Pith. sign in

REVIEW 2 major objections 5 minor 43 references

LOTUSim: Multi-Domain Simulator for Marine Robotics

T0 review · 2 major / 5 minor · reviewed 2026-07-12 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read LOTUSim keeps multi-user maritime drone swarms interactive in real time and models wind-driven underwater currents more accurately than the usual stochastic baselines.

desk verdict Solid open multi-domain HITL maritime simulator with concrete real-time scaling numbers; the Ekman current improvement is real but only locally validated. read the letter →

arxiv 2607.03072 v1 pith:RW7U3YZC submitted 2026-07-03 cs.MA cs.RO

Cédric Buche (CROSSING , IMT Atlantique) , Juliette Grosset (CROSSING) , Hélène Lechêne , Marie Dubromel (CROSSING) , Pierig Havez-Bodivit , Malcom Neo , Julien Prodhon This is my paper · ORCID
classification cs.MAcs.RO
keywords maritimeroboticshuman-in-the-loopsimulationmulti-domainmulti-agentsystemsEkmancurrentsreal-timeswarmsim-to-realtransferunderwatervehicledynamics
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Maritime robotics needs simulators that let human operators rehearse missions with aerial, surface and underwater vehicles under realistic currents, yet most existing tools either ignore people in the loop or use oversimplified ocean physics. This paper presents LOTUSim, an open-source multi-domain platform that keeps strict real-time execution and high visual frame rates while hundreds of heterogeneous drones share one interactive world. Its second claim is a lightweight three-layer current model, drawn from classical Ekman theory, that tracks wind-driven flow that turns with depth and measurably outperforms the Gauss–Markov processes common in earlier simulators when checked against ocean reanalysis data. A sympathetic reader cares because operator training, multi-vehicle coordination and sim-to-real control transfer all become cheaper and safer once both interactivity and depth-dependent currents are available at scale.

What carries the argument

The Ekman-inspired three-layer current model (surface spiral driven by wind stress and Coriolis, geostrophic interior, bottom spiral) together with the distributed Gazebo–ROS2–Unity server-client loop that randomises update order and offloads physics and rendering.

What would settle it

Re-run the identical MAE/RMSE comparison on an independent open-ocean or high-latitude reanalysis set, or close the loop with a physical BlueROV under measured currents; if the reported 40–85 percent error reduction disappears or vehicle trajectories diverge, the central accuracy claim fails.

Watch

Extended reading notes

Core claim

LOTUSim simultaneously delivers human-in-the-loop real-time performance for large heterogeneous maritime fleets and a computationally cheap Ekman-layered underwater current model whose depth-dependent velocities reduce absolute and root-mean-square error by roughly 40–85 percent relative to standard Gauss–Markov currents when validated on coastal reanalysis observations.

Load-bearing premise

The three-layer steady-state Ekman formulation with fixed drag breakpoints and layer depths, tuned on five days near one coastal site, is assumed accurate enough for other ocean regimes and for closed-loop vehicle control without further local calibration.

