arxiv: 2604.19946 · v1 · submitted 2026-04-21 · 💻 cs.RO

Recognition: unknown

SL(C)AMma: Simultaneous Localisation, (Calibration) and Mapping With a Magnetometer Array

Thomas Edridge , Manon Kok

Authors on Pith no claims yet

Pith reviewed 2026-05-10 01:53 UTC · model grok-4.3

classification 💻 cs.RO

keywords magnetometer arraySLAMindoor localizationsensor calibrationdrift reductionmagnetic mappingfiltering algorithms

0 comments

The pith

Magnetometer arrays with joint calibration enable reliable indoor SLAM and cut drift by over 80% versus proprioceptive sensors.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces SLAMma, a magnetic field-based SLAM method using a magnetometer array, and SLCAMma, which additionally estimates calibration parameters. It shows via Monte Carlo simulations that calibration can be accurately recovered with enough orientation changes, and that the array provides consistent measurements across sensors no matter the motion. Validation on ten real datasets confirms that these approaches succeed where single-magnetometer SLAM fails, delivering good trajectories with substantial drift reduction.

Core claim

A magnetometer array allows direct estimation of odometry for SLAM while resolving inconsistencies between sensors; joint estimation of calibration parameters succeeds when the platform undergoes sufficient orientation excitation, yielding trajectory estimates with more than 80% less drift than pure integration of proprioceptive sensors in cases where single-magnetometer SLAM breaks down.

What carries the argument

The SLAMma and SLCAMma filtering algorithms that process array measurements for simultaneous localization, mapping, and optional calibration.

If this is right

Calibration parameters are accurately estimated under sufficient orientation excitation.
Inter-sensor measurement consistency is achieved regardless of motion type.
Trajectory estimates remain accurate in scenarios where single-magnetometer SLAM fails.
Drift is reduced by more than 80% compared to integration of proprioceptive sensors alone.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Array-based magnetic SLAM could support extended indoor exploration without external positioning references.
Joint calibration might reduce the need for separate pre-calibration procedures in deployed robotic systems.
Similar array methods could help resolve measurement inconsistencies in other multi-sensor indoor navigation setups.

Load-bearing premise

Sufficient orientation excitation must be present during operation for the calibration parameters to be estimated accurately.

What would settle it

Monte Carlo simulations showing inaccurate calibration estimates when orientation excitation is limited, or experimental datasets where SLAMma and SLCAMma do not achieve the reported drift reduction.

Figures

Figures reproduced from arXiv: 2604.19946 by Manon Kok, Thomas Edridge.

**Figure 1.** Figure 1: Example of a 30-magnetometer array moving in a square in an indoor environment with a spatially disturbed magnetic field. The background colour corresponds to the norm of the magnetic field, and the transparency is proportional to the uncertainty of the magnetic field predictions. magnetometer is low-cost, compact, and does not require line of sight. The magnetometer, which measures a 3D vector of the ambi… view at source ↗

**Figure 2.** Figure 2: Magnetometer array of size 345mm × 245mm, consisting of 30 magnetometers. The body frame is defined in the centre of the sensor board. We illustrate the measurement y b 1,k and position s b 1 of the first magnetometer. The relative positions of the magnetometers s b i are assumed to be known in the body frame as shown in [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Figure 3: MAE and 1 STD of the estimated bias and scaling parameters across 30 MC simulations for six types of motion. are calculated as MAE: µθ,k = 1 NmcNmag N Xmag i=1 X Nmc j=1 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

**Figure 4.** Figure 4: Inter-sensor measurement consistency across the magnetometers is critical for our SLAM algorithm. Specifically, magnetometers in the same position and orientation should measure the same values. We validate this consistency across the six different motion types by comparing the measurements norms of the uncalibrated magnetometers against the calibrated magnetometers using a new planar motion dataset. We … view at source ↗

**Figure 5.** Figure 5: (top) shows the Tiny no rot (only translation) and tiny yaw rot (90◦ corner turns) datasets. Both were recorded in a tiny square in close proximity to magnetic field disturbances, causing a high magnetic field norm range of [20, 50]µT. The estimated trajectories for Tiny no rot demonstrate that the magnetometer array (SLAMma/SLCAMma), reduces positional and rotational drift significantly by approximately … view at source ↗

