arxiv: 2511.01219 · v6 · submitted 2025-11-03 · 💻 cs.RO

Tackling the Kidnapped Robot Problem via Sparse Feasible Hypothesis Sampling and Reliable Batched Multi-Stage Inference

Muhua Zhang , Lei Ma , Ying Wu , Kai Shen , Deqing Huang , Henry Leung This is my paper

Pith reviewed 2026-05-18 01:40 UTC · model grok-4.3

classification 💻 cs.RO

keywords kidnapped robot problemglobal relocalizationLiDARRRThypothesis samplingpose estimationoccupancy grid

0 comments

The pith

The proposed framework estimates the global pose of a kidnapped robot efficiently and reliably from a single LiDAR scan and an occupancy grid map while the robot remains stationary.

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

This paper addresses the Kidnapped Robot Problem by proposing a passive 2-D global relocalization framework. It casts the problem as non-convex and solves it using sparse feasible hypothesis sampling via RRT under traversability constraints from an occupancy grid map. The hypotheses are ordered and evaluated using new metrics SMAD and TAM in a batched multi-stage inference process with early termination. A reader would care if this enables reliable relocalization without the robot needing to move, supporting better long-term autonomy for mobile robots in known environments.

Core claim

The framework solves the kidnapped robot problem by generating sparse, uniformly distributed feasible positional hypotheses with RRT under traversability constraints, preliminarily ordering them with SMAD for early termination, and using TAM for reliable orientation selection and final pose evaluation in multi-stage inference.

What carries the argument

The combination of RRT-based sparse hypothesis generation under traversability constraints and batched multi-stage inference with SMAD and TAM metrics to balance completeness and efficiency in non-convex global relocalization.

Load-bearing premise

The RRT sampling under traversability constraints will produce hypotheses close enough to the true pose that the subsequent inference can identify and refine it.

What would settle it

If experiments show frequent failure to recover the true pose because it was not generated in the initial hypothesis set from the RRT, that would indicate the method does not reliably solve the problem.

Figures

Figures reproduced from arXiv: 2511.01219 by Deqing Huang, Henry Leung, Kai Shen, Lei Ma, Muhua Zhang, Ying Wu.

**Figure 1.** Figure 1: The entire pipeline of the proposed framework. Algorithm 1 Traversability-Constrained Hypothesis Sampling 1: Input: Map origin 𝒑଴ , RRT tree 𝑻ோோ், RRT sampling boundary 𝓑, grid map 𝓜, map width 𝑊ℳ, map height 𝐻ℳ, map resolution 𝑟ℳ, sampling spacing 𝜌, and gain threshold 𝜀௚௔௜௡ 2: Output: Sparse hypotheses 𝓟 = {𝒑௝ }௝ୀଵ ே𝒑 with 𝒑௝ ∈ ℛ(𝒑଴) 3: Initialize: 𝑻𝑹𝑹𝑻 ← {𝒑0 }, 𝓟 ← ∅, 𝓟௟௔௦௧ ← ∅ 4: 𝑁𝒔𝒂𝒎𝒑𝒍𝒊𝒏𝒈 ← ⌈𝑊ℳ𝐻ℳ𝑟ℳ ଶ … view at source ↗

**Figure 3.** Figure 3: Accelerated SMAD calculation via prefix sum of ranges from the panoramic map-synthesized scan. 𝑀𝑖𝑛𝐷𝑖𝑠𝑡𝑇𝑜𝑂𝑏𝑠𝑡𝑎𝑐𝑙𝑒(𝒑, 𝓜) . When 𝑑௢௕௦ ≥ 2𝑟௥௢௕௢௧ , 𝜂 = 𝜂௠௔௫ , the maximum expansion distance of RRT. When 𝑑௢௕௦ ≤ 𝑟௥௢௕௢௧ , 𝜂 = 𝑟௥௢௕௢௧ .When 𝑟௥௢௕௢௧ ≤ 𝑑௢௕௦ ≤ 2𝑟௥௢௕௢௧, 𝜂 decreases linearly as 𝑑௢௕௦ decreases. 6) TraversabilityCheck: Verifies whether all points 𝒑 on the line segment [𝒑଴ , 𝒑ଵ ] satisfy 𝑑௢௕௦(𝒑) ≥ 𝑟௥௢௕௢௧. 7)… view at source ↗

