A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data

arXiv:2606.19711 · cs.RO, cs.LG, cs.SY, eess.SY · Submitted 2026-06-18 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: We're looking at the paper: "A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data".

Title and authors: Rosa: Let's look at what the full title tells us, "A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data." Dev It’s essentially proposing a framework where you treat the polynomial part and the data-driven part as differentiable pieces of one predictor.

Taro: That sounds like they are trying to find a middle ground between those two extremes, which is usually the problem when you're dealing with complex nonlinear systems in practice.

Rosa: Right. Conventional models use predefined polynomial bases where you fit unknown hydrodynamic coefficients to experimental data, but this new approach lets the polynomial and the data-adaptive basis component be calibrated together because they share a single prediction objective Dev That means they are optimized simultaneously instead of fitting them separately, which should lead to better results when you only have limited sea-trial data.

Taro: If you're fitting them jointly, does that mean the polynomial part gets more stable because it’s being guided by the neural component?

Rosa: It does. And this paper focuses on developing a gradient-based co-calibration method for full-scale AUV maneuvering prediction, which is how they achieve that joint optimization Dev The implication is that we can get a more accurate prediction of the vehicle's motion when it's actually operating in the water, rather than relying on models that only work perfectly in a controlled setting.

Taro: But what about when things go wrong? If one part of the model has to compensate for something unexpected, how does this joint optimization handle those surprises gracefully?

The paper's summary: Dev: So, summarizing the core idea of "A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data," it’s about using field-based measurements to construct an AUV maneuvering dynamics model that actually mirrors how the vehicle operates when it's in the water Rosa They define a learning target in terms of water-relative velocity, and then they reintroduce the current component during rollout for comparison with those sea-trial measurements Taro That means they aren't just modeling the vehicle itself; they are explicitly accounting for the ocean current which is crucial for real operating conditions.

Rosa: It’s not just about fitting numbers to data, though. They are using a composite predictor where one part uses explicit polynomial hydrodynamic basis functions, and the other part uses long short-term memory networks to learn history-dependent nonlinear features Dev That's the core structure: a compact, interpretable physical model coupled with a flexible, data-driven learning module.

Taro: I see. So you have the known physics modeled by polynomials and the unknown nonlinear stuff learned by the neural network, but they work together in one system Rosa The paper is interested in how to make this whole thing differentiable so you can optimize both parts at once, which is a key technical achievement here.

Dev: It’s about moving away from just polynomial-only or neural-only baselines, which are the standard ways we usually approach this problem Taro The paper suggests that by treating them as components of one predictor, we get better recursive trajectory and velocity prediction than those simpler methods.

The paper's improvements: Rosa: Now for the specific improvements they suggest, they focus on how the system is structured to handle these two components together Dev One big improvement is using a sensitivity-aware mechanism that regulates how the polynomial component updates its coefficients, keeping them within bounded correction variables Taro This prevents those polynomial coefficients from drifting wildly when it sees noisy field data.

Dev: That's smart. And they parameterize those polynomial coefficients using something like theta equals theta zero multiplied by one plus rho times tangent w, where rho is this prescribed maximum relative correction range Rosa It constrains the polynomial module to stay in a controlled neighborhood around its initial fit, which makes the results much more stable during recursive prediction.

Taro: So they are essentially saying that you don't want the polynomial part to completely rewrite itself based on every little measurement fluctuation, but rather have it adjust within a defined boundary Dev This speaks directly to robustness when the world misbehaves and gives us a more predictable system.

Rosa: And there’s another improvement they bring in regarding how they handle the current effect during rollout Taro Instead of just training on current-compensated data, they have this projection procedure where you estimate the current velocity and then project it into the body-fixed frame using an updated heading to reconstruct what the measured velocity should have been Dev This allows them to properly simulate what happens in a sea trial when they reintroduce that current effect.

Conclusion: Rosa: So, wrapping up on "A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data," the main implication is that by using this differentiable composite approximation, we can build a more robust and accurate AUV maneuvering model directly from real sea-trial data Dev It shows that combining explicit physical knowledge with learning methods in this way leads to better recursive prediction than using either approach alone.