**Figure 6.** Figure 6: Position and rotation errors for the Tiny square yaw rot experiment, using a 30-magnetometer array. The estimation uncertainties of three standard deviations are too small to be shown, except for dead reckoning. To verify whether the online calibration is successful and the magnetometers in the SLCAMma algorithm reach inter-sensor measurement consistency, similar to the results in [PITH_FULL_IMAGE:figures… view at source ↗

**Figure 7.** Figure 7: Magnetometer measurements of validation experiments using the final calibration parameters from the Small snake high experiment ( [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗

**Figure 8.** Figure 8: Estimated trajectories for six longer experiments, using a 30-magnetometer array. The background colour shows the predicted magnetic field norm, and the transparency is proportional to the prediction uncertainty. Acknowledgements This work is funded by the Sensor AI Lab, under the AI Labs program of Delft University of Technology. We thank Gustaf Hendeby from Linkoping University, and Isaac Skog and ¨ Chua… view at source ↗

read the original abstract

Indoor localisation techniques suffer from attenuated Global Navigation Satellite System (GNSS) signals and from the accumulation of unbounded drift by integration of proprioceptive sensors. Magnetic field-based Simultaneous Localisation and Mapping (SLAM) reduces drift through loop closures by revisiting previously seen locations, but extended exploration of unseen areas remains challenging. Recently, magnetometer arrays have demonstrated significant benefits over single magnetometers, as they can directly estimate the odometry. However, inconsistencies between magnetometer measurements negatively affect odometry estimates and complicate loop closure detection. We propose two filtering algorithms: The first focuses on magnetic field-based SLAM using a magnetometer array (SLAMma). The second extends this to jointly estimate the magnetometer calibration parameters (SLCAMma). We demonstrate, using Monte Carlo simulations, that the calibration parameters can be accurately estimated when there is sufficient orientation excitation, and that magnetometers achieve inter-sensor measurement consistency regardless of the type of motion. Experimental validation on ten datasets confirms these results, and we demonstrate that in cases where single magnetometer SLAM fails, SLAMma and SLCAMma provide good trajectory estimates with, more than 80% drift reduction compared to integration of proprioceptive sensors.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Magnetometer arrays with joint online calibration deliver strong drift reduction in indoor SLAM, but the method's success hinges on rotational motion that isn't quantified in the reported tests.

read the letter

The one thing to know is that this paper gives a filtering approach to magnetic SLAM that uses a whole array of magnetometers to estimate odometry directly while also handling calibration in one of the versions. That combination lets them get better trajectory estimates in extended indoor runs. What is new is the joint calibration inside the SLAM filter for the array case. Prior work had arrays for odometry or single sensors for SLAM, but putting the calibration parameters into the state and estimating them alongside the map and pose is the step forward. The Monte Carlo tests back up that the parameters can be recovered when rotation is present, and the ten real datasets show the array keeps the measurements consistent across different motions. They do well on the experimental side by testing cases where single-magnetometer SLAM breaks down and showing clear improvement over proprioceptive dead reckoning. The drift reduction number is substantial and directly relevant to the problem they set out to solve. On the soft side, the observability of the calibration depends on having enough orientation changes, and while the abstract mentions this, there is no reported measure of how much excitation was in the datasets or what the minimum requirement is. If a dataset had mostly linear motion, the calibration could stay uncertain and the consistency result might not generalize. The lack of specific error metrics or ablation studies in the summary also leaves the exact performance a bit open until the full methods are checked. This paper is for people working on magnetic or multi-sensor indoor localization in robotics. A reader who needs a method that works without external infrastructure and can self-calibrate will get concrete algorithms and supporting evidence from it. I think it deserves a serious referee. The core idea is sound and the validation targets the right failure modes, even if the excitation condition needs more detail in the write-up.

Referee Report

2 major / 2 minor

Summary. The paper proposes two filtering algorithms for indoor localization: SLAMma, which performs magnetic field-based SLAM with a magnetometer array to estimate odometry and reduce drift via loop closures, and SLCAMma, which extends this by jointly estimating magnetometer calibration parameters (biases, scale factors, soft-iron). Monte Carlo simulations are used to show that calibration parameters are accurately estimated under sufficient orientation excitation and that inter-sensor consistency is achieved regardless of motion type. Validation on ten real datasets demonstrates that both methods yield good trajectory estimates with more than 80% drift reduction versus proprioceptive integration, succeeding in cases where single-magnetometer SLAM fails.

