Article(id=1154016879124402200, tenantId=1146029695717560320, journalId=1152916057816748034, issueId=1154016878675611672, articleNumber=null, orderNo=null, doi=10.3969/j.issn.2095–1469.2024.01.01, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1650988800000, receivedDateStr=2022-04-27, revisedDate=1656432000000, revisedDateStr=2022-06-29, acceptedDate=null, acceptedDateStr=null, onlineDate=1753068664579, onlineDateStr=2025-07-21, pubDate=null, pubDateStr=null, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1753068664579, onlineIssueDateStr=2025-07-21, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1753068664579, creator=13701087609, updateTime=1753068664579, updator=13701087609, issue=Issue{id=1154016878675611672, tenantId=1146029695717560320, journalId=1152916057816748034, year='2024', volume='14', issue='1', pageStart='1', pageEnd='153', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=0, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1753068664473, creator=13701087609, updateTime=1757481546563, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1172525847715136459, tenantId=1146029695717560320, journalId=1152916057816748034, issueId=1154016878675611672, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1172525847715136460, tenantId=1146029695717560320, journalId=1152916057816748034, issueId=1154016878675611672, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=1, endPage=13, ext={EN=ArticleExt(id=1154016879506083866, articleId=1154016879124402200, tenantId=1146029695717560320, journalId=1152916057816748034, language=EN, title=A Review of LiDAR-Based Simultaneous Localization and Mapping Methods for Autonomous Driving, columnId=1153813376129028490, journalTitle=Chinese Journal of Automotive Engineering, columnName=Review and Rrospect, runingTitle=null, highlight=null, articleAbstract=

Simultaneous localization and mapping (SLAM) technology enables autonomous vehicles to estimate their own poses and establish the map of an unknown environment according to the data collected by onboard sensors. SLAM can provide localization information to the decisionmaking module for vehicle planning, and has become one of the research hotspots of autonomous driving technology in recent years. Based on the point cloud data collected by LiDAR, this paper focuses on the SLAM technology applied in autonomous driving. The related research at home and abroad has been reviewed including the frontend odometry, the backend optimization and loop closure detection. Due to the limitations of a single sensor, the opportunities and challenges of multisensor fusion SLAM technology for autonomous driving are discussed based on the research hotspots and difficulties in the field of multisensor fusion.

, 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=Geng ZHANG, Chao YANG, Weida WANG, Ying LI), CN=ArticleExt(id=1154016896291688501, articleId=1154016879124402200, tenantId=1146029695717560320, journalId=1152916057816748034, language=CN, title=基于激光雷达的自动驾驶同步定位与建图方法综述, columnId=1153756964954825490, journalTitle=汽车工程学报, columnName=综述与展望, runingTitle=null, highlight=null, articleAbstract=

同步定位与建图(Simultaneous Localization and Mapping,SLAM)技术可使自动驾驶车辆在未知环境中根据车载传感器采集到的数据估计自身位姿,建立环境地图,为车辆的规划、决策提供定位信息,是近年来自动驾驶技术研究的热点之一。基于车载激光雷达的点云数据,聚焦SLAM技术在自动驾驶领域的应用,围绕前端里程计、后端优化和回环检测技术,对国内外相关研究进行综述。考虑到单一传感器的局限性,结合目前多传感器融合研究的热点与难点,展望了自动驾驶多传感器融合SLAM技术在自动驾驶领域的机遇与挑战。

, correspAuthors=null, authorNote=null, correspAuthorsNote=

李颖(1991-),女,安徽池州人,预聘助理教授,主要研究方向为自动驾驶感知、多传感器融合感知、移动激光扫描、点云处理、几何和语义建模和高精度地图。Tel: 18626876417 E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=xGBruUoONqqXHnGoQS0S1g==, magXml=ELma8/e0wiq4bNaroHk2Mw==, pdfUrl=null, pdf=ZloVy1Cz3iff21dlbqW0qw==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=z2XYcSNDqZjUcVQkGdsoXw==, mapNumber=null, authorCompany=null, fund=null, authors=

张庚(1997-),男,北京市人,硕士研究生,主要研究方向为自动驾驶同步定位与建图。Tel: 18201625972 E-mail:

, authorsList=张庚, 杨超, 王伟达, 李颖)}, authors=[Author(id=1154016898086850618, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=3120200430@bit.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154016898179125308, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898086850618, language=EN, stringName=Geng ZHANG, firstName=Geng, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Technology Beijing 100081 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154016898254622781, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898086850618, language=CN, stringName=张庚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京理工大学 北京 100081, bio={"img":"i9lLnFWlYlilybvOyxsKHA==","content":"

张庚(1997-),男,北京市人,硕士研究生,主要研究方向为自动驾驶同步定位与建图。Tel: 18201625972 E-mail:

"}, bioImg=i9lLnFWlYlilybvOyxsKHA==, bioContent=

张庚(1997-),男,北京市人,硕士研究生,主要研究方向为自动驾驶同步定位与建图。Tel: 18201625972 E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154016897919078454, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, xref=null, ext=[AuthorCompanyExt(id=1154016897952632887, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Technology Beijing 100081 China), AuthorCompanyExt(id=1154016897956827192, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京理工大学 北京 100081)])]), Author(id=1154016898309148735, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, 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=1154016898380451905, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898309148735, language=EN, stringName=Chao YANG, firstName=Chao, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Technology Beijing 100081 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154016898434977858, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898309148735, language=CN, stringName=杨超, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京理工大学 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154016897919078454, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, xref=null, ext=[AuthorCompanyExt(id=1154016897952632887, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Technology Beijing 100081 China), AuthorCompanyExt(id=1154016897956827192, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京理工大学 北京 100081)])]), Author(id=1154016898485309508, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, 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=1154016898615332934, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898485309508, language=EN, stringName=Weida WANG, firstName=Weida, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Technology Beijing 100081 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154016898795688007, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898485309508, language=CN, stringName=王伟达, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京理工大学 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154016897919078454, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, xref=null, ext=[AuthorCompanyExt(id=1154016897952632887, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Technology Beijing 100081 China), AuthorCompanyExt(id=1154016897956827192, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京理工大学 北京 100081)])]), Author(id=1154016898837631049, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=ying.li@bit.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154016898887962699, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898837631049, language=EN, stringName=Ying LI, firstName=Ying, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Beijing Institute of Technology Beijing 100081 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154016898980237388, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, authorId=1154016898837631049, language=CN, stringName=李颖, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=北京理工大学 北京 100081, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154016897919078454, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, xref=null, ext=[AuthorCompanyExt(id=1154016897952632887, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Technology Beijing 100081 China), AuthorCompanyExt(id=1154016897956827192, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京理工大学 北京 100081)])])], keywords=[Keyword(id=1154016899521302605, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, orderNo=1, keyword=autonomous driving), Keyword(id=1154016901249355854, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, orderNo=2, keyword=SLAM), Keyword(id=1154016901316464719, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, orderNo=3, keyword=LiDAR), Keyword(id=1154016901404545104, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, orderNo=4, keyword=front-end odometry), Keyword(id=1154016901505208401, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, orderNo=5, keyword=back-end optimization), Keyword(id=1154016901547151442, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, orderNo=6, keyword=loop closure detection), Keyword(id=1154016901584900179, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, orderNo=1, keyword=自动驾驶), Keyword(id=1154016901647814740, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, orderNo=2, keyword=同步定位与建图), Keyword(id=1154016901693952085, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, orderNo=3, keyword=激光雷达), Keyword(id=1154016901735895126, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, orderNo=4, keyword=前端里程计), Keyword(id=1154016901832364119, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, orderNo=5, keyword=后端优化), Keyword(id=1154016901878501464, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, orderNo=6, keyword=回环检测)], refs=[Reference(id=1154016903300370529, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2020, volume=69, issue=3, pageStart=2463, pageEnd=2476, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=CHIANG KW, TSAI GJ, CHU HJ, journalName=IEEE Transactions on Vehicular Technology, refType=null, unstructuredReference=CHIANG KW, TSAI GJ, CHU HJ, et al. Performance Enhancement of INS/GNSS/Refreshed-SLAM Integration for Acceptable Lane-Level Navigation Accuracy[J]. IEEE Transactions on Vehicular Technology, 2020,69(3):2463-2476., articleTitle=Performance Enhancement of INS/GNSS/Refreshed-SLAM Integration for Acceptable Lane-Level Navigation Accuracy, refAbstract=null), Reference(id=1154016903359090786, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=6418, pageEnd=6423, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=CHEN Yunxia, WU Yaohua, XING Han, journalName=Proceedings of the 2017 Chinese Automation Congress (CAC), refType=null, unstructuredReference=CHEN Yunxia, WU Yaohua, XING Han. A Complete Solution for AGV SLAM Integrated with Navigation in Modern Warehouse Environment[C]// Proceedings of the 2017 Chinese Automation Congress (CAC), 2017:6418-6423., articleTitle=A Complete Solution for AGV SLAM Integrated with Navigation in Modern Warehouse Environment, refAbstract=null), Reference(id=1154016903417811043, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2019, volume=16, issue=2, pageStart=1, pageEnd=13, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=KOIDE K, MIURA J, MENEGATTI E, journalName=International Journal of Advanced Robotic Systems, refType=null, unstructuredReference=KOIDE K, MIURA J, MENEGATTI E. A Portable Three-Dimensional LIDAR-Based System for Long-Term and Wide-Area People Behavior Measurement[J]. International Journal of Advanced Robotic Systems, 2019,16(2):1-13., articleTitle=A Portable Three-Dimensional LIDAR-Based System for Long-Term and Wide-Area People Behavior Measurement, refAbstract=null), Reference(id=1154016903463948388, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=2, pageStart=327, pageEnd=332, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=危双丰, 庞帆, 刘振彬, journalName=计算机应用研究, refType=null, unstructuredReference=危双丰, 庞帆, 刘振彬, 等. 基于激光雷达的同时定位与地图构建方法综述[J]. 计算机应用研究, 2020(2):327-332., articleTitle=基于激光雷达的同时定位与地图构建方法综述, refAbstract=null), Reference(id=1154016903535251557, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=2, pageStart=327, pageEnd=332, url=null, language=null, rfNumber=[4], rfOrder=4, authorNames=WEI Shuangfeng, PANG Fan, LIU Zhenbin, journalName=Application Research of Computers, refType=null, unstructuredReference=WEI Shuangfeng, PANG Fan, LIU Zhenbin, et al. Survey of LiDAR-Based SLAM Algorithm[J]. Application Research of Computers, 2020(2):327-332. (in Chinese), articleTitle=Survey of LiDAR-Based SLAM Algorithm, refAbstract=null), Reference(id=1154016903610749030, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2019, volume=2, issue=3, pageStart=173, pageEnd=190, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=REID TG, HOUTS SE, CAMMARATA R, journalName=SAE International Journal of Connected and Automated Vehicles, refType=null, unstructuredReference=REID TG, HOUTS SE, CAMMARATA R, et al. Localization Requirements for Autonomous Vehicles[J]. SAE International Journal of Connected and Automated Vehicles, 2019,2(3):173-190., articleTitle=Localization Requirements for Autonomous Vehicles, refAbstract=null), Reference(id=1154016903669469287, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=满增光, journalName=null, refType=null, unstructuredReference=满增光. 基于激光雷达的室内AGV地图创建与定位方法研究[D]. 南京: 南京航空航天大学, 2014., articleTitle=基于激光雷达的室内AGV地图创建与定位方法研究, refAbstract=null), Reference(id=1154016903715606632, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=MAN Zengguang, journalName=null, refType=null, unstructuredReference=MAN Zengguang. Research on Mapping and Localization of Indoor AGV Based on LADAR[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2014. (in Chinese), articleTitle=Research on Mapping and Localization of Indoor AGV Based on LADAR, refAbstract=null), Reference(id=1154016903761743977, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=1, pageStart=7, pageEnd=9, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=任兴明, journalName=电子工程师, refType=null, unstructuredReference=任兴明. GPS/GLONASS双星定位差分解算失效分析及解决措施[J]. 电子工程师, 2008(1):7-9., articleTitle=GPS/GLONASS双星定位差分解算失效分析及解决措施, refAbstract=null), Reference(id=1154016903845630058, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=1, pageStart=7, pageEnd=9, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=REN Xingming, journalName=Informatization Research, refType=null, unstructuredReference=REN Xingming. Failure Analysis and Solution of GPS/GLONASS Double Satellite Differential Calculation[J]. Informatization Research, 2008(1):7-9. (in Chinese), articleTitle=Failure Analysis and Solution of GPS/GLONASS Double Satellite Differential Calculation, refAbstract=null), Reference(id=1154016903904350315, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=4, pageStart=61, pageEnd=69, url=null, language=null, rfNumber=[8], rfOrder=10, authorNames=李晨曦, 张军, 靳欣宇, journalName=北京联合大学学报, refType=null, unstructuredReference=李晨曦, 张军, 靳欣宇, 等. 激光雷达SLAM技术及其在无人车中的应用研究进展[J]. 北京联合大学学报, 2017(4):61-69., articleTitle=激光雷达SLAM技术及其在无人车中的应用研究进展, refAbstract=null), Reference(id=1154016903958876268, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=4, pageStart=61, pageEnd=69, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=LI Chenxi, ZHANG Jun, JIN Xinyu, journalName=Journal of Beijing Union University, refType=null, unstructuredReference=LI Chenxi, ZHANG Jun, JIN Xinyu, et al. Review of LIDAR SLAM Technology and its Application in Autonomous Vehicles[J]. Journal of Beijing Union University, 2017(4):61-69. (in Chinese), articleTitle=Review of LIDAR SLAM Technology and its Application in Autonomous Vehicles, refAbstract=null), Reference(id=1154016904013402221, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=李昀泽, journalName=null, refType=null, unstructuredReference=李昀泽. 基于激光雷达的室内机器人SLAM研究[D]. 广州: 华南理工大学, 2016., articleTitle=基于激光雷达的室内机器人SLAM研究, refAbstract=null), Reference(id=1154016904072122478, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=LI Yunze, journalName=null, refType=null, unstructuredReference=LI Yunze. Research on SLAM of Indoor Robot Based in LIDAR[D]. Guangzhou: South China University of Technology, 2016. (in Chinese), articleTitle=Research on SLAM of Indoor Robot Based in LIDAR, refAbstract=null), Reference(id=1154016904143425647, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2010, volume=null, issue=null, pageStart=22, pageEnd=29, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=KONOLIGE K, GRISETTI G, KUEMMERLE R, journalName=Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, refType=null, unstructuredReference=KONOLIGE K, GRISETTI G, KUEMMERLE R, et al. Efficient Sparse Pose Adjustment for 2D Mapping[C]// Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2010:22-29., articleTitle=Efficient Sparse Pose Adjustment for 2D Mapping, refAbstract=null), Reference(id=1154016904214728816, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=15, authorNames=ZHANG Ji, SINGH S, journalName=null, refType=null, unstructuredReference=ZHANG Ji, SINGH S. LOAM: Lidar Odometry and Mapping in Real-Time[R]. 