Article(id=1322506697441137531, tenantId=1146029695717560320, journalId=1281212831689347082, issueId=1322506567065383670, articleNumber=null, orderNo=null, doi=10.14075/j.jgg.2024.06.288, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1718726400000, receivedDateStr=2024-06-19, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1793239767656, onlineDateStr=2026-10-29, pubDate=1752508800000, pubDateStr=2025-07-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1793239767656, onlineIssueDateStr=2026-10-29, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1793239767656, creator=13701087609, updateTime=1793239767656, updator=13701087609, issue=Issue{id=1322506567065383670, tenantId=1146029695717560320, journalId=1281212831689347082, year='2025', volume='45', issue='7', pageStart='661', pageEnd='770', issueExtLink='null', onlineDate='null', pubDate='1752508800000', pubDateStr='2025-07-15', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1793239736572, creator='13701087609', updateTime=1793239736572, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext=null, issueFiles=null, downloadFileDto=null}, startPage=732, endPage=737, ext={EN=ArticleExt(id=1322506697629881212, articleId=1322506697441137531, tenantId=1146029695717560320, journalId=1281212831689347082, language=EN, title=Robust Total Least-Squares Estimator Based on Equation Residuals, columnId=null, journalTitle=Journal of Geodesy and Geodynamics, columnName=null, runingTitle=null, highlight=null, articleAbstract=

When designing a robust total least-squares(RTLS) estimation algorithm based on M-estimation theory, the total least-squares(TLS) estimation is not suitable for residual prediction, and reweighting based on individual random variable residuals has theoretical flaws. We derive the TLS estimation iteration formula based on multiplicative error model, theoretically explain the equation residuals or total residuals, and explain that this residual has the properties of least-squares(LS) estimation residuals. Then, considering the robustness in both observation space and structural space, an RTLS estimation algorithm is designed based on equation residuals or total residuals. Finally, through Monte Carlo experiments, the RTLS algorithm is compared with other reweighting strategies in terms of robustness, efficiency, and iteration speed, demonstrating the advantages of the proposed RTLS algorithm.

, authors=Zhipeng LÜ1, 2, 3, authorsList=Zhipeng LÜ, authorCompany=null, correspAuthors=null, authorNote=

LV Zhipeng, PhD, lecturer, majors in theory and method of geodetic data processing, E-mail:

, correspAuthorsNote=null, copyrightStatement=All rights reserved. Unauthorized reproduction is prohibited., 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, fund=null), CN=ArticleExt(id=1322506698443576193, articleId=1322506697441137531, tenantId=1146029695717560320, journalId=1281212831689347082, language=CN, title=基于等式残差的抗差总体最小二乘估计算法, columnId=1322506169105633839, journalTitle=大地测量与地球动力学, columnName=现代大地测量, runingTitle=null, highlight=null, articleAbstract=

基于M估计理论设计抗差总体最小二乘(robust total least-squares,RTLS)估计算法时,由于总体最小二乘(total least-squares,TLS)估计不适用于残差预测,根据单个随机量残差进行重定权存在理论缺陷。本文基于乘性误差模型推导TLS估计迭代公式,对等式残差或总残差进行理论解释,说明该残差具有最小二乘(least-squares,LS)估计残差的性质;然后顾及观测空间和结构空间的抗差性,基于等式残差或总残差设计相应的RTLS估计算法;最后通过Monte Carlo实验,从抗差性、有效性和迭代效率等方面与其他重定权策略进行比较,证明该RTLS估计算法的优势。

, authors=吕志鹏1, 2, 3, authorsList=吕志鹏, authorCompany=null, correspAuthors=null, authorNote=

吕志鹏,博士,讲师,主要研究方向为大地测量数据处理理论和方法,E-mail:

, correspAuthorsNote=null, copyrightStatement=版权所有,未经授权,不得转载。, copyrightOwner=大地测量与地球动力学编辑部, extLink=null, articleAbsUrl=null, sourceXml=SUil1P9RwPhf7ZIdRE7fAw==, magXml=zzvI8whXtNZ0yCop9MRZRw==, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=j/2SH7cez7SS6SHsn4h3Yg==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=wV2yzC18ZG7tkB1KDCGwxw==, mapNumber=null, fund=null)}, authors=[Author(id=1322506699001418636, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lvzhipeng1989@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322506699093693328, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, authorId=1322506699001418636, language=EN, stringName=Zhipeng LÜ, firstName=Zhipeng, middleName=null, lastName=LÜ, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1 School of Transportation Engineering, East China Jiaotong University, 808 East-Shuanggang Street, Nanchang 330013, China
2 Xi'an Institute of Surveying and Mapping, 1 Mid-Yanta Road, Xi'an 710054, China
3 Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, 808 East-Shuanggang Street, Nanchang 330013, China, bio={"content":"

LV Zhipeng, PhD, lecturer, majors in theory and method of geodetic data processing, E-mail:

"}, bioImg=null, bioContent=

LV Zhipeng, PhD, lecturer, majors in theory and method of geodetic data processing, E-mail:

, aboutCorrespAuthor=null), CN=AuthorExt(id=1322506699181773713, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, authorId=1322506699001418636, language=CN, stringName=吕志鹏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013
2 西安测绘研究所, 西安市雁塔路中段1号, 710054
3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013, bio={"content":"

吕志鹏,博士,讲师,主要研究方向为大地测量数据处理理论和方法,E-mail:

"}, bioImg=null, bioContent=

吕志鹏,博士,讲师,主要研究方向为大地测量数据处理理论和方法,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322506698678457218, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698686845827, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698678457218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Transportation Engineering, East China Jiaotong University, 808 East-Shuanggang Street, Nanchang 330013, China), AuthorCompanyExt(id=1322506698703623044, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698678457218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013)]), AuthorCompany(id=1322506698791703429, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698800092038, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698791703429, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Xi'an Institute of Surveying and Mapping, 1 Mid-Yanta Road, Xi'an 710054, China), AuthorCompanyExt(id=1322506698808480647, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698791703429, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 西安测绘研究所, 西安市雁塔路中段1号, 710054)]), AuthorCompany(id=1322506698900755336, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698904949641, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698900755336, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, 808 East-Shuanggang Street, Nanchang 330013, China), AuthorCompanyExt(id=1322506698913338250, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698900755336, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013)])])], keywords=[Keyword(id=1322506699290825618, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, orderNo=1, keyword=equation residuals), Keyword(id=1322506699366323091, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, orderNo=2, keyword=total least-squares(TLS)), Keyword(id=1322506699446014868, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, orderNo=3, keyword=robust estimation), Keyword(id=1322506699517318037, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, orderNo=4, keyword=multiplicative error), Keyword(id=1322506699576038294, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, orderNo=1, keyword=等式残差), Keyword(id=1322506699676701591, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, orderNo=2, keyword=总体最小二乘), Keyword(id=1322506699752199064, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, orderNo=3, keyword=抗差估计), Keyword(id=1322506699827696537, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, orderNo=4, keyword=乘性误差)], refs=[Reference(id=1322506701799019433, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s00190-013-0643-2, pmid=null, pmcid=null, year=2013, volume=87, issue=8, pageStart=733, pageEnd=749, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=Fang X, journalName=Journal of Geodesy, refType=null, unstructuredReference= Fang X . Weighted Total Least Squares: Necessary and Sufficient Conditions, Fixed and Random Parameters[J]. Journal of Geodesy, 2013, 87 (8): 733- 749, articleTitle=Weighted Total Least Squares: Necessary and Sufficient Conditions, Fixed and Random Parameters, refAbstract=null), Reference(id=1322506701870322602, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s00190-015-0799-z, pmid=null, pmcid=null, year=2015, volume=89, issue=6, pageStart=573, pageEnd=576, url=null, language=null, rfNumber=2, rfOrder=1, authorNames=Chang G B, journalName=Journal of Geodesy, refType=null, unstructuredReference= Chang G B . On Least-Squares Solution to 3D Similarity Transformation Problem under Gauss-Helmert Model[J]. Journal of Geodesy, 2015, 89 (6): 573- 576, articleTitle=On Least-Squares Solution to 3D Similarity Transformation Problem under Gauss-Helmert Model, refAbstract=null), Reference(id=1322506701937431467, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s40328-014-0046-8, pmid=null, pmcid=null, year=2014, volume=49, issue=2, pageStart=189, pageEnd=207, url=null, language=null, rfNumber=3, rfOrder=2, authorNames=Fang X, journalName=Acta Geodaetica et Geophysica, refType=null, unstructuredReference= Fang X . A Total Least Squares Solution for Geodetic Datum Transformations[J]. Acta Geodaetica et Geophysica, 2014, 49 (2): 189- 207, articleTitle=A Total Least Squares Solution for Geodetic Datum Transformations, refAbstract=null), Reference(id=1322506702012928940, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2017, volume=46, issue=8, pageStart=978, pageEnd=987, url=null, language=null, rfNumber=4, rfOrder=3, authorNames=王乐洋, 许光煜, 温贵森, journalName=测绘学报, refType=null, unstructuredReference=王乐洋, 许光煜, 温贵森. 一种相关观测的Partial EIV模型求解方法[J]. 测绘学报, 2017, 46 (8): 978- 987, articleTitle=一种相关观测的Partial EIV模型求解方法, refAbstract=null), Reference(id=1322506702101009325, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2017, volume=46, issue=8, pageStart=978, pageEnd=987, url=null, language=null, rfNumber=4, rfOrder=4, authorNames=Wang Leyang, Xu Guangyu, Wen Guisen, journalName=null, refType=null, unstructuredReference= Wang Leyang , Xu Guangyu , Wen Guisen . A Method for Partial EIV Model with Correlated Observations[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46 (8): 978- 987, articleTitle=null, refAbstract=null), Reference(id=1322506702172312494, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s00190-007-0190-9, pmid=null, pmcid=null, year=2008, volume=82, issue=7, pageStart=415, pageEnd=421, url=null, language=null, rfNumber=5, rfOrder=5, authorNames=Schaffrin B, Wieser A, journalName=Journal of Geodesy, refType=null, unstructuredReference= Schaffrin B , Wieser A . On Weighted Total Least-Squares Adjustment for Linear Regression[J]. Journal of Geodesy, 2008, 82 (7): 415- 421, articleTitle=On Weighted Total Least-Squares Adjustment for Linear Regression, refAbstract=null), Reference(id=1322506702243615663, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2016, volume=45, issue=8, pageStart=890, pageEnd=894, url=null, language=null, rfNumber=6, rfOrder=6, authorNames=曾文宪, 方兴, 刘经南, journalName=测绘学报, refType=null, unstructuredReference=曾文宪, 方兴, 刘经南, 等. 通用EIV平差模型及其加权整体最小二乘估计[J]. 测绘学报, 2016, 45 (8): 890- 894, articleTitle=通用EIV平差模型及其加权整体最小二乘估计, refAbstract=null), Reference(id=1322506702340084656, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2016, volume=45, issue=8, pageStart=890, pageEnd=894, url=null, language=null, rfNumber=6, rfOrder=7, authorNames=Zeng Wenxian, Fang Xing, Liu Jingnan, journalName=null, refType=null, unstructuredReference= Zeng Wenxian , Fang Xing , Liu Jingnan , et al. Weighted Total Least Squares of Universal EIV Adjustment Model[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45 (8): 890- 894, articleTitle=null, refAbstract=null), Reference(id=1322506702423970737, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s11200-012-0671-z, pmid=null, pmcid=null, year=2014, volume=58, issue=1, pageStart=1, pageEnd=16, url=null, language=null, rfNumber=7, rfOrder=8, authorNames=Fang X, journalName=Studia Geophysica et Geodaetica, refType=null, unstructuredReference= Fang X . A Structured and Constrained Total Least-Squares Solution with Cross-Covariances[J]. Studia Geophysica et Geodaetica, 2014, 58 (1): 1- 16, articleTitle=A Structured and Constrained Total Least-Squares Solution with Cross-Covariances, refAbstract=null), Reference(id=1322506702512051122, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2018, volume=144, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=9, authorNames=Wang L Y, Zhao Y W, journalName=Journal of Surveying Engineering, refType=null, unstructuredReference= Wang L Y , Zhao Y W . Scaled Unscented Transformation of Nonlinear Error Propagation: Accuracy, Sensitivity, and Applications[J]. Journal of Surveying Engineering, 2018, 144 (1), articleTitle=Scaled Unscented Transformation of Nonlinear Error Propagation: Accuracy, Sensitivity, and Applications, refAbstract=null), Reference(id=1322506702621103027, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2012, volume=41, issue=5, pageStart=715, pageEnd=722, url=null, language=null, rfNumber=9, rfOrder=10, authorNames=陈义, 陆珏, journalName=测绘学报, refType=null, unstructuredReference=陈义, 陆珏. 以三维坐标转换为例解算稳健总体最小二乘方法[J]. 测绘学报, 2012, 41 (5): 715- 722, articleTitle=以三维坐标转换为例解算稳健总体最小二乘方法, refAbstract=null), Reference(id=1322506702696600500, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2012, volume=41, issue=5, pageStart=715, pageEnd=722, url=null, language=null, rfNumber=9, rfOrder=11, authorNames=Chen Yi, Lu Jue, journalName=null, refType=null, unstructuredReference= Chen Yi , Lu Jue . Performing 3D Similarity Transformation by Robust Total Least Squares[J]. Acta Geodaetica et Cartographica Sinica, 2012, 41 (5): 715- 722, articleTitle=null, refAbstract=null), Reference(id=1322506702780486581, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2023, volume=149, issue=2, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=12, authorNames=Gong X Q, Wang H Y, Lu T D, journalName=Journal of Surveying Engineering, refType=null, unstructuredReference= Gong X Q , Wang H Y , Lu T D , et al. Combined Prediction Model for High-Speed Railway Bridge Pier Settlement Based on Robust Weighted Total Least-Squares Autoregression and Adaptive Dynamic Cubic Exponential Smoothing[J]. Journal of Surveying Engineering, 2023, 149 (2), articleTitle=Combined Prediction Model for High-Speed Railway Bridge Pier Settlement Based on Robust Weighted Total Least-Squares Autoregression and Adaptive Dynamic Cubic Exponential Smoothing, refAbstract=null), Reference(id=1322506702876955574, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s11200-015-0916-8, pmid=null, pmcid=null, year=2016, volume=60, issue=2, pageStart=177, pageEnd=194, url=null, language=null, rfNumber=11, rfOrder=13, authorNames=Wang B, Li J C, Liu C, journalName=Studia Geophysica et Geodaetica, refType=null, unstructuredReference= Wang B , Li J C , Liu C . A Robust Weighted Total Least Squares Algorithm and Its Geodetic Applications[J]. Studia Geophysica et Geodaetica, 2016, 60 (2): 177- 194, articleTitle=A Robust Weighted Total Least Squares Algorithm and Its Geodetic Applications, refAbstract=null), Reference(id=1322506702965035959, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1007/s00190-017-1043-9, pmid=null, pmcid=null, year=2018, volume=92, issue=1, pageStart=47, pageEnd=58, url=null, language=null, rfNumber=12, rfOrder=14, authorNames=Chang G B, Xu T H, Wang Q X, journalName=Journal of Geodesy, refType=null, unstructuredReference= Chang G B , Xu T H , Wang Q X . M-Estimator for the 3D Symmetric Helmert Coordinate Transformation[J]. Journal of Geodesy, 2018, 92 (1): 47- 58, articleTitle=M-Estimator for the 3D Symmetric Helmert Coordinate Transformation, refAbstract=null), Reference(id=1322506703036339128, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2016, volume=45, issue=5, pageStart=552, pageEnd=559, url=null, language=null, rfNumber=13, rfOrder=15, authorNames=赵俊, 归庆明, journalName=测绘学报, refType=null, unstructuredReference=赵俊, 归庆明. 部分变量误差模型的整体抗差最小二乘估计[J]. 测绘学报, 2016, 45 (5): 552- 559, articleTitle=部分变量误差模型的整体抗差最小二乘估计, refAbstract=null), Reference(id=1322506703103447993, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=null, pmid=null, pmcid=null, year=2016, volume=45, issue=5, pageStart=552, pageEnd=559, url=null, language=null, rfNumber=13, rfOrder=16, authorNames=Zhao Jun, Gui Qingming, journalName=null, refType=null, unstructuredReference= Zhao Jun , Gui Qingming . Total Robustified Least Squares Estimation in Partial Errors-in-Variables Model[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45 (5): 552- 559, articleTitle=null, refAbstract=null), Reference(id=1322506703178945466, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, doi=10.1080/03610918.2019.1691225, pmid=null, pmcid=null, year=2022, volume=51, issue=4, pageStart=1946, pageEnd=1966, url=null, language=null, rfNumber=14, rfOrder=17, authorNames=Wang L Y, Li Z Q, Yu F B, journalName=Communications in Statistics-Simulation and Computation, refType=null, unstructuredReference= Wang L Y , Li Z Q , Yu F B . Jackknife Method for the Location of Gross Errors in Weighted Total Least Squares[J]. Communications in Statistics-Simulation and Computation, 2022, 51 (4): 1946- 1966, articleTitle=Jackknife Method for the Location of Gross Errors in Weighted Total Least Squares, refAbstract=null)], funds=[Fund(id=1322506701614470055, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, awardId=42204047, language=EN, fundingSource=National Natural Science Foundation of China(42204047), fundOrder=null, country=null), Fund(id=1322506701673190312, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, awardId=42204047, language=CN, fundingSource=国家自然科学基金(42204047), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1322506698678457218, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698686845827, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698678457218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Transportation Engineering, East China Jiaotong University, 808 East-Shuanggang Street, Nanchang 330013, China), AuthorCompanyExt(id=1322506698703623044, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698678457218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013)]), AuthorCompany(id=1322506698791703429, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698800092038, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698791703429, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Xi'an Institute of Surveying and Mapping, 1 Mid-Yanta Road, Xi'an 710054, China), AuthorCompanyExt(id=1322506698808480647, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698791703429, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 西安测绘研究所, 西安市雁塔路中段1号, 710054)]), AuthorCompany(id=1322506698900755336, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698904949641, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698900755336, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, 808 East-Shuanggang Street, Nanchang 330013, China), AuthorCompanyExt(id=1322506698913338250, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698900755336, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013)])], figs=[ArticleFig(id=1322506699991274394, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, label=Fig. 1, caption=Multilinear regression simulation data, figureFileSmall=xY7vr2M3ynqZRtnCO/8Gww==, figureFileBig=j/2SH7cez7SS6SHsn4h3Yg==, tableContent=null), ArticleFig(id=1322506700079354779, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, label=图1, caption=多线性回归模拟数据, figureFileSmall=xY7vr2M3ynqZRtnCO/8Gww==, figureFileBig=j/2SH7cez7SS6SHsn4h3Yg==, tableContent=null), ArticleFig(id=1322506700343595932, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, label=Fig. 2, caption=Comparison of average iteration time among 5 algorithms under different contamination situations, figureFileSmall=siRc0tndY+pQ52BSK+JBgw==, figureFileBig=JMWTaQRfoGQ1gpZiMS2kwg==, tableContent=null), ArticleFig(id=1322506700410704797, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, label=图2, caption=不同污染模式下5种算法平均迭代时间比较, figureFileSmall=siRc0tndY+pQ52BSK+JBgw==, figureFileBig=JMWTaQRfoGQ1gpZiMS2kwg==, tableContent=null), ArticleFig(id=1322506700498785182, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, label=Tab. 1, caption=