Significance. If the central claims hold with proper quantification, the work would advance array-based magnetic SLAM for GNSS-denied environments by providing direct odometry from magnetometer inconsistencies and joint calibration, potentially enabling longer explorations with reduced drift. Strengths include the use of Monte Carlo simulations for observability analysis and validation across ten datasets; however, the absence of quantitative excitation metrics and absolute error values limits immediate impact assessment.

major comments (2)

[Abstract] Abstract: The central claim that 'calibration parameters can be accurately estimated when there is sufficient orientation excitation' is load-bearing for the SLCAMma results and the >80% drift reduction, yet no quantitative metric is defined (e.g., minimum angular-velocity variance, rotation-matrix condition number, or observability rank). The ten datasets are not checked against any such threshold, leaving open the possibility that some successful runs have insufficient excitation and that the joint estimation is under-determined.
[Abstract] Abstract: The headline performance result ('more than 80% drift reduction compared to integration of proprioceptive sensors') is presented without absolute error metrics (RMSE, final position error), ablation tables isolating array odometry versus calibration, or filter equations. This makes the claim difficult to reproduce or compare to prior single-magnetometer SLAM work and weakens substantiation of the inter-sensor consistency result.

minor comments (2)

[Abstract] The acronym SL(C)AMma is introduced in the title but not expanded or motivated in the abstract; a brief parenthetical definition would improve readability.
[Abstract] The abstract states 'Experimental validation on ten datasets confirms these results' but provides no summary statistics or reference to a results table; adding a high-level quantitative overview would strengthen the summary paragraph.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed review of our manuscript. The comments highlight important areas for improving clarity, reproducibility, and substantiation of our claims. We address each major comment point by point below and have prepared revisions to incorporate the suggested quantitative elements and supporting analyses.

read point-by-point responses

Referee: [Abstract] Abstract: The central claim that 'calibration parameters can be accurately estimated when there is sufficient orientation excitation' is load-bearing for the SLCAMma results and the >80% drift reduction, yet no quantitative metric is defined (e.g., minimum angular-velocity variance, rotation-matrix condition number, or observability rank). The ten datasets are not checked against any such threshold, leaving open the possibility that some successful runs have insufficient excitation and that the joint estimation is under-determined.

Authors: We agree that a quantitative metric for sufficient orientation excitation would strengthen the manuscript and allow readers to independently verify the conditions under which calibration parameters remain observable. Our Monte Carlo simulations systematically varied rotational excitation levels to illustrate the transition from under-determined to well-conditioned estimation, but we did not formalize a threshold or apply it to the real datasets. In the revised manuscript we will define a concrete metric (e.g., the condition number of the accumulated rotation matrix or the variance of angular velocity over each trajectory segment) and report its value for every dataset. This addition will confirm that all reported successful runs satisfy the excitation requirement and will directly address the concern about potential under-determination. revision: yes
Referee: [Abstract] Abstract: The headline performance result ('more than 80% drift reduction compared to integration of proprioceptive sensors') is presented without absolute error metrics (RMSE, final position error), ablation tables isolating array odometry versus calibration, or filter equations. This makes the claim difficult to reproduce or compare to prior single-magnetometer SLAM work and weakens substantiation of the inter-sensor consistency result.

Authors: The filter equations for both SLAMma and SLCAMma are already provided in Section III of the full manuscript. We acknowledge, however, that the abstract and results section would benefit from explicit absolute error metrics and an ablation study to isolate the contributions of array-based odometry and joint calibration. In the revised version we will add a table reporting RMSE and final position error for all ten datasets across the compared methods (proprioceptive integration, single-magnetometer SLAM, SLAMma, and SLCAMma). We will also include an ablation analysis that quantifies the incremental benefit of the magnetometer array and of the online calibration step. These changes will improve reproducibility and enable direct comparison with prior single-magnetometer magnetic SLAM literature. revision: yes

Circularity Check

0 steps flagged

No significant circularity; claims rest on external simulations and datasets

full rationale

The paper proposes SLAMma and SLCAMma filters for magnetometer-array SLAM with optional joint calibration. Central results (accurate calibration under sufficient excitation, inter-sensor consistency, >80% drift reduction on ten datasets) are obtained via Monte Carlo simulations and experimental validation rather than any derivation that reduces outputs to fitted inputs by construction. No equations, self-citations, or ansatzes are shown to make predictions tautological; the work remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The approach rests on standard domain assumptions about magnetic field stability and sensor models; no new free parameters or invented entities are introduced in the abstract.