Taro: I think the main point is that you get the interpretability of the polynomial coefficients while still getting the flexibility of a neural network to capture those tricky nonlinear interactions Rosa It’s a practical way to get an accurate model for things like deadreckoning support and digital twin construction, where stable prediction matters more than trying to assign a unique physical meaning to every single fitted coefficient.

Dev: Yeah, it sets a new standard for how we combine mechanistic and data-driven models under one approximation perspective Taro It’s not about replacing the physics model; it's about making the whole system differentiable so that the learning process improves both parts simultaneously.

Rosa: Before we wrap up this talk, Taro, what's your final thought on this framework?

Taro: I just want to stress that while it improves prediction compared to those baselines, you still need richer excitation and repeated-seed robustness for it to be truly reliable in the field Dev The paper points out that they should look at multi-step training objectives next because real-world maneuvering is often more complex than a single step Rosa That’s where the real work continues.

Dev: Right, we need to see how this holds up under more challenging conditions and if the safety case for these kinds of models is solid as they move from simulation to actual deployment Taro It’s a necessary next step for any system that relies on these models for critical tasks.

Rosa: Well, that covers what this paper does. We've talked about how this differentiable composite approximation framework allows us to build AUV maneuvering models directly from sea-trial data, and it sets a path forward by jointly calibrating polynomial and neural components with improved stability mechanisms like the bounded updates for those coefficients Dev That’s where we'll leave it for today.

College of Shipbuilding Engineering, Harbin Engineering University · China Academy of Aerospace Aerodynamics, Beijing · Institute of Artificial Intelligence, Shanghai University · China Ship Scientific Research Center

cs.RO, cs.LG, cs.SY, eess.SY

Submitted: 2026-06-18

Updated: 2026-11-18

Comments: Accepted for publication in Ocean Engineering

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 83/100

The gist: Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics The proposed method improves recursive…

Key concepts

Polynomial Hydrodynamic Component
This part of the model uses explicitly defined polynomial basis functions to represent known hydrodynamic responses like surge, sway, and yaw rate. It provides a compact and interpretable physical representation of how the vehicle's motion is fundamentally driven by its physical characteristics in water.
Neural Residual Component
This component, implemented via LSTM networks, learns complex, history-dependent nonlinear features from the maneuvering sequence. It captures the deviations or 'residuals' that the simple polynomial model cannot explain on its own, allowing for more nuanced prediction of vehicle behavior.
Differentiable Composite Approximation
The framework links these two components through a differentiable coupling mechanism. This allows the entire system to be optimized end-to-end using standard gradient descent techniques, enabling joint calibration of the polynomial and neural parts under a single predictive objective.
Sensitivity-Aware Mechanism
This mechanism regulates the updates to the polynomial coefficients by constraining them within bounded correction variables. This prevents uncontrolled parameter drift during training on noisy field data, ensuring that the physical model remains stable while allowing necessary adjustments.

Terminology

Summary

Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics The proposed method improves recursive trajectory and velocity prediction compared with polynomial-only, neural-only, and frozen-prior hybrid baselines

Problem Formulation

The objective of this study is to construct an AUV maneuvering dynamics model from in-field free-running sea-trial data The modeling problem is to jointly calibrate a polynomial hydrodynamic component and a neural residual component within a single predictive model

How the training signal is built

The learning target is defined in the water-relative velocity frame, while the current component is reintroduced during rollout for comparison with sea-trial measurements The relative state-input vector is represented by sr,k = ν⊤ r,k a⊤ k⊤ = ur,k vr,k rk nk δk⊤

Where the gains come from

The composite predictor is constructed as νbr,k+1 = C (pθ, Rϕ, Hr,k), where C(·) denotes a differentiable coupling of the polynomial prior and the data-adaptive module The polynomial module uses explicitly specified polynomial hydrodynamic basis functions, which provide a compact and interpretable representation of the dominant surge, sway, and yaw-rate responses The data-adaptive basis module is implemented using long short-term memory (LSTM) networks, which learn history-dependent nonlinear features from the maneuvering sequence

Pruning redundant reasoning tokens

The framework employs a sensitivity-aware mechanism regulating the bounded updates of the polynomial component This results in effective polynomial coefficients being generated from bounded correction variables: θ(w) = θ0 ⊙ (1 + ρ tanh w), where w is trainable, ρ is the prescribed maximum relative correction range, and ⊙ denotes element-wise multiplication