Statistical results obtained by different algorithms under the influence of random error

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 1.600 122 41 1.599 771 54 1.600 227 91 1.600 224 03 1.599 789 54
0.013 217 71 0.014 338 53 0.014 302 42 0.013 991 95 0.013 865 24
0.013 217 14 0.014 336 71 0.014 300 60 0.013 990 16 0.013 863 64
β 0.200 162 40 0.199 811 37 0.200 187 74 0.200 179 93 0.199 821 25
0.012 155 11 0.013 201 55 0.013 166 75 0.012 919 35 0.012 778 80
0.012 154 02 0.013 200 20 0.013 165 42 0.012 918 10 0.012 777 55
γ 1.000 488 61 0.999 295 38 1.000 635 74 1.000 515 52 0.999 597 36
0.038 572 34 0.041 906 08 0.041 811 70 0.041 019 34 0.040 623 91
0.038 569 24 0.041 900 16 0.041 806 87 0.041 016 10 0.040 621 91
), ArticleFig(id=1322506700582671263, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, label=表1, caption=

随机误差影响下不同算法得到的统计结果

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 1.600 122 41 1.599 771 54 1.600 227 91 1.600 224 03 1.599 789 54
0.013 217 71 0.014 338 53 0.014 302 42 0.013 991 95 0.013 865 24
0.013 217 14 0.014 336 71 0.014 300 60 0.013 990 16 0.013 863 64
β 0.200 162 40 0.199 811 37 0.200 187 74 0.200 179 93 0.199 821 25
0.012 155 11 0.013 201 55 0.013 166 75 0.012 919 35 0.012 778 80
0.012 154 02 0.013 200 20 0.013 165 42 0.012 918 10 0.012 777 55
γ 1.000 488 61 0.999 295 38 1.000 635 74 1.000 515 52 0.999 597 36
0.038 572 34 0.041 906 08 0.041 811 70 0.041 019 34 0.040 623 91
0.038 569 24 0.041 900 16 0.041 806 87 0.041 016 10 0.040 621 91
), ArticleFig(id=1322506700666557344, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, label=Tab. 2, caption=

Statistical results obtained by different algorithms with small outliers

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 1.598 385 10 1.599 322 86 1.601 480 49 1.599 354 58 1.600 609 40
0.034 552 22 0.025 942 37 0.023 489 34 0.023 302 07 0.022 716 78
0.034 514 46 0.025 933 55 0.023 442 64 0.023 293 13 0.022 708 60
β 0.198 333 22 0.200 381 27 0.201 909 96 0.200 377 13 0.200 317 34
0.032 050 38 0.023 447 33 0.021 459 62 0.021 262 08 0.020 766 71
0.032 007 00 0.023 444 23 0.021 374 45 0.021 258 73 0.020 764 29
γ 0.894 793 05 0.995 663 50 1.006 323 68 1.003 030 12 0.997 860 84
0.143 099 87 0.076 164 57 0.069 402 96 0.069 000 33 0.066 861 63
0.097 000 36 0.076 041 02 0.069 114 26 0.068 933 76 0.066 827 40
), ArticleFig(id=1322506700842718113, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, label=表2, caption=

小粗差污染下不同算法得到的统计结果

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 1.598 385 10 1.599 322 86 1.601 480 49 1.599 354 58 1.600 609 40
0.034 552 22 0.025 942 37 0.023 489 34 0.023 302 07 0.022 716 78
0.034 514 46 0.025 933 55 0.023 442 64 0.023 293 13 0.022 708 60
β 0.198 333 22 0.200 381 27 0.201 909 96 0.200 377 13 0.200 317 34
0.032 050 38 0.023 447 33 0.021 459 62 0.021 262 08 0.020 766 71
0.032 007 00 0.023 444 23 0.021 374 45 0.021 258 73 0.020 764 29
γ 0.894 793 05 0.995 663 50 1.006 323 68 1.003 030 12 0.997 860 84
0.143 099 87 0.076 164 57 0.069 402 96 0.069 000 33 0.066 861 63
0.097 000 36 0.076 041 02 0.069 114 26 0.068 933 76 0.066 827 40
), ArticleFig(id=1322506700943381410, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, label=Tab. 3, caption=

Statistical results obtained by different algorithms with middle outliers

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 1.635 778 36 1.600 344 92 1.600 457 81 1.592 143 39 1.600 247 73
0.437 114 47 0.016 323 26 0.016 195 10 0.023 348 52 0.015 191 27
0.435 647 76 0.016 319 62 0.016 188 63 0.021 986 97 0.015 189 25
β 0.222 200 67 0.200 509 29 0.200 580 63 0.194 487 29 0.200 376 29
0.349 593 95 0.014 929 98 0.014 780 68 0.020 757 72 0.013 865 49
0.348 888 32 0.014 921 29 0.014 769 27 0.020 012 32 0.013 860 38
γ 0.287 359 94 1.002 447 78 1.002 397 53 0.980 113 27 1.002 071 05
1.173 344 41 0.047 909 63 0.047 334 14 0.068 001 19 0.042 272 73
0.932 138 00 0.047 847 05 0.047 273 38 0.065 028 31 0.042 221 97
), ArticleFig(id=1322506701027267491, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, label=表3, caption=

中粗差污染下不同算法得到的统计结果

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 1.635 778 36 1.600 344 92 1.600 457 81 1.592 143 39 1.600 247 73
0.437 114 47 0.016 323 26 0.016 195 10 0.023 348 52 0.015 191 27
0.435 647 76 0.016 319 62 0.016 188 63 0.021 986 97 0.015 189 25
β 0.222 200 67 0.200 509 29 0.200 580 63 0.194 487 29 0.200 376 29
0.349 593 95 0.014 929 98 0.014 780 68 0.020 757 72 0.013 865 49
0.348 888 32 0.014 921 29 0.014 769 27 0.020 012 32 0.013 860 38
γ 0.287 359 94 1.002 447 78 1.002 397 53 0.980 113 27 1.002 071 05
1.173 344 41 0.047 909 63 0.047 334 14 0.068 001 19 0.042 272 73
0.932 138 00 0.047 847 05 0.047 273 38 0.065 028 31 0.042 221 97
), ArticleFig(id=1322506701136319396, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=EN, label=Tab. 4, caption=