axioms (1)

domain assumption Magnetometer array measurements can be rendered consistent through calibration under sufficient orientation change
Invoked to justify joint estimation in SLCAMma and consistency claims.

pith-pipeline@v0.9.0 · 5512 in / 1246 out tokens · 32217 ms · 2026-05-10T01:53:49.165603+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

75 extracted references · 7 canonical work pages · 2 internal anchors

[1]

and Hendeby, G

Skog, I. and Hendeby, G. and Gustafsson, F. , title =. Proc. 21st Int. Conf. Inf. Fusion (FUSION) , year =
[2]

and Hendeby, G

Skog, I. and Hendeby, G. and Trulsson, F. , title =. Proc. Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[3]

and Kok, M

Skog, I. and Kok, M. and Hendeby, G. and Huang, C. and Edridge, T. , title =. Proc. IEEE/ION Position, Location Navig. Symp. (PLANS) , year =
[4]

and Kok, M

Edridge, T. and Kok, M. , title =. Proc. 26th Int. Conf. Inf. Fusion (FUSION) , year =
[5]

and Kok, M

Edridge, T. and Kok, M. , title =. IEEE J. Indoor Seamless Positioning Navig. , year =
[6]

and Hendeby, G

Huang, C. and Hendeby, G. and Skog, I. , title =. Proc. 25th Int. Conf. Inf. Fusion (FUSION) , year =
[7]

and Hendeby, G

Huang, C. and Hendeby, G. and Prieur, C. and Fourati, H. and Skog, I. , title =. IEEE Sensors J. , volume =. 2024 , publisher =

2024
[8]

Joint Magnetometer-IMU Calibration via Maximum A Posteriori Estimation

Huang, C. and Hendeby, G. and Skog, I. , title =. arXiv preprint arXiv:2505.16662 , year =

work page internal anchor Pith review Pith/arXiv arXiv
[9]

Inertial Magnetic SLAM Systems Using Low-Cost Sensors

Huang, C. and Hendeby, G. and Skog, I. , title =. arXiv preprint arXiv:2512.10128 , year =

work page internal anchor Pith review Pith/arXiv arXiv
[10]

Using distributed magnetometers to increase

Vissi. Using distributed magnetometers to increase. Proc. 46th IEEE Conf. Decis. Control (CDC) , year =
[11]

and Kok, M

Solin, A. and Kok, M. and Wahlstr. Modeling and interpolation of the ambient magnetic field by. IEEE Trans. Robot. , volume =. 2018 , publisher =

2018
[12]

Modeling magnetic fields using

Wahlstr. Modeling magnetic fields using. Proc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP) , year =
[13]

and Solin, A

Kok, M. and Solin, A. , title =. Proc. 21st Int. Conf. Inf. Fusion (FUSION) , year =
[14]

Kok, M. and Sch. Magnetometer calibration using inertial sensors , journal =. 2016 , publisher =

2016
[15]

and Hol, J

Kok, M. and Hol, J. and Sch. Using Inertial Sensors for Position and Orientation Estimation , journal =. 2017 , publisher =

2017
[16]

and Helmons, R

Viset, F. and Helmons, R. and Kok, M. , title =. Sensors , volume =. 2022 , publisher =

2022
[17]

and Helmons, R

Viset, F. and Helmons, R. and Kok, M. , title =. Proc. 26th Int. Conf. Inf. Fusion (FUSION) , year =
[18]

, title =

Dorveaux, E. , title =
[19]

and Boudot, T

Dorveaux, E. and Boudot, T. and Hillion, M. and Petit, N. , title =. Proc. Amer. Control Conf. (ACC) , year =
[20]

and Petit, N

Dorveaux, E. and Petit, N. , title =. Proc. Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[21]

and Sand, S

Siebler, B. and Sand, S. and Hanebeck, U. D. , title =. IEEE Access , volume =. 2022 , publisher =