2014 . 10.15607/RSS.2014.Χ.007 ., articleTitle=LOAM: Lidar Odometry and Mapping in Real-Time, refAbstract=null), Reference(id=1154016904260866161, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=1374, pageEnd=1381, url=null, language=null, rfNumber=[12], rfOrder=16, authorNames=LING Yonggen, SHEN Shaojie, journalName=Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), refType=null, unstructuredReference=LING Yonggen, SHEN Shaojie. Building Maps for Autonomous Navigation Using Sparse Visual SLAM Features[C]// Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017:1374-1381., articleTitle=Building Maps for Autonomous Navigation Using Sparse Visual SLAM Features, refAbstract=null), Reference(id=1154016904340557938, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2012, volume=null, issue=null, pageStart=3354, pageEnd=3361, url=null, language=null, rfNumber=[13], rfOrder=17, authorNames=GEIGER A, LENZ P, URTASUN R, journalName=Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=GEIGER A, LENZ P, URTASUN R. Are We Ready for Autonomous Driving? The KITTI Vision Benchmark Suite[C]// Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, 2012:3354-3361., articleTitle=Are We Ready for Autonomous Driving? The KITTI Vision Benchmark Suite, refAbstract=null), Reference(id=1154016904395083891, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=2443, pageEnd=2451, url=null, language=null, rfNumber=[14], rfOrder=18, authorNames=SUN Pei, KRETZSCHMAR H, DOTIWALLA X, journalName=Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, refType=null, unstructuredReference=SUN Pei, KRETZSCHMAR H, DOTIWALLA X, et al. Scalability in Perception for Autonomous Driving: Waymo open dataset[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020:2443-2451., articleTitle=Scalability in Perception for Autonomous Driving: Waymo open dataset, refAbstract=null), Reference(id=1154016904449609844, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2017, volume=36, issue=1, pageStart=3, pageEnd=15, url=null, language=null, rfNumber=[15], rfOrder=19, authorNames=MADDERN W, PASCOE G, LINEGAR C, journalName=The International Journal of Robotics Research, refType=null, unstructuredReference=MADDERN W, PASCOE G, LINEGAR C, et al. 1 Year, 1 000 km: The Oxford RobotCar Dataset[J]. The International Journal of Robotics Research, 2017,36(1):3-15., articleTitle=1 Year, 1 000 km: The Oxford RobotCar Dataset, refAbstract=null), Reference(id=1154016904495747189, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=1110, pageEnd=1116, url=null, language=null, rfNumber=[16], rfOrder=20, authorNames=JIANG Peng, OSTEEN P, WIGNESS M, journalName=Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), refType=null, unstructuredReference=JIANG Peng, OSTEEN P, WIGNESS M, et al. RELLIS-3D Dataset: Data, Benchmarks and Analysis[C]// Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), 2021:1110-1116., articleTitle=RELLIS-3D Dataset: Data, Benchmarks and Analysis, refAbstract=null), Reference(id=1154016904554467446, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=6, pageStart=914, pageEnd=930, url=null, language=null, rfNumber=[17], rfOrder=21, authorNames=宗文鹏, 李广云, 李明磊, journalName=中国光学, refType=null, unstructuredReference=宗文鹏, 李广云, 李明磊, 等. 激光扫描匹配方法研究综述[J]. 中国光学, 2018(6):914-930., articleTitle=激光扫描匹配方法研究综述, refAbstract=null), Reference(id=1154016904600604791, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=6, pageStart=914, pageEnd=930, url=null, language=null, rfNumber=[17], rfOrder=22, authorNames=ZONG Wenpeng, LI Guangyun, LI Minglei, journalName=Chinese Optics, refType=null, unstructuredReference=ZONG Wenpeng, LI Guangyun, LI Minglei, et al. A Survey of Laser Scan Matching Methods[J]. Chinese Optics, 2018(6):914-930. (in Chinese), articleTitle=A Survey of Laser Scan Matching Methods, refAbstract=null), Reference(id=1154016904697073786, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=1987, volume=PAMI-9, issue=5, pageStart=698, pageEnd=700, url=null, language=null, rfNumber=[18], rfOrder=23, authorNames=ARUN KS, HUANGT S, BLOSTEIN S D, journalName=IEEE Transactions on Pattern Analysis and Machine Intelligence, refType=null, unstructuredReference=ARUN KS, HUANGT S, BLOSTEIN S D. Least-Squares Fitting of Two 3-D Point Sets[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1987,PAMI-9(5):698-700., articleTitle=Least-Squares Fitting of Two 3-D Point Sets, refAbstract=null), Reference(id=1154016904768376956, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=null, pageStart=19, pageEnd=25, url=null, language=null, rfNumber=[19], rfOrder=24, authorNames=CENSI A, journalName=Proceedings of the 2008 IEEE International Conference on Robotics and Automation, refType=null, unstructuredReference=CENSI A. An ICP Variant Using a Point-to-Line Metric[C]// Proceedings of the 2008 IEEE International Conference on Robotics and Automation, 2008:19-25., articleTitle=An ICP Variant Using a Point-to-Line Metric, refAbstract=null), Reference(id=1154016904864845951, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=742, pageEnd=749, url=null, language=null, rfNumber=[20], rfOrder=25, authorNames=SERAFIN J, GRISETTI G, journalName=Proceedings of the 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), refType=null, unstructuredReference=SERAFIN J, GRISETTI G. NICP: Dense Normal Based Point Cloud Registration[C]// Proceedings of the 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2015:742-749., articleTitle=NICP: Dense Normal Based Point Cloud Registration, refAbstract=null), Reference(id=1154016904915177601, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2006, volume=null, issue=null, pageStart=2291, pageEnd=2296, url=null, language=null, rfNumber=[21], rfOrder=26, authorNames=CENSI A, journalName=Proceedings of the 2006 IEEE International Conference on Robotics and Automation, ICRA 2006, refType=null, unstructuredReference=CENSI A. Scan Matching in a Probabilistic Framework[C]// Proceedings of the 2006 IEEE International Conference on Robotics and Automation, ICRA 2006, 2006:2291-2296., articleTitle=Scan Matching in a Probabilistic Framework, refAbstract=null), Reference(id=1154016904973897859, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=14, issue=8, pageStart=3436, pageEnd=3446, url=null, language=null, rfNumber=[22], rfOrder=27, authorNames=LYU Chen, XING Yang, ZHANG Junzhi, journalName=IEEE Transactions on Industrial Informatics, refType=null, unstructuredReference=LYU Chen, XING Yang, ZHANG Junzhi, et al. Levenberg-Marquardt Backpropagation Training of Multi-layer Neural Networks for State Estimation of a Safety-Critical Cyber-Physical System[J]. IEEE Transactions on Industrial Informatics, 2018,14(8):3436-3446., articleTitle=Levenberg-Marquardt Backpropagation Training of Multi-layer Neural Networks for State Estimation of a Safety-Critical Cyber-Physical System, refAbstract=null), Reference(id=1154016905028423813, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=4390, pageEnd=4396, url=null, language=null, rfNumber=[23], rfOrder=28, authorNames=WANG Han, WANG Chen, CHEN Chunlin, journalName=Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), refType=null, unstructuredReference=WANG Han, WANG Chen, CHEN Chunlin, et al. F-LOAM: Fast LiDAR Odometry and Mapping[C]// Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021:4390-4396., articleTitle=F-LOAM: Fast LiDAR Odometry and Mapping, refAbstract=null), Reference(id=1154016905112309896, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=4758, pageEnd=4765, url=null, language=null, rfNumber=[24], rfOrder=29, authorNames=SHAN Tixiao, ENGLOT B, journalName=Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), refType=null, unstructuredReference=SHAN Tixiao, ENGLOT B. LeGO-LOAM: Lightweight and Ground-Optimized Lidar Odometry and Mapping on Variable Terrain[C]// Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018:4758-4765., articleTitle=LeGO-LOAM: Lightweight and Ground-Optimized Lidar Odometry and Mapping on Variable Terrain, refAbstract=null), Reference(id=1154016905171030154, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=6764, pageEnd=6769, url=null, language=null, rfNumber=[25], rfOrder=30, authorNames=FOSSEL J, TUYLS K, SCHNIEDERS B, journalName=Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), refType=null, unstructuredReference=FOSSEL J, TUYLS K, SCHNIEDERS B, et al. NOctoSLAM: Fast Octree Surface Normal Mapping and Registration[C]// Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017:6764-6769., articleTitle=NOctoSLAM: Fast Octree Surface Normal Mapping and Registration, refAbstract=null), Reference(id=1154016905225556107, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=11619, pageEnd=11625, url=null, language=null, rfNumber=[26], rfOrder=31, authorNames=YOKOZUKA M, KOIDE K, OISHI S, journalName=Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), refType=null, unstructuredReference=YOKOZUKA M, KOIDE K, OISHI S, et al. LiTAMIN2: Ultra Light LiDAR-Based SLAM Using Geometric Approximation Applied with KL-Divergence[C]// Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), 2021:11619-11625., articleTitle=LiTAMIN2: Ultra Light LiDAR-Based SLAM Using Geometric Approximation Applied with KL-Divergence, refAbstract=null), Reference(id=1154016905280082060, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2021, volume=6, issue=2, pageStart=3317, pageEnd=3324, url=null, language=null, rfNumber=[27], rfOrder=32, authorNames=PAN Yue, XIAO Pengchuan, HE Yujie, journalName=IEEE Robotics and Automation Letters, refType=null, unstructuredReference=PAN Yue, XIAO Pengchuan, HE Yujie, et al. MULLS: Inertial Odometry Package by Tightly-Coupled Iterated Kalman Filter[J]. IEEE Robotics and Automation Letters, 2021,6(2):3317-3324., articleTitle=MULLS: Inertial Odometry Package by Tightly-Coupled Iterated Kalman Filter, refAbstract=null), Reference(id=1154016905334608013, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2022, volume=38, issue=4, pageStart=2053, pageEnd=2073, url=null, language=null, rfNumber=[28], rfOrder=33, authorNames=XU Wei, CAI Yixi, HE Dongjiao, journalName=IEEE Transactions on Robotics, refType=null, unstructuredReference=XU Wei, CAI Yixi, HE Dongjiao, et al. FAST-LIO2: Fast Direct LiDAR-Inertial Odometry[J]. IEEE Transactions on Robotics, 2022,38(4):2053-2073., articleTitle=FAST-LIO2: Fast Direct LiDAR-Inertial Odometry, refAbstract=null), Reference(id=1154016905384939662, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=10672, pageEnd=10678, url=null, language=null, rfNumber=[29], rfOrder=34, authorNames=LIN Jiarong, ZHANG Fu, journalName=Proceeding of the 2022 International Conference on Robotics and Automation (ICRA), refType=null, unstructuredReference=LIN Jiarong, ZHANG Fu. R3LIVE: A Robust, Real-Time, RGB-Colored, LiDAR-Inertial-Visual Tightly-Coupled State Estimation and Mapping Package[C]// Proceeding of the 2022 International Conference on Robotics and Automation (ICRA), 2022:10672-10678., articleTitle=R3LIVE: A Robust, Real-Time, RGB-Colored, LiDAR-Inertial-Visual Tightly-Coupled State Estimation and Mapping Package, refAbstract=null), Reference(id=1154016905431077007, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=35, authorNames=DAVISON A J, journalName=null, refType=null, unstructuredReference=DAVISON A J. FutureMapping: The Computational Structure of Spatial AI Systems[R]. 2018 . 