On an Improved Iterative Reweighted Least Squares Algorithm in Robust Estimation

doi:10.11947/j.AGCS.2018.20170576

Abstract

Abstract: In geodesy,classical least squares (LS) estimation methods rely heavily on assumptions which are often not met in practice.In particular,it is often assumed that the data errors are zero mean distributed,at least appproximately.Unfortunately,when there are outliers in the data,the classical LS estimators frequently have meaningless performance.In this case,robust estimation such as M-type estimation is usually applied,which is numerically implemented by a so called iterative reweighted least squares algorithm.In the current reweighting process,however,the equivalent normal matrix is required to be inverted in every iteration,which needs an expensive computation demand,especially when the number of the unknown parameters is large.Therefore,in this contribution,the numerical process of the iterative reweighted least squares algorithm is essentially improved,which is mainly represented by avoiding the inversion of the equivalent normal matrix.The numerical example shows that the improved version is performed much superior to the previous one.

Key words: robust estimation, iterative reweighted least-squares, matrix inversion

CLC Number:

P207

FANG Xing, HUANG Lixiong, ZENG Wenxian, WU Yun. On an Improved Iterative Reweighted Least Squares Algorithm in Robust Estimation[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(10): 1301-1306.

References

[1] HUBER P J. Robust Statistics[M]. New York:John Wiley and Sons, 1981.
[2] AMIRI-SIMKOOEI A. Formulation of L₁ Norm Minimization in Gauss-Markov Models[J]. Journal of Surveying Engineering, 2003, 129(1):37-43.
[3] 彭军还, 李淑慧, 师芸, 等. 不等式约束M估计的均方误差矩阵和解的改善条件[J]. 测绘学报, 2010, 39(2):129-134. PENG Junhuan, LI Shuhui, SHI Yun, et al. The Mean Squares Error Matrix of Inequality Constrained M-estimate and the Conditions for Improving Solutions[J]. Acta Geodaetica et Cartographica Sinica, 2010, 39(2):129-134.
[4] 欧吉坤. 一种三步抗差方案的设计[J]. 测绘学报, 1996, 25(3):173-179. OU Jikun. Design of a New Scheme of Robust Estimation by Three Steps[J]. Acta Geodaetica et Cartographica Sinica, 1996, 25(3):173-179.
[5] GUO Jianfeng, OU Jikun, WANG Haitao. Robust Estimation for Correlated Observations:Two Local Sensitivity-based Downweighting Strategies[J]. Journal of Geodesy, 2010, 84(4):243-250.
[6] 王新洲, 陶本藻, 邱卫宁, 等. 高等测量平差[M]. 北京:测绘出版社, 2006. WANG Xinzhou, TAO Benzao, QIU Weining. Advanced Surveying Adjustment[M]. Beijing:Surveying and Mapping Press, 2006.
[7] TUKEY J W. A Survey of Sampling from Contaminated Distribution[M]//GLESER L J, PERLMAN M D, PRESS S J, et al. Contributions to Probability and Statistics. California:Stanford University Press, 1960.
[8] BOX G E P. Non-normality and Tests on Variances[J]. Biometrika, 1953, 40(3-4):318-335.
[9] HUBER P J. Robust Estimation of a Location Parameter[J]. The Annals of Mathematical Statistics, 1964, 35(1):73-101.
[10] HAMPEL F R. Contributions to the Theory of Robust Estimation[D]. Berkeley:University of California, 1968.
[11] KRARUP T, JUHL J, KUBIK K. Gotterdammerung over Least Squares Adjustment[C]//Proceedings of the 14th Congress of International Archives of Photogrammetry. Hamburg:[s.n.], 1980.
[12] 周江文. 经典误差理论与抗差估计[J]. 测绘学报, 1989, 18(2):115-120. ZHOU Jiangwen. Classical Theory of Errors and Robust Estimation[J]. Acta Geodaetica et Cartographica Sinica, 1989, 18(2):115-120.
[13] 杨元喜. 自适应抗差最小二乘估计[J]. 测绘学报, 1996, 25(3):206-211. YANG Yuanxi. Adaptively Robust Least Squares Estimation[J]. Acta Geodaetica et Cartographica Sinica, 1996, 25(3):206-211.
[14] BASELGA S. Global Optimization Solution of Robust Estimation[J]. Journal of Surveying Engineering, 2007, 133(3):123-128.
[15] 杨元喜, 柴洪洲, 宋力杰. 污染分布的逼近及应用[J]. 测绘学报, 1999, 28(3):209-214. DOI:10.3321/j.issn:1001-1595.1999.03.005. YANG Yuanxi, CHAI Hongzhou, SONG Lijie. Approximation for Contaminated Distribution and Its Applications[J]. Acta Geodaetica et Cartographica Sinica, 1999, 28(3):209-214.
[16] GUI Q, ZHANG J. Robust Biased Estimation and Its Applications in Geodetic Adjustments[J]. Journal of Geodesy, 1998, 72(7-8):430-435.
[17] YANG Y, SONG L, XU T. Robust Estimator for Correlated Observations Based on Bifactor Equivalent Weights[J]. Journal of Geodesy, 2002, 76(6-7):353-358.
[18] KNIGHT N L, WANG Jinling. A Comparison of Outlier Detection Procedures and Robust Estimation Methods in GPS Positioning[J]. The Journal of Navigation, 2009, 62(4):699-709.
[19] 李德仁. 利用选择权迭代法进行粗差定位[J]. 武汉测绘学院学报, 1984(1):46-48. LI Deren. Gross Error Location by Means of the Iteration Method with Variable Weights[J]. Journal of Wuhan Technical University of Surveying and Mapping, 1984(1):46-48.
[20] 葛永慧. 再生权最小二乘法稳健估计[M]. 北京:科学出版社, 2015. GE Yonghui. Self-born Weighted Robust Least Squares Estimation[M]. Beijing:Science Press, 2015.
[21] 杨元喜. 抗差估计理论及其应用[M]. 北京:八一出版社, 1993. YANG Yuanxi. Theory of Robust Estimation and Its Application[M]. Beijing:"August 1st" Publishing House. 1993.
[22] 杨元喜, 宋力杰, 徐天河. 双因子方差膨胀抗差估计[J]. 解放军测绘研究所学报, 2001, 21(2):1-5. YANG Yuanxi, SONG Lijie, XU Tianhe. Robust Parameter Estimation for Correlated Observations by Variance-Covariance Inflating[J]. Journal of the Institute of Surveying and Mapping of the PLA, 2001, 21(2):1-5.
[23] YANG Yuanxi. Robust Estimation for Dependent Observations[J]. Manuscripta Geodaetica, 1994, 19(1):10-17.
[24] 李德仁, 袁修孝. 误差处理与可靠性理论[M]. 2版. 武汉:武汉大学出版社, 2012. LI Deren, YUAN Xiuxiao. Error Processing and Reliability Theory[M]. 2nd ed. Wuhan:Press of Wuhan University, 2012.
[25] KOCH K R. Parameter Estimation and Hypothesis Testing in Linear Models[M]. Berlin:Springer-Verlag, 1999.
[26] XU Peiliang. On Robust Estimation with Correlated Observations[J]. Bulletin Géodésique, 1989, 63(3):237-252.