2022
[22]

and Lehner, A

Siebler, B. and Lehner, A. and Sand, S. and Hanebeck, U. D. , title =. Proc. 34th Int. Tech. Meeting Satell. Div. Inst. Navig. (ION GNSS+) , year =
[23]

and Heirich, O

Siebler, B. and Heirich, O. and Sand, S. and Hanebeck, U. , title =. Proc. IEEE/ION Position, Location Navig. Symp. (PLANS) , year =
[24]

and Gerstewitz, T

Siebler, B. and Gerstewitz, T. and Sand, S. and Hanebeck, U. D. , title =. Proc. 13th Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[25]

and Siebler, B

Dammann, A. and Siebler, B. and Sand, S. , title =. Sensors , volume =. 2024 , publisher =

2024
[26]

and Dong, Y

Vallivaara, I. and Dong, Y. and Arslan, T. , title =. Proc. 14th Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[27]

and Fourati, H

Zmitri, M. and Fourati, H. and Prieur, C. , title =. IEEE Sensors J. , volume =. 2021 , publisher =

2021
[28]

and Fourati, H

Zmitri, M. and Fourati, H. and Prieur, C. , title =. Sensors , volume =. 2020 , publisher =

2020
[29]

Chesneau, C. I. and Robin, R. and Meier, H. and Hillion, M. and Prieur, C. , title =. Asian J. Control , volume =. 2019 , publisher =

2019
[30]

Chesneau, C. I. and Hillion, M. and Prieur, C. , title =. Proc. Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[31]

and Abed-Meraim, K

Ouyang, G. and Abed-Meraim, K. , title =. Electronics , volume =. 2022 , publisher =

2022
[32]

and Afzal, M

Renaudin, V. and Afzal, M. H. and Lachapelle, G. , title =. J. Sensors , volume =. 2010 , publisher =

2010
[33]

and Qu, X

Wang, C. and Qu, X. and Zhang, X. and Zhu, W. and Fang, G. , title =. IEEE Trans. Instrum. Meas. , volume =. 2017 , publisher =

2017
[34]

and Wang, C

Mu, Y. and Wang, C. and Zhang, X. and Xie, W. , title =. IEEE Trans. Instrum. Meas. , volume =. 2018 , publisher =

2018
[35]

and Zhang, X

Mu, Y. and Zhang, X. and Xie, W. , title =. IEEE Magn. Lett. , volume =. 2019 , publisher =

2019
[36]

and Frassl, M

Angermann, M. and Frassl, M. and Doniec, M. and Julian, B. J. and Robertson, P. , title =. Proc. Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[37]

and Gallagher, T

Li, B. and Gallagher, T. and Dempster, A. G. and Rizos, C. , title =. Proc. Int. Conf. Indoor Positioning and Indoor Navig. (IPIN) , year =
[38]

and Shockley, J

Storms, W. and Shockley, J. and Raquet, J. , title =. Proc. Ubiquitous Positioning Indoor Navig. Locat. Based Service (UPINLBS) , year =
[39]

and Jald

Skog, I. and Jald. Position and orientation estimation of a permanent magnet using a small-scale sensor array , booktitle =. 2018 , pages =

2018
[40]

and Rantakokko, J

Nilsson, J.-O. and Rantakokko, J. and H. Accurate indoor positioning of firefighters using dual foot-mounted inertial sensors and inter-agent ranging , booktitle =. 2014 , pages =

2014
[41]

Bell, B. M. and Cathey, Frederick W. , journal =. The iterated. 1993 , month =

1993
[42]

The International Journal of Robotics Research , volume =

Simultaneous localization and mapping with sparse extended information filters , author =. The International Journal of Robotics Research , volume =. 2004 , publisher =

2004
[43]

Kay, S. M. , title =. 1993 , publisher =

1993
[44]

2010 , publisher =

Statistical Sensor Fusion , author =. 2010 , publisher =

2010
[45]

Barfoot, T. D. , title =. 2024 , publisher =

2024
[46]

Quaternion kinematics for the error-state Kalman filter

Sola, J. , title =. arXiv preprint arXiv:1711.02508 , year =

work page Pith review arXiv
[47]

and Rasmussen, C

Williams, C. and Rasmussen, C. , title =. Adv. Neural Inf. Process. Syst. , volume =
[48]

and Carlone, L

Cadena, C. and Carlone, L. and Carrillo, H. and Latif, Y. and Scaramuzza, D. and Neira, J. and Reid, I. and Leonard, J. J. , title =. IEEE Trans. Robot. , volume =. 2017 , publisher =

2017
[49]