10.48550/arXiv.1803.11288 ., articleTitle=FutureMapping: The Computational Structure of Spatial AI Systems, refAbstract=null), Reference(id=1154016905481408656, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=4530, pageEnd=4537, url=null, language=null, rfNumber=[31], rfOrder=36, authorNames=CHEN Xieyuanli, MILIOTO A, PALAZZOLO E, journalName=Proceedings of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), refType=null, unstructuredReference=CHEN Xieyuanli, MILIOTO A, PALAZZOLO E, et al. SuMa++: Efficient LiDAR-Based Semantic SLAM[C]// Proceedings of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019:4530-4537., articleTitle=SuMa++: Efficient LiDAR-Based Semantic SLAM, refAbstract=null), Reference(id=1154016905582071953, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=1362, pageEnd=1371, url=null, language=null, rfNumber=[32], rfOrder=37, authorNames=WANG Jingwen, RYUNZ M, AGAPITO L, journalName=Proceedings of the 2021 International Conference on 3D Vision (3DV), refType=null, unstructuredReference=WANG Jingwen, RYUNZ M, AGAPITO L. DSP-SLAM: Object Oriented SLAM with Deep Shape Priors[C]// Proceedings of the 2021 International Conference on 3D Vision (3DV), 2021:1362-1371., articleTitle=DSP-SLAM: Object Oriented SLAM with Deep Shape Priors, refAbstract=null), Reference(id=1154016905632403602, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=5144, pageEnd=5151, url=null, language=null, rfNumber=[33], rfOrder=38, authorNames=SHIN Y S, PARK Y S, KIM A, journalName=Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), refType=null, unstructuredReference=SHIN Y S, PARK Y S, KIM A. Direct Visual SLAM Using Sparse Depth for Camera-LiDAR System[C]// Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), 2018:5144-5151., articleTitle=Direct Visual SLAM Using Sparse Depth for Camera-LiDAR System, refAbstract=null), Reference(id=1154016905678540947, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=null, pageStart=115, pageEnd=130, url=null, language=null, rfNumber=[34], rfOrder=39, authorNames=SHIN Y S, PARK Y S, KIM A, journalName=Autonomous Robots, refType=null, unstructuredReference=SHIN Y S, PARK Y S, KIM A. DVL-SLAM: Sparse Depth Enhanced Direct Visual-LiDAR SLAM[J]. Autonomous Robots, 2020,44:115-130., articleTitle=DVL-SLAM: Sparse Depth Enhanced Direct Visual-LiDAR SLAM, refAbstract=null), Reference(id=1154016905728872596, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2022, volume=23, issue=9, pageStart=14509, pageEnd=14523, url=null, language=null, rfNumber=[35], rfOrder=40, authorNames=CHOU CC, CHOU C F, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=CHOU CC, CHOU C F. Efficient and Accurate Tightly-Coupled Visual-Lidar SLAM[J]. IEEE Transactions on Intelligent Transportation Systems, 2022,23(9):14509-14523., articleTitle=Efficient and Accurate Tightly-Coupled Visual-Lidar SLAM, refAbstract=null), Reference(id=1154016905775009941, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=1626, pageEnd=1631, url=null, language=null, rfNumber=[36], rfOrder=41, authorNames=TIAN Yang, MA Shugen, journalName=Proceedings of the 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO), refType=null, unstructuredReference=TIAN Yang, MA Shugen. An Analysis About the Problem of SLAM in Tunnel-Like Environment with Serpentine Locomotion and Its Solution[C]// Proceedings of the 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO), 2018:1626-1631., articleTitle=An Analysis About the Problem of SLAM in Tunnel-Like Environment with Serpentine Locomotion and Its Solution, refAbstract=null), Reference(id=1154016905829535894, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=404, pageEnd=417, url=null, language=null, rfNumber=[37], rfOrder=42, authorNames=WENZEL P, WANG Rui, YANG Nan, journalName=Proceedings of the DAGM German Conference on Pattern Recognition, Lecture Notes in Computer Science, refType=null, unstructuredReference=WENZEL P, WANG Rui, YANG Nan, et al. 4Seasons: A Cross-Season Dataset for Multi-Weather SLAM in Autonomous Driving[C]// Proceedings of the DAGM German Conference on Pattern Recognition, Lecture Notes in Computer Science, 2020:404-417., articleTitle=4Seasons: A Cross-Season Dataset for Multi-Weather SLAM in Autonomous Driving, refAbstract=null)], funds=[Fund(id=1154016903178735711, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, awardId=52102449, language=CN, fundingSource=国家自然科学基金项目(52102449), fundOrder=null, country=null), Fund(id=1154016903229067360, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, awardId=2021M690394, language=CN, fundingSource=中国博士后科学基金项目(2021M690394), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1154016897919078454, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, xref=null, ext=[AuthorCompanyExt(id=1154016897952632887, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Beijing Institute of Technology Beijing 100081 China), AuthorCompanyExt(id=1154016897956827192, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, companyId=1154016897919078454, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=北京理工大学 北京 100081)])], figs=[ArticleFig(id=1154016902700585049, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, label=null, caption=null, figureFileSmall=NBsXAvw5PxflziX6bOmUPg==, figureFileBig=+yHi3Iklds2K51PLfLyIfg==, tableContent=null), ArticleFig(id=1154016902759305306, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, label=图 1, caption=自动驾驶车辆常用定位手段, figureFileSmall=NBsXAvw5PxflziX6bOmUPg==, figureFileBig=+yHi3Iklds2K51PLfLyIfg==, tableContent=null), ArticleFig(id=1154016902838997083, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, label=null, caption=null, figureFileSmall=B+0o54bKpp2Rqjxutcs+cg==, figureFileBig=i4RSSAsG7jj/BdfyNRa+PQ==, tableContent=null), ArticleFig(id=1154016902889328732, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, label=图 2, caption=SLAM框架图, figureFileSmall=B+0o54bKpp2Rqjxutcs+cg==, figureFileBig=i4RSSAsG7jj/BdfyNRa+PQ==, tableContent=null), ArticleFig(id=1154016902943854685, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=EN, label=null, caption=null, figureFileSmall=null, figureFileBig=null, tableContent=
传感器 水平视场角/ (°) 垂直视场角/ (°) 优点 缺点
相机 单目相机 90 60 成本低,前装量产易 存在尺度不确定性
双目相机 69 42 室内外应用不受限制 计算资源消耗大
深度相机 70 60 深度信息准确 探测距离近, 室外应用难
激光雷达 机械式旋转激光雷达 单线雷达 360 成本低, 视场角大 无垂直方向信息
多线雷达 360 30 点云数据规整, 视场角大 体积大,前装量产难
固态激光雷达 40 36 体积小, 稳定性高 探测距离近
半固态激光雷达 70 70 视场覆盖率随积分时间增大而提高 高速场景点云畸变严重
), ArticleFig(id=1154016903002574942, tenantId=1146029695717560320, journalId=1152916057816748034, articleId=1154016879124402200, language=CN, label=表 1, caption=SLAM 核心传感器优缺点对比, figureFileSmall=null, figureFileBig=null, tableContent=
传感器 水平视场角/ (°) 垂直视场角/ (°) 优点 缺点
相机 单目相机 90 60 成本低,前装量产易 存在尺度不确定性
双目相机 69 42 室内外应用不受限制 计算资源消耗大
深度相机 70 60 深度信息准确 探测距离近, 室外应用难
激光雷达 机械式旋转激光雷达 单线雷达 360 成本低, 视场角大 无垂直方向信息
多线雷达 360 30 点云数据规整, 视场角大 体积大,前装量产难
固态激光雷达 40 36 体积小, 稳定性高 探测距离近
半固态激光雷达 70 70 视场覆盖率随积分时间增大而提高 高速场景点云畸变严重
)], attaches=null, journal=Journal(id=1149663730306166798, delFlag=0, nameCn=汽车工程学报, nameEn=Chinese Journal of Automotive Engineering, nameHistory1=null, nameHistory2=null, issn=2095-1469, eissn=null, cn=50-1206/U, coden=null, periodic=1, 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=jNBfZbtw0oWUAJwGjDXBiw==, journalPrice=null, startedYear=null, abbrevIsoEn=, journalRemark=null, publicationField=null, createdTime=1752030793026, updatedTime=1761730209297, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=Q, firstLetterEn=Q, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=jNBfZbtw0oWUAJwGjDXBiw==, picEn=AzxyR8a1Jh9nr4is1i+eyQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1190346030903296709, language=CN, name=汽车工程学报, 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=1761730209327, updatedTime=1761730209327, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://tougao.ijournals.cn/ch/index.aspx, submissionEditorUrl=http://tougao.ijournals.cn/ch/index.aspx, submissionReviewUrl=http://tougao.ijournals.cn/ch/index.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190346030945239750, language=EN, name=Chinese Journal of Automotive Engineering, 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=1761730209337, updatedTime=1761730209337, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://tougao.ijournals.cn/ch/index.aspx, submissionEditorUrl=http://tougao.ijournals.cn/ch/index.aspx, submissionReviewUrl=http://tougao.ijournals.cn/ch/index.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1152916057816748034, websiteList=[Website(id=1153018156433002908, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1152916057816748034, 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/qcgcxb/CN, language=CN, createTime=1752830550535, createBy=18614031015, updateTime=1752831104570, updateBy=18614031015, name=汽车工程学报, tplId=1146099689490845704, title=汽车工程学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1154109936599552608, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156433002908, code=articleTextType, value=kx, createTime=1753090851211, updateTime=1753090851211, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109936578581085, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156433002908, code=banner, value=null, createTime=1753090851206, updateTime=1753090851206, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109936561803868, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156433002908, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=AAkd3aJi43D2QHSdjNHtbQ==, createTime=1753090851202, updateTime=1753090851202, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109936591163999, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156433002908, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753090851209, updateTime=1753090851209, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109936586969694, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156433002908, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753090851208, updateTime=1753090851208, creator=18614031015, updator=18614031015)]), Website(id=1153018156516888991, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1152916057816748034, 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/qcgcxb/EN, language=EN, createTime=1752830550556, createBy=18614031015, updateTime=1752831098136, updateBy=18614031015, name=汽车工程学报, tplId=1146101810881728533, title=汽车工程学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1154109989737189989, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156516888991, code=articleTextType, value=hep, createTime=1753090863880, updateTime=1753090863880, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109989716218466, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156516888991, code=banner, value=null, createTime=1753090863875, updateTime=1753090863875, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109989695246945, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156516888991, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic?fileId=AAkd3aJi43D2QHSdjNHtbQ==, createTime=1753090863870, updateTime=1753090863870, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109989732995684, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156516888991, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic, createTime=1753090863879, updateTime=1753090863879, creator=18614031015, updator=18614031015), WebsiteProps(id=1154109989724607075, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1153018156516888991, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1753090863877, updateTime=1753090863877, creator=18614031015, updator=18614031015)])], journalTitle=汽车工程学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=1, seqNo=null, journalTitleEn=Chinese Journal of Automotive Engineering, journalPhotoCn=jNBfZbtw0oWUAJwGjDXBiw==, journalPhotoEn=AzxyR8a1Jh9nr4is1i+eyQ==, journalFirstLetter=Q, 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/qcgcxb/CN/10.3969/j.issn.2095–1469.2024.01.01, detailUrlEn=https://castjournals.cast.org.cn/joweb/qcgcxb/EN/10.3969/j.issn.2095–1469.2024.01.01, pdfUrlCn=https://castjournals.cast.org.cn/joweb/qcgcxb/CN/PDF/10.3969/j.issn.2095–1469.2024.01.01, pdfUrlEn=https://castjournals.cast.org.cn/joweb/qcgcxb/EN/PDF/10.3969/j.issn.2095–1469.2024.01.01, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于激光雷达的自动驾驶同步定位与建图方法综述
收藏切换
PDF下载
张庚 , 杨超 , 王伟达 , 李颖
汽车工程学报 | 综述与展望 2024,14(1): 1-13
收起
收藏切换
汽车工程学报 | 综述与展望 2024, 14(1): 1-13
基于激光雷达的自动驾驶同步定位与建图方法综述
全屏
张庚 , 杨超, 王伟达, 李颖
作者信息
  • 北京理工大学 北京 100081
  • 张庚(1997-),男,北京市人,硕士研究生,主要研究方向为自动驾驶同步定位与建图。Tel: 18201625972 E-mail:

通讯作者:


李颖(1991-),女,安徽池州人,预聘助理教授,主要研究方向为自动驾驶感知、多传感器融合感知、移动激光扫描、点云处理、几何和语义建模和高精度地图。Tel: 18626876417 E-mail:
A Review of LiDAR-Based Simultaneous Localization and Mapping Methods for Autonomous Driving
Geng ZHANG , Chao YANG, Weida WANG, Ying LI
Affiliations
  • Beijing Institute of Technology Beijing 100081 China
doi: 10.3969/j.issn.2095–1469.2024.01.01
文章导航
收藏切换

同步定位与建图(Simultaneous Localization and Mapping,SLAM)技术可使自动驾驶车辆在未知环境中根据车载传感器采集到的数据估计自身位姿,建立环境地图,为车辆的规划、决策提供定位信息,是近年来自动驾驶技术研究的热点之一。基于车载激光雷达的点云数据,聚焦SLAM技术在自动驾驶领域的应用,围绕前端里程计、后端优化和回环检测技术,对国内外相关研究进行综述。考虑到单一传感器的局限性,结合目前多传感器融合研究的热点与难点,展望了自动驾驶多传感器融合SLAM技术在自动驾驶领域的机遇与挑战。

自动驾驶  /  同步定位与建图  /  激光雷达  /  前端里程计  /  后端优化  /  回环检测

Simultaneous localization and mapping (SLAM) technology enables autonomous vehicles to estimate their own poses and establish the map of an unknown environment according to the data collected by onboard sensors. SLAM can provide localization information to the decisionmaking module for vehicle planning, and has become one of the research hotspots of autonomous driving technology in recent years. Based on the point cloud data collected by LiDAR, this paper focuses on the SLAM technology applied in autonomous driving. The related research at home and abroad has been reviewed including the frontend odometry, the backend optimization and loop closure detection. Due to the limitations of a single sensor, the opportunities and challenges of multisensor fusion SLAM technology for autonomous driving are discussed based on the research hotspots and difficulties in the field of multisensor fusion.

