pith. sign in

arxiv: 2607.01739 · v1 · pith:TINUKT76new · submitted 2026-07-02 · 📡 eess.SY · cs.SY

Development and Identification of a Linear Low-Speed Ship Maneuvering Model from Full-Scale Data

Agnes N. Mwange , Taichi Kambara , Kouki Wakita , Kazuyoshi Hosogaya , Atsuo Maki This is my paper

Pith reviewed 2026-07-03 08:02 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords ship maneuveringsystem identificationlinear state-space modellow-speed dynamicsCMA-ESfull-scale dataautonomous berthing
0
0 comments X

The pith

A linear time-invariant continuous-time state-space model can represent low-speed ship maneuvering when its parameters are identified from full-scale data.

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

The paper shows that ship dynamics at low speeds, such as during berthing, can be captured by a single linear state-space model that does not change with time. Parameters of this model are found by applying the CMA-ES optimizer directly to recorded trajectories from real ship maneuvers. When the resulting model is simulated, its predicted paths closely track the measured data in validation tests. This outcome indicates that the usual need for elaborate nonlinear models may be relaxed for practical identification and control at low speeds.

Core claim

Low-speed ship maneuvering motion can be modeled as a time-invariant continuous-time linear state-space system whose parameters are estimated from full-scale maneuvering data using the CMA-ES algorithm, producing outputs that agree closely with the observed trajectories.

What carries the argument

Time-invariant continuous-time linear state-space model whose coefficients are fitted to full-scale data by CMA-ES optimization.

If this is right

  • Controller design for automated berthing can proceed with standard linear control techniques rather than nonlinear methods.
  • Model parameters can be obtained directly from operational data without separate hydrodynamic coefficient calculations.
  • Simplified linear models may suffice for real-time prediction in low-speed regimes where nonlinear effects were previously assumed dominant.
  • The identification procedure can be repeated on new data sets to update the model as ship conditions change.

Where Pith is reading between the lines

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

  • The same linear identification pipeline could be tested on other vessel types to check transferability without re-deriving hull-specific equations.
  • If the linear approximation holds, hybrid controllers might switch between this low-speed model and existing high-speed models at a single speed threshold.
  • Embedding the model in a Kalman filter would allow online state estimation from noisy sensor readings during actual berthing operations.

Load-bearing premise

The complex nonlinear dynamics of low-speed ship maneuvering can nevertheless be captured well enough by one fixed linear state-space model.

What would settle it

Run the identified model on a fresh set of full-scale maneuvering trials from the same vessel and measure whether the predicted position and heading errors remain within the same bounds as the original validation set.

Figures

Figures reproduced from arXiv: 2607.01739 by Agnes N. Mwange, Atsuo Maki, Kazuyoshi Hosogaya, Kouki Wakita, Taichi Kambara.

Figure 1
Figure 1. Figure 1: Subject ship used in the study [33] [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Coordinate systems. ship-fixed coordinate system, O0, is set at the ship’s center of gravity. The origin of the inertial coordinate system, O, coincides with O0 when the ship is at the berth. The relationship between the two coordinate systems is governed by:   x˙0 y˙0 ψ˙   =   cosψ −sinψ 0 sinψ cosψ 0 0 0 1     us vm r   (1) where ˙x0,y˙0,ψ˙ denote the time derivative of the ship’s posi￾tion i… view at source ↗
Figure 4
Figure 4. Figure 4: Correlation between state variables and control in [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of state variables and control inputs [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Distribution of state variables and control inputs in [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 8
Figure 8. Figure 8: Port 2 - Comparison between the full-scale ship [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: Port 4 - Comparison between the full-scale ship [PITH_FULL_IMAGE:figures/full_fig_p010_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Port 5 - Comparison between the full-scale ship [PITH_FULL_IMAGE:figures/full_fig_p011_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Comparison of model-predicted trajectories under varying re-initialization intervals against actual ship trajectories [PITH_FULL_IMAGE:figures/full_fig_p012_12.png] view at source ↗
read the original abstract

Despite significant technological progress, the realization of fully autonomous berthing and unberthing remains a significant challenge. One of the primary obstacles is the complex, non-linear nature of low-speed ship dynamics, which are difficult to model and control and often necessitate equally complex maneuvering models and control systems. This study proposes a simplified approach to bridge this gap by modeling the ship dynamics in the form of a time-invariant, continuous-time linear state-space system. The model parameters are estimated through system identification using the Covariance Adaptation Strategy Evolution Strategy (CMA-ES) applied to full-scale maneuvering data. Validation results demonstrate a strong agreement between the model output and empirical data. This outcome demonstrates the significant potential of simplified models to effectively define the maneuvering motion of a ship at low speeds.