Model construction and evaluation

The current compensation procedure involves estimating the current velocity and compensating for its effect, thereby obtaining the water-relative motion states of the vehicle The current-compensated velocity sequence used for training is constructed by removing the estimated body-frame current component The proposed model achieves the lowest errors in surge, sway, yaw, and trajectory prediction

Discussion

The role of the unified differentiable architecture is to optimize these basis components under one prediction objective This framework provides a differentiable way to view mechanistic and data-driven maneuvering models from the same approximation perspective The current treatment is another necessary part of the modeling problem Furthermore, the calibrated coefficients should be interpreted as effective prediction coefficients rather than independently identified hydrodynamic derivatives The model remains suitable for deadreckoning support, maneuvering simulation, and digital-twin construction where stable recursive prediction is more relevant than assigning a unique physical interpretation to each fitted coefficient The formulation is extendable, and the full-scale experimental results reported in this study provide supporting evidence for its practical applicability The paper presents a differentiable composite approximation framework for AUV maneuvering prediction from full-scale sea-trial data The paper presents a differentiable composite approximation framework for AUV maneuvering prediction from full-scale sea-trial data>

Conclusions

The paper presented a differentiable composite approximation framework for AUV maneuvering prediction from full-scale sea-trial data The proposed model improves average recursive prediction compared to the standalone polynomial prior and neural-only LSTM baselines Future work should examine richer excitation, repeated-seed robustness, multi-step training objectives, and closed-loop validation The current treatment is another necessary part of the modeling problem The paper presents a differentiable composite approximation framework for AUV maneuvering prediction from full-scale sea-trial data The paper presents a differentiable composite approximation framework for AUV maneuvering prediction from full-scale sea-trial data>

References

Ahmed, F., Xiang, X., Jiang, C., Xiang, G., Yang, S., 2023. Survey on traditional and ai based estimation techniques for hydrodynamic coefficients of autonomous underwater vehicle. p. 0

Where the paper says it

A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data Aobo Wanga,b, Aifei Xiac , Zihao Wangc,∗ , Lizhu Haod aCollege of Shipbuilding Engineering, Harbin Engineering University, Harbin, 150001, China bChina Academy of Aerospace Aerodynamics, Beijing, 100074, China c Institute of Artificial Intelligence, Shanghai University, Shanghai, 200444, China dChina Ship Scientific Research Center, Wuxi, 214082, China Abstract Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics. From an approximation perspective, conventional maneuvering models use predefined constraint polynomial bases, whereas datadriven models use data-adaptive bases.

Ocean Engineering 268, 113300

Busse, C., Reich, J.E., Renner, B.C., 2022. In situ damping parameter estimation for an underwater vehicle using onboard sensors, in: 2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV), IEEE

Du, X., Zheng, Z., Guan, S., 2018. Numerical calculation of hydrodynamic interactions of submarine flow on auv, in: 2018 OCEANS-MTS/IEEE Kobe Techno-Oceans (OTO), IEEE

Fossen, T.I., 2021. Handbook of Marine Craft Hydrodynamics and Motion Control. p. 8

Where the paper says it

A feasible solution is to estimate the current velocity and compensate for its effect, thereby obtaining the water-relative motion states of the vehicle. This approach follows the relative-velocity formulation for marine craft under ocean currents (Fossen, 2021).

2nd ed., Wiley, Hoboken, NJ

Go, G., Ahn, H.T., 2019. Hydrodynamic derivative determination based on cfd and motion simulation for a tow-fish. p. 1

Where the paper says it

By capturing the nonlinear characteristics of underwater motion, these models form the foundation for the agent to understand how it interacts with its environment and serve as a key enabler for reliable state prediction, informed decision-making, modelaided navigation, and model-based control (Smallwood and Whitcomb, 2004; Fossen, 2021). Conventional AUV dynamic models are often derived from polynomial approximations of hydrodynamic forces and moments, where pre-defined polynomial basis functions involving velocity states and control inputs are fitted to experimental data.