Statistical results obtained by different algorithms with large outliers

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 0.855 264 89 1.254 026 21 1.293 172 04 1.336 222 25 1.354 067 60
16.539 651 11 0.838 727 56 1.361 969 02 0.626 470 75 0.608 003 45
16.491 948 61 0.764 045 85 1.326 957 50 0.568 231 38 0.556 044 47
β 1.139 716 49 1.988 698 41 0.029 380 46 0.021 837 58 0.046 808 39
10.910 477 61 0.484 241 68 0.658 659 06 0.434 072 98 0.390 585 73
10.858 836 06 0.435 708 21 0.636 176 65 0.395 825 10 0.359 290 34
γ -1.346 682 11 0.307 753 31 0.472 363 13 0.407 233 40 0.496 216 18
15.843 486 55 1.470 468 19 1.674 706 74 1.403 420 83 1.239 304 56
15.668 731 58 1.297 332 35 1.589 415 62 1.272 091 89 1.132 288 69
), ArticleFig(id=1322506701274731429, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, language=CN, label=表4, caption=

大粗差污染下不同算法得到的统计结果

, figureFileSmall=null, figureFileBig=null, tableContent=
参数 TLS RTLS_ObV RTLS_CfM RTLS_Obs RTLS_Eqn
α 0.855 264 89 1.254 026 21 1.293 172 04 1.336 222 25 1.354 067 60
16.539 651 11 0.838 727 56 1.361 969 02 0.626 470 75 0.608 003 45
16.491 948 61 0.764 045 85 1.326 957 50 0.568 231 38 0.556 044 47
β 1.139 716 49 1.988 698 41 0.029 380 46 0.021 837 58 0.046 808 39
10.910 477 61 0.484 241 68 0.658 659 06 0.434 072 98 0.390 585 73
10.858 836 06 0.435 708 21 0.636 176 65 0.395 825 10 0.359 290 34
γ -1.346 682 11 0.307 753 31 0.472 363 13 0.407 233 40 0.496 216 18
15.843 486 55 1.470 468 19 1.674 706 74 1.403 420 83 1.239 304 56
15.668 731 58 1.297 332 35 1.589 415 62 1.272 091 89 1.132 288 69
)], attaches=null, journal=Journal(id=1281211523964080133, delFlag=0, nameCn=大地测量与地球动力学, nameEn=Journal of Geodesy and Geodynamics, nameHistory1=null, nameHistory2=null, issn=1671-5942, eissn=null, cn=42-1655/P, 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=BeWo6OWOseISH6uPOmc2hA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Geodesy and Geodynamics, journalRemark=null, publicationField=null, createdTime=1783394231074, updatedTime=1788951966472, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=D, firstLetterEn=D, subjectCode=Science, subjectName=null, subjectCodeEn=Science, subjectNameEn=null, picCn=BeWo6OWOseISH6uPOmc2hA==, picEn=YHLfVa8gITbOo0FeIjbLDg==, jcr=null, cjcr=null, exts=[JournalExt(id=1304522356412666698, language=CN, name=大地测量与地球动力学, 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=1788951966622, updatedTime=1788951966622, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.jgg09.com/journalx_ddcl/authorLogOn.action, submissionEditorUrl=http://www.jgg09.com/journalx_ddcl/editorLogOn.action, submissionReviewUrl=http://www.jgg09.com/journalx_ddcl/expertLogOn.action, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1304522356576244555, language=EN, name=Journal of Geodesy and Geodynamics, 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=1788951966661, updatedTime=1788951966661, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.jgg09.com/journalx_ddcl/authorLogOn.action, submissionEditorUrl=http://www.jgg09.com/journalx_ddcl/editorLogOn.action, submissionReviewUrl=http://www.jgg09.com/journalx_ddcl/expertLogOn.action, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1281212831689347082, websiteList=[Website(id=1281216327125799907, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1281212831689347082, 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/ddcl/CN, language=CN, createTime=1783395376235, createBy=18614031015, updateTime=1783395439358, updateBy=18614031015, name=大地测量与地球动力学-中文, tplId=1146099689490845704, title=大地测量与地球动力学, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1281216715371548758, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=articleTextType, value=kx, createTime=1783395468800, updateTime=1783395468800, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715346382931, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=banner, value=null, createTime=1783395468794, updateTime=1783395468794, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715400908889, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=grayFlag, value=0, createTime=1783395468807, updateTime=1783395468807, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715333800018, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=logo, value=https://castjournals.cast.org.cn/joweb/ddcl/CN/file/pic?fileId=87adUdwolRb6UqGo0QjlJQ==, createTime=1783395468791, updateTime=1783395468791, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715421880411, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=minRunFlag, value=0, createTime=1783395468812, updateTime=1783395468812, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715363160149, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/ddcl/CN/file/pic, createTime=1783395468798, updateTime=1783395468798, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715409297498, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=silenceFlag, value=0, createTime=1783395468809, updateTime=1783395468809, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715358965844, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1783395468797, updateTime=1783395468797, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715384131671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=themeColor, value=null, createTime=1783395468803, updateTime=1783395468803, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216715392520280, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327125799907, code=themeStyle, value=null, createTime=1783395468805, updateTime=1783395468805, creator=18614031015, updator=18614031015)]), Website(id=1281216327713002474, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1281212831689347082, 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/ddcl/EN, language=EN, createTime=1783395376375, createBy=18614031015, updateTime=1783395435863, updateBy=18614031015, name=大地测量与地球动力学-英文, tplId=1146101810881728533, title=Journal of Geodesy and Geodynamics, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1281216680483328066, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=articleTextType, value=kx, createTime=1783395460482, updateTime=1783395460483, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680453967935, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=banner, value=null, createTime=1783395460475, updateTime=1783395460475, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680521076805, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=grayFlag, value=0, createTime=1783395460491, updateTime=1783395460491, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680437190718, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=logo, value=https://castjournals.cast.org.cn/joweb/ddcl/EN/file/pic?fileId=87adUdwolRb6UqGo0QjlJQ==, createTime=1783395460471, updateTime=1783395460471, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680542048327, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=minRunFlag, value=0, createTime=1783395460496, updateTime=1783395460496, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680474939457, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/ddcl/EN/file/pic, createTime=1783395460480, updateTime=1783395460480, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680533659718, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=silenceFlag, value=0, createTime=1783395460494, updateTime=1783395460494, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680462356544, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1783395460477, updateTime=1783395460477, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680495910979, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=themeColor, value=null, createTime=1783395460485, updateTime=1783395460485, creator=18614031015, updator=18614031015), WebsiteProps(id=1281216680508493892, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1281216327713002474, code=themeStyle, value=null, createTime=1783395460488, updateTime=1783395460488, creator=18614031015, updator=18614031015)])], journalTitle=大地测量与地球动力学, weixinUrl=null, journalUrl=http://www.jgg09.com/CN/home, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Geodesy and Geodynamics, journalPhotoCn=BeWo6OWOseISH6uPOmc2hA==, journalPhotoEn=YHLfVa8gITbOo0FeIjbLDg==, journalFirstLetter=D, 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, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/ddcl/CN/10.14075/j.jgg.2024.06.288, detailUrlEn=https://castjournals.cast.org.cn/joweb/ddcl/EN/10.14075/j.jgg.2024.06.288, pdfUrlCn=https://castjournals.cast.org.cn/joweb/ddcl/CN/PDF/10.14075/j.jgg.2024.06.288, pdfUrlEn=https://castjournals.cast.org.cn/joweb/ddcl/EN/PDF/10.14075/j.jgg.2024.06.288, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1752508800000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于等式残差的抗差总体最小二乘估计算法
收藏切换
吕志鹏 1, 2, 3
大地测量与地球动力学 | 现代大地测量 2025,45(7): 732-737
收起
收藏切换
大地测量与地球动力学 |现代大地测量 2025 , 45 (7) : 732 -737
基于等式残差的抗差总体最小二乘估计算法
全屏
[Author(id=1322506699001418636, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lvzhipeng1989@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1322506699093693328, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, authorId=1322506699001418636, language=EN, stringName=Zhipeng LÜ, firstName=Zhipeng, middleName=null, lastName=LÜ, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1 School of Transportation Engineering, East China Jiaotong University, 808 East-Shuanggang Street, Nanchang 330013, China
2 Xi'an Institute of Surveying and Mapping, 1 Mid-Yanta Road, Xi'an 710054, China
3 Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, 808 East-Shuanggang Street, Nanchang 330013, China, bio={"content":"

LV Zhipeng, PhD, lecturer, majors in theory and method of geodetic data processing, E-mail:

"}, bioImg=null, bioContent=

LV Zhipeng, PhD, lecturer, majors in theory and method of geodetic data processing, E-mail:

, aboutCorrespAuthor=null), CN=AuthorExt(id=1322506699181773713, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, authorId=1322506699001418636, language=CN, stringName=吕志鹏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013
2 西安测绘研究所, 西安市雁塔路中段1号, 710054
3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013, bio={"content":"

吕志鹏,博士,讲师,主要研究方向为大地测量数据处理理论和方法,E-mail:

"}, bioImg=null, bioContent=

吕志鹏,博士,讲师,主要研究方向为大地测量数据处理理论和方法,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1322506698678457218, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698686845827, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698678457218, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Transportation Engineering, East China Jiaotong University, 808 East-Shuanggang Street, Nanchang 330013, China), AuthorCompanyExt(id=1322506698703623044, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698678457218, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013)]), AuthorCompany(id=1322506698791703429, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698800092038, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698791703429, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Xi'an Institute of Surveying and Mapping, 1 Mid-Yanta Road, Xi'an 710054, China), AuthorCompanyExt(id=1322506698808480647, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698791703429, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 西安测绘研究所, 西安市雁塔路中段1号, 710054)]), AuthorCompany(id=1322506698900755336, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, xref=null, ext=[AuthorCompanyExt(id=1322506698904949641, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698900755336, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, 808 East-Shuanggang Street, Nanchang 330013, China), AuthorCompanyExt(id=1322506698913338250, tenantId=1146029695717560320, journalId=1281212831689347082, articleId=1322506697441137531, companyId=1322506698900755336, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013)])])]
吕志鹏1, 2, 3
作者信息
  • 1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013
  • 2 西安测绘研究所, 西安市雁塔路中段1号, 710054
  • 3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013
作者简介:

吕志鹏,博士,讲师,主要研究方向为大地测量数据处理理论和方法,E-mail:

Robust Total Least-Squares Estimator Based on Equation Residuals
Zhipeng LÜ1, 2, 3
Affiliations
  • 1 School of Transportation Engineering, East China Jiaotong University, 808 East-Shuanggang Street, Nanchang 330013, China
  • 2 Xi'an Institute of Surveying and Mapping, 1 Mid-Yanta Road, Xi'an 710054, China
  • 3 Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, 808 East-Shuanggang Street, Nanchang 330013, China
  • LV Zhipeng, PhD, lecturer, majors in theory and method of geodetic data processing, E-mail:

出版时间: 2025-07-15 doi: 10.14075/j.jgg.2024.06.288
文章导航
收藏切换

基于M估计理论设计抗差总体最小二乘(robust total least-squares,RTLS)估计算法时,由于总体最小二乘(total least-squares,TLS)估计不适用于残差预测,根据单个随机量残差进行重定权存在理论缺陷。本文基于乘性误差模型推导TLS估计迭代公式,对等式残差或总残差进行理论解释,说明该残差具有最小二乘(least-squares,LS)估计残差的性质;然后顾及观测空间和结构空间的抗差性,基于等式残差或总残差设计相应的RTLS估计算法;最后通过Monte Carlo实验,从抗差性、有效性和迭代效率等方面与其他重定权策略进行比较,证明该RTLS估计算法的优势。

等式残差  /  总体最小二乘  /  抗差估计  /  乘性误差

When designing a robust total least-squares(RTLS) estimation algorithm based on M-estimation theory, the total least-squares(TLS) estimation is not suitable for residual prediction, and reweighting based on individual random variable residuals has theoretical flaws. We derive the TLS estimation iteration formula based on multiplicative error model, theoretically explain the equation residuals or total residuals, and explain that this residual has the properties of least-squares(LS) estimation residuals. Then, considering the robustness in both observation space and structural space, an RTLS estimation algorithm is designed based on equation residuals or total residuals. Finally, through Monte Carlo experiments, the RTLS algorithm is compared with other reweighting strategies in terms of robustness, efficiency, and iteration speed, demonstrating the advantages of the proposed RTLS algorithm.

