模型进行消融实验,结果如表2图9所示。
消融实验结果显示,从STGTF中去除不同部分后,得到的模型普遍存在不同程度的性能退化。其中,STTF(Spatio-Temporal Transformer)为去掉历史轨迹特征提取模块的模型,将历史轨迹数据经过一层全连接层直接输入时空交互特征提取模块,STGTF与之相比,RMSE降低约5.2%,说明提取车辆的历史轨迹时间序列特征和运动状态特征能够提升模型的预测能力。TGTF(Temporal GRU-Transformer)为只考虑待预测车辆历史轨迹而不考虑空间交互作用的模型,STGTF与之相比,RMSE降低约10.2%。结果表明,随着预测时长的增加,未考虑车辆交互的模型性能退化愈发明显,说明车辆间交互作用影响车辆的轨迹预测,在长期预测问题中格外关键。SGTF(Spatial GRU-Transformer)只考虑交互作用的空间相关性而不考虑时间依赖性的模型,STGTF与之相比,RMSE降低约2.4%,说明考虑交互作用的时间依赖性是有必要的。综上可知,本文模型各模块对提升模型性能均有效。
对模型进行定性分析,选择具有代表性的2种场景,并对其预测结果进行可视化,如图10所示。
图10a可知,本文模型比较准确地预测出车辆直行时的未来轨迹,同时,该场景下周围车辆的换道行为,并未影响对目标车辆轨迹的预测,说明本文模型具有一定的抗干扰性。图10b中,本文模型虽然在换道点附近的预测结果稍有偏差,但仍可准确地预测车辆的行驶趋势。
本文模型与对比模型在换道场景中的预测轨迹对比结果如图11所示。在变道的复杂场景中,本文模型同样能够更准确地预测未来轨迹,且更接近真实轨迹。
为提升自动驾驶技术在复杂交通环境下的安全性,本文建立了基于GRU和Transformer的车辆轨迹预测方法STGTF,实现了对车辆未来行驶轨迹的预测。通过RMSE比较预测轨迹与真实轨迹之间的差异,验证了本文模型的性能及各组成部分的有效性。定量试验结果表明,本文模型与现有模型相比,在1~5 s内的预测精度均有所提升,且提升效果随时间增长明显。定性试验结果表明,本文模型能够得到较为接近真实轨迹的预测结果。
未来,将考虑引入道路环境数据信息,使预测结果符合道路的几何特征,增强预测结果的可靠性,进一步提升自动驾驶汽车路径规划的安全性。
  • *国家自然科学基金项目(71961016)
  • 国家自然科学基金项目(72161024)
  • 甘肃省教育厅“双一流”重大研究项目(GSSYLXM-04)
参考文献 引证文献
排序方式:
[1]
HU L, ZHOU X Q, ZHANG X, et al. A Review on Key Challenges in InIntelligent Vehicles: Safety and Driver-Oriented Features[J]. IET Intelligent Transport Systems, 2021, 15(9): 1093-1105.
[2]
LIU J B, MAO X Y, FANG Y Q, et al. A Survey on Deep-Learning Approaches for Vehicle Trajectory Prediction in Autonomous Driving[C]// 2021 IEEE International Conference on Robotics and Biomimetics (ROBIO). Sanya, China: IEEE, 2021: 978-985.
[3]
LEFÈVRE S, VASQUEZ D, LAUGIER C. A Survey on Motion Prediction and Risk Assessment for Intelligent Vehicles[J]. Robomech Journal, 2014, 1(1): 1-14.
[4]
CHEN W H, WANG F F, SUN H B. S2TNet: Spatio-Temporal Transformer Networks for Trajectory Prediction in Autonomous Driving[C]// Asian Conference on Machine Learning. Online: PMLR, 2021: 454-469.
[5]
高振海, 鲍明喜, 高菲, 等. 智能驾驶系统对周边交通车辆预期轨迹的单模态网络预测方法[J]. 汽车技术, 2022(11): 1-9.
GAO Z H, BAO M X, GAO F, et al. A Uni-Modal Network Prediction Method for Surrounding Vehicle Expected Trajectory in Intelligent Driving System[J]. Automobile Technology, 2022(11): 1-9.
[6]
DEO N, TRIVEDI M M. Convolutional Social Pooling for Vehicle Trajectory Prediction[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. Salt Lake City, UT, USA: IEEE, 2018: 1468-1476.
[7]
ALAHI A, GOEL K, RAMANATHAN V, et al. Social LSTM: Human Trajectory Prediction in Crowded Spaces[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 961-971.
[8]
李文礼, 韩迪, 石晓辉, 等. 基于时-空注意力机制的车辆轨迹预测[J]. 中国公路学报, 2023, 36(1): 226-239.
LI W L, HAN D, SHI X H, et al. Vehicle Trajectory Prediction Based on Spatial-Temporal Attention Mechanism[J]. China Journal of Highway and Transport, 2023, 36(1): 226-239.
[9]
AGRIM G, JUSTIN J, LI F F, et al. Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 2255-2264.
[10]
LI L, ZHAO W Z, WANG C Y, et al. BRAM-ED: Vehicle Trajectory Prediction Considering the Change of Driving Behavior[J]. IEEE/ASME Transactions on Mechatronics, 2022, 27(6): 5690-5700.
[11]
MESSAOUD K, YAHIAOUI A, VERROUST-BLONDET A, et al. Attention Based Vehicle Trajectory Prediction[J]. IEEE Transactions on Intelligent Vehicles, 2021, 6(1): 175-185.
[12]
LIU Y C, ZHANG J H, FANG L J, et al. Multimodal Motion Prediction with Stacked Transformers[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville, TN, USA: IEEE, 2021: 7573-7582.
[13]
赵树恩, 苏天彬, 赵东宇. 基于图神经网络的交互车辆驾驶意图识别及轨迹预测[J]. 汽车技术, 2023(7): 24-30.
ZHAO S E, SU T B, ZHAO D Y. Interactive Vehicle Driving Intention Recognition and Trajectory Prediction Based on Graph Neural Network[J]. Automobile Technology, 2023(7): 24-30.
[14]
CHOI S, KIM J, YEO H. Attention-Based Recurrent Neural Network for Urban Vehicle Trajectory Prediction[J]. Procedia Computer Science, 2019, 151: 327-334.
[15]
ZHANG Z J. Improved Adam Optimizer for Deep Neural Networks[C]// 2018 IEEE/ACM 26th International Symposium on Quality of Service (IWQoS). Banff, AB, Canada: IEEE, 2018: 1-2.
[16]
SONG H R, DING W C, CHEN Y X, et al. PiP: Planning-Informed Trajectory Prediction for Autonomous Driving[C]// European Conference on Computer Vision. Glasgow, United Kingdom: Springer, Cham, 2020: 598-614.
2024年第卷第7期
PDF下载
211
88
引用本文
BibTeX
文章信息
doi: 10.19620/j.cnki.1000-3703.20230877
  • 首发时间:2025-12-22
  • 出版时间:2024-07-24
补充材料
相关文章
文章信息
作者
出版历史
基金
*国家自然科学基金项目(71961016)
国家自然科学基金项目(72161024)
甘肃省教育厅“双一流”重大研究项目(GSSYLXM-04)
作者信息
    1 兰州交通大学,电子与信息工程学院,兰州 730070
    2 兰州交通大学,交通运输学院,兰州 730070
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/qcjs/CN/10.19620/j.cnki.1000-3703.20230877
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