Editorial extensions

If this is right

  • Operator training and multi-user naval mission rehearsal can scale to hundreds of mixed aerial-surface-underwater vehicles while staying at RTF = 1 and FPS > 140.
  • Controllers and path planners trained against the Ekman currents should transfer more faithfully to real vehicles than those trained on constant or purely stochastic currents.
  • Sim-to-real fault-tolerant control loops already demonstrated on a BlueROV2 can be stress-tested under depth-varying flow at accelerated real-time factors.
  • Distributed multi-agent scheduling with randomised update order becomes a practical template for other real-time multi-domain robot simulators.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the current model is analytic rather than CFD, the same layered formulation could be dropped into other open simulators with only modest engineering effort.
  • The multi-user Photon networking layer already present implies that geographically separated crews could rehearse joint naval scenarios without co-locating hardware.
  • If the Ekman layers prove robust, they offer a lightweight way to inject wind-driven bias into reinforcement-learning reward landscapes for underwater navigation.
  • The same architecture could later host acoustic or optical attenuation models that also vary with the computed current shear.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. LOTUSim is presented as an open-source, real-time multi-domain maritime simulator (aerial/surface/underwater) built on ROS2, Gazebo and Unity, with multi-user immersive HITL support via Photon networking and VR/Leap Motion/eye-tracking. The first claim is that the distributed architecture maintains strict real-time execution (RTF=1), high visual fidelity (FPS>140) and interactive responsiveness while scaling to large heterogeneous fleets (e.g., 750 LRAUVs under a 200 ms HITL threshold, or a mixed swarm of ~187 vehicles). The second claim is a three-layer steady-state Ekman-inspired underwater current model (surface spiral + geostrophic interior + bottom spiral, Eqs. 1–4) that is computationally cheap enough for real-time use and that reduces MAE/RMSE by roughly 40–85 % relative to a Gauss–Markov baseline when compared against Copernicus reanalysis data at a Brest coastal site (Table VIII). Supporting material includes sensor suites, open vehicle models, a prior BlueROV sim-to-real transfer, and RTF>1 scaling for AI training.

Significance. If the claims hold, LOTUSim fills a genuine gap: most maintained marine simulators (HoloOcean, Stonefish, MarineGym, OceanSim) remain autonomy- and sensor-centric and lack native multi-user immersive HITL. Concrete scaling numbers on a stated high-end laptop (Table V) and an open-source release are valuable for operator training and multi-domain mission rehearsal. The Ekman-layer current model is a useful, lightweight alternative to pure stochastic or pre-computed CFD fields; the open code and the independent Copernicus comparison strengthen reproducibility. The work is therefore of clear practical interest to the marine-robotics and naval-simulation communities even if the current-model generalisation remains local.

major comments (2)
  1. Section IV-B / Eqs. (1)–(4) and Table VIII: the headline accuracy claim (40–85 % MAE/RMSE reduction versus Gauss–Markov) rests entirely on 2 800 points from a single coastal box off Brest (lon [−6.25°, −6°], lat [46.6°, 47°]), five selected days, and depths 0.5–1000 m. The formulation assumes steady-state flow, fixed CD breakpoints (Eq. 1), and constant layer depths Ds, Db that are never numerically stated or fitted; the middle layer is pure geostrophy (Eq. 3) with no stratification or tidal residual. Consequently the reported relative-error ratios are local empirical results, not a demonstration that the same parameterisation remains superior (or even well-behaved) under different wind regimes, latitudes or bathymetry. Without at least one additional site, a sensitivity study on Ds/Db, or a closed-loop vehicle-control experiment, the claim of “sufficient physical fidelity for large-scal
  2. Table V and the associated thresholds (200 ms HITL responsiveness, 30 ms physics-loop limit) are reported only for a single high-end laptop configuration (i9-13980HX + RTX 4090). No variance, no multi-machine distributed measurements, and no ablation of the randomised update-order / network-latency scheduler are given. While the absolute numbers are useful, the paper’s claim of “robust o scalable multi-user interaction” would be stronger if the same metrics were shown under more modest hardware or under concurrent multi-client load; otherwise the operational envelope remains incompletely characterised.
minor comments (5)
  1. Ds and Db appear in Eqs. (2) and (4) but are never assigned numerical values or a fitting procedure; a short paragraph or table entry would make the model fully reproducible.
  2. Table VIII reports IMAE/IRMSE percentages without error bars or statistical tests; given N=700 per time slice, even simple standard errors would strengthen the comparison.
  3. The multi-agent scheduler claim (randomised update order + network latency yields non-determinism) is asserted without a quantitative fairness or bias metric; a short experiment would be welcome.
  4. Minor typographical issues: “operatorin-the-loop”, inconsistent spacing around citations, and a few missing accents in author names.
  5. Figure 5 (RTF vs. agents) would benefit from error bars or multiple runs; the single-curve presentation leaves the variability unknown.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: real-time metrics are direct hardware measurements and Ekman current errors are scored against independent Copernicus reanalysis, not against model-generated data.