Petersen, K. B. and Pedersen, M. S. , title =. 2008 , number =

2008
[50]

and Thomas, Jonathan D

Krnjaic, Aleksandar and Steleac, Raul D. and Thomas, Jonathan D. and Papoudakis, Georgios and Sch. Scalable multi-agent reinforcement learning for warehouse logistics with robotic and human co-workers , booktitle =. 2024 , pages =

2024
[51]

and Barrau, A

Goffin, S. and Barrau, A. and Bonnabel, S. and Br. IEEE Transactions on Automatic Control , year =
[52]

and Gkelias, A

Zafari, F. and Gkelias, A. and Leung, K. K. , title =. IEEE Commun. Surv. Tut. , volume =. 2019 , publisher =

2019
[53]

and Pich

Davidson, P. and Pich. A survey of selected indoor positioning methods for smartphones , journal =. 2016 , publisher =

2016
[54]

and Junginger, S

Huang, J. and Junginger, S. and Liu, H. and Thurow, K. , title =. Robotics , volume =. 2023 , publisher =

2023
[55]

and Torres-Sospedra, J

Mendoza-Silva, G. and Torres-Sospedra, J. and Huerta, J. , title =. Sensors , volume =. 2019 , publisher =

2019
[56]

, title =

Foxlin, E. , title =. IEEE Comput. Graph. Appl. , volume =. 2005 , publisher =

2005
[57]

and Lazzari, G

Smieszek, T. and Lazzari, G. and Salath. Assessing the dynamics and control of droplet-and aerosol-transmitted influenza using an indoor positioning system , journal =. 2019 , publisher =

2019
[58]

Location-aware access to hospital information and services , journal =

Rodr. Location-aware access to hospital information and services , journal =. 2004 , publisher =

2004
[59]

and Mazzenga, F

Giuliano, R. and Mazzenga, F. and Petracca, M. and Vari, M. , title =. Proc. 9th Int. Wireless Commun. Mobile Comput. Conf. (IWCMC) , year =
[60]

and Yang, G

Yu, X. and Yang, G. and Jones, S. and Saniie, J. , title =. Proc. Int. Conf. Electro/Inf. Technol. (EIT) , year =
[61]

and Wang, L

Xu, H. and Wang, L. and Zhang, Y. and Qiu, K. and Shen, S. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[62]

and Li, J

Yang, B. and Li, J. and Zhang, H. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[63]

and Yang, C

Cao, Y. and Yang, C. and Li, R. and Knoll, A. and Beltrame, G. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[64]

Nguyen, T. H. and Nguyen, T.-M. and Xie, L. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[65]

and Zhang, F

Lin, J. and Zhang, F. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[66]

and Kohler, D

Hess, W. and Kohler, D. and Rapp, H. and Andor, D. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[67]

and Xiao, P

Pan, Y. and Xiao, P. and He, Y. and Shao, Z. and Li, Z. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[68]

and Englot, B

Shan, T. and Englot, B. and Ratti, C. and Rus, D. , title =. Proc. IEEE Int. Conf. Robot. Autom. (ICRA) , year =
[69]

and Zhao, H

Yoo, U. and Zhao, H. and Altamirano, A. and Yuan, W. and Feng, C. , title =. arXiv preprint arXiv:2303.04307 , year =

work page arXiv
[70]

and Liu, F

Lu, J. and Liu, F. and Girerd, C. and Yip, M. C. , title =. arXiv preprint arXiv:2302.14039 , year =

work page arXiv
[71]

and Tseng, Y.-C

Chung, C.-M. and Tseng, Y.-C. and Hsu, Y.-C. and Shi, X.-Q. and Hua, Y.-H. and Yeh, J.-F. and Chen, W.-C. and Chen, Y.-T. and Hsu, W. H. , title =. arXiv preprint arXiv:2209.13274 , year =

work page arXiv
[72]

and He, F

Yan, G. and He, F. and Shi, C. and Cai, X. and Li, Y. , title =. arXiv preprint arXiv:2202.13708 , year =

work page arXiv
[73]

Robotics Applications , howpublished =
[74]

2026 , eprint=

Lyu, Qiyang and Wu, Zhenyu and Wang, Wei and Shen, Hongming and Wang, Danwei , journal=. 2026 , eprint=

2026
[75]

Modeling and calibration of inertial and vision sensors , journal =

Hol, Jeroen and Sch. Modeling and calibration of inertial and vision sensors , journal =. 2010 , publisher =

2010