arxiv: 2604.11861 · v1 · submitted 2026-04-13 · 💻 cs.RO · cs.MA

Recognition: unknown

BIND-USBL: Bounding IMU Navigation Drift using USBL in Heterogeneous ASV-AUV Teams

Pranav Kedia , Rajini Makam , Heiko Hamann , Suresh Sundaram

Authors on Pith no claims yet

Pith reviewed 2026-05-10 15:51 UTC · model grok-4.3

classification 💻 cs.RO cs.MA

keywords cooperative localizationAUV navigationUSBL positioningASV-AUV teamsdrift boundingacoustic schedulingdead-reckoning correctionheterogeneous marine robots

0 comments

The pith

A fleet of surface vessels can bound underwater vehicle drift by supplying scheduled acoustic fixes that overcome sparsity in coverage rather than relying on fix accuracy alone.

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

The paper introduces BIND-USBL to keep AUVs from accumulating unbounded error during long GPS-denied missions. Instead of treating each USBL measurement in isolation, the approach models how ASV formations create acoustic coverage over a survey area and then schedules shared-channel fixes so that no AUV waits too long for a position update. Evaluation in simulation shows that error stays controlled when scale, team size, and geometry are matched to the mission, and that a graph-based scheduler raises the rate of delivered fixes without causing collisions or high latency. A sympathetic reader sees this as a way to let underwater vehicles complete extended lawnmower or coverage tasks while staying localized through coordinated surface support.

Core claim

Long-duration navigation failure in AUVs stems primarily from the temporal sparsity and geometric gaps in external position fixes rather than from the precision of any single measurement. BIND-USBL counters this by coupling a multi-ASV formation model that ties survey scale and anchor placement to acoustic availability, a conflict-graph TDMA scheduler that reuses the channel for multiple vehicles, and delayed fusion of the resulting fixes with drift-prone inertial dead-reckoning. In simulated heterogeneous teams running coverage missions, the resulting localization performance depends on the interplay of survey extent, acoustic footprint, team composition, and formation shape, while the re-s

What carries the argument

The conflict-graph-based TDMA uplink scheduler combined with a multi-ASV formation model that maps survey scale and anchor placement onto acoustic coverage and fix latency.

If this is right

Dead-reckoning drift remains bounded for the full mission duration provided the ASV formation maintains continuous acoustic coverage over the operating area.
The spatial-reuse scheduler increases the number of USBL fixes delivered to each AUV per unit time without violating the no-collision constraint on the shared acoustic channel.
End-to-end latency from measurement to fusion stays low enough that delayed updates still correct inertial error before it exceeds mission tolerances.
Heterogeneous team composition and ASV geometry directly determine the fraction of the survey area that receives timely fixes.

Where Pith is reading between the lines

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

The same scheduling logic could be adapted to other acoustic or optical positioning modalities that share a medium with collision constraints.
Dynamic re-formation of the ASV fleet during a mission might further reduce gaps when AUV paths deviate from the planned lawnmower grid.
Integration with surface-wave or current models could tighten the mapping from formation geometry to expected fix availability.
The framework suggests a design pattern for any team of mobile anchors that must service multiple drifting agents over a large workspace.

Load-bearing premise

The HoloOcean simulator reproduces real acoustic propagation, vehicle motion, and multi-vehicle interference closely enough that measured performance gains will appear in actual ocean deployments.

What would settle it

Field trials in which AUV position error grows faster than predicted once acoustic interference or surface-wave effects exceed simulator levels, even when the same ASV formations and scheduler are used.

Figures

Figures reproduced from arXiv: 2604.11861 by Heiko Hamann, Pranav Kedia, Rajini Makam, Suresh Sundaram.