Applied Ocean Research 82, 191–209

Hegrenaes, O., Hallingstad, O., Jalving, B., 2007a. Comparison of mathematical models for the HUGIN 4500 AUV based on experimental data, in: 2007 Symposium on Underwater Technology and Workshop on Scientific Use of Submarine Cables and Related Technologies, IEEE

Hegrenaes, O., Hallingstad, O., Jalving, B., 2007b. A framework for obtaining steady-state maneuvering characteristics of underwater vehicles using seatrial data, in: 2007 Mediterranean Conference on Control & Automation, IEEE

Lei, L., Liu, X.W., Yang, G., 2022a. Incremental residual learning-based dynamic modeling and stability analysis for multipower underwater vehicles. p. 2

Where the paper says it

(2022b) used a physics-guided neural network in which a theoretical model provides the baseline and an RBF neural network compensates for modeling errors. Incremental residual learning has also been investigated for underwater vehicles to compensate for the physical-modeling gap caused by environmental uncertainty (Lei et al.

Mechanism and Machine Theory 178, 105088

Lei, L., Yang, G., Guo, J., 2022b. Physics-guided neural network for underwater glider flight modeling. p. 2

Where the paper says it

(2022b) used a physics-guided neural network in which a theoretical model provides the baseline and an RBF neural network compensates for modeling errors. Incremental residual learning has also been investigated for underwater vehicles to compensate for the physical-modeling gap caused by environmental uncertainty (Lei et al.

Applied Ocean Research 121, 103082

Nielsen, R.E., Papageorgiou, D., Nalpantidis, L., Jensen, B.T., Blanke, M., 2022. Machine learning enhancement of manoeuvring prediction for ship digital twin using full-scale recordings. p. 3

Where the paper says it

Such a biased baseline may fit the training data but yield coefficients with reduced physical reliability and interpretability, forcing the neural residual to compensate for both unmodeled dynamics and baseline error, which may degrade recursive maneuvering prediction. To address field-data-based AUV maneuvering modeling using full-scale sea-trial data, this study reformulates hybrid maneuvering prediction as a unified composite approximation problem.

Ocean Engineering 257, 111579

Nouri, N., Mostafapour, K., Bahadori, R., 2016. An apparatus to estimate the hydrodynamic coefficients of autonomous underwater vehicles using water tunnel testing. p. 1

Where the paper says it

Introduction Autonomous Underwater Vehicles (AUVs) play a vital role in marine applications such as exploration, monitoring, and defense. As missions grow in complexity, AUVs are expected to operate autonomously over extended periods with minimal human intervention.

Review of Scientific Instruments 87, 065106

Ramirez, W.A., Kocijan, J., Leong, Z.Q., Nguyen, H.D., Jayasinghe, S.G., 2021. Dynamic system identification of underwater vehicles using multioutput gaussian processes. p. 2

Where the paper says it

For example, neural networks construct highly flexible function approximators through cascaded nonlinear transformations, enabling them to capture complex nonlinear relationships and interaction effects without explicitly specifying the functional form in advance. Related approaches have been explored for marine vehicle dynamics, including multi-output Gaussian processes (Ramirez et al.

International Journal of Automation and Computing 18, 681–693

Sabet, M.T., Daniali, H.M., Fathi, A., Alizadeh, E., 2017. Identification of an autonomous underwater vehicle hydrodynamic model using the extended, cubature, and transformed unscented kalman filter. p. 1

Where the paper says it

Introduction Autonomous Underwater Vehicles (AUVs) play a vital role in marine applications such as exploration, monitoring, and defense. As missions grow in complexity, AUVs are expected to operate autonomously over extended periods with minimal human intervention.

IEEE Journal of Oceanic Engineering 43, 457–467

Sajedi, Y., Bozorg, M. p. 1

Where the paper says it

, 2017; Busse et al. , 2022), and H-infinity filter (Sajedi and Bozorg, 2019).

, 2019. Robust estimation of hydrodynamic coefficients of an auv using kalman and h∞ filters. p. 0
Where the paper says it

To account for ocean-current effects in field data, a turning-motion-based current estimation and compensation procedure is incorporated to construct currentcompensated learning targets for training and rollout. The framework is evaluated using sea-trial data collected from a 7-meter AUV under multiple maneuvering conditions.