**Figure 4.** Figure 4: Typical degradation cases of likelihood-field-based scan-to-map alignment metrics for relocalization tasks. 𝑆[𝑡] = ෍𝓏̂௝ ᇱ(𝒑) ௧ ௝ୀଵ , 𝑡 ∈ [1,2𝑁௦ ], (17) where {𝓏̂௝ ᇱ(𝒑)} is duplicated once to form the cyclic extension. Then, the mean range of the synthesized scan under orientation 𝜃௠ is computed in constant time, as shown in [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗

**Figure 5.** Figure 5: Experimental robot and configurations of real-world experimental scenarios [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗

**Figure 6.** Figure 6: Typical degradation cases of likelihood-field-based scan-to-map alignment metrics for relocalization tasks [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗

**Figure 7.** Figure 7: Spatial SMAD Heatmaps of four relocalization cases in the scenario of [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗

read the original abstract

This paper addresses the Kidnapped Robot Problem (KRP), a core localization challenge of relocalizing a robot in a known map without prior pose estimate upon localization loss or at SLAM initialization. For this purpose, a passive 2-D global relocalization framework is proposed. It estimates the global pose efficiently and reliably from a single LiDAR scan and an occupancy grid map while the robot remains stationary, thereby enhancing the long-term autonomy of mobile robots. The proposed framework casts global relocalization as a non-convex problem and solves it via the multi-hypothesis scheme with batched multi-stage inference and early termination, balancing completeness and efficiency. The Rapidly-exploring Random Tree (RRT), under traversability constraints, asymptotically covers the reachable space to generate sparse, uniformly distributed feasible positional hypotheses, fundamentally reducing the sampling space. The hypotheses are preliminarily ordered by the proposed Scan Mean Absolute Difference (SMAD), a coarse beam-error level metric that facilitates the early termination by prioritizing high-likelihood candidates. The SMAD computation is optimized for limited scan measurements. The Translation-Affinity Scan-to-Map Alignment Metric (TAM) is proposed for reliable orientation selection at hypothesized positions and accurate final global pose evaluation to mitigate degradation in conventional likelihood-field metrics under translational uncertainty induced by sparse hypotheses, as well as non-panoramic LiDAR scan and environmental changes. Real-world experiments on a resource-constrained mobile robot with non-panoramic LiDAR scans show that the proposed framework achieves competitive performance in success rate, robustness under measurement uncertainty, and computational efficiency.

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

The paper gives a practical RRT-constrained sampling method plus new SMAD and TAM metrics for single-scan kidnapped robot relocalization, but finite-sample coverage of the true pose remains an unquantified risk.

read the letter

The main thing here is a concrete way to do passive global relocalization from one non-panoramic LiDAR scan against an occupancy grid. The authors use RRT under traversability constraints to produce a sparse set of position hypotheses, order them quickly with their Scan Mean Absolute Difference metric for early termination, and then apply the Translation-Affinity Metric to pick orientations and score final poses more reliably than standard likelihood fields when scans are partial or the environment has shifted a little. Real hardware tests on a resource-limited robot show competitive success rates and run times while the robot stays put. That combination of constrained sampling and the two custom metrics is the actual new piece; it is not just another multi-hypothesis wrapper. The experiments are grounded in physical runs rather than simulation only, which adds weight. The soft spot is exactly the one the stress test flags. The method assumes the RRT tree will land close enough to the true pose for the later stages to recover it. The paper notes asymptotic coverage, but the practical sparse sampling used for speed has no probabilistic bounds, minimum density, or worst-case checks for narrow corridors, map discretization, or nearby obstacles. That link is the least secure and could produce silent failures the current results do not fully expose. This work is aimed at people building long-term indoor autonomy stacks who already have a map and need a reliable stationary relocalization fallback. A reader in that niche will find usable implementation ideas in the metrics and the batched inference pipeline. It is coherent enough on its own terms to deserve a serious referee, even if the coverage question needs tightening in revision.