autonomous driving  /  SLAM  /  LiDAR  /  front-end odometry  /  back-end optimization  /  loop closure detection
张庚, 杨超, 王伟达, 李颖. 基于激光雷达的自动驾驶同步定位与建图方法综述. 汽车工程学报, 2024 , 14 (1) : 1 -13 . DOI: 10.3969/j.issn.2095–1469.2024.01.01
Geng ZHANG, Chao YANG, Weida WANG, Ying LI. A Review of LiDAR-Based Simultaneous Localization and Mapping Methods for Autonomous Driving[J]. Chinese Journal of Automotive Engineering, 2024 , 14 (1) : 1 -13 . DOI: 10.3969/j.issn.2095–1469.2024.01.01
SLAM技术在自动驾驶领域能辅助车辆在未知环境中根据传感器采集的数据实时估计自身位姿并绘制环境地图,为车辆提供用于自主规划与决策 [ 1 - 4 ] 的定位信息。常规的车辆定位手段包括全球定位导航系统(Global Navigation Satellite System, GNSS)、SLAM、惯性测量单元(Inertial Measurement Unit, IMU)定位、轮速计定位等,如 图 1 所示。 根据中国工业和信息化部 2021 年发布的 GB/T 40429-2021 《汽车驾驶自动化分级》,将自动驾驶按照自动化程度由低到高分为 0~5 级,共 6 个级别。从 3 级自动驾驶开始,自动驾驶车辆能在其设计运行条件内持续地执行全部动态驾驶任务,即自动驾驶系统可以接管驾驶员。REID 等 [ 5 ] 根据可靠性原理, 计算了满足 3 级以上的自动驾驶的定位需求。在高速公路场景下, 自动驾驶车辆的纵向定位精度需达到 ${1.40}\mathrm{\;m}$ ,水平定位精度需达到 ${0.57}\mathrm{\;m}$ ; 城市公路场景下的定位精度需求更高, 需达到 0.29 m。然而,自动驾驶场景复杂多变,在城市楼宇间、隧道中, 由于建筑物遮挡导致的卫星信号丢失、轮式里程计飘移等问题,仅依靠 GNSS 难以达到上述精度要求。而 SLAM 技术充分利用车载传感器采集的数据,可达到厘米级的定位精度 [ 6 ] ,弥补传统 GNSS 在特定场景下定位失效的缺点 [ 7 ] 。因此, SLAM技术在自动驾驶领域具有很高的研究价值与应用潜力, 对车辆智能化发展必将产生深远影响 [ 8 ]
SLAM 技术路线多样, 但宏观上具备相同的框架。SLAM框架包括数据采集、前端里程计、回环检测、后端优化和建图,共 5 个模块, 如 图 2 所示。 各模块主要任务为:数据采集模块负责激光雷达、 相机、IMU等传感器数据的读取和同步; 前端里程计模块估计相邻数据帧间车辆的运动, 记录局部地图; 回环检测模块判断车辆是否到达过先前经过的位置, 如果检测到回环, 会把信息提供给后端进行处理; 后端优化模块接收不同时刻里程计测量的车辆位姿及回环检测信息, 使用优化方法得到全局一致的轨迹;建图模块根据后端优化估计的轨迹,建立与任务要求对应的地图。
本文以该框架为指导, 系统地对基于激光雷达的自动驾驶同步定位与建图方法展开综述。文章结构如下: 首先, 梳理 SLAM 的技术路线, 包括发展过程、输出地图分类以及常用数据集与评价指标; 第 2 节介绍 SLAM 的前端里程计, 列举常用的点云预处理方法以及配准方法; 第 3 节介绍 SLAM 的后端优化手段; 第 4 节介绍 SLAM 的回环检测, 列举近年来提出的场景识别算法; 第 5 节通过横向比较 SLAM 常用传感器的性能,引入多传感器融合 SLAM技术, 分析提高 SLAM算法精度的途径; 第 6 节聚焦 SLAM 研究的热点与难点, 对 SLAM 的未来做出展望; 最后, 对 SLAM 技术在新时代车辆领域的应用进行总结。
SLAM 的发展大致可被分为 3 个阶段。早期 ( 1986 ~ 2010 年 ),中期( 2010 ~ 2014 年 ),近期 (2014 年至今)。在 1986 年举办的 IEEE Robotics and Automation Conference 上, 概率 SLAM 被首次提出, 标志着 SLAM 研究的开始。SLAM 的本质是状态估计问题。在早期,基于卡尔曼滤波 (Karman Filter, KF) 的方法占据明显的主导地位。 但是, 由于车辆往往是非线性非高斯系统, KF法不能很好地估计需要的参数, 扩展卡尔曼滤波 (Extended Karman Filter, EKF) 法逐渐成为主流, 后续又出现了基于粒子滤波 (Particle Filter, PF) 的方法 [ 9 ] ; 2010 年,基于图优化方法的 Karto SLAM [ 10 ] ,标志着基于优化的 SLAM 时代来临。 Karto SLAM 之后, 基于优化的方法比基于滤波方法表现出更好的效果。2014 年, ZHANG Ji 等 [ 11 ] 的研究标志着激光 SLAM 已经基本成熟, 如何提高 SLAM的鲁棒性成为新的研究热点。由于自动驾驶车辆需要在复杂多变的环境中定位, 所以相关学者开始关注 GNSS 信号受阻、以及环境中的动态障碍物对 SLAM 的影响。
SLAM 的输出不仅包括定位信息, 还包括周边环境的地图。常见的地图包括拓扑地图、度量地图。前者强调地图元素之间的关系, 后者则能精确描述环境中各物体的位置, 可以作为自动驾驶规划与决策的输入信息。度量地图可以是稀疏或稠密的, 激光雷达建立的一般是稀疏地图, 这类地图只能用于定位, 不能用于导航避障之类的工作。因此, 很多 SLAM 系统中会额外进行稠密地图, 甚至语义地图的构建。LING Yonggen 等 [ 12 ] 基于相机 SLAM的稀疏特征, 建立了稠密 SLAM 框架, 由姿态跟踪、局部捆集优化、回环检测、三维建图、自由空间体积提取 5 个线程组成。姿态跟踪线程估计出每一帧的相机姿态, 并决定是否插入关键帧。回环检测部分和全局优化在另外的线程中执行。三维建图线程将优化过的稀疏特征点转化成稠密的体积表达。自由空间体积提取线程会定期输出最新的自由空间量, 用于规划和决策。
相比于精确的度量地图, 拓扑地图不擅长表达复杂的地形结构。它将地图抽象为图论中的 “点” 和 “边”,只考虑点 $\mathrm{A}$ 和点 $\mathrm{B}$ 之间的连通性,而不考虑如何从 $\mathrm{A}$ 点到达 $\mathrm{B}$ 点。如何对度量地图进行分割, 形成具有节点与边的拓扑地图, 又如何使用拓扑地图进行导航与路径规划, 仍是有待研究和解决的问题。
SLAM在自动驾驶等领域应用广泛,在对比不同算法之间的性能时, 往往需要一个数据基准。但室外场景的数据收集工作规模大而繁琐, 涉及各种传感器间的标定和时空同步等。公开数据集的出现节省了数据准备工作, 提供了一个算法验证与对比的平台。目前主流的自动驾驶场景激光雷达数据集包括:KITTI 数据集 [ 13 ] ,包含市区、乡村和高速公路等场景采集的真实图像数据, 其里程计基准包含了 22 个序列, 可同时用于 SLAM 算法的训练与评估; Waymo 数据集 [ 14 ] ,自动驾驶公司 Waymo 的数据开放项目,包含 1150 个场景,且对激光雷达和相机的边界框进行了仔细的标注, 并在各帧之间使用了一致的标识符;Oxford RobotCar 数据集 [ 15 ] 是牛津大学实验室提出的公开数据集, 是为了研究自动驾驶车辆在真实城市环境中的长期定位和建图问题而采集的, 包含了雨、雪、阳光直射等各种天气条件;RELLIS 数据集 [ 16 ] 特别针对越野环境, 在得克萨斯农工大学校区采集了超过 6000 张图片、 13000 帧雷达数据, 为自动驾驶车辆的自主导航功能全方面地提供相机、雷达点云、立体视觉、高精度GPS和IMU数据。
用于评价 SLAM 算法精度常见的指标为输出位姿的精度、建立地图的精度等。对准确性的评价需要引入更准确的参考, 往往会采用价格昂贵的试验设备。考虑到本研究的对象为自动驾驶车辆, 车辆上的 GNSS-RTK 在信号条件良好的情况下可以提供高精度的定位信息,那么以平均平移误差 (Average Translational Error, ATE) 和平均旋转误差(Average Rotational Error, APE)为标准, 将 SLAM算法输出的轨迹与位姿与 GNSS 提供的基准进行比较, 可以间接完成算法准确性的定量评估。 在 KITTI 数据集中, 给出了相应的评判准则, 如式 (1)所示。
$ {E}_{\text{trans }}\left( F\right) = \frac{1}{\left| F\right| }\mathop{\sum }\limits_{{i, j \in F}}{\begin{Vmatrix}{\widehat{T}}_{j}{\widehat{T}}_{i}^{-1}{T}_{i}{T}_{j}^{-1}\end{Vmatrix}}_{2} $
$ {E}_{\text{rot }}\left( F\right) = \frac{1}{\left| F\right| }\mathop{\sum }\limits_{{i, j \in F}}\angle {\left\lbrack {\widehat{T}}_{j}{\widehat{T}}_{i}^{-1}{T}_{i}{T}_{j}^{-1}\right\rbrack }^{ \circ } $
式中: $F$ 为两帧点云的集合; $\widehat{T}$$T$ 分别为车辆位姿的估计值和真值; $\angle \left\lbrack \cdot \right\rbrack$ 为旋转角。
激光 SLAM 中前端里程计的核心功能是根据激光数据帧估计激光雷达以及车辆的运动 [ 17 ] 。由于里程计处于 SLAM 算法信息流靠前的位置, 所以又被称作前端。目前的研究主要基于两类:迭代最近点 (Iterative Closet Point, ICP) 算法与正态分布变换 (Normal Distributions Transform, NDT) 算法。
ICP 算法是点云配准的主流方法。ICP 通过计算一帧点云中的所有点, 或者选取的某些特征点与另一帧点云的距离,求解出当前点云帧相对于参考帧的位置 [ 18 ] 。但是,当激光雷达的稀疏性与环境的随机性结合, 会导致两帧点云产生差异, 如形状各异的树叶、动态行进的车辆等, 给 ICP 算法引入较大的误差。针对这一现象, 相关学者在 ICP 点到点配准的基础上提出了点到线配准 (PL-ICP) [ 19 ] 、 点到面配准 ${\left( \mathrm{{NICP}}\right) }$ [ 20 ] 等改进方法,提高了 $\mathrm{{ICP}}$ 算法的性能。PL-ICP把每次扫描的数据看作对真实物理世界某曲面的采样, 采用分段线性法对实际曲面进行近似, 用激光点到最近两点连线的距离作为误差函数, 模拟实际激光点到曲面的距离。因此, PL-ICP 的收敛速度更快, 但更容易陷入局部极值, 所以在使用时,多采用全局 ICP 法 [ 21 ] ,即先进行粗匹配, 再使用 PL-ICP 进行精确计算。NICP 算法在误差项设计时考虑了更多因素, 要求待配准的两个点不仅要距离接近, 而且所在处的法向量方向也要相同。在匹配时, 根据距离、曲率以及法向量进行筛选,并在优化变换参数时优化增加了法向量的参数。但是, 上述方法的局限性体现在仅能对两帧独立的点云进行配准, 而应用于自动驾驶的 SLAM 需要连续处理传感器信息流。
基于上述研究, ZHANG Ji等 [ 11 ] 提出激光里程计 (Lidar Odometry and Mapping, LOAM)。该算法通过遍历点云, 计算每点附近的曲率, 将原始数据分为平面点与边缘点。以下一帧平面特征点到参考帧特征平面的距离、边缘特征点到特征线的距离为残差构建 Levenberg-Marquardt(LM)优化方程 [ 22 ] ,解算位姿矩阵。此外, LOAM创新性地将 SLAM 任务分为帧间配准以及地图拼接两个子任务,以 ${10}\mathrm{\;{Hz}}$ 的频率进行帧间配准,实时估计载体的状态,在累积了多帧点云后,以 $1\mathrm{\;{Hz}}$ 的频率进行地图拼接,输出精度更高的点云地图。自 2014 年至今, 仍然在 KITTI 数据集的里程计榜单上占据前 3 位。但 LOAM 的运算速度不高,每帧配准耗时约 ${90}\mathrm{\;{ms}}$ 。F-LOAM [ 23 ] 在LOAM的基础上,更注重于速度与精度的平衡, 在保证算法精度不下降的前提下, 将算法实时性提高了 18.3%。SHAN Tixiao 等 [ 24 ] 于 2018 年 提出 LeGO-LOAM,与 LOAM 相比,使用了地面标签匹配与两步 LM 优化法, 在维持里程计精度不变的前提下, 大幅降低了计算时间开销,仅需 ${10}\mathrm{\;{ms}}$ 。上述算法使用 $\mathrm{K}$ 维树进行位姿的存储, 其运行效率仍具有局限性。FOSSEL 等 [ 25 ] 提出的NOctoSLAM在前端使用NICP进行帧图位姿匹配,并使用八叉树替代传统的 $\mathrm{K}$ 维树构建地图, 从而可以快速地寻找数据关联和计算近似法向量。 与传统的基于 $\mathrm{K}$ 维树地图的 NICP 算法相比,效率提高了近 2 倍。除了使用八叉树地图来提高算法速度, YOKOZUKA 等 [ 26 ] 在 LiTAMIN2 中使用一种对称 $\mathrm{{KL}}$ 散度的 ICP 方法来显著提高 LiDAR SLAM 的速度, 该方法通过将点云从 LiDAR 投影到每个体素网格并将投影的子点云近似为一个正态分布, 从而大大减少了用于配准的点数, 在里程计帧速率上实现了 ${500} \sim {1000}\mathrm{{fps}}$ 的计算速度,在 KITTI 数据集上有良好的表现。
目前, 基于 ICP 的前端里程计的发展已经十分成熟, PAN Yue 等 [ 27 ] 提出的 MULLS 在前端采用了双阈值地面滤波和主成分分析, 从每一帧中提取大致分类的特征点用于配准, 在不同类型的激光雷达以及各种室内外场景下取得良好的效果, 是目前顶级的纯激光 SLAM 之一。
NDT算法是另一种点云配准的方法。NDT将参考点云储存到网格或体素中, 计算每个网格的多维正态分布参数, 对于待匹配的点云, 通过计算每个转换点的概率密度, 以概率密度的和作为优化项, 使用优化算法寻找使转换点的概率密度最大的位姿变换矩阵参数, 完成两帧点云的配准 [ 28 ]
北京理工大学龚建伟团队 [ 29 ] 研究了基于NDT 的越野环境 CPFG-SLAM, 并针对越野环境特征稀疏的特点, 提出帧间累加采样策略, 保证了帧间配准的精度。与 ICP 算法不同, NDT 将环境信息储存在三维体素中, 使用正态分布来描述, 当两帧点云重合时参考点的概率密度会很大, 如果通过优化的方法, 求解使概率密度最大的变换参数, 就可以快速地配准两帧点云。NDT的优势主要体现在: 通过对环境信息的划分处理, 能容忍环境的细微变化, 加速了计算过程, 支持更大且稠密的地图, 在当时的 KITTI 榜单中以 ${0.03}\mathrm{\;s}$ 的运行时间排名第 14 位。 但 CPFG-SLAM 的局限性体现在纯激光 SLAM 在越野环境下易产生漂移, 为了解决该问题, 相同的团队在一年后提出了LiDAR-IMU紧耦合的SLAM方法 [ 25 ] ,通过对 IMU 数据的预积分提高了定位的精度。
香港理工大学团队 [ 26 ] 分析了NDT在香港公路的效果, 得到了 NDT 算法与城市交通状况相关的结论, 在稀疏的城市环境下可以取得厘米级的定位, 而在稠密的城市中定位误差平均会增加约 10 m, 使可靠性显著降低。
前端里程计可以生成小范围的轨迹和地图, 但随着时间推移, 帧间配准的误差会累计, 最终导致大范围的地图出现错位。因此, 引入后端优化的目的是在更大的尺度上对轨迹和地图进行优化, 以输出高精度的结果。