**Figure 1.** Figure 1: shows the spatial layout of the mission. The IMUonly trajectory exhibits the characteristic piecewise quadratic drift predicted by eq. (23); the fused IMU+USBL trajectory tracks the planned path closely, demonstrating that a steady supply of USBL fixes is sufficient to bound inertial drift. Across five independent runs, the TDMA scheduler delivers an aggregate applied fix rate of 0.994 Hz (in average 298.… view at source ↗

**Figure 2.** Figure 2: Trajectory overview: L = 140 m, 3 AUVs, 1 ASV (star, α0 = 0◦). The dashed blue circle marks the HF uplink range, and the dotted orange circle marks the MF downlink range. The upper lawnmower strip lies almost entirely outside the HF footprint. TABLE I PER-AUV METRICS FOR THE COVERAGE-FAILURE AND RECOVERY CASES (L = 140 M, 3 AUVS, SINGLE REPRESENTATIVE RUN). Config AUV Fixes Cov. (%) CTE (m) Dist. (m) 1 ASV… view at source ↗

**Figure 3.** Figure 3: Fleet metrics for the single-ASV failure case ( [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

**Figure 4.** Figure 4: Per-AUV CTE, fix count, and HF coverage for [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗

**Figure 5.** Figure 5: Mean CTE (m) across all formation angles and [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗

**Figure 6.** Figure 6: Distance-normalised configuration sweep ( [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗

read the original abstract

Accurate and continuous localization of Autonomous Underwater Vehicles (AUVs) in GPS-denied environments is a persistent challenge in marine robotics. In the absence of external position fixes, AUVs rely on inertial dead-reckoning, which accumulates unbounded drift due to sensor bias and noise. This paper presents BIND-USBL, a cooperative localization framework in which a fleet of Autonomous Surface Vessels (ASVs) equipped with Ultra-Short Baseline (USBL) acoustic positioning systems provides intermittent fixes to bound AUV dead-reckoning error. The key insight is that long-duration navigation failure is driven not by the accuracy of individual USBL measurements, but by the temporal sparsity and geometric availability of those fixes. BIND-USBL combines a multi-ASV formation model linking survey scale and anchor placement to acoustic coverage, a conflict-graph-based TDMA uplink scheduler for shared-channel servicing, and delayed fusion of received USBL updates with drift-prone dead reckoning. The framework is evaluated in the HoloOcean simulator using heterogeneous ASV-AUV teams executing lawnmower coverage missions. The results show that localization performance is shaped by the interaction of survey scale, acoustic coverage, team composition, and ASV-formation geometry. Further, the spatial-reuse scheduler improves per-AUV fix delivery rate without violating the no-collision constraint, while maintaining low end-to-end fix latency.

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

BIND-USBL shows how ASV formations plus a conflict-graph TDMA scheduler can raise USBL fix rates for drifting AUVs, but the entire case rests on HoloOcean runs with no field checks.

read the letter

The paper's central point is that AUV dead-reckoning drift can be bounded more effectively when ASVs supply USBL fixes on a shared acoustic channel, and that the rate of those fixes depends on formation geometry and how the channel is scheduled. The authors link survey scale and anchor placement to coverage, then add a scheduler that reuses space without collisions while keeping latency low. They fuse the delayed fixes back into the AUV's dead-reckoning filter and test the whole stack on lawnmower missions in simulation. That integration is the actual new piece: prior work had USBL, formation control, and TDMA separately, but not this specific combination for heterogeneous teams with explicit coverage modeling and delayed fusion. The scheduler results are the clearest win; they show higher per-AUV fix delivery without breaking the no-collision rule. The framework description itself is coherent and avoids obvious circularity. The main limitation is that every performance trend comes from HoloOcean. There are no real acoustic recordings, no measured multipath or interference data, and no vehicle dynamics validation against tank or sea trials. Without those, it is hard to know whether the reported interactions between scale, coverage, and team size survive model mismatch. The abstract also gives no concrete position-error numbers or direct baselines, so the size of the improvement stays qualitative. This is useful reading for marine-robotics groups that already run ASV-AUV operations and need ideas for acoustic scheduling. A practitioner could take the scheduler logic and test it on their own hardware. It is worth sending to peer review because the synthesis is thoughtful and the problem is practical; referees will mainly press on adding validation or at least sensitivity checks against the simulator assumptions.