full rationale

The two headline claims rest on external or direct evidence. Real-time interactive performance (Table V) reports measured FPS (>140), RTF(=1) and update-rate thresholds on a stated hardware configuration; these quantities are not derived from any fitted parameter of the simulator itself. The Ekman-layered current model (Eqs. 1–4) is a fixed, literature-parameterised formulation (Coriolis, CD breakpoints taken from Curcic & Haus, classical surface/bottom spirals) whose outputs are compared, via MAE/RMSE, to an independent ocean-reanalysis product (Copernicus, 2800 points off Brest). The comparison is therefore a genuine external validation, not a self-prediction. The sole self-reference is the brief sim-to-real citation [38] for a BlueROV fault-tolerant controller; that experiment is presented only as an additional capability demonstration and is not used to justify either the scaling numbers or the current-model error reductions. No uniqueness theorem, ansatz smuggling, or definitional identity appears in the derivation chain. Score 1 reflects only the presence of a non-load-bearing self-citation.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The central claims rest on classical fluid-dynamics approximations (Ekman theory, Airy waves, Fossen equations), standard software stacks, and a small set of free parameters taken from literature or chosen for the three-layer idealisation. No new physical entities are postulated; the free parameters are the usual modelling knobs of an engineering simulator.

free parameters (3)
  • surface drag coefficient CD breakpoints = 0.79 + 0.08 U10 (U10<20.5); 2.43e-3 otherwise
    Piecewise formula (Eq. 1) with numerical coefficients 0.79, 0.08, 2.43 and threshold 20.5 m/s taken from Curcic & Haus; treated as fixed inputs that directly set surface stress.
  • surface and bottom Ekman layer depths Ds, Db
    Layer thicknesses that control the exponential decay of the spiral; not fitted in the paper but chosen as free modelling parameters that determine how much of the water column feels wind or bottom friction.
  • HITL responsiveness threshold (200 ms) and physics-loop threshold (30 ms) = 200 ms / 30 ms
    Perceptual cut-offs used to declare ‘real-time interactive’ performance; chosen by the authors rather than derived from first principles.
assumptions (4)
  • domain assumption Steady-state Ekman balance (Coriolis + friction + wind stress) adequately describes the vertical structure of coastal currents for the purposes of vehicle simulation.
    Invoked throughout Section IV-B; the three-layer closed-form solution follows directly once this balance is assumed.
  • domain assumption Airy linear wave theory plus Fossen’s equations of motion are sufficient for real-time surface-ship hydrodynamics.
    Stated in Section IV-A as the foundation of LOTUSim-Xdyn.
  • ad hoc to paper Randomising asset update order plus network latency yields a sufficiently non-deterministic multi-agent scheduler.
    Section III-B; used to claim fairness of the distributed MAS without a formal fairness proof.
  • domain assumption Copernicus reanalysis fields constitute ground truth for current-model validation.
    Section IV-C methodology; all MAE/RMSE ratios are computed against these fields.
invented entities (1)
  • LOTUSim three-layer Ekman current model independent evidence
    purpose: Provide a computationally cheap, depth-dependent, wind-driven current field suitable for real-time multi-vehicle simulation.
    The layered closed-form expressions (Eqs. 2–4) are an engineering idealisation of classical Ekman theory; they are not a new physical entity but a new software component.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LOTUSim: Multi-Domain Simulator for Marine Robotics." pith.science (2026). https://pith.science/paper/RW7U3YZC

@misc{pith2026260703072,
  author       = {Pith},
  title        = {Pith review of: LOTUSim: Multi-Domain Simulator for Marine Robotics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RW7U3YZC}},
  note         = {Machine review of arXiv:2607.03072}
}
read the original abstract