Referee Report

3 major / 3 minor

Summary. The manuscript proposes a passive 2-D global relocalization framework for the Kidnapped Robot Problem that casts the task as a non-convex optimization solved via sparse feasible positional hypotheses generated by RRT under traversability constraints from an occupancy grid map, preliminary ordering by the new Scan Mean Absolute Difference (SMAD) metric to enable early termination, and refinement via the Translation-Affinity Scan-to-Map Alignment Metric (TAM) within batched multi-stage inference. The approach claims to estimate the global pose reliably and efficiently from a single non-panoramic LiDAR scan while the robot remains stationary.

Significance. If the central claims hold, the work offers a practical advance for long-term autonomy of mobile robots by enabling stationary single-scan relocalization on resource-constrained platforms. The real-world experiments with non-panoramic LiDAR provide concrete evidence of competitive success rates and efficiency, while the SMAD and TAM metrics directly target challenges arising from sparse sampling, translational uncertainty, and environmental changes.

major comments (3)

[Hypothesis generation step] Hypothesis generation step using RRT under traversability constraints: the central reliability claim rests on the assumption that the finite, efficiency-driven sparse set of hypotheses will include or lie sufficiently close to the true pose for SMAD ordering plus TAM-based multi-stage inference to recover it. While asymptotic coverage is noted, no probabilistic coverage bounds, minimum sampling density requirements, or worst-case analysis for narrow corridors, map discretization artifacts, or non-traversable regions near the true pose are supplied; this is load-bearing for the performance claims given the non-panoramic LiDAR and environmental change factors.
[Real-world experiments] Real-world experiments section: the reported competitive success rate, robustness under measurement uncertainty, and computational efficiency lack quantitative baselines, error bars, statistical tests, or detailed failure-case analysis, leaving moderate gaps in substantiating the claims for resource-constrained robots with non-panoramic scans.
[TAM metric] TAM metric definition and evaluation: the motivation for TAM to mitigate degradation in conventional likelihood-field metrics under translational uncertainty is stated, but the manuscript provides no direct ablation or quantitative comparison demonstrating the improvement in orientation selection accuracy at hypothesized positions relative to standard metrics.

minor comments (3)

[Parameter selection] The free parameters (RRT sampling density/number of hypotheses and SMAD early-termination threshold) are identified but lack explicit guidance or sensitivity analysis on their selection for different environments.
[Metrics definitions] Notation for SMAD (optimized for limited scan measurements) and TAM should include explicit equations or pseudocode in the main text to improve clarity and reproducibility.
[Abstract] The abstract would benefit from at least one quantitative result (e.g., success rate or runtime) to support the 'competitive performance' claim.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments on our manuscript. We address each major comment point by point below, providing clarifications and indicating where revisions have been made to strengthen the paper.

read point-by-point responses

Referee: [Hypothesis generation step] Hypothesis generation step using RRT under traversability constraints: the central reliability claim rests on the assumption that the finite, efficiency-driven sparse set of hypotheses will include or lie sufficiently close to the true pose for SMAD ordering plus TAM-based multi-stage inference to recover it. While asymptotic coverage is noted, no probabilistic coverage bounds, minimum sampling density requirements, or worst-case analysis for narrow corridors, map discretization artifacts, or non-traversable regions near the true pose are supplied; this is load-bearing for the performance claims given the non-panoramic LiDAR and environmental change factors.

Authors: We appreciate the referee's emphasis on the foundational role of the hypothesis generation step. The RRT under traversability constraints is intended to asymptotically cover the reachable space, producing a sparse yet uniformly distributed set of feasible poses that the subsequent SMAD ordering and TAM refinement can reliably recover, as supported by our real-world results. We agree that explicit probabilistic coverage bounds, minimum density requirements, or worst-case analysis for edge cases like narrow corridors or discretization artifacts would provide stronger theoretical grounding. In the revised manuscript, we have added a dedicated discussion subsection on empirical sampling density and coverage, including new experiments in narrow-corridor and near-obstacle scenarios. However, deriving general probabilistic bounds for RRT in arbitrary constrained environments is a complex theoretical question that lies beyond the scope of this applied work; we instead rely on the combination of asymptotic guarantees and extensive empirical validation. revision: partial
Referee: [Real-world experiments] Real-world experiments section: the reported competitive success rate, robustness under measurement uncertainty, and computational efficiency lack quantitative baselines, error bars, statistical tests, or detailed failure-case analysis, leaving moderate gaps in substantiating the claims for resource-constrained robots with non-panoramic scans.