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.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes modeling low-speed ship maneuvering dynamics as a time-invariant continuous-time linear state-space system whose parameters are identified via CMA-ES from full-scale trial data; it reports that validation shows strong agreement between model predictions and measurements, suggesting simplified linear models suffice for this regime.

Significance. A well-supported linear model would reduce complexity for autonomous berthing controllers relative to nonlinear hydrodynamic models. The use of CMA-ES on real data is a positive methodological choice, but the absence of any reported quantitative fit metrics, cross-validation protocol, or baseline comparisons leaves the central claim untestable from the provided information.

major comments (2)
  1. [Abstract] Abstract: the claim of 'strong agreement between the model output and empirical data' is unsupported by any numerical metrics (RMSE, R², cross-validation error, or comparison to nonlinear baselines), preventing evaluation of whether the linear structure is dynamically adequate or merely overfit.
  2. [Validation] Validation procedure (implicit in the abstract and results description): it is not stated whether the validation trajectories are disjoint from the CMA-ES identification set. If they are the same trials, the reported agreement can be achieved by parameter adjustment without the linear time-invariant assumption capturing the quadratic drag and lift terms the abstract itself identifies as dominant at low speed.
minor comments (2)
  1. The state-space realization (choice of states, inputs, and output equations) is not detailed; explicit matrices or a diagram would clarify the model order and observability assumptions.
  2. No mention is made of the sampling rate, sensor noise characteristics, or preprocessing of the full-scale data; these details are needed to assess identifiability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address the major points below and indicate where revisions will be made.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim of 'strong agreement between the model output and empirical data' is unsupported by any numerical metrics (RMSE, R², cross-validation error, or comparison to nonlinear baselines), preventing evaluation of whether the linear structure is dynamically adequate or merely overfit.

    Authors: We agree that the abstract's claim would be strengthened by quantitative metrics. The revised manuscript will report RMSE and R² values computed on the validation trajectories, along with a brief statement on cross-validation approach. A direct comparison to nonlinear hydrodynamic baselines is outside the paper's scope, which centers on the viability of the linear structure, but we will note this limitation explicitly. revision: yes

  2. Referee: [Validation] Validation procedure (implicit in the abstract and results description): it is not stated whether the validation trajectories are disjoint from the CMA-ES identification set. If they are the same trials, the reported agreement can be achieved by parameter adjustment without the linear time-invariant assumption capturing the quadratic drag and lift terms the abstract itself identifies as dominant at low speed.

    Authors: We acknowledge that the manuscript does not explicitly state whether validation trajectories are disjoint from the identification set. The revised text will clarify that validation used separate full-scale maneuvers not included in the CMA-ES optimization, thereby testing generalization rather than in-sample fit. This directly addresses the concern about whether the linear model captures the relevant dynamics beyond parameter tuning. revision: yes

Circularity Check

0 steps flagged

No circularity: standard external-data parameter estimation

full rationale

The paper's core procedure is system identification of a linear time-invariant state-space model via CMA-ES optimization against full-scale maneuvering measurements, followed by validation against the same empirical data. No derivation chain reduces any claimed prediction or result to a quantity defined by the fitted parameters themselves; the model structure is posited as an approximation and the parameters are obtained from independent external observations. No self-citations appear in the provided text as load-bearing for uniqueness or ansatz choices, and the approach does not rename known results or smuggle assumptions via prior author work. The derivation is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Only the abstract is available; the ledger therefore records the central modeling assumption stated there and the data-driven fitting step.

free parameters (1)
  • linear state-space parameters
    Estimated by CMA-ES from full-scale maneuvering data; exact count and values not given in abstract.
axioms (1)
  • domain assumption Low-speed ship dynamics can be represented by a time-invariant linear continuous-time state-space system
    Explicitly proposed in the abstract as the modeling choice to simplify the acknowledged non-linear problem.

pith-pipeline@v0.9.1-grok · 5677 in / 1227 out tokens · 25054 ms · 2026-07-03T08:02:24.622118+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    Abkowitz, Lectures on ship hydrodynamics–steering and manoeuvrability

    M.A. Abkowitz, Lectures on ship hydrodynamics–steering and manoeuvrability. Tech. rep. (1964)

    work page 1964

  2. [2]

    Strom-Tejsen, A digital computer technique for prediction of standard maneuvers of surface ships

    J. Strom-Tejsen, A digital computer technique for prediction of standard maneuvers of surface ships. Tech. rep. (1965)

    work page 1965

  3. [3]