Referee Report

2 major / 2 minor

Summary. The paper presents BIND-USBL, a cooperative localization framework for AUVs in GPS-denied environments. A fleet of ASVs equipped with USBL systems provides intermittent position fixes to bound IMU dead-reckoning drift. The framework incorporates a multi-ASV formation model relating survey scale and anchor placement to acoustic coverage, a conflict-graph TDMA scheduler for shared-channel uplink servicing, and delayed fusion of USBL updates with drift-prone dead reckoning. It is evaluated exclusively in the HoloOcean simulator on heterogeneous ASV-AUV teams performing lawnmower coverage missions. Results indicate that localization performance depends on interactions among survey scale, acoustic coverage, team composition, and ASV-formation geometry; the spatial-reuse scheduler increases per-AUV fix delivery rate without violating no-collision constraints while keeping end-to-end latency low.

Significance. If the simulation results transfer, the work offers a practical synthesis for designing heterogeneous marine teams to extend AUV mission duration through bounded navigation error. The explicit modeling of formation geometry, scheduler conflict graphs, and delayed fusion provides reusable engineering insights for acoustic multi-vehicle coordination. However, the simulation-only evaluation limits the strength of claims about real-world performance shaping factors.

major comments (2)

[Evaluation / Results] Evaluation (results discussion and abstract): The central claims that localization performance is shaped by survey scale, acoustic coverage, team composition, and ASV-formation geometry, and that the spatial-reuse scheduler improves fix delivery rate, are supported only by qualitative trends. No quantitative error metrics (e.g., position RMSE, drift rates), baseline comparisons (single-ASV USBL, unscheduled TDMA, or pure dead-reckoning), or statistical significance tests are reported, making it impossible to assess the magnitude or reliability of the reported interactions.
[Methods / Evaluation] Simulator validation (methods and evaluation): The framework's conclusions rest on HoloOcean reproducing acoustic propagation, multi-vehicle interference, and vehicle hydrodynamics. No sensitivity analysis to model mismatch, no comparison of simulated vs. measured multipath or USBL error statistics, and no real-world experiments are provided. This directly undermines transfer of the claimed performance-shaping interactions and scheduler benefits to field deployments.

minor comments (2)

[Framework description] The description of the delayed-fusion implementation lacks explicit equations or pseudocode for how USBL measurements are time-aligned with the IMU propagation step.
[Scheduler section] Notation for the conflict-graph scheduler (e.g., edge weights, reuse factor) should be defined once and used consistently across text and figures.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive and detailed comments. We address each major comment below, indicating where we will revise the manuscript to strengthen the presentation of results and methods.

read point-by-point responses

Referee: [Evaluation / Results] Evaluation (results discussion and abstract): The central claims that localization performance is shaped by survey scale, acoustic coverage, team composition, and ASV-formation geometry, and that the spatial-reuse scheduler improves fix delivery rate, are supported only by qualitative trends. No quantitative error metrics (e.g., position RMSE, drift rates), baseline comparisons (single-ASV USBL, unscheduled TDMA, or pure dead-reckoning), or statistical significance tests are reported, making it impossible to assess the magnitude or reliability of the reported interactions.

Authors: We agree that the current results rely on qualitative trends. In the revised manuscript we will add quantitative metrics including position RMSE and drift rates across parameter sweeps. Baseline comparisons against single-ASV USBL, unscheduled TDMA, and pure dead-reckoning will be included, together with statistical significance tests to quantify the magnitude and reliability of the reported interactions and scheduler benefits. revision: yes
Referee: [Methods / Evaluation] Simulator validation (methods and evaluation): The framework's conclusions rest on HoloOcean reproducing acoustic propagation, multi-vehicle interference, and vehicle hydrodynamics. No sensitivity analysis to model mismatch, no comparison of simulated vs. measured multipath or USBL error statistics, and no real-world experiments are provided. This directly undermines transfer of the claimed performance-shaping interactions and scheduler benefits to field deployments.