基于滤波的方法源于贝叶斯估计理论, 是早期解决SLAM问题的方法。1988 年, SMITH 等 [ 32 ] 提出基于EKF的SLAM方案,使用最大似然算法进行数据关联, 但是该方法计算量复杂, 鲁棒性差, 构建的地图是特征地图,无法应用于导航避障。 MONTEMERLO 等 [ 33 ] 于 2002 年 提出 FastSLAM, 在 EKF-SLAM 的基础上进行改进, 将 SLAM 问题划分为定位问题和已知位姿的建图问题, 主要采用 PF 代替 EKF 来估计位姿。在运动方程上进行粒子的传播, 利用观测方程计算粒子的权重, 且根据估计的位姿来构建地图。但是, 在大规模场景下, 粒子滤波需要大量的粒子表示环境, 从而导致该方法的计算量非常庞大, 同时, 频繁的重采样将导致粒子耗散, 严重影响后续的建图问题。针对上述缺陷, GRISETTI 等 [ 34 ] 于 2007 年提出了 Gmapping, 以 FastSLAM 为基本原理,是室内 2D 激光 SLAM 的里程碑式研究成果, 现在仍被广泛应用。基于滤波的方法直到现在仍具有活力, CHEN Xieyuanli 等 [ 35 ] 与 2021 年提出的 Range-MCL 就是基于蒙特卡罗定位 (Monte Carlo Localization, MCL) 和粒子滤波来估计车辆的姿态, 以实现全局定位。该研究使用三角网格地图以及距离图像投影为 MCL 提供了一种新的观测模型, 在动态的室外大范围环境中取得良好的全局定位结果, 并且能适配多种型号的 LiDAR。
基于优化的方法与基于滤波的方法相比会表现出更好的性能。图优化法是利用图的均值来表示地图, 利用大量的节点表示车辆轨迹的每个位置点和传感器测量数据。每个新节点加入, 就会进行计算更新。基于图优化的 SLAM 后端优化法可概括分为 4类:基于最小二乘法、基于随机梯度下降、基于松弛迭代和基于流形迭代。GUTMANN 等 [ 10 ] 于 2010 年提出的 Karto SLAM 中首次应用了图优化框架, 具备前端扫描匹配、全局优化以及回环检测模块; SANTOS 等 [ 36 ] 对当时 5 种具有代表性的 2D 激光SLAM算法进行了测试评估和总结, 结果表明, Gmapping 和 Karto SLAM 在定位和建图的准确性和效率上要更加优越。
谷歌公司提出的 Cartographer [ 37 ] 对 Karto SLAM 进行了优化, 提出了用分支定界法解决子地图的构建以及与全局地图的匹配问题, 实现了回环检测, 在 2D 和 3D SLAM 中均取得了较好的效果, 是目前较为成熟的开源激光 SLAM 技术之一。
目前基于图优化的开源优化库有 GTSAM [ 38 ] 、 G2O [ 39 ] 、Ceres、BA等。
回环检测在 SLAM 中是很重要的一部分, 本质上是一种数据相似性检测算法。其功能是确保车辆在重复经过相同地点时能正确识别, 并以该地点为基准优化整张地图。但是, 由于环境往往是动态变化的, 行人等动态障碍物给场景识别带来了很大的困难。
BELONGIE 等 [ 40 ] 针对激光雷达的数据特性, 把点云的局部特征点附近的点云形状编码进图像中, 提出了 Shape Context。但是, 该算法对于点云数据量有要求, 待匹配的点云数据量应该大于参考点云, 否则容易引起误匹配。车载激光雷达一般以固定频率运行, 不能满足该算法对点云数据量的要求。KIM 等 [ 41 ] 提出 Scan Context,受到了 Shape Context 的启发, 使用非直方图的全局描述符, 根据点云的深度和角度信息, 把无序的三维点云投影到二维平面上, 用于回环检测以及场景识别。将 Scan Context 的粗配准与 ICP 的细配准结合, 可以在快速的前提下保证更高的精度。但是, 在编码过程中, 仅记录了点云高度方向的最高点, 仍有改进空间。相同的团队在 2021 年发布了最新的研究成果 Scan Context++ [ 42 ] ,在之前 Scan Context 的旋转不变性基础上, 提出了新的描述子, 使其对于旋转和平移都具有不变性, 并通过引入总结描述符大幅提升了计算速度。该研究结合了大量的试验, 证明了在城市环境中的车辆虽然存在大量的平移和旋转变换, Scan Context++仍能取得较好的效果。针对 Scan Context 的改进还有许多, 例如 WANG Ying 等 [ 43 ] 在 2020 年提出的 LiDAR IRIS,使用一个具有旋转不变性的全局描述符, 把场景的信息通过 LoG-Gabor 滤波器的二值特征提取后, 获得二进制编码的 IRIS 图像, 充分利用了点云的大部分信息, 而且该全局描述符也具有旋转不变性, 避免了暴力搜索,节约了计算资源。BEHLEY 等 [ 44 ] 提出的 SuMa方法首次在激光雷达上利用 Surfel 地图高效地生成投影数据关联, 并实现回环检测, 该方法只基于激光点云就可以建立大尺度环境下的全局一致性地图。ZHANG Ruitao 等 [ 45 ] 提出一种新颖的神经网络架构 VPD-Map, 利用视觉点云为快速的回环检测提供了全局的三维视觉描述符, 还能为前端的位姿估计提供特征图的先验信息。
SLAM 使用的传感器种类繁多, 其中核心传感器除了上述的激光雷达外还有相机, 分别对应了激光与视觉两种主流的SLAM技术路线。
按照机械结构, 激光雷达可以被划分为机械式旋转激光雷达、固态激光雷达、半固态激光雷达等类别。机械式旋转激光雷达按照线数分为单线 LiDAR、多线 LiDAR。前者扫描速度快、分辨率高、测距精确、可靠性强。但是, 单线 LiDAR 只能平面式扫描, 不能测量物体高度, 有一定局限性, 主要应用于室内或封闭环境, 如地下矿车 [ 46 ] ; 多线激光雷达又被称作 $3\mathrm{D}$ 激光雷达,场景还原能力更强, 使用 16 线、32 线、64 线或更高线数的激光同时扫描, 可以识别物体的高度信息, 输出空间点云地图, 在自动驾驶领域经常被采用 [ 47 ] 。早期激光雷达内部的电机和轴承等机械部件精密且昂贵, 导致其无法通过车规级测试, 影响了前装量产市场占有率。但随着制造技术的发展,相继出现了固态、半固态激光雷达。其中, 固态激光雷达使用 Flash 技术或光学相控阵技术 (Optical Phased Array, OPA), 内部无任何移动部件, 在可靠性和耐用性上有很大的优势, 缺点是扫描角度有限,且信噪比差 [ 48 ] ,通过在平坦区域使用激光水准仪对车载固态激光雷达进行标定, 可以在保证角度和距离精度为 ${0.063}^{ \circ }$${0.0316}\mathrm{\;m}$ 的前提下,减少激光雷达标定的时间成本 [ 49 ] 。半固态激光雷达通过微机电技术(Micro-Electro-Mechanical System, MEMS), 将机械式旋转激光雷达的结构集成到硅基芯片上, 通过内部可旋转的微振镜改变单个发射器的发射角度, 从而达到不用旋转外部结构就能扫描的效果, MEMS半固态激光雷达目前发展得最成熟, 体积适中, 是当下乘用车前装量产的首选方案,代表产品有浩界 (Horizon)、觅道 (Mid) 系列 [ 50 ] 等。LIN Jiarong 等 [ 51 ] 对半固态激光雷达的应用进行了探索, 在研究中使用小角度非均匀扫描的 Livox 激光雷达实现了周围环境体征的提取、载体的位姿估计以及地图的更新。通过逐点计算非均匀点云的曲率, 提取边缘与平面特征, 然后以 20 $\mathrm{{Hz}}$ 的频率提供里程计输出。
视觉 SLAM 使用相机完成定位与建图, 传感器主要为单目相机、双目相机和深度相机等。相机的一个重要参数为视场角, 直接影响其探测范围。相机视场角的大小由焦距决定, 不同种类的相机间视场角范围相差不大。不同激光雷达的优缺点对比见 表 1
自动驾驶车辆常用的传感器除了激光雷达、相机,还有组合导航、毫米波雷达、超声波雷达等。 传感器的工作原理决定了其适用范围, 如果仅在智能无人车辆上安装一种传感器, 其使用场景必定受限于该种传感器的性能。例如, 只在车上安装相机, 那么在低光照或者缺乏纹理的环境下, 智能无人车辆将无法采集到有效的环境信息。因此, 自动驾驶车辆系统往往采用传感器融合方案, 将多种传感器的信息和数据, 在一定的准则下加以综合分析,在不同环境下表现出高鲁棒性 [ 52 ]
多传感器融合框架分为松耦合与紧耦合。松耦合中激光运动估计和惯导运动估计系统是两个独立的模块, 分别对参数做出估计, 再将每个模块输出的参数结果进行融合, 常用的融合方法有加权平均、KF 等。WAN Guowei 等 [ 53 ] 将 LiDAR、RTK 与 $\mathrm{{IMU}}$ 数据松融合,并使用一个误差状态 $\mathrm{{KF}}$ ,使定位精度可以达到 RMS 平均值 $5 \sim {10}\mathrm{\;{cm}}$ ,具备高可靠性和鲁棒性。此外, LOAM 和 LeGO-LOAM 去除点云畸变的过程也属于 LiDAR 和 IMU 的松耦合。
与松耦合相比, 紧耦合使用两个传感器的原始数据共同估计一组变量, 可以更充分地利用传感器数据, 实现更好的效果, 是目前研究的重点。YE Haoyang 等 [ 54 ] 在 LOAM 特征提取的框架上引入了 IMU数据, 提出了 LIO-Mapping, 使用滑窗管理进行特征点的残差计算, 后端采用基于优化的方法实现两种传感器的紧耦合, 由于计算量较大无法满足实时运行的要求。QIN Chao 等 [ 55 ] 提出 LINS,使用了基于误差状态的迭代扩展卡尔曼滤波, 以车辆中心重新定义了状态变量, 和 LIO-Mapping 进行了对比,在精度相同的情况下取得了更高的效率。 ZHAO Shibo 等 [ 56 ] 也提出了类似的紧耦合SLAM架构 LIOM, 由 4 个模块组成, 数据预处理模块使用惯性测量来补偿运动失真,动态目标检测模块用 $\mathrm{{CNN}}$ 网络从每次激光扫描中检测和去除动态目标, 激光惯性里程计是核心部分, 实现激光 IMU 紧耦合,建图模块负责利用里程计的输出建立全局地图, 然后使用当前帧激光点云和全局地图进行正态分布变换来实现定位结果的精校, LIOM、LINS 的后端是采用基于滤波的方法实现的两种传感器紧耦合。SHAN Tixiao 等 [ 57 ] 提出了一种基于因子图优化的紧耦合激光-惯性里程计方法 LIO-SAM, 综合考虑 LiDAR 里程计因子、IMU 预积分因子、GPS 因子和回环因子来得到车辆全局一致的位姿, 建图结果优于 LIOM。相同的研究团队在 LIO-SAM 的基础上加入了相机,提出 LVI-SAM [ 58 ] ,由视觉惯性子系统(LIS)和激光惯性子系统(VIS)组成,两个系统间存在交互, 如 VIS 为 LIS 提供匹配初值, LIS 为 VIS 提供初始化初值和深度先验。但是, 紧耦合的缺点是传感器噪声可能互相影响, 在处理时需要考虑。
LI Kailai 等 [ 59 ] 提出了 LiLi-OM,一种基于关键帧的紧耦合激光-惯性融合方案,使用半固态激光雷达 DJI Livox 进行实时建图。
XU Wei 等 [ 60 ] 提出了一种紧耦合的 FAST-LIO, 不但使用迭代 $\mathrm{{KF}}$ 融合激光雷达和 $\mathrm{{IMU}}$ ,还采用反向传播过程来补偿运动失真, 用大量试验证明了在无人机飞行平台上即使存在快速运动或强烈振动噪声时, FAST-LIO 仍具有鲁棒性。相同的团队在一年后又提出 FAST-LIO2 [ 61 ] 。与 FAST-LIO 相比,直接将原始点注册到地图而不提取特征, 提高了地图的准确性,而且通过增量 $\mathrm{K}$ 维树数据结构 $\mathrm{{ikd}}$ -Tree 维护映射, 该结构支持增量更新和动态重新平衡, 在约 ${17}\mathrm{{rad}}/\mathrm{s}$ 的角速度下维持高鲁棒性,且适配了旋转机械式和固态等多种激光雷达。在LiLi-OM、 LIO-SAM、LINS 失败的场景下,仍能高质量地完成建图。
结合目前的研究热点, SLAM 有以下 3 个发展方向。
(1)轻量 SLAM。此类 SLAM 能在轻型车辆, 如自动物流车上鲁棒地运行, 并以此为基础开发上层应用。SLAM是一种定位手段, 能为上层应用提供自身的估计位姿,从本质上说应该服务于导航、 娱乐 [ 62 ] 和医疗等。这些应用场景下的计算资源有限, SLAM不能占用所有计算资源,所以对 SLAM 的轻量化有强烈的需求。
(2)语义 SLAM。单目视觉 SLAM 研究奠基人 DAVISON [ 63 ] 认为语义信息对构建更大更好的 SLAM 系统是必需的。语义能帮助 SLAM。传统的物体识别、分割算法往往只考虑一帧数据, 而 SLAM 过程能得到连续不断的数据流。如果能把运动过程中的每一帧数据都进行标注, 就能得到一个带有语义标签的地图, 可以被回环检测和后端优化利用,进而提高算法精度。SLAM 也能帮助语义。 物体识别和分割都需要大量的训练数据。要让分类器识别各个角度的物体, 需要从不同视角采集该物体的数据, 然后进行人工标定。而 SLAM 中, 由于可以估计相机的运动, 自动计算物体在图像中的位置, 节省人工标定的成本。如果有自动生成的带高质量标注的样本数据, 能在很大程度上加速分类器的训练过程。SuMa++[64]是基于SuMa框架提出来的, 利用每帧点云中逐点的语义标号构建全局语义地图, 使用 RangeNet++ 滤除场景中的动态物体, 提高了构建地图的精度。WANG Jingwen 等 [ 65 ] 提出的以物体级语义 DSP-SLAM 为前景对象构建了一个丰富而精确的稠密 $3\mathrm{D}$ 模型的联合地图,并用稀疏的地标点来表示背景。可以在单目、双目或双目+激光雷达 3 种不同的输入模式下以每秒 10 帧的频率运行。语义 SLAM 的研究通常会利用纹理更丰富的图像作为语义分割的基础, 因此, 激光 SLAM 和视觉 SLAM 融合就成为了语义 SLAM 未来发展的趋势。SHIN 等 [ 66 ] 为了同时保证 SLAM 系统的实时性和高精度, 利用滑动窗口法以及严格的位姿边缘化进行帧间配准, 在 KITTI 数据集实现了 0.98% 的误差。相同的团队还将直接法的视觉 SLAM 与激光 SLAM 结合, 利用 KITTI 数据集和自行录制的便携型数据集证明了视觉-激光 SLAM 系统在窄视场下的高鲁棒性 [ 67 ] 。基于直接法的视觉SLAM精度高, 实时性强, CHOU 等 [ 68 ] 同样提出视觉-激光耦合的 TVL-SLAM 系统, 将 ORB-SLAM2 的工作与激光 SLAM 组合, 令两种前端状态估计模块独立运行, 在 KITTI 数据集上仅有 0.5% 的误差。
(3)新的评价指标。SLAM适用性广泛,不同应用领域下,需求输出和性能要求不同,难以通过像数据集评估之类的方法来确定一成不变的 SLAM 基准。因此,有必要针对不同领域的 SLAM 制定不同的评价指标。举例来说,对于 $\mathrm{{AR}}/\mathrm{{VR}}$ 设备等室内环境的应用, 感兴趣的指标为新探索区域中的局部位姿准确性、能良好映射的位姿估计重复性区域、场景变化的检测精度、重定位的鲁棒性等。对于自动驾驶车辆等复杂而开阔的室外场景, 对物体分割的准确性有了更高的要求; 对于无人机等领域的应用, 则更侧重系统延迟和系统能量使用率等指标。
得益于激光雷达传感器视场范围大、分辨率高等优点, 本文聚焦基于激光雷达的 SLAM 技术在自动驾驶领域的应用, 围绕激光 SLAM 的核心传感器、前端里程计、后端优化和回环检测等方面, 对近年来国内外相关研究成果进行了系统的综述。 SLAM 技术经过 30 余年的发展, 已经可以在特定场景下实时且高质量地完成建图。但是, 由于自动驾驶场景复杂多样, 某些环境下的 SLAM 效果仍需提高, 比如隧道中的长距离定位问题和雨雪环境下的建图问题等。不过, 随着新型传感器的推出及算法的迭代, 未来的 SLAM一定会在自动驾驶领域发挥重要作用。
  • 国家自然科学基金项目(52102449)
  • 中国博士后科学基金项目(2021M690394)
参考文献 引证文献
排序方式:
[1]
CHIANG KW, TSAI GJ, CHU HJ, et al. Performance Enhancement of INS/GNSS/Refreshed-SLAM Integration for Acceptable Lane-Level Navigation Accuracy[J]. IEEE Transactions on Vehicular Technology, 2020,69(3):2463-2476.
[2]