Authors: We thank the referee for identifying these gaps in the experimental presentation. To better substantiate the claims of competitive success rates, robustness, and efficiency on resource-constrained platforms, the revised manuscript now includes quantitative baselines against additional state-of-the-art global localization methods, error bars computed over repeated trials, and appropriate statistical significance tests. We have also expanded the failure-case analysis to explicitly discuss scenarios involving measurement uncertainty, non-panoramic scan limitations, and environmental changes. These additions appear in the updated Experiments and Discussion sections. revision: yes
Referee: [TAM metric] TAM metric definition and evaluation: the motivation for TAM to mitigate degradation in conventional likelihood-field metrics under translational uncertainty is stated, but the manuscript provides no direct ablation or quantitative comparison demonstrating the improvement in orientation selection accuracy at hypothesized positions relative to standard metrics.

Authors: We acknowledge that a direct ablation study would more clearly demonstrate the advantages of TAM. In the revised manuscript, we have incorporated a new ablation experiment in the Experiments section that quantitatively compares TAM against standard likelihood-field metrics for orientation selection at hypothesized positions. The results confirm improved accuracy under translational uncertainty, non-panoramic scans, and environmental variations, directly supporting the motivation for TAM. This comparison is presented both in the main text and supplementary material. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on novel algorithmic proposals and external experiments

full rationale

The paper introduces new components including RRT-based sparse hypothesis sampling under traversability constraints, the SMAD coarse metric for ordering, and the TAM alignment metric for pose evaluation. These are defined and motivated directly from the problem setup and standard RRT coverage properties rather than reducing to fitted parameters, self-referential equations, or load-bearing self-citations. The central claim of reliable single-scan relocalization is supported by real-world experiments on resource-constrained robots, making the framework self-contained against external benchmarks with no evident reduction of outputs to inputs by construction.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 2 invented entities

The method adds new algorithmic entities (SMAD and TAM) and relies on standard robotics assumptions about map accuracy and stationarity; no free parameters are explicitly quantified in the abstract but typical tuning thresholds for sampling density and termination are implied.

free parameters (2)

RRT sampling density or number of hypotheses
Controls how many feasible positions are generated and directly affects coverage and runtime
SMAD early-termination threshold
Determines when to stop the multi-stage inference on high-likelihood candidates

axioms (2)

domain assumption The occupancy grid map is static and accurately represents the environment
All scan-to-map comparisons and traversability constraints depend on this map
domain assumption The robot remains stationary during the single-scan relocalization
Enables direct use of one scan without motion compensation

invented entities (2)

SMAD metric no independent evidence
purpose: Coarse beam-error metric for preliminary hypothesis ordering and early termination
Newly proposed to prioritize candidates before expensive alignment
TAM metric no independent evidence
purpose: Translation-affinity scan-to-map alignment for orientation selection and final evaluation
Introduced to handle translational uncertainty and non-panoramic scans

pith-pipeline@v0.9.0 · 5830 in / 1633 out tokens · 48338 ms · 2026-05-18T01:40:21.904623+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

The Rapidly-exploring Random Tree (RRT), under traversability constraints, asymptotically covers the reachable space to generate sparse, uniformly distributed feasible positional hypotheses... The Scan Mean Absolute Difference (SMAD)... The Translation-Affinity Scan-to-Map Alignment Metric (TAM)... batched multi-stage inference with early termination
IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

RRT under robot’s traversability constraints... VoxelDownsample... AdaptiveExpandDist

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Offline-Online Hierarchical 3D Global Relocalization With Synthetic LiDAR Sensing and Descriptor-Space Retrieval
cs.RO 2026-05 unverdicted novelty 4.0