equation residuals  /  total least-squares(TLS)  /  robust estimation  /  multiplicative error
吕志鹏. 基于等式残差的抗差总体最小二乘估计算法. 大地测量与地球动力学, 2025 , 45 (7) : 732 -737 . DOI: 10.14075/j.jgg.2024.06.288
Zhipeng LÜ. Robust Total Least-Squares Estimator Based on Equation Residuals[J]. Journal of Geodesy and Geodynamics, 2025 , 45 (7) : 732 -737 . DOI: 10.14075/j.jgg.2024.06.288
在变量误差(errors-in-variables,EIV)模型下,总体最小二乘(TLS)估计被证明具有弱一致性和渐近无偏性。在测绘领域,TLS估计得到广泛而深入的研究,并逐步形成相对完善的理论和方法体系[1-2]。其中,具有最小二乘(LS)估计形式的TLS估计算法因其简洁的公式表达、优良的收敛特性和丰富的扩展性而受到广泛关注[3-8]。
现代测量数据采集过程具有自动化、高精度和海量性等特点,导致观测数据不可避免地受到粗差污染。然而,TLS估计的崩溃污染率为0,缺乏抗差性。因此,抗差总体最小二乘(RTLS)估计成为TLS拓展方法研究的热点问题。在测绘领域,通常基于M估计采用选权迭代法设计相应的RTLS估计算法。陈义等[9]基于非线性高斯-赫尔默特(Gauss-Helmert,GH)模型,采用Huber权函数给出一种三维坐标转换的RTLS估计算法;Gong等[10]进一步使用IGG权函数替代Huber权函数,以提高算法的抗差性能;Wang等[11]基于非线性GH模型,利用中位数法得到具有抗差性的验后单位权中误差,并顾及结构抗差性计算标准化总残差,采用IGGⅢ权函数设计出相应的RTLS估计算法;Chang等[12]系统性研究三维对称赫尔默特变换的M估计算法;赵俊等[13]仅对系数矩阵中的随机量进行重定权,提出一种RTLS估计算法。然而,在杠杆观测值/权、多粗差和模式化粗差等不利情况下,M估计并不具备良好的抗差性。近年来,诸多学者开始探讨具有高崩溃污染率的RTLS估计算,如Wang等[14]采用刀切法设计相应的RTLS估计算法。
需要注意的是,与LS估计不同,TLS估计不适用于残差预测[5]。此外,对于单个观测方程而言,无论是系数还是观测值被粗差污染,都会对该观测方程的其他随机量产生极大影响,因此基于单个随机量进行重定权并不合理。本文基于乘性误差模型推导TLS估计算法,从理论上说明等式残差具有LS估计残差的性质[1],进而基于等式残差进行重定权,设计相应的RTLS估计算法,并从抗差性、有效性和迭代效率等多个方面与其他重定权策略进行比较,得出若干有益结论。
测量数据在采集过程中不可避免地受到随机误差影响,如果系数矩阵和观测向量均包含随机误差,经典的GM模型将变形为如下EIV模型[1, 4]:
$\text { 函数模型:} \boldsymbol{y}=\left(\boldsymbol{A}-\boldsymbol{E}_{\boldsymbol{A}}\right) \boldsymbol{x}+\boldsymbol{e}_{\boldsymbol{y}}$
$\text { 随机模型 }: \boldsymbol{e}=\left[\begin{array}{c}\boldsymbol{e}_{\boldsymbol{A}} \\\boldsymbol{e}_{\boldsymbol{y}}\end{array}\right]=\left[\begin{array}{c}\operatorname{vec}\left(\boldsymbol{E}_{\boldsymbol{A}}\right) \\\boldsymbol{e}_{\boldsymbol{y}}\end{array}\right] \sim N\left(0, \sigma_0^2 \boldsymbol{Q}\right)$
式中,x为m阶待估参数向量;A和y分别为n×m阶系数矩阵和n阶观测向量;EA和ey分别为A和y对应的随机误差;eA =vec(EA),其中,vec(·)表示矩阵拉直变换算子;σ0为验前单位权中误差;Q为A和y对应的n(m+1)×n(m+1)阶协因数矩阵。需要注意的是,A和y在某些测绘应用中存在结构特征,即A和/或y中存在常量或相关随机量,此时对随机模型施加任意观测方程均包含随机量的约束,即可采用加权总体最小二乘(weighted total least-squares,WTLS)估计进行求解。
从经典测量平差理论和方法出发,EIV模型(1)和(2)可以等价为非线性GH模型,其对应的函数模型如下:
$f(\boldsymbol{Y}-\boldsymbol{e}, \boldsymbol{x})=0$
式中,$ \boldsymbol{Y}=\left[\begin{array}{c}\operatorname{vec}(\boldsymbol{A}) \\\boldsymbol{y}\end{array}\right]$是由A和y中元素组成的列向量。TLS问题可以转变为如下非线性约束优化问题:
$\boldsymbol{e}^{\mathrm{T}} \boldsymbol{Q}^{-} \boldsymbol{e}=\min \text { s. t. } f(\boldsymbol{Y}-\boldsymbol{e}, \boldsymbol{x})=0$
式中,Q-表示Q的广义逆矩阵,这表明A和y中存在结构特征[7]。假定待估参数x和随机误差e在第j次迭代得到的估计值和预测值分别为$ \hat{\boldsymbol{x}}^j$和$ \tilde{\boldsymbol{e}}^j$,进一步有$ \boldsymbol{x}^j=\hat{\boldsymbol{x}}^j-\tilde{\mathbf{0}}$和$ \boldsymbol{e}^j=\tilde{\boldsymbol{e}}^i-\tilde{\mathbf{0}}$,其中,$ \tilde{\mathbf{0}}$表示对应参数阶数相同的零均值随机向量,用于消除迭代过程中对应参数随机性的递归影响。
将非线性GH模型(3)中x和e在其近似值xj和ej处进行泰勒级数展开并保留至一次项,可得线性化的GH模型[6-7]:
$\left\{\begin{array}{l}f\left(\boldsymbol{Y}^j, \boldsymbol{x}^j\right)+\left.\frac{\partial f}{\partial \boldsymbol{x}}\right|_{\boldsymbol{y}^j, \boldsymbol{x}^j}\left(\boldsymbol{x}-\boldsymbol{x}^j\right)+ \\\left.\quad \frac{\partial f}{\partial \boldsymbol{e}}\right|_{\boldsymbol{y}^j, \boldsymbol{x}^j}\left(\boldsymbol{e}-\boldsymbol{e}^j\right)=0 \\\Leftrightarrow \boldsymbol{A}^j\left(\boldsymbol{x}-\boldsymbol{x}^j\right)+\boldsymbol{B}^j \boldsymbol{e}=\boldsymbol{l}^j\end{array}\right.$
式中,
$\left\{\begin{array}{l}\boldsymbol{A}^j=\left.\frac{\partial f}{\partial \boldsymbol{x}}\right|_{\boldsymbol{Y}^j, \boldsymbol{x}^j}=\boldsymbol{A}-\boldsymbol{E}_{\boldsymbol{A}}^j \\\boldsymbol{B}^j=\left.\frac{\partial f}{\partial \boldsymbol{e}}\right|_{\boldsymbol{Y}^j, \boldsymbol{x}^j}=\left[\boldsymbol{x}^{j^{\mathrm{T}}} \otimes \boldsymbol{I}_n-\boldsymbol{I}_n\right] \\\boldsymbol{l}^j=f\left(\boldsymbol{Y}^j, \boldsymbol{x}^j\right)-\boldsymbol{B}^j \boldsymbol{e}^j=\boldsymbol{y}-\boldsymbol{A} \boldsymbol{x}^j \\\boldsymbol{Y}^j=\boldsymbol{Y}-\boldsymbol{e}^j\end{array}\right.$
GH模型(5)可以视为经典测量平差中的概括平差模型。通过去参数随机化操作,式(5)中闭合差lj可视为乘性误差模型,对其应用误差传播定律有$ \boldsymbol{D}\left(\boldsymbol{l}^j\right)=\sigma_0^2 \boldsymbol{B}^j \boldsymbol{Q} \boldsymbol{B}^{j^{\mathrm{T}}}=\sigma_0^2 \boldsymbol{Q}_l^j \text {, 其中, } \boldsymbol{Q}_l^j=\boldsymbol{B}^j \boldsymbol{Q} \boldsymbol{B}^{j^{\mathrm{T}}}$。上述对随机模型施加任意观测方程均包含随机量的约束,以保证Qlj凯利逆存在。对于概括平差模型:
$\left\{\begin{array}{l}\boldsymbol{A}^j\left(\boldsymbol{x}-\boldsymbol{x}^j\right)+\boldsymbol{B}^j \boldsymbol{e}=\boldsymbol{l}^j \\\boldsymbol{e} \sim N\left(0, \sigma_0^2 \boldsymbol{Q}\right)\end{array}\right.$
采用LS估计进行求解可得参数增量估计值$ \delta \hat{\boldsymbol{x}}^{j+1}$和参数估计值$ \hat{\boldsymbol{x}}^{j+1}$:
$\delta \hat{\boldsymbol{x}}^{j+1}=\left[\boldsymbol{A}^{j^{\mathrm{T}}}\left(\boldsymbol{Q}_l^j\right)^{-1} \boldsymbol{A}^j\right]^{-1} \boldsymbol{A}^j\left(\boldsymbol{Q}_l^j\right)^{-1}\left(\boldsymbol{y}-\boldsymbol{A} \boldsymbol{x}^j\right)$
$\begin{gathered}\hat{\boldsymbol{x}}^{j+1}=\hat{\boldsymbol{x}}^j+\delta \hat{\boldsymbol{x}}^{j+1}= \\{\left[\boldsymbol{A}^{j^{\mathrm{T}}}\left(\boldsymbol{Q}_l^j\right)^{-1} \boldsymbol{A}^j\right]^{-1} \boldsymbol{A}^j\left(\boldsymbol{Q}_l^j\right)^{-1}\left(\boldsymbol{y}-\boldsymbol{E}_A^j \boldsymbol{x}^j\right)}\end{gathered}$
以及残差预测值$ \tilde{\boldsymbol{e}}^{j+1}$:
$\tilde{\boldsymbol{e}}^{j+1}=\boldsymbol{Q} \boldsymbol{B}^{j^{\mathrm{T}}}\left(\boldsymbol{Q}_l^j\right)^{-1}\left(\boldsymbol{l}^j-\boldsymbol{A}^j \delta \hat{\boldsymbol{x}}^{j+1}\right)$