    Abkowitz, Measurement of hydrodynamic characteristics from ship maneuvering trials by system identification (1980)

    M.A. Abkowitz, Measurement of hydrodynamic characteristics from ship maneuvering trials by system identification (1980)

    work page 1980

  4. [4]

    Ogawa, H

    A. Ogawa, H. Kasai, On the mathematical model of manoeu- vring motion of ships, International Shipbuilding Progress25, 306 (1978)

    work page 1978

  5. [5]

    Yoshimura, K

    Y . Yoshimura, K. Nomoto, Modeling of manoeuvring behaviour of ships with a propeller idling, boosting and reversing, Journal of the Society of Naval Architects of Japan1978, 57 (1978)

    work page 1978

  6. [6]

    Inoue, M

    S. Inoue, M. Hirano, K. Kijima, Hydrodynamic derivatives on ship manoeuvring (1981)

    work page 1981

  7. [7]

    K. Kose, H. HINATA, Y . Hashizume, E. Futagawa, On a math- ematical model of maneuvering motions of ships in low speeds (1984)

    work page 1984

  8. [8]

    Yoshimura, I

    Y . Yoshimura, I. Nakao, A. Ishibashi, Unified mathematical model for ocean and harbour manoeuvring (2009)

    work page 2009

  9. [9]

    Yasukawa, Y

    H. Yasukawa, Y . Yoshimura, Introduction of mmg standard method for ship maneuvering predictions, Journal of Marine Sci- ence and Technology (2015)

    work page 2015

  10. [10]

    FUJINO, A

    M. FUJINO, A. KIRITA, On the manoeuvrability of ships while stopping by adverse rotation of propeller : 1st report, Journal of the Kansai Society of Naval Architects, Japan169, 57 (1978)

    work page 1978

  11. [11]

    K. Kose, T. Saeki, On a new mathematical model of manoeuvring motions of a ship, Journal of the Society of Naval Architects of Japan1979, 229 (1979)

    work page 1979

  12. [12]

    Y . YOSHIMURA, Yasuo, Mathematical model for the manoeu- vring ship motion in shallow water : 2nd report : Mathematical model at slow forward speed, Journal of the Kansai Society of Naval Architects, Japan210, 77 (1988)

    work page 1988

  13. [13]

    YOSHIMURA, H

    Y . YOSHIMURA, H. SAKURAI, Mathematical model for the ma- noeuvring ship motion in shallow water (3rd report) : Manoeuvra- bility of a twin-propeller twin rudder ship, Journal of the Kansai Society of Naval Architects, Japan211, 115 (1989) Development and Identification of a Linear Low-Speed Ship Maneuvering Model from Full-Scale Data 13

    work page 1989

  14. [14]

    S.K. Lee, M. Fujino, T. Fukasawa, A study on the manoeuvring mathematical model for a twin-propeller twin-rudder ship, Journal of the Society of Naval Architects of Japan1988, 109 (1988)

    work page 1988

  15. [15]

    KOBAY ASHI, A

    H. KOBAY ASHI, A. ISHIBASHI, K. SHIMOKAW A, Y . SHIMURA, A study on mathematical model for the ma- neuvering motions of twin-propeller twin-rudder ship : In reference to the maneuvering motion from ordinary speed range to low speed range, The Journal of Japan Institute of Navigation 91, 263 (1994)

    work page 1994

  16. [16]

    Dubbioso, M

    G. Dubbioso, M. Viviani, Aspects of twin screw ships semi- empirical maneuvering models, Ocean Engineering (2012)

    work page 2012

  17. [17]

    Crane, Maneuvering trials of the 278,000 dwt esso osaka in shallow and deep waters

    C.L. Crane, Maneuvering trials of the 278,000 dwt esso osaka in shallow and deep waters. inSNAME Transactions, vol. 87 (1979), vol. 87, pp. 251–283

    work page 1979

  18. [18]

    Abkowitz, Measurement of hydrodynamic characteristics from ship maneuvering trials by system identification

    M.A. Abkowitz, Measurement of hydrodynamic characteristics from ship maneuvering trials by system identification. in Transactions of Society of Naval Architects and Marine Engineers 88 (1980), p. 283–318

    work page 1980

  19. [19]

    HIRANO, J

    M. HIRANO, J. TAKASHINA, S. MORIY A, Y . NAKAMURA, An experimental study on maneuvering hydrodynamic forces in shallow water, TRANSACTIONS OF THE WEST-JAPAN SOCI- ETY OF NA V AL ARCHITECTS69, 101 (1985)

    work page 1985

  20. [20]