Authors: HoloOcean was selected because it has been used and partially validated in prior marine-robotics literature for acoustic propagation and hydrodynamics. We will add a sensitivity analysis to key acoustic and hydrodynamic parameters and compare simulated USBL error statistics against published field measurements. We acknowledge that the absence of new real-world experiments limits direct transfer claims; this will be stated explicitly as a limitation with real-world validation identified as future work. revision: partial

standing simulated objections not resolved

Real-world experimental validation, as the present study is simulation-only and new field trials cannot be conducted within the revision timeline.

Circularity Check

0 steps flagged

No circularity in framework derivation or simulation evaluation

full rationale

The paper defines BIND-USBL via three independent components (multi-ASV formation model linking scale to coverage, conflict-graph TDMA scheduler, delayed USBL/dead-reckoning fusion) and reports outcomes from HoloOcean lawnmower simulations. No load-bearing step reduces by construction to a fitted input, self-definition, or self-citation chain; performance claims about scale/coverage/geometry interactions and scheduler benefits are direct simulation results, not tautological renamings or predictions forced by the inputs. This is a self-contained engineering synthesis evaluated against an external simulator benchmark.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The abstract relies on standard robotics assumptions about IMU bias and noise models plus acoustic ranging; no free parameters, invented entities, or ad-hoc axioms are explicitly introduced in the provided text.

axioms (2)

domain assumption IMU dead-reckoning accumulates unbounded drift due to sensor bias and noise
Stated in the opening sentence of the abstract as the core problem to be solved.
domain assumption USBL provides intermittent but usable position fixes when geometrically available
Implicit in the description of how ASVs bound AUV error.

pith-pipeline@v0.9.0 · 5553 in / 1399 out tokens · 27830 ms · 2026-05-10T15:51:48.386098+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

32 extracted references · 1 canonical work pages

[1]

A survey of underwater multi-robot systems,

Z. Zhou, J. Liu, and J. Yu, “A survey of underwater multi-robot systems,” IEEE/CAA Journal of Automatica Sinica, vol. 9, no. 1, pp. 1–19, 2022

2022
[2]

A comprehensive study on modelling and control of autonomous underwater vehicle,

R. Makam, P. Mane, S. Sundaram, and P. Sujit, “A comprehensive study on modelling and control of autonomous underwater vehicle,” in Assistive Robotics. CRC Press, 2023, pp. 264–296

2023
[3]

Auv navigation and localization: A review,

L. Paull, S. Saeedi, M. Seto, and H. Li, “Auv navigation and localization: A review,”IEEE Journal of Oceanic Engineering, vol. 39, no. 1, pp. 131–149, 2014

2014
[4]

Cooperative asv/auv system exploiting active acoustic localization,

M. Bresciani, G. Peralta, F. Ruscio, L. Bazzarello, A. Caiti, and R. Costanzi, “Cooperative asv/auv system exploiting active acoustic localization,” inProceedings of the IEEE/MTS OCEANS Conference, Sep. 2021, pp. 1–7

2021
[5]

One-way travel- time inverted ultra-short baseline localization for low-cost autonomous underwater vehicles,

N. R. Rypkema, E. M. Fischell, and H. Schmidt, “One-way travel- time inverted ultra-short baseline localization for low-cost autonomous underwater vehicles,” inIEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 4920–4926

2017
[6]

Passive inverted ultra-short baseline (piUSBL) localization: An experimental evaluation of accuracy,

N. R. Rypkema and H. Schmidt, “Passive inverted ultra-short baseline (piUSBL) localization: An experimental evaluation of accuracy,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019, pp. 7197–7204

2019
[7]

Synchronous-clock range-angle relative acoustic navigation: A unified approach to multi- AUV localization, command, control, and coordination,

N. R. Rypkema, H. Schmidt, and E. M. Fischell, “Synchronous-clock range-angle relative acoustic navigation: A unified approach to multi- AUV localization, command, control, and coordination,”Field Robotics, vol. 2, pp. 774–806, 2022