Simulation is essential for maritime robotics, supporting operator training, mission rehearsal, and human-vehicle interaction in environments where real-world testing is costly or hazardous. Existing simulators focus primarily on autonomy systems and often lack human-in-the-loop interaction and realistic environmental physics. This paper introduces LOTUSim, an open-source, real-time maritime simulator supporting multi-user interaction across aerial, surface, and underwater robotic systems for coordinated naval-style operations. The first contribution of this work is enabling real-time interactive performance for users while ensuring scalability to large fleets operating within a shared interactive simulation environment. Validation demonstrates robust human-in-the-loop performance, maintaining strict real-time execution and high visual fidelity while scaling to large heterogeneous maritime drone swarms. The second contribution is a computationally efficient, Ekman-inspired layered, underwater current model that captures wind-driven, depth-dependent flow dynamics with sufficient physical fidelity for large-scale simulations. Validation against ocean reanalysis data demonstrates substantially improved accuracy compared to commonly used stochastic Gauss-Markov current models. These results confirm LOTUSim's suitability as a simulation platform for operatorin-the-loop maritime robotics research.

Figures

Figures reproduced from arXiv: 2607.03072 by the authors.

Figure 1
Figure 1. LOTUSim, an open multi-domain simulator The main contributions of LOTUSim are: • A framework enabling real-time interaction, demon￾strating robust interactive performance and scalability for large heterogeneous fleets while maintaining real￾time responsiveness. • A high-fidelity underwater current model for real￾time simulation, with an Ekman-inspired layered cur￾rent model, validated against ocean reanalysis data a… view at source ↗
Figure 2
Figure 2. System architecture of LOTUSim When using Gazebo as the asset orchestrator, each of the main client types is associated with a dedicated Gazebo plugin responsible for querying the corresponding client to update the asset state. During each simulation update cycle in Gazebo, where the timestep length is user-configurable, each asset may issue a query to its respective client module, depending on simulation requiremen… view at source ↗
Figure 3
Figure 3. Left: user experimenting with VR in LOTUSim. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: RTF vs. number of agents (30ms) VII. CONCLUSION REFERENCES [1] H. Chen, Y. Xu, Y. Ren, Y. Ye, X. Li, N. Ding, P. Cong, Z. Wang, B. Liu, Y. Chen, Z. Dou, X. Leng, M. Li, Y. Ma, and C. Tu, “Symbiosim: Human-in-the-loop simulation platform for bidirectional continuing lea…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 9 canonical work pages

  1. [1]

    Symbiosim: Human-in-the-loop simulation platform for bidirectional continuing learning in human–robot interaction,

    H. Chen, Y . Xu, Y . Ren, Y . Ye, X. Li, N. Ding, P. Cong, Z. Wang, B. Liu, Y . Chen, Z. Dou, X. Leng, M. Li, Y . Ma, and C. Tu, “Symbiosim: Human-in-the-loop simulation platform for bidirectional continuing learning in human–robot interaction,” arXiv preprint arXiv:2502.07358 , 2025. [Online]. Available: https: //arxiv.org/abs/2502.07358

    arXiv 2025

  2. [4]

    Marinegym: A high-performance reinforcement learning platform for underwater robotics,

    S. Chu, Z. Huang, Y . Li, M. Lin, D. Li, I. Carlucho, Y . R. Petillot, and C. Yang, “Marinegym: A high-performance reinforcement learning platform for underwater robotics,” in 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , 2025, pp. 17 146–17 153

    2025

  3. [5]

    Stonefish: Supporting machine learning research in marine robotics,

    M. Grimaldi, P. Cieslak, E. Ochoa, V . Bharti, H. Rajani, I. Carlucho, M. Koskinopoulou, Y . R. Petillot, and N. Gracias, “Stonefish: Supporting machine learning research in marine robotics,” 2025. [Online]. Available: https://arxiv.org/abs/2502.11887

    arXiv 2025

  4. [6]

    HoloOcean: An underwater robotics simulator,

    E. Potokar, S. Ashford, M. Kaess, and J. Mangelson, “HoloOcean: An underwater robotics simulator,” in Proc. IEEE Intl. Conf. on Robotics and Automation, ICRA , Philadelphia, PA, USA, May 2022