Ocean Engineering 182, 386–394

Skulstad, R., Li, G., Fossen, T. p. 1

Where the paper says it

To support such capabilities, precise dynamic modeling becomes essential. By capturing the nonlinear characteristics of underwater motion, these models form the foundation for the agent to understand how it interacts with its environment and serve as a key enabler for reliable state prediction, informed decision-making, modelaided navigation, and model-based control (Smallwood and Whitcomb, 2004; Fossen, 2021).

I., Vik, B., Zhang, H. p. 1
Where the paper says it

Prior studies have identified surge or yaw dynamics using least squares (Hegrenaes et al. , 2007b; Zhang et al.

, 2021. A hybrid approach to motion prediction for ship docking—integration of a neural network model into the ship dynamic model. p. 3
Where the paper says it

ideas have also been explored in marine ship-motion prediction and digitaltwin modeling (Skulstad et al. , 2021; Nielsen et al.

IEEE Transactions on Instrumentation and Measurement 70, 1–11

Smallwood, D.A., Whitcomb, L. p. 1

Where the paper says it

To support such capabilities, precise dynamic modeling becomes essential. By capturing the nonlinear characteristics of underwater motion, these models form the foundation for the agent to understand how it interacts with its environment and serve as a key enabler for reliable state prediction, informed decision-making, modelaided navigation, and model-based control (Smallwood and Whitcomb, 2004; Fossen, 2021).

L., 2004. Model-based dynamic positioning of underwater robotic vehicles: theory and experiment. p. 1
Where the paper says it

To support such capabilities, precise dynamic modeling becomes essential. By capturing the nonlinear characteristics of underwater motion, these models form the foundation for the agent to understand how it interacts with its environment and serve as a key enabler for reliable state prediction, informed decision-making, modelaided navigation, and model-based control (Smallwood and Whitcomb, 2004; Fossen, 2021).

IEEE Journal of oceanic engineering 29, 169–186

Tešnar, M., Wehbe, B., Valdenegro-Toro, M., 2025. Modeling of auv dynamics with limited resources: Efficient online learning using uncertainty. p. 2

Where the paper says it

(2021) employed multi-output Gaussian processes to model the dynamics of an underactuated AUV with limited data, while Tešnar et al. (2025) constructed an ensemble-MLP dynamics model and used predictive uncertainty to guide data selection for rehearsal in storagelimited online learning.

arXiv preprint arXiv:2504. p. 0
Where the paper says it

∗Corresponding author: zihaowang@shu. cn arXiv:2606.

04583 URL: https://arxiv. p. 0
Where the paper says it

∗Corresponding author: zihaowang@shu. cn arXiv:2606.

org/abs/2504.04583

Tiano, A., Sutton, R., Lozowicki, A., Naeem, W., 2007. Observer kalman filter identification of an autonomous underwater vehicle. p. 0

Where the paper says it

A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data Aobo Wanga,b, Aifei Xiac , Zihao Wangc,∗ , Lizhu Haod aCollege of Shipbuilding Engineering, Harbin Engineering University, Harbin, 150001, China bChina Academy of Aerospace Aerodynamics, Beijing, 100074, China c Institute of Artificial Intelligence, Shanghai University, Shanghai, 200444, China dChina Ship Scientific Research Center, Wuxi, 214082, China Abstract Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics. From an approximation perspective, conventional maneuvering models use predefined constraint polynomial bases, whereas datadriven models use data-adaptive bases.

Control engineering practice 15, 727–739

Zhang, H., Tong, X., Guo, H., Xia, S., 2020. The parameter identification of the autonomous underwater vehicle based on multi-innovation least squares identification algorithm. p. 0

Where the paper says it

A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data Aobo Wanga,b, Aifei Xiac , Zihao Wangc,∗ , Lizhu Haod aCollege of Shipbuilding Engineering, Harbin Engineering University, Harbin, 150001, China bChina Academy of Aerospace Aerodynamics, Beijing, 100074, China c Institute of Artificial Intelligence, Shanghai University, Shanghai, 200444, China dChina Ship Scientific Research Center, Wuxi, 214082, China Abstract Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics. From an approximation perspective, conventional maneuvering models use predefined constraint polynomial bases, whereas datadriven models use data-adaptive bases.