2022
[8]

Factor-graph-based passive acoustic navigation for decentralized cooperative localization using bearing elevation depth difference,

K. Velasco, T. W. McLain, and J. G. Mangelson, “Factor-graph-based passive acoustic navigation for decentralized cooperative localization using bearing elevation depth difference,” inIEEE ICRA Workshop on Field Robotics, 2025

2025
[9]

Raspi2USBL: An open-source Raspberry Pi-based passive inverted ultra-short baseline positioning system for underwater robotics,

J. Huang, Y . Wang, and Y . Chen, “Raspi2USBL: An open-source Raspberry Pi-based passive inverted ultra-short baseline positioning system for underwater robotics,”arXiv preprint arXiv:2511.06998, 2025

work page arXiv 2025
[10]

Developing SailSwarm: Small uncrewed sailing vessels for maritime environments,

P. Kedia, C. Apolinsky, and H. Hamann, “Developing SailSwarm: Small uncrewed sailing vessels for maritime environments,” Oct. 2024, extended Abstract; presented as a poster at IROS 2024 (MHURS workshop)

2024
[11]

Development of autonomous sailboat sails and future perspectives: A review,

Z. Sun, A. Feng, J. Yu, W. Zhao, and Y . Huang, “Development of autonomous sailboat sails and future perspectives: A review,” Renewable and Sustainable Energy Reviews, vol. 207, p. 114918, 2025. [Online]. Available: https://www.sciencedirect.com/science/article/pii/ S1364032124006440

2025
[12]

Autonomous sailboat design: A review from the performance perspective,

Y . An, J. Yu, and J. Zhang, “Autonomous sailboat design: A review from the performance perspective,”Ocean Engineering, vol. 238, p. 109753, 2021

2021
[13]

Hybrid tracking controller for an asv providing mission support for an auv,

T. R. Fyrvik, J. E. Bremnes, and A. J. Sørensen, “Hybrid tracking controller for an asv providing mission support for an auv,”IFAC PapersOnLine, vol. 55, no. 31, pp. 91–97, 2022

2022
[14]

Design of a switching controller for tracking auvs with an asv,

J. E. Bremnes, T. R. Fyrvik, T. R. Krogstad, and A. J. Sørensen, “Design of a switching controller for tracking auvs with an asv,”IEEE Transactions on Control Systems Technology, vol. 32, no. 5, pp. 1785– 1800, Sep. 2024

2024
[15]

Heteroge- neous cross domain coordinated control of asv-auv system for maritime search and rescue,

H. Wang, S. Tong, A. Wang, W. Zhang, Z. Hu, and Z. Peng, “Heteroge- neous cross domain coordinated control of asv-auv system for maritime search and rescue,”Ocean Engineering, vol. 306, p. 117950, 2024

2024
[16]

Predefined-time terminal sliding mode cooperative trajectory tracking control for usv-auv under weak communication,

R. Hu, D. Wu, Z. You, D. Wu, and W. Tu, “Predefined-time terminal sliding mode cooperative trajectory tracking control for usv-auv under weak communication,”Ocean Engineering, vol. 333, p. 121459, 2025

2025
[17]

On field experience on underwater acoustic localization through usbl modems,

R. Costanzi, N. Monnini, A. Ridolfi, B. Allotta, and A. Caiti, “On field experience on underwater acoustic localization through usbl modems,” inOCEANS 2017 - Aberdeen, 2017, pp. 1–5

2017
[18]

Experimental analysis of deep-sea auv based on multi-sensor integrated navigation and position- ing,

Y . Liu, Y . Sun, B. Li, X. Wang, and L. Yang, “Experimental analysis of deep-sea auv based on multi-sensor integrated navigation and position- ing,”Remote Sensing, vol. 16, no. 1, p. 199, 2024

2024
[19]

