DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM

Back-end optimization is a key process to eliminate the cumulative error in Visual Simultaneous Localization and Mapping (VSLAM). Existing VSLAM frameworks often use kernel function-based back-end optimization methods. However, these methods typically rely on fixed kernel parameters based on the chi...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhimin Hu, Lan Cheng, Jiangxia Wei, Xinying Xu, Zhe Zhang, Gaowei Yan
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/8/2529
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850155011763339264
author Zhimin Hu
Lan Cheng
Jiangxia Wei
Xinying Xu
Zhe Zhang
Gaowei Yan
author_facet Zhimin Hu
Lan Cheng
Jiangxia Wei
Xinying Xu
Zhe Zhang
Gaowei Yan
author_sort Zhimin Hu
collection DOAJ
description Back-end optimization is a key process to eliminate the cumulative error in Visual Simultaneous Localization and Mapping (VSLAM). Existing VSLAM frameworks often use kernel function-based back-end optimization methods. However, these methods typically rely on fixed kernel parameters based on the chi-square test, assuming Gaussian-distributed reprojection errors. In practice, though, reprojection errors are not always Gaussian, which can reduce robustness and accuracy. Therefore, we propose a data-adaptive iteratively reweighted robust kernel (DA-IRRK) approach, which combines median absolute deviation (MAD) with iteratively reweighted strategies. The robustness parameters are adaptively adjusted according to the MAD of reprojection errors, and the Huber kernel function is used to demonstrate the implementation of the back-end optimization process. The method is compared with other robust function-based approaches via the EuRoC dataset and the KITTI dataset, showing adaptability across different VSLAM frameworks and demonstrating significant improvements in trajectory accuracy on the vast majority of dataset sequences. The statistical analysis of the results from the perspective of reprojection error indicates DA-IRRK can tackle non-Gaussian noises better than the compared methods.
format Article
id doaj-art-04ab8ebe5fe94c77819f59f1cd6c7bc1
institution OA Journals
issn 1424-8220
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-04ab8ebe5fe94c77819f59f1cd6c7bc12025-08-20T02:25:07ZengMDPI AGSensors1424-82202025-04-01258252910.3390/s25082529DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAMZhimin Hu0Lan Cheng1Jiangxia Wei2Xinying Xu3Zhe Zhang4Gaowei Yan5Electrical and Power Engineering, Yingxi Campus, Taiyuan University of Technology, No. 79 Yingze West Street, Wanbailin District, Taiyuan 030024, ChinaElectrical and Power Engineering, Yingxi Campus, Taiyuan University of Technology, No. 79 Yingze West Street, Wanbailin District, Taiyuan 030024, ChinaElectrical and Power Engineering, Yingxi Campus, Taiyuan University of Technology, No. 79 Yingze West Street, Wanbailin District, Taiyuan 030024, ChinaElectrical and Power Engineering, Yingxi Campus, Taiyuan University of Technology, No. 79 Yingze West Street, Wanbailin District, Taiyuan 030024, ChinaElectrical and Power Engineering, Yingxi Campus, Taiyuan University of Technology, No. 79 Yingze West Street, Wanbailin District, Taiyuan 030024, ChinaElectrical and Power Engineering, Yingxi Campus, Taiyuan University of Technology, No. 79 Yingze West Street, Wanbailin District, Taiyuan 030024, ChinaBack-end optimization is a key process to eliminate the cumulative error in Visual Simultaneous Localization and Mapping (VSLAM). Existing VSLAM frameworks often use kernel function-based back-end optimization methods. However, these methods typically rely on fixed kernel parameters based on the chi-square test, assuming Gaussian-distributed reprojection errors. In practice, though, reprojection errors are not always Gaussian, which can reduce robustness and accuracy. Therefore, we propose a data-adaptive iteratively reweighted robust kernel (DA-IRRK) approach, which combines median absolute deviation (MAD) with iteratively reweighted strategies. The robustness parameters are adaptively adjusted according to the MAD of reprojection errors, and the Huber kernel function is used to demonstrate the implementation of the back-end optimization process. The method is compared with other robust function-based approaches via the EuRoC dataset and the KITTI dataset, showing adaptability across different VSLAM frameworks and demonstrating significant improvements in trajectory accuracy on the vast majority of dataset sequences. The statistical analysis of the results from the perspective of reprojection error indicates DA-IRRK can tackle non-Gaussian noises better than the compared methods.https://www.mdpi.com/1424-8220/25/8/2529visual SLAMback-end optimizationmedian absolute deviationdata-adaptiveiteratively reweighted robust kernel
spellingShingle Zhimin Hu
Lan Cheng
Jiangxia Wei
Xinying Xu
Zhe Zhang
Gaowei Yan
DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM
Sensors
visual SLAM
back-end optimization
median absolute deviation
data-adaptive
iteratively reweighted robust kernel
title DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM
title_full DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM
title_fullStr DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM
title_full_unstemmed DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM
title_short DA-IRRK: Data-Adaptive Iteratively Reweighted Robust Kernel-Based Approach for Back-End Optimization in Visual SLAM
title_sort da irrk data adaptive iteratively reweighted robust kernel based approach for back end optimization in visual slam
topic visual SLAM
back-end optimization
median absolute deviation
data-adaptive
iteratively reweighted robust kernel
url https://www.mdpi.com/1424-8220/25/8/2529
work_keys_str_mv AT zhiminhu dairrkdataadaptiveiterativelyreweightedrobustkernelbasedapproachforbackendoptimizationinvisualslam
AT lancheng dairrkdataadaptiveiterativelyreweightedrobustkernelbasedapproachforbackendoptimizationinvisualslam
AT jiangxiawei dairrkdataadaptiveiterativelyreweightedrobustkernelbasedapproachforbackendoptimizationinvisualslam
AT xinyingxu dairrkdataadaptiveiterativelyreweightedrobustkernelbasedapproachforbackendoptimizationinvisualslam
AT zhezhang dairrkdataadaptiveiterativelyreweightedrobustkernelbasedapproachforbackendoptimizationinvisualslam
AT gaoweiyan dairrkdataadaptiveiterativelyreweightedrobustkernelbasedapproachforbackendoptimizationinvisualslam