    2022

  5. [7]

    A preview of holoocean 2.0,

    B. Romrell, A. Austin, B. Meyers, R. Anderson, C. Noh, and J. G. Mangelson, “A preview of holoocean 2.0,” 2025. [Online]. Available: https://arxiv.org/abs/2510.06160

    arXiv 2025

  6. [8]

    Unmanned surface vehicle simulator with realistic environmental disturbances,

    M. Paravisi, D. H. Santos, V . Jorge, G. Heck, L. M. Gonc ¸alves, and A. Amory, “Unmanned surface vehicle simulator with realistic environmental disturbances,” Sensors, vol. 19, no. 5, p. 1068, Mar

  7. [9]

    Available: http://dx.doi.org/10.3390/s19051068

    [Online]. Available: http://dx.doi.org/10.3390/s19051068

    work page doi:10.3390/s19051068

  8. [11]

    Available: https://arxiv.org/abs/2311.10377

    [Online]. Available: https://arxiv.org/abs/2311.10377

    arXiv

Show all 43 references
  1. [12]

    Uuv simulator: A gazebo-based package for underwater intervention and multi-robot simulation,

    M. M. M. Manhaes, S. A. Scherer, M. V oss, L. R. Douat, and T. Rauschenbach, “Uuv simulator: A gazebo-based package for underwater intervention and multi-robot simulation,” in OCEANS 2016 MTS/IEEE Monterey . IEEE, Sep. 2016, p. 1–8. [Online]. Available: http://dx.doi.org/10.11...

    2016 doi

  2. [13]

    Dave aquatic virtual environment: Toward a general underwater robotics simulator,

    M. M. Zhang, W.-S. Choi, J. Herman, D. Davis, C. V ogt, M. McCarrin, Y . Vijay, D. Dutia, W. Lew, S. Peters, and B. Bingham, “Dave aquatic virtual environment: Toward a general underwater robotics simulator,”

  3. [14]

    Available: https://arxiv.org/abs/2209.02862

    [Online]. Available: https://arxiv.org/abs/2209.02862

    arXiv

  4. [15]

    Airsim: High-fidelity visual and physical simulation for autonomous vehicles,

    S. Shah, D. Dey, C. Lovett, and A. Kapoor, “Airsim: High-fidelity visual and physical simulation for autonomous vehicles,” 2017. [Online]. Available: https://arxiv.org/abs/1705.05065

    2017 arXiv

  5. [16]

    Design and use paradigms for gazebo, an open-source multi-robot simulator,

    N. Koenig and A. Howard, “Design and use paradigms for gazebo, an open-source multi-robot simulator,” in 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566), ser. IROS-04, vol. 3. IEEE, 2004, p. 2149–2154. [Online]. Availabl...

    2004 doi arXiv

  6. [17]

    A tensorial approach to computational continuum mechanics using object-oriented techniques,

    H. G. Weller, G. Tabor, H. Jasak, and C. Fureby, “A tensorial approach to computational continuum mechanics using object-oriented techniques,” Computers in Physics , vol. 12, no. 6, p. 620–631, Nov

  7. [18]

    Available: http://dx.doi.org/10.1063/1.168744

    [Online]. Available: http://dx.doi.org/10.1063/1.168744

    doi

  8. [19]

    Quadrotor flight simulation in a cfd-generated urban wind field,

    N. Kakavitsas, A. Willis, R. Jacobik, M. Uddin, and A. Wolek, “Quadrotor flight simulation in a cfd-generated urban wind field,” Mar. 2024. [Online]. Available: http://dx.doi.org/10.36227/techrxiv. 171085159.95135995/v1

    2024 doi

  9. [20]

    Oceansim: A gpu-accelerated underwater robot perception simulation framework,

    J. Song, H. Ma, O. Bagoren, A. V . Sethuraman, Y . Zhang, and K. A. Skinner, “Oceansim: A gpu-accelerated underwater robot perception simulation framework,” 2025. [Online]. Available: https://arxiv.org/abs/2503.01074