International Journal of Advanced Robotic Systems 17, 1729881420921016

Zhao, Y., Hu, Z., Du, W., Geng, L., Yang, Y., 2024. Research on modeling method of autonomous underwater vehicle based on a physics-informed neural network. p. 0

Where the paper says it

A Differentiable Composite Approximation Framework for Autonomous Underwater Vehicle Maneuvering Modeling from Sea-Trial Data Aobo Wanga,b, Aifei Xiac , Zihao Wangc,∗ , Lizhu Haod aCollege of Shipbuilding Engineering, Harbin Engineering University, Harbin, 150001, China bChina Academy of Aerospace Aerodynamics, Beijing, 100074, China c Institute of Artificial Intelligence, Shanghai University, Shanghai, 200444, China dChina Ship Scientific Research Center, Wuxi, 214082, China Abstract Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics. From an approximation perspective, conventional maneuvering models use predefined constraint polynomial bases, whereas datadriven models use data-adaptive bases.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI system can perform joint calibration of a polynomial hydrodynamic component and a data-adaptive basis component by optimizing them under a common prediction loss, enabling joint calibration rather than separate-stage fitting. This allows for simultaneous optimization of both basis components and improves field-data-based AUV maneuvering prediction.

  2. Improved AI system incorporates ocean current effects by using a turning-motion-based current estimation and compensation procedure to construct currentcompensated learning targets for training and rollout. This accounts for ocean currents in field data, which is crucial for accurate modeling in real operating conditions.

  3. Improved AI system employs a sensitivity-aware mechanism that regulates bounded polynomial updates while the neural residual captures remaining nonlinear discrepancies under a shared prediction objective. This mechanism ensures the polynomial coefficients remain adjustable within bounded ranges, preventing parameter drift and improving stability during recursive prediction.

  4. Improved AI system uses a differentiable composite approximation architecture where the polynomial module and the LSTM adaptive-basis module are coupled through a skip connection, allowing the adaptive-basis module to learn deviations conditioned on the polynomial hydrodynamic prediction. This serial-parallel composite approximation remains differentiable with respect to both components, ensuring stable recursive rollout.

  5. Improved AI system implements a bounded coefficient parameterization by generating effective coefficients from bounded correction variables theta(w) = theta0 dot (1 + rho tanh w), where rho is the prescribed maximum relative correction range. This parameterization constrains the polynomial basis module to stay within a controlled neighborhood of the initial polynomial fit, enhancing stability under noisy field data.

  6. Improved AI system optimizes training using a loss function L(w, phi) that penalizes one-step prediction error and also includes an adaptive correction penalty term lambda delta nu T q D nu-2 delta nu q. This objective discourages the adaptive-basis module from unnecessarily replacing the polynomial basis module, leading to a more interpretable error decomposition.

  7. Improved AI system performs current reintroduction during rollout by projecting the estimated current into the body-fixed frame using an updated heading and reconstructing measured velocity as nu b m = nu b r + [Vc cos(beta c - psi) Vc sin(beta c - psi) 0]. This allows for the prediction of the current-affected vehicle motion observed in sea trials.

Abstract

Field-based modeling from onboard measurements can produce autonomous underwater vehicle (AUV) maneuvering models that reflect real operating characteristics. From an approximation perspective, conventional maneuvering models use predefined constraint polynomial bases, whereas data-driven models use data-adaptive bases. Motivated by this basis-function view, this paper presents a differentiable composite approximation formulation, in which the polynomial basis component and the data-adaptive basis component are treated as differentiable parts of a unified predictor and calibrated jointly. A gradient-based co-calibration method is developed for full-scale AUV maneuvering prediction, where a sensitivity-aware mechanism regulates bounded polynomial updates while the neural residual captures remaining nonlinear discrepancies under a shared prediction objective. To account for ocean-current effects in field data, a turning-motion-based current estimation and compensation procedure is incorporated to construct current-compensated learning targets for training and rollout. The framework is evaluated using sea-trial data collected from a 7-meter AUV under multiple maneuvering conditions. Results show that the proposed method improves recursive trajectory and velocity prediction compared with polynomial-only, neural-only, and frozen-prior hybrid baselines, demonstrating its applicability to field-data-based AUV maneuvering modeling.

Sources

Related papers