进而可通过$ \boldsymbol{e}^{j+1}=\tilde{\boldsymbol{e}}^{j+1}-\tilde{\mathbf{0}}$重构Ej+1和更新Aj+1。同时,也可以得到闭合差lj的残差预测值$ \tilde{\boldsymbol{e}}_l^{j+1}$:
$\tilde{\boldsymbol{e}}_l^{j+1}=\boldsymbol{y}-\boldsymbol{A} \hat{\boldsymbol{x}}^{j+1}$
需要注意的是,$\tilde{\boldsymbol{e}}^{j+1} $是系数矩阵和观测向量中单个随机量的残差预测值,而TLS估计不适用于残差预测[5]。$ \tilde{\boldsymbol{e}}^{j+1}$由系数矩阵和观测向量的线性组合构成,具有LS估计预测残差的形式,本质上是非线性GH模型(3)中观测方程的预测残差,被形象地称为等式残差。
与LS估计类似,应用误差传播律可得$ \hat{\boldsymbol{x}}^{j+1}$和$ \tilde{\boldsymbol{e}}^{j+1}$的协因数矩阵以及验后单位权中误差$ \hat{\sigma}_0$:
$\boldsymbol{Q}_x^{j+1}=\left[\boldsymbol{A}^{j+1^{\mathrm{T}}}\left(\boldsymbol{Q}_l^{j+1}\right)^{-1} \boldsymbol{A}^{j+1}\right]^{-1}$
$\boldsymbol{Q}_{\bar{e}_l}^{j+1}=\boldsymbol{Q}_l^{j+1}-\boldsymbol{A}^{j+1}\left[\boldsymbol{A}^{j+1^{\mathrm{T}}}\left(\boldsymbol{Q}_l^{j+1}\right)^{-1} \boldsymbol{A}^{j+1}\right]^{-1} \boldsymbol{A}^{j+1^{\mathrm{T}}}$
$\hat{\sigma}_0=\sqrt{\frac{\tilde{\boldsymbol{e}}_l^{j+1^{\mathrm{T}}}\left(\boldsymbol{Q}_l^{j+1}\right)^{-1} \tilde{\boldsymbol{e}}_l^{j+1}}{m-n}}$
式中,$\boldsymbol{Q}_l^{j+1}=\boldsymbol{B}^{j+1} \boldsymbol{Q} \boldsymbol{B}^{j+1^{\mathrm{T}}} $。
EIV模型任一观测方程的观测量和系数中可能存在粗差,会对该观测方程中其他随机量的预测残差产生影响,同时TLS估计并不适用于残差预测,因此基于单个随机量的重定权策略存在缺陷。本文§1.1基于乘性误差模型推导出等式残差,并已说明该类型残差具有LS估计预测残差的形式,可以设计相应的RTLS估计算法用于残差预测。
选权迭代法是M估计算法设计中常用的方法,等价权函数是选权迭代法实现的核心。回降权函数能够有效抑制较大粗差的不利影响,呈现出更优的抗差性。IGGⅢ权函数是一种典型的回降权函数,其对应的协因数因子表达式为[9]:
$R_i=\left\{\begin{array}{l}1, \left|\tilde{v}_{l i}\right| \leqslant k_0 \\\frac{\left|\tilde{v}_{l i}\right|}{k_0}\left(\frac{k_1-k_0}{k_1-\left|\tilde{v}_{l i}\right|}\right)^2, k_0 <\left|\tilde{v}_{l i}\right| \leqslant k_1 \\10^{10}, \left|\tilde{v}_{l i}\right|>k_1\end{array}\right.$
式中,Ri为协因数因子矩阵R的第i个对角线元素;$\tilde{v}_{l i}=\frac{\tilde{e}_{l i}^{j+1}}{\hat{\sigma}_0 Q_{\tilde{e}_{l i}}^{j+1}} $为第i个标准化的等式残差,用于实现M估计的结构抗差性,其中,$ \tilde{e}_{l i}^{j+1} \text { 和 } Q_{\tilde{e}_{l i}}^{j+1}$分别表示等式残差向量的第i个分量和等式残差协因数矩阵的第i个对角线元素;k0和k1可以根据不同污染情况进行预设,用于调节M估计的抗差性能。进一步可得等价协因数矩阵如下:
$\overline{\boldsymbol{Q}}_l^{j+1}=\sqrt{\boldsymbol{R}} \boldsymbol{Q}_l^{j+1} \sqrt{\boldsymbol{R}}$
式中,R中对角线元素由式(15)给出,非对角线元素为0,$ \sqrt{\boldsymbol R}$表示对R中对角线元素求平方根。当观测值受到粗差污染时,参数估值将产生偏差,导致验后单位权中误差畸变,影响RTLS估计的结构抗差性。一种有效的解决方法是采用中位数法计算具有抗差性的验后单位权中误差:
$\hat{\sigma}_0=1.4826 \cdot \mathop {{\rm{med}}}\limits_{i = 1}^m\left(\left|\frac{\tilde{e}_{l i}^{j+1}}{\sqrt{Q_{\tilde{e}_{l i}}^{j+1}}}\right|\right)$
式中,$ \mathop {{\rm{med}}}\limits_{i = 1}^m(\bullet)$表示求取m个元素的中位数。
RTLS估计算法通常采用LS估值作为初值,其算法流程如下:
1) 设置迭代收敛于阈值δ;
2)初值$ \hat{\boldsymbol{x}}^0=\left(\boldsymbol{A}^{\mathrm{T}} \boldsymbol{Q}_y^{-1} \boldsymbol{A}\right)^{-1} \boldsymbol{A}^{\mathrm{T}} \boldsymbol{Q}_y^{-1} \boldsymbol{y} \text { 和 } \boldsymbol{R}=\boldsymbol{I}_m$;
3) 更新$ \boldsymbol{B}^1=\left[\hat{\boldsymbol{x}}^{1^{\mathrm{T}}} \otimes \boldsymbol{I}_n-\boldsymbol{I}_n\right] \text { 和 } \boldsymbol{Q}_l^1=\boldsymbol{B}^1 \boldsymbol{Q} \boldsymbol{B}^{1^{\mathrm{T}}}$;
4) 设置循环迭代计数j=1,进入循环;
5) 更新$ \overline{\boldsymbol{Q}}_l^j=\sqrt{\boldsymbol{R}} \boldsymbol{Q}_l^j \sqrt{\boldsymbol{R}} \text { 和 } \boldsymbol{l}^j=\boldsymbol{y}-\boldsymbol{A} \hat{\boldsymbol{x}}^j$;
6) 参数估计$ \delta \hat{\boldsymbol{x}}^{j+1}=\left[\boldsymbol{A}^{j^{\mathrm{T}}}\left(\overline{\boldsymbol{Q}}_l^j\right)^{-1} \boldsymbol{A}^j\right]^{-1} \boldsymbol{A}^j\left(\overline{\boldsymbol{Q}}_l^j\right)^{-1}\left(\boldsymbol{y}-\boldsymbol{A} \boldsymbol{x}^j\right)^{-1} \text { 和 } \hat{\boldsymbol{x}}^{j+1}=\hat{\boldsymbol{x}}^j+\delta \hat{\boldsymbol{x}}^{j+1}$;
7) 残差预测$\tilde{\boldsymbol{e}}^{j+1}=Q \boldsymbol{B}^{\mathrm{T}^{\mathrm{T}}}\left(\overline{\boldsymbol{Q}}_l^j\right)^{-1}\left(\boldsymbol{l}^j-\boldsymbol{A}^j \delta \hat{\boldsymbol{x}}^{j+1}\right) $;
8) 重构$ \boldsymbol{E}_A^{j+1}=\operatorname{ivec}\left(\tilde{\boldsymbol{e}}^{j+1}\right)$,其中,ivec(·)表示向量矩阵化算子,更新$ \boldsymbol{A}^{j+1}=\boldsymbol{A}-\boldsymbol{E}_A^{j+1}$;
9) 等式残差预测$\tilde{\boldsymbol{e}}_l^{j+1}=\boldsymbol{y}-\boldsymbol{A} \hat{\boldsymbol{x}}^{j+1} $;
10) 更新$ \boldsymbol{B}^{j+1}=\left[\hat{\boldsymbol{x}}^{j+1^{\mathrm{T}}} \otimes \boldsymbol{I}_n-\boldsymbol{I}_n\right]$和Qlj+1 = Bj+1QBj+1T;
11) 更新$ Q_{\tilde{e}_l}^{j+1}=Q_l^{j+1}-A^{j+1}\left[A^{j+1^{\mathrm{T}}}\left(Q_l^{j+1}\right)^{-1}\boldsymbol{A}^{j+1}\right]^{-1} \boldsymbol{A}^{j+1^{\mathrm{T}}}$;
12) 计算$ \hat{\sigma}_0=1.4826 \cdot \mathop {{\rm{med}}}\limits_{i = 1}^m\left(\left|\frac{\tilde{e}_{l i}^{j+1}}{\sqrt{Q_{\tilde{e}_{l i}}^{j+1}}}\right|\right)$和$ \tilde{v}_{l i}=\frac{\tilde{e}_{l i}^{j+1}}{\hat{\sigma}_0 \sqrt{Q_{\tilde{e}_{l i}}^{j+1}}}(i=1, \cdots, m)$;
13) 通过式(15)计算R;
14) 当$ \left\|\hat{\boldsymbol{x}}^{j+1}-\hat{\boldsymbol{x}}^j\right\|<\delta$终止迭代;否则,j=j+1,重复循环;
15) 计算$\hat{\sigma}_0=\sqrt{\frac{\tilde{\boldsymbol{e}}_l^{j+1^{\mathrm{T}}}\left(\overline{\boldsymbol{Q}}_l^{j+1}\right)^{-1} \tilde{\boldsymbol{e}}_l^{j+1}}{n-m-p}} $其中,p表示落入拒绝区的观测方程数量。
本文采用多线性拟合问题验证基于等式残差设计的RTLS估计算法的优势,该算例对应的EIV模型如下[5]:
$T(t)=\alpha G(t)+\beta S(t)+\gamma$
式中,G(t)、S(t)和T(t)表示与气象现象相关的参数,本文忽略其对应的具体科学和工程问题,仅考察其数值意义。假设式(18)中参数的真值分别为:
$\left[\begin{array}{lll}\alpha & \beta & \gamma\end{array}\right]=\left[\begin{array}{lll}1.6 & 2.2 & 1.0\end{array}\right]$
G(t)、S(t)和T(t)的时间序列如图 1所示。