Design and experimental results of passive iusbl for small auv navigation,

Y . Wang, S. H. Huang, Z. Wang, R. Hu, M. Feng, P. Du, W. Yang, and Y . Chen, “Design and experimental results of passive iusbl for small auv navigation,”Ocean Engineering, vol. 248, p. 110793, 2022

2022
[20]

An in-situ sound speed profile correction scheme for the tight-coupling integration of sins/usbl in deep-sea arv navigation,

H. Liu, S. Zhao, Z. Wang, J. Zhou, K. Du, and R. Shan, “An in-situ sound speed profile correction scheme for the tight-coupling integration of sins/usbl in deep-sea arv navigation,”Satellite Navigation, vol. 6, no. 1, p. 31, 2025

2025
[21]

Localization uncertainty estimation for autonomous underwater vehicle navigation,

J. Park and S. Cho, “Localization uncertainty estimation for autonomous underwater vehicle navigation,”Journal of Marine Science and Engi- neering, vol. 11, no. 8, p. 1540, Aug. 2023

2023
[22]

Robust cooperative navigation for auvs using the student’s t distribution,

Q. Li, S. M. Naqvi, J. Neasham, and J. Chambers, “Robust cooperative navigation for auvs using the student’s t distribution,” inProceedings of the IEEE International Conference, 2022

2022
[23]

Optimal geometric configuration of sensors for received signal strength based cooperative localization of submerged auvs,

X. Bo, A. A. Razzaqi, X. Wang, and G. Farid, “Optimal geometric configuration of sensors for received signal strength based cooperative localization of submerged auvs,”Ocean Engineering, vol. 214, p. 107785, 2020

2020
[24]

Investigation of the usv-auv cooperative environment via reinforcement learning and its impact on data collection and energy efficiency,

M. M. T ¨urko˘glu and E. Akyuz, “Investigation of the usv-auv cooperative environment via reinforcement learning and its impact on data collection and energy efficiency,”Ocean Engineering, vol. 348, p. 124004, 2026

2026
[25]

Cooperative formation control of autonomous underwater vehicles: An overview,

B. Das, B. Subudhi, and B. B. Pati, “Cooperative formation control of autonomous underwater vehicles: An overview,”International Journal of Automation and Computing, vol. 13, no. 3, pp. 199–225, Jun. 2016

2016
[26]

Development of formation control system for multiple auvs with sonar interference avoidance function,

A. Okamoto, M. Sasano, K. Kim, and T. Fujiwara, “Development of formation control system for multiple auvs with sonar interference avoidance function,”Journal of Robotics and Mechatronics, vol. 36, no. 2, 2024

2024
[27]

Cooperative coverage control for heterogeneous auvs based on control barrier functions and consensus theory,

F. Mao, D. Zhang, L. Xu, and R. Wang, “Cooperative coverage control for heterogeneous auvs based on control barrier functions and consensus theory,”Sensors, vol. 26, no. 3, p. 822, 2026

2026
[28]

Cooperative target state estimation of multiple auvs based on an enhanced imm-ukf approach,

J. Hu, L. Guo, G. Chen, Y . Chen, and J. Gao, “Cooperative target state estimation of multiple auvs based on an enhanced imm-ukf approach,” IFAC PapersOnLine, vol. 59, no. 22, pp. 770–775, 2025

2025
[29]

HoloOcean: An underwater robotics simulator,

E. Potokar, S. Ashford, M. Kaess, and J. G. Mangelson, “HoloOcean: An underwater robotics simulator,” in2022 International Conference on Robotics and Automation (ICRA), 2022

2022
[30]

T. I. Fossen,Handbook of marine craft hydrodynamics and motion control. John wiley & sons, 2011

2011
[31]

Navigation of a miniaturized autonomous underwater ve- hicle exploring waters under ice,

M. Nitsch, “Navigation of a miniaturized autonomous underwater ve- hicle exploring waters under ice,” Ph.D. dissertation, RWTH Aachen, Germany, 06 2024

2024
[32]

Diestel,Graph Theory, 5th ed

R. Diestel,Graph Theory, 5th ed. Springer, 2017

2017