    Biancardi, A simplified mathematical model for an onboard maneuvering simulator, Simulation (1988)

    C.G. Biancardi, A simplified mathematical model for an onboard maneuvering simulator, Simulation (1988)

    work page 1988

  21. [21]

    M. Ueno, T. Nimura, H. Miyazaki, T. Fujiwara, K. Nonaka, H. Yabuki, Model experiment and sea trial for investigating ma- noeuvrability of a training ship, Journal of the Society of Naval Architects of Japan2001, 71 (2001)

    work page 2001

  22. [22]

    M. Ueno, Y . Tsukada, Y . Kitagawa, Rudder effectiveness cor- rection for scale model ship testing, Ocean Engineering92, 267 (2014)

    work page 2014

  23. [23]

    M. Ueno, Y . Tsukada, Rudder effectiveness and speed correction for scale model ship testing, Ocean Engineering109, 495 (2015)

    work page 2015

  24. [24]

    M. Ueno, R. Suzuki, Y . Tsukada, Estimation of stopping ability of full-scale ship using free-running model, Ocean Engineering130, 260 (2017)

    work page 2017

  25. [25]

    Y . Liu, L. Zou, Z. Zou, H. Guoa, Predictions of ship maneuverabil- ity based on virtual captive model tests, Engineering Applications of Computational Fluid Mechanics12, 334 (2018)

    work page 2018

  26. [26]

    Sakamoto, K

    N. Sakamoto, K. Ohashi, M. Araki, K. ichi Kume, H. Kobayashi, Identification of kvlcc2 manoeuvring parameters for a modular- type mathematical model by rans method with an overset ap- proach, Ocean Engineering188, 106257 (2019)

    work page 2019

  27. [27]

    yuan Hwang, Application of system identification to ship ma- neuvering

    W. yuan Hwang, Application of system identification to ship ma- neuvering. Ph.D. thesis (1980)

    work page 1980

  28. [28]

    Hwang, Cancellation effect and parameter identifiability of ship steering dynamics, International Shipbuilding Progress29, 90 (1982)

    W.Y . Hwang, Cancellation effect and parameter identifiability of ship steering dynamics, International Shipbuilding Progress29, 90 (1982)

    work page 1982

  29. [29]

    Miyauchi, A

    Y . Miyauchi, A. Maki, N. Umeda, D.M. Rachman, Y . Akimoto, System parameter exploration of ship maneuvering model for au- tomatic docking/berthing using cma-es, Journal of Marine Science and Technology (2022)

    work page 2022

  30. [30]

    Deogaonkar, A.K

    V . Deogaonkar, A.K. Jadhav, K. Ramachandran, A.S. Somayajula, Data driven identification of ship maneuvering coefficients, Ocean Engineering (2023)

    work page 2023

  31. [31]

    Hasan, Discovering ship maneuvering models from data, Jour- nal of Marine Science and Technology (2025)

    A. Hasan, Discovering ship maneuvering models from data, Jour- nal of Marine Science and Technology (2025)

    work page 2025

  32. [32]

    W. Luo, X. Li, Measures to diminish the parameter drift in the modeling of ship manoeuvring using system identification, Ap- plied Ocean Research (2017)

    work page 2017

  33. [33]

    Mwange, Y

    A.N. Mwange, Y . Miyauchi, T. Kambara, H. Koike, K. Hosogaya, A. Ishibashi, A. Maki, Quantitative evaluation of full-scale ship maneuvering characteristics during berthing and unberthing, Jour- nal of Marine Science and Technology (2025)

    work page 2025

  34. [34]

    Rachman, Y

    D.M. Rachman, Y . Aoki, Y . Miyauchi, N. Umeda, A. Maki, Exper- imental low-speed positioning system with vectwin rudder for au- tomatic docking (berthing), Journal of Marine Science and Tech- nology28(3), 689 (2023)

    work page 2023

  35. [35]

    Fossen, Handbook of marine craft hydrodynamics and motion control (2011)

    T.I. Fossen, Handbook of marine craft hydrodynamics and motion control (2011)

    work page 2011

  36. [36]

    Sakamoto, Y

    N. Sakamoto, Y . Akimoto, Modified box constraint handling for the covariance matrix adaptation evolution strategy. in Proceedings of the Genetic and Evolutionary Computation Conference Companion (2017), pp. 183–184

    work page 2017

  37. [37]

    A. Maki, N. Sakamoto, Y . Akimoto, H. Nishikawa, N. Umeda, Application of optimal control theory based on the evolution strat- egy (cma-es) to automatic berthing, Journal of Marine Science and Technology25, 221 (2020)

    work page 2020