A hierarchical offline-online framework for 3D global relocalization using synthetic LiDAR and descriptor retrieval achieves 3-second average time and 8 cm accuracy with order-of-magnitude efficiency gains over prior methods.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages · cited by 1 Pith paper

[1]

An Automatic Analog Instrument Reading System Using Computer Vision and Inspection Robot,

J. Huang, J. Wang, Y. Tan, D. Wu and Y. Cao, "An Automatic Analog Instrument Reading System Using Computer Vision and Inspection Robot," in IEEE Transactions on Instrumentation and Measurement , vol. 69, no. 9, pp. 6322-6335, Sept. 2020

work page 2020
[2]

Graph-Based SLAM for Under- Train Inspection Robots in Narrow Site,

M. Zhang, L. Ma, K. Shen and Y. Sun, "Graph-Based SLAM for Under- Train Inspection Robots in Narrow Site," 2024 30th International Conference on Mechatronics and Machine Vision in Practice (M2VIP) , Leeds, United Kingdom, 2024, pp. 1-5

work page 2024
[3]

A Hybrid- Dimensional Laser SLAM Framework for Indoor Quadruped Inspection Robots,

J. Cheng, M. Zhang, L. Ma, H. Chen, Y. Gan and D. Huang, "A Hybrid- Dimensional Laser SLAM Framework for Indoor Quadruped Inspection Robots," in IEEE Sensors Journal, vol. 24, no. 10, pp. 16935-16942, 15 May15, 2024

work page 2024
[4]

A Survey on Active Simultaneous Localization and Mapping: State of the Art and New Frontiers,

J. A. Placed et al., "A Survey on Active Simultaneous Localization and Mapping: State of the Art and New Frontiers," in IEEE Transactions on Robotics, vol. 39, no. 3, pp. 1686-1705, June 2023

work page 2023
[5]

Finite Memory- Simultaneous Localization and Calibration With Application to Mobile Robots,

D. K. Lee, J. M. Pak, P. Shi and C. K. Ahn, "Finite Memory- Simultaneous Localization and Calibration With Application to Mobile Robots," in IEEE Transactions on Industrial Electronics , vol. 72, no. 6, pp. 6145-6154, June 2025

work page 2025
[6]

Indoor Localization Uncertainty Control Based on Wireless Ranging for Robots Path Planning,

F. Shamsfakhr, A. Antonucci, L. Palopoli, D. Macii and D. Fontanelli, "Indoor Localization Uncertainty Control Based on Wireless Ranging for Robots Path Planning," in IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1-11, 2022

work page 2022
[7]

High- Traversability and Precise Navigation for Mobile Robots in Constrained Environments,

M. Zhang, L. Ma, Y. Wu, K. Shen, Y. Sun and H. Leung, "High- Traversability and Precise Navigation for Mobile Robots in Constrained Environments," in IEEE Sensors Journal , vol. 25, no. 12, pp. 22815- 22826, 15 June15, 2025

work page 2025
[8]

Model-Based Deep Learning for Low-Cost IMU Dead Reckoning of Wheeled Mobile Robot,

F. Guo, H. Yang, X. Wu, H. Dong, Q. Wu and Z. Li, "Model-Based Deep Learning for Low-Cost IMU Dead Reckoning of Wheeled Mobile Robot," in IEEE Transactions on Industrial Electronics , vol. 71, no. 7, pp. 7531- 7541, July 2024

work page 2024
[9]

Robust Vision-Aided Inertial Navigation System for Protection Against Ego-Motion Uncertainty of Unmanned Ground Vehicle,

C. Zhai, M. Wang, Y. Yang and K. Shen, "Robust Vision-Aided Inertial Navigation System for Protection Against Ego-Motion Uncertainty of Unmanned Ground Vehicle," in IEEE Transactions on Industrial Electronics, vol. 68, no. 12, pp. 12462-12471, Dec. 2021

work page 2021
[10]

