Robust pose-graph loop-closures with expectation-maximization

Gim Hee Lee, Friedrich Fraundorfer, Marc Pollefeys

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

In this paper, we model the robust loop-closure pose-graph SLAM problem as a Bayesian network and show that it can be solved with the Classification Expectation-Maximization (EM) algorithm. In particular, we express our robust pose-graph SLAM as a Bayesian network where the robot poses and constraints are latent and observed variables. An additional set of latent variables is introduced as weights for the loop-constraints. We show that the weights can be chosen as the Cauchy function, which are iteratively computed from the errors between the predicted robot poses and observed loop-closure constraints in the Expectation step, and used to weigh the cost functions from the pose-graph loop-closure constraints in the Maximization step. As a result, outlier loop-closure constraints are assigned low weights and exert less influences in the pose-graph optimization within the EM iterations. To prevent the EM algorithm from getting stuck at local minima, we perform the EM algorithm multiple times where the loop constraints with very low weights are removed after each EM process. This is repeated until there are no more changes to the weights. We show proofs of the conceptual similarity between our EM algorithm and the M-Estimator. Specifically, we show that the weight function in our EM algorithm is equivalent to the robust residual function in the M-Estimator. We verify our proposed algorithm with experimental results from multiple simulated and real-world datasets, and comparisons with other existing works.

Original languageEnglish
Title of host publicationIROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems
Pages556-563
Number of pages8
DOIs
Publication statusPublished - 2013
Event2013 26th IEEE/RSJ International Conference on Intelligent Robots and Systems: New Horizon, IROS 2013 - Tokyo, Japan
Duration: 3 Nov 20138 Nov 2013

Conference

Conference2013 26th IEEE/RSJ International Conference on Intelligent Robots and Systems: New Horizon, IROS 2013
CountryJapan
CityTokyo
Period3/11/138/11/13

Fingerprint

Bayesian networks
Robots
Cost functions

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Cite this

Lee, G. H., Fraundorfer, F., & Pollefeys, M. (2013). Robust pose-graph loop-closures with expectation-maximization. In IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems (pp. 556-563). [6696406] https://doi.org/10.1109/IROS.2013.6696406

Robust pose-graph loop-closures with expectation-maximization. / Lee, Gim Hee; Fraundorfer, Friedrich; Pollefeys, Marc.

IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems. 2013. p. 556-563 6696406.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Lee, GH, Fraundorfer, F & Pollefeys, M 2013, Robust pose-graph loop-closures with expectation-maximization. in IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems., 6696406, pp. 556-563, 2013 26th IEEE/RSJ International Conference on Intelligent Robots and Systems: New Horizon, IROS 2013, Tokyo, Japan, 3/11/13. https://doi.org/10.1109/IROS.2013.6696406
Lee GH, Fraundorfer F, Pollefeys M. Robust pose-graph loop-closures with expectation-maximization. In IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems. 2013. p. 556-563. 6696406 https://doi.org/10.1109/IROS.2013.6696406
Lee, Gim Hee ; Fraundorfer, Friedrich ; Pollefeys, Marc. / Robust pose-graph loop-closures with expectation-maximization. IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems. 2013. pp. 556-563
@inproceedings{0fd0f81f20ea46ae9f3436cbd0239c0d,
title = "Robust pose-graph loop-closures with expectation-maximization",
abstract = "In this paper, we model the robust loop-closure pose-graph SLAM problem as a Bayesian network and show that it can be solved with the Classification Expectation-Maximization (EM) algorithm. In particular, we express our robust pose-graph SLAM as a Bayesian network where the robot poses and constraints are latent and observed variables. An additional set of latent variables is introduced as weights for the loop-constraints. We show that the weights can be chosen as the Cauchy function, which are iteratively computed from the errors between the predicted robot poses and observed loop-closure constraints in the Expectation step, and used to weigh the cost functions from the pose-graph loop-closure constraints in the Maximization step. As a result, outlier loop-closure constraints are assigned low weights and exert less influences in the pose-graph optimization within the EM iterations. To prevent the EM algorithm from getting stuck at local minima, we perform the EM algorithm multiple times where the loop constraints with very low weights are removed after each EM process. This is repeated until there are no more changes to the weights. We show proofs of the conceptual similarity between our EM algorithm and the M-Estimator. Specifically, we show that the weight function in our EM algorithm is equivalent to the robust residual function in the M-Estimator. We verify our proposed algorithm with experimental results from multiple simulated and real-world datasets, and comparisons with other existing works.",
author = "Lee, {Gim Hee} and Friedrich Fraundorfer and Marc Pollefeys",
year = "2013",
doi = "10.1109/IROS.2013.6696406",
language = "English",
isbn = "9781467363587",
pages = "556--563",
booktitle = "IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems",

}

TY - GEN

T1 - Robust pose-graph loop-closures with expectation-maximization

AU - Lee, Gim Hee

AU - Fraundorfer, Friedrich

AU - Pollefeys, Marc

PY - 2013

Y1 - 2013

N2 - In this paper, we model the robust loop-closure pose-graph SLAM problem as a Bayesian network and show that it can be solved with the Classification Expectation-Maximization (EM) algorithm. In particular, we express our robust pose-graph SLAM as a Bayesian network where the robot poses and constraints are latent and observed variables. An additional set of latent variables is introduced as weights for the loop-constraints. We show that the weights can be chosen as the Cauchy function, which are iteratively computed from the errors between the predicted robot poses and observed loop-closure constraints in the Expectation step, and used to weigh the cost functions from the pose-graph loop-closure constraints in the Maximization step. As a result, outlier loop-closure constraints are assigned low weights and exert less influences in the pose-graph optimization within the EM iterations. To prevent the EM algorithm from getting stuck at local minima, we perform the EM algorithm multiple times where the loop constraints with very low weights are removed after each EM process. This is repeated until there are no more changes to the weights. We show proofs of the conceptual similarity between our EM algorithm and the M-Estimator. Specifically, we show that the weight function in our EM algorithm is equivalent to the robust residual function in the M-Estimator. We verify our proposed algorithm with experimental results from multiple simulated and real-world datasets, and comparisons with other existing works.

AB - In this paper, we model the robust loop-closure pose-graph SLAM problem as a Bayesian network and show that it can be solved with the Classification Expectation-Maximization (EM) algorithm. In particular, we express our robust pose-graph SLAM as a Bayesian network where the robot poses and constraints are latent and observed variables. An additional set of latent variables is introduced as weights for the loop-constraints. We show that the weights can be chosen as the Cauchy function, which are iteratively computed from the errors between the predicted robot poses and observed loop-closure constraints in the Expectation step, and used to weigh the cost functions from the pose-graph loop-closure constraints in the Maximization step. As a result, outlier loop-closure constraints are assigned low weights and exert less influences in the pose-graph optimization within the EM iterations. To prevent the EM algorithm from getting stuck at local minima, we perform the EM algorithm multiple times where the loop constraints with very low weights are removed after each EM process. This is repeated until there are no more changes to the weights. We show proofs of the conceptual similarity between our EM algorithm and the M-Estimator. Specifically, we show that the weight function in our EM algorithm is equivalent to the robust residual function in the M-Estimator. We verify our proposed algorithm with experimental results from multiple simulated and real-world datasets, and comparisons with other existing works.

UR - http://www.scopus.com/inward/record.url?scp=84893815683&partnerID=8YFLogxK

U2 - 10.1109/IROS.2013.6696406

DO - 10.1109/IROS.2013.6696406

M3 - Conference contribution

SN - 9781467363587

SP - 556

EP - 563

BT - IROS 2013: New Horizon, Conference Digest - 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems

ER -