为研究不同重定权策略对RTLS估计算法的影响,5种参数估计算法被用于数据分析:1)TLS估计,作为抗差估计算法有效性和抗差性的参考[1];2)RTLS_ObV估计,仅对观测向量中的随机量进行重定权;3)RTLS_CfM估计,仅对系数矩阵中的随机量进行重定权[13];4)RTLS_Obs估计,分别对系数矩阵和观测向量中的随机量进行重定权[11];5)RTLS_Eqn估计,对每个观测方程对应的等式残差进行重定权,即本文算法。上述4种抗差估计算法均顾及结构空间和观测空间抗差性,并采用IGGⅢ权函数,在实验中给定k0=2.0和k1=4.5。此外,为全面分析不同重定权策略的性能,对图 1中G(t)、S(t)和T(t)施加N(0, 0.052)的正态随机误差,同时,假设G(t)、S(t)和T(t)受到污染率为10%的粗差污染。通过均匀分布引入3种不同尺度的粗差,取值区间分别为:1)[0.2,0.5),定义为小粗差;2)[0.5,5),定义为中粗差;3)[5,30],定义为大粗差。在不受粗差影响以及受到3种不同尺度粗差影响条件下,进行1 000次Monte Carlo实验,采用均值(mean)、均方根误差(RMSE)和标准差(STD)等3种统计量对不同参数估计方法进行评估。4种不同误差模式下不同参数估计方法的统计结果如表 1~4所示,图 2为不同参数估计方法的平均迭代时间。
表 1为仅受随机误差影响下,5种不同参数估计方法的统计结果,由于观测值不受粗差污染,TLS方法估计值最优,且估计偏差最小。4种RTLS估计方法的有效性均有所降低,估计偏差均有所增加,说明均存在误判的情况。结合图 2中不同估计方法的平均迭代时间可知,对于无粗差的观测数据,宜采用TLS估计。RTLS_Eqn估计在4种抗差方法中统计特性最佳,表明对单个随机量进行重定权存在更大的误判可能性,从而造成参数估计有效性下降和估计偏差变大。
表 2为小粗差污染条件下,5种不同参数估计方法的统计结果。由表可知,TLS估计崩溃污染率为0,对各种模式的粗差污染均不具有抗差性,TLS估计值存在明显的估计偏差。4种RTLS估计方法均呈现出不同程度的抗差性,相对完全重定权的RTLS_Obs估计,部分重定权的RTLS_ObV估计和RTLS_CfM估计的有效性更差,估计偏差更大。RTLS_CfM估计相对RTLS_ObV估计呈现出更优的估计有效性,但估计偏差更大。结合图 2可知,RTLS_Eqn估计在4种抗差估计方法中表现最佳。
由表 3可知,在中粗差污染情况下,由于未对粗差采取抑制措施,TLS估计存在更加显著的估计偏差,估计有效性也极大下降。4种RTLS估计方法均呈现出不同程度的抗差性。与小粗差污染情况不同的是,RTLS_ObV估计和RTLS_CfM估计相较RTLS_Obs估计呈现出更好的统计特性,即更优的估计有效性和更小的估计偏差。如图 2所示,RTLS_ObV估计的平均迭代时间相对RTLS_Obs估计也显著减小,RTLS_Eqn估计在4种抗差统计方法中表现最佳。需要注意的是,相对小粗差污染情况,4种RTLS估计算法在中粗差污染情况下的抗差性和有效性均有所提升,说明小尺度的粗差更难被发现,这将影响RTLS估计的性能。
在大粗差污染条件下极易形成杠杆观测,如表 4所示,TLS估计明显崩溃,估计偏差显著,估计有效性降低。由图 2可知,TLS估计的平均迭代时间变大,在不存在重定权的情况下,说明算法收敛性变差。4种RTLS估计算法均为M估计方法,在非模式化粗差情况下,M估计抗差性能良好。但对于模式化粗差,M估计抗差性将减弱甚至崩溃。如表 4所示,相对TLS估计,4种RTLS估计算法仍呈现出抗差性,但也不同程度地受到粗差影响,产生估计偏差;RTLS_Eqn估计仍是4种RTLS估计中抗差性能最佳的方法。
本文首先基于乘性误差模型,从理论上阐释等式残差的数值意义,并说明在最小二乘估计理论框架下,该残差更适用于残差预测;然后基于等式残差进行重定权,设计相应的抗差总体最小二乘估计算法——RTLS_Eqn,该估计算法具有观测空间和结构空间抗差性;最后对不同的重定权策略进行比较,得到以下结论:
1) 在仅存在随机误差的情况下,RTLS估计算法可能会错误地识别粗差,这不仅会降低估计的有效性,还会产生估计偏差。TLS估计崩溃污染率为0,表明其对模式化和非模式化的粗差均不具备抗差性。
2) M估计是最重要的抗差估计方法之一,其对非模式化粗差具有良好的抗差性。在不同尺度粗差污染情况下,基于等式残差或总残差进行重定权的RTLS_Eqn估计表现更佳。RTLS_CfM估计和RTLS_Obs估计在不同尺度粗差污染情况下的估计偏差明显恶化,而RTLS_ObV估计的抗差性能则更为稳定。对于小尺度的粗差污染,RTLS估计仍存在误判的可能性,导致估计有效性降低和估计偏差增大。
3) 当存在杠杆观测粗差时,各种RTLS估计算法均出现不同程度的偏差,对于此类模式化粗差,M估计无法有效抑制系统性的崩溃。一种有效的处理方法就是设计相应的M估计算法。同时,RTLS估计算法的收敛性、收敛速率问题也有待进一步研究。
  • 国家自然科学基金(42204047)
参考文献 引证文献
排序方式:
1
Fang X . Weighted Total Least Squares: Necessary and Sufficient Conditions, Fixed and Random Parameters[J]. Journal of Geodesy, 2013, 87 (8): 733- 749
2
Chang G B . On Least-Squares Solution to 3D Similarity Transformation Problem under Gauss-Helmert Model[J]. Journal of Geodesy, 2015, 89 (6): 573- 576
3
Fang X . A Total Least Squares Solution for Geodetic Datum Transformations[J]. Acta Geodaetica et Geophysica, 2014, 49 (2): 189- 207
4
王乐洋, 许光煜, 温贵森. 一种相关观测的Partial EIV模型求解方法[J]. 测绘学报, 2017, 46 (8): 978- 987
Wang Leyang , Xu Guangyu , Wen Guisen . A Method for Partial EIV Model with Correlated Observations[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46 (8): 978- 987
5
Schaffrin B , Wieser A . On Weighted Total Least-Squares Adjustment for Linear Regression[J]. Journal of Geodesy, 2008, 82 (7): 415- 421
6
曾文宪, 方兴, 刘经南, 等. 通用EIV平差模型及其加权整体最小二乘估计[J]. 测绘学报, 2016, 45 (8): 890- 894
Zeng Wenxian , Fang Xing , Liu Jingnan , et al. Weighted Total Least Squares of Universal EIV Adjustment Model[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45 (8): 890- 894
7
Fang X . A Structured and Constrained Total Least-Squares Solution with Cross-Covariances[J]. Studia Geophysica et Geodaetica, 2014, 58 (1): 1- 16
8
Wang L Y , Zhao Y W . Scaled Unscented Transformation of Nonlinear Error Propagation: Accuracy, Sensitivity, and Applications[J]. Journal of Surveying Engineering, 2018, 144 (1)
9
陈义, 陆珏. 以三维坐标转换为例解算稳健总体最小二乘方法[J]. 测绘学报, 2012, 41 (5): 715- 722
Chen Yi , Lu Jue . Performing 3D Similarity Transformation by Robust Total Least Squares[J]. Acta Geodaetica et Cartographica Sinica, 2012, 41 (5): 715- 722
10
Gong X Q , Wang H Y , Lu T D , et al. Combined Prediction Model for High-Speed Railway Bridge Pier Settlement Based on Robust Weighted Total Least-Squares Autoregression and Adaptive Dynamic Cubic Exponential Smoothing[J]. Journal of Surveying Engineering, 2023, 149 (2)
11
Wang B , Li J C , Liu C . A Robust Weighted Total Least Squares Algorithm and Its Geodetic Applications[J]. Studia Geophysica et Geodaetica, 2016, 60 (2): 177- 194
12
Chang G B , Xu T H , Wang Q X . M-Estimator for the 3D Symmetric Helmert Coordinate Transformation[J]. Journal of Geodesy, 2018, 92 (1): 47- 58
13
赵俊, 归庆明. 部分变量误差模型的整体抗差最小二乘估计[J]. 测绘学报, 2016, 45 (5): 552- 559
Zhao Jun , Gui Qingming . Total Robustified Least Squares Estimation in Partial Errors-in-Variables Model[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45 (5): 552- 559
14
Wang L Y , Li Z Q , Yu F B . Jackknife Method for the Location of Gross Errors in Weighted Total Least Squares[J]. Communications in Statistics-Simulation and Computation, 2022, 51 (4): 1946- 1966
2025年第45卷第7期
35
15
引用本文
BibTeX
文章信息
doi: 10.14075/j.jgg.2024.06.288
  • 接收时间:2024-06-19
  • 首发时间:2026-10-29
  • 出版时间:2025-07-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-06-19
基金
National Natural Science Foundation of China(42204047)
国家自然科学基金(42204047)
作者信息
    1 华东交通大学交通运输工程学院, 南昌市双港东大街808号, 330013
    2 西安测绘研究所, 西安市雁塔路中段1号, 710054
    3 综合立体交通信息感知与融合江西省重点实验室, 南昌市双港东大街808号, 330013
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/ddcl/CN/10.14075/j.jgg.2024.06.288
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
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
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