Stochastic Physics-Informed Neural Networks on Lie Groups for Learning Underwater Vehicle Dynamics

• URL: http://arxiv.org/abs/2608.08356v1
• Date: Sat, 08 Aug 2026 22:48:21 GMT
• Title: Stochastic Physics-Informed Neural Networks on Lie Groups for Learning Underwater Vehicle Dynamics
• Abstract summary: This paper presents a novel data-driven framework for learning underwater vehicle dynamics.<n>We develop a physics-informed network architecture that respects the physical and geometric constraints of underwater vehicles.<n>The results demonstrate that our method learns accurate and robust dynamics models, enabling safe model-based control in challenging marine environments.
• Score: 10.900663835109425
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate models of underwater vehicle motion are needed for autonomous execution of marine tasks like infrastructure inspection and scientific sampling. However, such motion is challenging to characterize using traditional physics-based methods. This paper presents a novel data-driven framework for learning stochastic underwater vehicle dynamics. Using Euler-Poincaré dynamics and the geometry of Lie groups, we develop a stochastic physics-informed neural network architecture that respects the physical and geometric constraints of underwater vehicles. Our approach leverages structure-preserving stochastic integration and builds upon moment matching and finite dimensional matching to ensure geometrically-consistent training. We evaluate our approach in simulation and on an underwater vehicle navigating dock pylons in a harbor environment. The results demonstrate that our method learns accurate and robust dynamics models, enabling safe model-based control in challenging marine environments.

Related papers

• Geometry-Aware Surrogate for Real-Time Hydrodynamics Estimation of Autonomous Ground Vehicles in Amphibious Environments [0.0]
  This work presents a per-surface neural network surrogate that bridges the gap by predicting geometry- hydrodynamic forces at real-time rates.<n>A vehicle specific Signed Distance Field (SDF) provides per-surface submergence inputs, allowing the model to resolve how loading varies with vehicle geometry, depth, and flow direction.<n>To evaluate the model under real-world conditions, water wading trials of a full-scale vehicle at different submersion depths are used.
  arXiv  Detail & Related papers  (2026-05-18T15:24:23Z)
• Physics-informed Deep Mixture-of-Koopmans Vehicle Dynamics Model with Dual-branch Encoder for Distributed Electric-drive Trucks [19.506735422845182]
  We propose a fully data-driven dynamics modeling method tailored for complex distributed electric-drive trucks (DETs)<n>We first propose a novel dual-branch encoder which encodes dynamic states and provides a powerful basis for the proposed Koopman-based methods entitled KODE.<n>A physics-informed supervision mechanism is incorporated into the training process to facilitate effective learning of both the encoder and the Koopman operator.
  arXiv  Detail & Related papers  (2026-03-18T06:47:40Z)
• Simple Models, Real Swimming: Digital Twins for Tendon-Driven Underwater Robots [8.744009526780692]
  We present a tendon-driven fish robot modeled in an efficient underwater swimmer environment.<n>Our results show that even simple, stateless models can serve as effective digital twins for soft underwater robots.
  arXiv  Detail & Related papers  (2026-02-26T17:55:22Z)
• PhysRVG: Physics-Aware Unified Reinforcement Learning for Video Generative Models [100.65199317765608]
  Physical principles are fundamental to realistic visual simulation, but remain a significant oversight in transformer-based video generation.<n>We introduce a physics-aware reinforcement learning paradigm for video generation models that enforces physical collision rules directly in high-dimensional spaces.<n>We extend this paradigm to a unified framework, termed Mimicry-Discovery Cycle (MDcycle), which allows substantial fine-tuning.
  arXiv  Detail & Related papers  (2026-01-16T08:40:10Z)
• Forecasting Continuous Non-Conservative Dynamical Systems in SO(3) [51.510040541600176]
  We propose a novel approach to modeling the rotation of moving objects in computer vision.<n>Our approach is agnostic to energy and momentum conservation while being robust to input noise.<n>By learning to approximate object dynamics from noisy states during training, our model attains robust extrapolation capabilities in simulation and various real-world settings.
  arXiv  Detail & Related papers  (2025-08-11T09:03:10Z)
• AI-Enhanced Automatic Design of Efficient Underwater Gliders [60.45821679800442]
  Building an automated design framework is challenging due to the complexities of representing glider shapes and the high computational costs associated with modeling complex solid-fluid interactions.<n>We introduce an AI-enhanced automated computational framework designed to overcome these limitations by enabling the creation of underwater robots with non-trivial hull shapes.<n>Our approach involves an algorithm that co-optimizes both shape and control signals, utilizing a reduced-order geometry representation and a differentiable neural-network-based fluid surrogate model.
  arXiv  Detail & Related papers  (2025-04-30T23:55:44Z)
• Physics Encoded Blocks in Residual Neural Network Architectures for Digital Twin Models [2.8720819157502344]
  Physics Informed Machine Learning has emerged as a popular approach for modeling and simulation in digital twins.<n>This paper presents a generic approach based on a novel physics-encoded residual neural network architecture.<n>Our method integrates differentiable physics blocks-implementing mathematical operators from physics-based models with feed-forward learning blocks.
  arXiv  Detail & Related papers  (2024-11-18T11:58:20Z)
• Learning Adaptive Hydrodynamic Models Using Neural ODEs in Complex Conditions [9.392180262607921]
  Reinforcement learning-based quadruped robots excel across various terrains but still lack the ability to swim in water. This paper presents the development and evaluation of a data-driven hydrodynamic model for amphibious quadruped robots.
  arXiv  Detail & Related papers  (2024-10-01T08:18:36Z)
• Deep Dynamics: Vehicle Dynamics Modeling with a Physics-Constrained Neural Network for Autonomous Racing [0.0]
  This paper introduces Deep Dynamics, a physics-constrained neural network (PCNN) for vehicle dynamics modeling of an autonomous racecar.<n>It combines physics coefficient estimation and dynamical equations to accurately predict vehicle states at high speeds.<n>Open-loop and closed-loop performance assessments, using a physics-based simulator and full-scale autonomous Indy racecar data, highlight Deep Dynamics as a promising approach for modeling racecar vehicle dynamics.
  arXiv  Detail & Related papers  (2023-12-07T15:44:56Z)
• Physics-Inspired Temporal Learning of Quadrotor Dynamics for Accurate Model Predictive Trajectory Tracking [76.27433308688592]
  Accurately modeling quadrotor's system dynamics is critical for guaranteeing agile, safe, and stable navigation. We present a novel Physics-Inspired Temporal Convolutional Network (PI-TCN) approach to learning quadrotor's system dynamics purely from robot experience. Our approach combines the expressive power of sparse temporal convolutions and dense feed-forward connections to make accurate system predictions.
  arXiv  Detail & Related papers  (2022-06-07T13:51:35Z)
• Constructing Neural Network-Based Models for Simulating Dynamical Systems [59.0861954179401]
  Data-driven modeling is an alternative paradigm that seeks to learn an approximation of the dynamics of a system using observations of the true system. This paper provides a survey of the different ways to construct models of dynamical systems using neural networks. In addition to the basic overview, we review the related literature and outline the most significant challenges from numerical simulations that this modeling paradigm must overcome.
  arXiv  Detail & Related papers  (2021-11-02T10:51:42Z)
• Physics-Integrated Variational Autoencoders for Robust and Interpretable Generative Modeling [86.9726984929758]
  We focus on the integration of incomplete physics models into deep generative models. We propose a VAE architecture in which a part of the latent space is grounded by physics. We demonstrate generative performance improvements over a set of synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2021-02-25T20:28:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.