Robust Lifelong Indoor LiDAR Localization Using the Area Graph,

F. Xie and S. Schwertfeger, "Robust Lifelong Indoor LiDAR Localization Using the Area Graph," in IEEE Robotics and Automation Letters, vol. 9, no. 1, pp. 531-538, Jan. 2024

work page 2024
[11]

Long-Term Localization Using Semantic Cues in Floor Plan Maps,

N. Zimmerman, T. Guadagnino, X. Chen, J. Behley and C. Stachniss, "Long-Term Localization Using Semantic Cues in Floor Plan Maps," in IEEE Robotics and Automation Letters , vol. 8, no. 1, pp. 176-183, Jan. 2023

work page 2023
[12]

Robust Onboard Localization in Changing Environments Exploiting Text Spotting,

N. Zimmerman, L. Wiesmann, T. Guadagnino, T. Läbe, J. Behley and C. Stachniss, "Robust Onboard Localization in Changing Environments Exploiting Text Spotting," 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, 2022, pp. 917-924

work page 2022
[13]

A Novel UWB/IMU/Odometer-Based Robot Localization System in LOS/NLOS Mixed Environments,

J. Sun, W. Sun, J. Zheng, Z. Chen, C. Tang and X. Zhang, "A Novel UWB/IMU/Odometer-Based Robot Localization System in LOS/NLOS Mixed Environments," in IEEE Transactions on Instrumentation and Measurement, vol. 73, pp. 1-13, 2024

work page 2024
[14]

Sensors applied to automated guided vehicle position control: A systematic literature review,

W. P. N. dos Reis and O. Morandin Junior, "Sensors applied to automated guided vehicle position control: A systematic literature review," The International Journal of Advanced Manufacturing Technology, vol. 113, no. 1, pp. 21–34, 2021

work page 2021
[15]

ERPoT: Effective and Reliable Pose Tracking for Mobile Robots Using Lightweight Polygon Maps,

H. Gao, Q. Qiu, H. Liu, D. Liang, C. Wang and X. Zhang, "ERPoT: Effective and Reliable Pose Tracking for Mobile Robots Using Lightweight Polygon Maps," in IEEE Transactions on Robotics, vol. 41, pp. 3799-3819, 2025

work page 2025
[16]

Monte Carlo localization for mobile robots,

F. Dellaert, D. Fox, W. Burgard and S. Thrun, "Monte Carlo localization for mobile robots," Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C) , Detroit, MI, USA, 1999, pp. 1322-1328 vol.2

work page 1999
[17]

Development of a new technique in ROS for mobile robots localization in known-based 2D environments,

I. Hatem and M. A. A. Khalil, “Development of a new technique in ROS for mobile robots localization in known-based 2D environments,” Tishreen Univ. J. Res. Sci. Stud. Eng. Sci. Ser., vol. 43, pp. 1–19, 2021

work page 2021
[18]

An Improved Localization of Mobile Robotic System Based on AMCL Algorithm,

M. -A. Chung and C. -W. Lin, "An Improved Localization of Mobile Robotic System Based on AMCL Algorithm," in IEEE Sensors Journal, vol. 22, no. 1, pp. 900-908, 1 Jan.1, 2022

work page 2022
[19]

IR-MCL: Implicit Representation-Based Online Global Localization,

H. Kuang, X. Chen, T. Guadagnino, N. Zimmerman, J. Behley and C. Stachniss, "IR-MCL: Implicit Representation-Based Online Global Localization," in IEEE Robotics and Automation Letters , vol. 8, no. 3, pp. 1627-1634, March 2023. 11

work page 2023
[20]

Research on Mobile Robot Localization Method Based on Adaptive Motion Model and Double-Threshold Relocation Strategy,

X. Xu et al., "Research on Mobile Robot Localization Method Based on Adaptive Motion Model and Double-Threshold Relocation Strategy," in IEEE Transactions on Instrumentation and Measurement, vol. 74, pp. 1- 12, 2025

work page 2025
[21]

Real-time loop closure in 2D LIDAR SLAM,

W. Hess, D. Kohler, H. Rapp and D. Andor, "Real-time loop closure in 2D LIDAR SLAM," 2016 IEEE International Conference on Robotics and Automation (ICRA), Stockholm, Sweden, 2016, pp. 1271-1278

work page 2016
[22]

Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration,

J. Yang, H. Li, D. Campbell and Y. Jia, "Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration," in IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 38, no. 11, pp. 2241- 2254, 1 Nov. 2016

work page 2016
[23]

rviz/UserGuide,

“rviz/UserGuide,” ROS Wiki, last modified Aug. 31, 2020. [Online]. Available: https://wiki.ros.org/rviz/UserGuide. Accessed: Sep. 30, 2025

work page 2020
[24]

Extracting Statistical Signatures of Geometry and Structure in 2D Occupancy Grid Maps for Global Localization,

S. -Y. An and J. Kim, "Extracting Statistical Signatures of Geometry and Structure in 2D Occupancy Grid Maps for Global Localization," in IEEE Robotics and Automation Letters , vol. 7, no. 2, pp. 4291-4298, April 2022

work page 2022
[25]

Active exploration for feature based global localization,

M. Seiz, P. Jensfelt and H. I. Christensen, "Active exploration for feature based global localization," Proceedings. 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2000) (Cat. No.00CH37113), Takamatsu, Japan, 2000, pp. 281-287 vol.1

work page 2000
[26]

Active global localization for a mobile robot using multiple hypothesis tracking,

P. Jensfelt and S. Kristensen, "Active global localization for a mobile robot using multiple hypothesis tracking," in IEEE Transactions on Robotics and Automation, vol. 17, no. 5, pp. 748-760, Oct. 2001

work page 2001
[27]

A Hybrid Active Global Localisation Algorithm for Mobile Robots,

A. Gasparri, S. Panzieri, F. Pascucci and G. Ulivi, "A Hybrid Active Global Localisation Algorithm for Mobile Robots," Proceedings 2007 IEEE International Conference on Robotics and Automation , Rome, Italy, 2007, pp. 3148-3153

work page 2007
[28]

Efficient active global localization for mobile robots operating in large and cooperative environments,

A. C. Murtra, J. M. Mirats Tur and A. Sanfeliu, "Efficient active global localization for mobile robots operating in large and cooperative environments," 2008 IEEE International Conference on Robotics and Automation, Pasadena, CA, USA, 2008, pp. 2758-2763

work page 2008
[29]

Active global localization based on localizability for mobile robots,

Y. Wang, W. Chen, J. Wang, and H. Wang, “Active global localization based on localizability for mobile robots,” Robotica, vol. 33, no. 8, pp. 1609–1627, 2015

work page 2015
[30]

Delight: An Efficient Descriptor for Global Localisation Using LiDAR Intensities,

K. P. Cop, P. V. K. Borges and R. Dubé, "Delight: An Efficient Descriptor for Global Localisation Using LiDAR Intensities," 2018 IEEE International Conference on Robotics and Automation (ICRA) , Brisbane, QLD, Australia, 2018, pp. 3653-3660

work page 2018
[31]

Efficient and Reliable LiDAR-Based Global Localization of Mobile Robots Using Multiscale/Resolution Maps,

J. Meng et al ., "Efficient and Reliable LiDAR-Based Global Localization of Mobile Robots Using Multiscale/Resolution Maps," in IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1- 15, 2021

work page 2021
[32]

Passive global localisation of mobile robot via 2D Fourier-Mellin invariant matching,

A. Filotheou, A. Tzitzis, E. Tsardoulias, A. Dimitriou, A. Symeonidis, G. Sergiadis, and L. Petrou, “Passive global localisation of mobile robot via 2D Fourier-Mellin invariant matching,” Journal of Intelligent & Robotic Systems, vol. 104, no. 2, p. 26, 2022

work page 2022
[33]

CBGL: Fast Monte Carlo Passive Global Localisation of 2D LIDAR Sensor,

A. Filotheou, "CBGL: Fast Monte Carlo Passive Global Localisation of 2D LIDAR Sensor," 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , Abu Dhabi, United Arab Emirates, 2024, pp. 3268-3275. Muhua Zhang (Graduate Student Member, IEEE) received the bachelor’s degree in mechanical design, manufacturing, and automation from So...

work page 2024