    2025 arXiv

  10. [21]

    Unav-sim: A visually realistic underwater robotics simulator and synthetic data-generation framework,

    A. Amer, O. ´Alvarez Tu ˜n´on, H. I. Ugurlu, J. le Fevre Sejersen, Y . Brodskiy, and E. Kayacan, “Unav-sim: A visually realistic underwater robotics simulator and synthetic data-generation framework,” 2023. [Online]. Available: https://arxiv.org/abs/2310. 11927

    2023

  11. [22]

    Open source simulator for unmanned underwater vehicles using ros and unity3d,

    P. Katara, M. Khanna, H. Nagar, and A. Panaiyappan, “Open source simulator for unmanned underwater vehicles using ros and unity3d,” in 2019 IEEE Underwater Technology (UT) . IEEE, Apr. 2019, p. 1–7. [Online]. Available: http://dx.doi.org/10.1109/UT.2019.8734309

    2019 doi

  12. [23]

    Ferber and G

    J. Ferber and G. Weiss, Multi-agent systems: an introduction to distributed artificial intelligence . Addison-wesley Reading, 1999, vol. 1

    1999

  13. [24]

    Ros2 and ignition: Towards scalable simulation of distributed robotic systems,

    J. Ko et al. , “Ros2 and ignition: Towards scalable simulation of distributed robotic systems,” in IEEE/SICE International Symposium on System Integration (SII) , 2020, pp. 378–383. [Online]. Available: https://ieeexplore.ieee.org/document/9075123

    2020

  14. [25]

    Effect of frame rate on user experience, performance and simulator sickness in virtual reality,

    S. Weech, S. Kenny, and M. Barnett-Cowan, “Effect of frame rate on user experience, performance and simulator sickness in virtual reality,” Human Factors, vol. 65, no. 4, p. 678–692, 2023

    2023

  15. [26]

    Perceptual thresholds for display lag in a real visual environment,

    A. e. a. Moss, “Perceptual thresholds for display lag in a real visual environment,” Human Factors, 2010

    2010

  16. [27]

    Embracing the lag: Real- time challenges in multi-agent systems,

    S. Dumnich, W. Birmingham, and B. Wolfe, “Embracing the lag: Real- time challenges in multi-agent systems,” in Proceedings of the AAAI Fall Symposium Series (FSS-23) , 2023

    2023

  17. [28]

    Ekman drift currents in the arctic ocean,

    K. Hunkins, “Ekman drift currents in the arctic ocean,” Deep Sea Research and Oceanographic Abstracts , vol. 13, no. 4, p. 607–620, Aug. 1966. [Online]. Available: http://dx.doi.org/10.1016/ 0011-7471(66)90592-4

    1966

  18. [29]

    POND and G

    S. POND and G. L. PICKARD, Currents with Friction; Wind- driven Circulation. Elsevier, 1983, p. 100–162. [Online]. Available: http://dx.doi.org/10.1016/B978-0-08-057054-9.50015-8

    1983 doi

  19. [30]

    T. I. Fossen, Handbook of Marine Craft Hydrodynamics and Motion Control . Wiley, Apr. 2011. [Online]. Available: http: //dx.doi.org/10.1002/9781119994138

    2011 doi

  20. [31]

    Cn-aeromodels: A c++ implementation of aerodynamic models for wind propulsion systems of cargo ships,

    A. Babarit and M. Charlou, “Cn-aeromodels: A c++ implementation of aerodynamic models for wind propulsion systems of cargo ships,” Journal of Open Source Software , vol. 9, no. 102, p. 6940, Oct

  21. [32]

    Available: http://dx.doi.org/10.21105/joss.06940

    [Online]. Available: http://dx.doi.org/10.21105/joss.06940

    doi

  22. [33]

    Goda, Random Seas and Design of Maritime Structures

    Y . Goda, Random Seas and Design of Maritime Structures . WORLD SCIENTIFIC, Jun. 2010. [Online]. Available: http: //dx.doi.org/10.1142/7425