CHEN Yunxia, WU Yaohua, XING Han. A Complete Solution for AGV SLAM Integrated with Navigation in Modern Warehouse Environment[C]// Proceedings of the 2017 Chinese Automation Congress (CAC), 2017:6418-6423.
[3]
KOIDE K, MIURA J, MENEGATTI E. A Portable Three-Dimensional LIDAR-Based System for Long-Term and Wide-Area People Behavior Measurement[J]. International Journal of Advanced Robotic Systems, 2019,16(2):1-13.
[4]
危双丰, 庞帆, 刘振彬, 等. 基于激光雷达的同时定位与地图构建方法综述[J]. 计算机应用研究, 2020(2):327-332.
WEI Shuangfeng, PANG Fan, LIU Zhenbin, et al. Survey of LiDAR-Based SLAM Algorithm[J]. Application Research of Computers, 2020(2):327-332. (in Chinese)
[5]
REID TG, HOUTS SE, CAMMARATA R, et al. Localization Requirements for Autonomous Vehicles[J]. SAE International Journal of Connected and Automated Vehicles, 2019,2(3):173-190.
[6]
满增光. 基于激光雷达的室内AGV地图创建与定位方法研究[D]. 南京: 南京航空航天大学, 2014.
MAN Zengguang. Research on Mapping and Localization of Indoor AGV Based on LADAR[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2014. (in Chinese)
[7]
任兴明. GPS/GLONASS双星定位差分解算失效分析及解决措施[J]. 电子工程师, 2008(1):7-9.
REN Xingming. Failure Analysis and Solution of GPS/GLONASS Double Satellite Differential Calculation[J]. Informatization Research, 2008(1):7-9. (in Chinese)
[8]
李晨曦, 张军, 靳欣宇, 等. 激光雷达SLAM技术及其在无人车中的应用研究进展[J]. 北京联合大学学报, 2017(4):61-69.
LI Chenxi, ZHANG Jun, JIN Xinyu, et al. Review of LIDAR SLAM Technology and its Application in Autonomous Vehicles[J]. Journal of Beijing Union University, 2017(4):61-69. (in Chinese)
[9]
李昀泽. 基于激光雷达的室内机器人SLAM研究[D]. 广州: 华南理工大学, 2016.
LI Yunze. Research on SLAM of Indoor Robot Based in LIDAR[D]. Guangzhou: South China University of Technology, 2016. (in Chinese)
[10]
KONOLIGE K, GRISETTI G, KUEMMERLE R, et al. Efficient Sparse Pose Adjustment for 2D Mapping[C]// Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2010:22-29.
[11]
ZHANG Ji, SINGH S. LOAM: Lidar Odometry and Mapping in Real-Time[R]. 2014 . 10.15607/RSS.2014.Χ.007 .
[12]
LING Yonggen, SHEN Shaojie. Building Maps for Autonomous Navigation Using Sparse Visual SLAM Features[C]// Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017:1374-1381.
[13]
GEIGER A, LENZ P, URTASUN R. Are We Ready for Autonomous Driving? The KITTI Vision Benchmark Suite[C]// Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, 2012:3354-3361.
[14]
SUN Pei, KRETZSCHMAR H, DOTIWALLA X, et al. Scalability in Perception for Autonomous Driving: Waymo open dataset[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020:2443-2451.
[15]
MADDERN W, PASCOE G, LINEGAR C, et al. 1 Year, 1 000 km: The Oxford RobotCar Dataset[J]. The International Journal of Robotics Research, 2017,36(1):3-15.
[16]
JIANG Peng, OSTEEN P, WIGNESS M, et al. RELLIS-3D Dataset: Data, Benchmarks and Analysis[C]// Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), 2021:1110-1116.
[17]
宗文鹏, 李广云, 李明磊, 等. 激光扫描匹配方法研究综述[J]. 中国光学, 2018(6):914-930.
ZONG Wenpeng, LI Guangyun, LI Minglei, et al. A Survey of Laser Scan Matching Methods[J]. Chinese Optics, 2018(6):914-930. (in Chinese)
[18]
ARUN KS, HUANGT S, BLOSTEIN S D. Least-Squares Fitting of Two 3-D Point Sets[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1987,PAMI-9(5):698-700.
[19]
CENSI A. An ICP Variant Using a Point-to-Line Metric[C]// Proceedings of the 2008 IEEE International Conference on Robotics and Automation, 2008:19-25.
[20]
SERAFIN J, GRISETTI G. NICP: Dense Normal Based Point Cloud Registration[C]// Proceedings of the 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2015:742-749.
[21]
CENSI A. Scan Matching in a Probabilistic Framework[C]// Proceedings of the 2006 IEEE International Conference on Robotics and Automation, ICRA 2006, 2006:2291-2296.
[22]
LYU Chen, XING Yang, ZHANG Junzhi, et al. Levenberg-Marquardt Backpropagation Training of Multi-layer Neural Networks for State Estimation of a Safety-Critical Cyber-Physical System[J]. IEEE Transactions on Industrial Informatics, 2018,14(8):3436-3446.
[23]
WANG Han, WANG Chen, CHEN Chunlin, et al. F-LOAM: Fast LiDAR Odometry and Mapping[C]// Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021:4390-4396.
[24]
SHAN Tixiao, ENGLOT B. LeGO-LOAM: Lightweight and Ground-Optimized Lidar Odometry and Mapping on Variable Terrain[C]// Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018:4758-4765.
[25]
FOSSEL J, TUYLS K, SCHNIEDERS B, et al. NOctoSLAM: Fast Octree Surface Normal Mapping and Registration[C]// Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017:6764-6769.
[26]
YOKOZUKA M, KOIDE K, OISHI S, et al. LiTAMIN2: Ultra Light LiDAR-Based SLAM Using Geometric Approximation Applied with KL-Divergence[C]// Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), 2021:11619-11625.
[27]
PAN Yue, XIAO Pengchuan, HE Yujie, et al. MULLS: Inertial Odometry Package by Tightly-Coupled Iterated Kalman Filter[J]. IEEE Robotics and Automation Letters, 2021,6(2):3317-3324.
[28]
XU Wei, CAI Yixi, HE Dongjiao, et al. FAST-LIO2: Fast Direct LiDAR-Inertial Odometry[J]. IEEE Transactions on Robotics, 2022,38(4):2053-2073.
[29]
LIN Jiarong, ZHANG Fu. R3LIVE: A Robust, Real-Time, RGB-Colored, LiDAR-Inertial-Visual Tightly-Coupled State Estimation and Mapping Package[C]// Proceeding of the 2022 International Conference on Robotics and Automation (ICRA), 2022:10672-10678.
[30]
DAVISON A J. FutureMapping: The Computational Structure of Spatial AI Systems[R]. 2018 . 10.48550/arXiv.1803.11288 .
[31]
CHEN Xieyuanli, MILIOTO A, PALAZZOLO E, et al. SuMa++: Efficient LiDAR-Based Semantic SLAM[C]// Proceedings of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019:4530-4537.
[32]
WANG Jingwen, RYUNZ M, AGAPITO L. DSP-SLAM: Object Oriented SLAM with Deep Shape Priors[C]// Proceedings of the 2021 International Conference on 3D Vision (3DV), 2021:1362-1371.
[33]
SHIN Y S, PARK Y S, KIM A. Direct Visual SLAM Using Sparse Depth for Camera-LiDAR System[C]// Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), 2018:5144-5151.
[34]
SHIN Y S, PARK Y S, KIM A. DVL-SLAM: Sparse Depth Enhanced Direct Visual-LiDAR SLAM[J]. Autonomous Robots, 2020,44:115-130.
[35]
CHOU CC, CHOU C F. Efficient and Accurate Tightly-Coupled Visual-Lidar SLAM[J]. IEEE Transactions on Intelligent Transportation Systems, 2022,23(9):14509-14523.
[36]
TIAN Yang, MA Shugen. An Analysis About the Problem of SLAM in Tunnel-Like Environment with Serpentine Locomotion and Its Solution[C]// Proceedings of the 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO), 2018:1626-1631.
[37]
WENZEL P, WANG Rui, YANG Nan, et al. 4Seasons: A Cross-Season Dataset for Multi-Weather SLAM in Autonomous Driving[C]// Proceedings of the DAGM German Conference on Pattern Recognition, Lecture Notes in Computer Science, 2020:404-417.
2024年第14卷第1期
PDF下载
265
101
引用本文
BibTeX
文章信息
doi: 10.3969/j.issn.2095–1469.2024.01.01
  • 接收时间:2022-04-27
  • 首发时间:2025-07-21
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2022-04-27
  • 修回日期:2022-06-29
基金
国家自然科学基金项目(52102449)
中国博士后科学基金项目(2021M690394)
作者信息
    北京理工大学 北京 100081

通讯作者:


李颖(1991-),女,安徽池州人,预聘助理教授,主要研究方向为自动驾驶感知、多传感器融合感知、移动激光扫描、点云处理、几何和语义建模和高精度地图。Tel: 18626876417 E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/qcgcxb/CN/10.3969/j.issn.2095–1469.2024.01.01
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