    2010 doi

  23. [34]

    An empirical model for random directional wave kinematics near the free surface,

    G. Rodenbusch and G. Forristall, “An empirical model for random directional wave kinematics near the free surface,” in Offshore Technology Conference , ser. 86OTC. OTC, May 1986. [Online]. Available: http://dx.doi.org/10.4043/5097-MS

    1986 doi

  24. [35]

    Stewart, Introduction to Physical Oceanography

    R. Stewart, Introduction to Physical Oceanography . University Press of Florida, 2009. [Online]. Available: https://open.umn.edu/ opentextbooks/textbooks/20

    2009

  25. [36]

    Physics across oceanography: Fluid mechanics and waves,

    S. Hautala, “Physics across oceanography: Fluid mechanics and waves,” University of Washington, 2020

    2020

  26. [37]

    Ekman-type solutions for shallow-water flows on a rotating sphere: A new perspective on a classical problem,

    A. Constantin and R. S. Johnson, “Ekman-type solutions for shallow-water flows on a rotating sphere: A new perspective on a classical problem,” Physics of Fluids , vol. 31, no. 2, Feb. 2019. [Online]. Available: http://dx.doi.org/10.1063/1.5083088

    2019 doi

  27. [38]

    Cushman-Roisin and J.-M

    B. Cushman-Roisin and J.-M. Beckers, Introduction to geophysical fluid dynamics: physical and numerical aspects , ser. International geophysics. Elsevier, 2011. [Online]. Available: https://www. elsevier.com/books/introduction-to-geophysical-fluid-dynamics/ cushman-roisin/978-...

    2011

  28. [39]

    Revised estimates of ocean surface drag in strong winds,

    M. Curcic and B. K. Haus, “Revised estimates of ocean surface drag in strong winds,” Geophysical Research Letters, vol. 47, no. 10, May

  29. [40]

    Available: http://dx.doi.org/10.1029/2020GL087647

    [Online]. Available: http://dx.doi.org/10.1029/2020GL087647

    doi

  30. [41]

    Predicting the ocean currents using deep learning,

    C. Bayindir, “Predicting the ocean currents using deep learning,”

  31. [42]

    Available: https://arxiv.org/abs/1906.08066

    [Online]. Available: https://arxiv.org/abs/1906.08066

    1906 arXiv

  32. [43]

    Estimating ocean surface currents with machine learning,

    A. Sinha and R. Abernathey, “Estimating ocean surface currents with machine learning,” Frontiers in Marine Science , vol. 8, Jun. 2021. [Online]. Available: http://dx.doi.org/10.3389/fmars.2021.672477

    2021 doi

  33. [44]

    Prediction of dominant ocean parameters for sustainable marine environment,

    D. Menaka and S. Gauni, “Prediction of dominant ocean parameters for sustainable marine environment,” IEEE Access , vol. 9, p. 146578–146591, 2021. [Online]. Available: http://dx.doi.org/10.1109/ ACCESS.2021.3122237

    2021

  34. [45]

    Asvsim (airsim for surface vehicles): A high-fidelity simulation framework for autonomous surface vehicle research,

    B. Lesy, S. Herremans, R. Kerstens, J. Steckel, W. Daems, S. Mercelis, and A. Anwar, “Asvsim (airsim for surface vehicles): A high-fidelity simulation framework for autonomous surface vehicle research,” 2025. [Online]. Available: https://arxiv.org/abs/2506.22174

    2025 arXiv

  35. [46]

    Control reallocation using deep reinforcement learning for actuator fault recovery of an autonomous underwater vehicle,

    K. Lagattu, E. Artusi, P. E. Santos, K. Sammut, G. Le Chenadec, and B. Clement, “Control reallocation using deep reinforcement learning for actuator fault recovery of an autonomous underwater vehicle,” in 2025 IEEE ICRA , 2025, pp. 2291–2297

    2025

Pith tools

Reviewed July 12, 2026 · model on record in the stance chip above.