MarineGym: A High-Performance Reinforcement Learning Platform for Underwater Robotics

• URL: http://arxiv.org/abs/2503.09203v1
• Date: Wed, 12 Mar 2025 09:47:58 GMT
• Title: MarineGym: A High-Performance Reinforcement Learning Platform for Underwater Robotics
• Abstract summary: MarineGym is a high-performance reinforcement learning (RL) platform specifically designed for underwater robotics.<n>MarineGym integrates a proposed GPU-accelerated hydrodynamic plugin based on Isaac Sim.<n>It also provides five models of unmanned underwater vehicles (UUVs), multiple propulsion systems, and a set of predefined tasks.
• Score: 9.015814050369107
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This work presents the MarineGym, a high-performance reinforcement learning (RL) platform specifically designed for underwater robotics. It aims to address the limitations of existing underwater simulation environments in terms of RL compatibility, training efficiency, and standardized benchmarking. MarineGym integrates a proposed GPU-accelerated hydrodynamic plugin based on Isaac Sim, achieving a rollout speed of 250,000 frames per second on a single NVIDIA RTX 3060 GPU. It also provides five models of unmanned underwater vehicles (UUVs), multiple propulsion systems, and a set of predefined tasks covering core underwater control challenges. Additionally, the DR toolkit allows flexible adjustments of simulation and task parameters during training to improve Sim2Real transfer. Further benchmark experiments demonstrate that MarineGym improves training efficiency over existing platforms and supports robust policy adaptation under various perturbations. We expect this platform could drive further advancements in RL research for underwater robotics. For more details about MarineGym and its applications, please visit our project page: https://marine-gym.com/.

Related papers

• From Foundation to Application: Improving VLA Models in Practice [62.1552185059214]
  We present LingBot-VLA 2.0, which advances LingBot-VLA through improvements in three functional domains.<n>Compared to the previous version, we revamp the data processing pipeline and curate around 60,000 hours of data for pretraining.<n>In particular, our system accommodates degrees of freedom for the heads, waists, mobile bases, and dexterous hands, thereby empowering the robots to tackle more complex tasks in practical scenarios.
  arXiv  Detail & Related papers  (2026-07-07T15:33:12Z)
• Marinarium: a New Arena to Bring Maritime Robotics Closer to Shore [13.103990961903408]
  The Marinarium is a stand-alone underwater research facility designed to provide a realistic testbed for maritime and space- analog robotic experimentation.<n>The Marinarium combines a fully instrumented underwater and aerial operational volume, extendable via a retractable roof for real-weather conditions, a digital twin in the SMaRCSim simulator and tight integration with a space robotics laboratory.
  arXiv  Detail & Related papers  (2026-02-26T14:40:20Z)
• Sim2Swim: Zero-Shot Velocity Control for Agile AUV Maneuvering in 3 Minutes [2.6511518230332753]
  Holonomic autonomous underwater vehicles (AUVs) have the hardware ability for agile maneuvering in both translational and rotational degrees of freedom (DOFs)<n>This paper presents the first general zero-shot sim2real deep reinforcement learning-based (DRL) velocity controller enabling path following and agile 6DOF maneuvering with a training duration of just 3 minutes.
  arXiv  Detail & Related papers  (2025-12-09T14:42:26Z)
• NAUTILUS: A Large Multimodal Model for Underwater Scene Understanding [60.76337064425815]
  We study the underwater scene understanding methods, which aim to achieve automated underwater exploration.<n>NautData is a dataset containing 1.45 M image-text pairs supporting eight underwater scene understanding tasks.<n>We propose a plug-and-play vision feature enhancement (VFE) module, which explicitly restores clear underwater information.
  arXiv  Detail & Related papers  (2025-10-31T14:00:35Z)
• DUViN: Diffusion-Based Underwater Visual Navigation via Knowledge-Transferred Depth Features [47.88998580611257]
  We propose a Diffusion-based Underwater Visual Navigation policy via knowledge-transferred depth features, named DUViN.<n>DuViN guides the vehicle to avoid obstacles and maintain a safe and perception awareness altitude relative to the terrain without relying on pre-built maps.<n> Experiments in both simulated and real-world underwater environments demonstrate the effectiveness and generalization of our approach.
  arXiv  Detail & Related papers  (2025-09-03T03:43:12Z)
• PICT -- A Differentiable, GPU-Accelerated Multi-Block PISO Solver for Simulation-Coupled Learning Tasks in Fluid Dynamics [59.38498811984876]
  We present our fluid simulator PICT, a differentiable pressure-implicit solver coded in PyTorch with Graphics-processing-unit (GPU) support.<n>We first verify the accuracy of both the forward simulation and our derived gradients in various established benchmarks.<n>We show that the gradients provided by our solver can be used to learn complicated turbulence models in 2D and 3D.
  arXiv  Detail & Related papers  (2025-05-22T17:55:10Z)
• Advancing Marine Research: UWSAM Framework and UIIS10K Dataset for Precise Underwater Instance Segmentation [110.02397462607449]
  We propose a large-scale underwater instance segmentation dataset, UIIS10K, which includes 10,048 images with pixel-level annotations for 10 categories.<n>We then introduce UWSAM, an efficient model designed for automatic and accurate segmentation of underwater instances.<n>We show that our model is effective, achieving significant performance improvements over state-of-the-art methods on multiple underwater instance datasets.
  arXiv  Detail & Related papers  (2025-05-21T14:36:01Z)
• Scaling Multi Agent Reinforcement Learning for Underwater Acoustic Tracking via Autonomous Vehicles [0.9217021281095907]
  Multi-Agent Reinforcement Learning is notoriously sample-inefficient.<n>High-fidelity simulators like Gazebo's LRAUV provide 100x faster-than-real-time single-robot simulations.<n>We propose an iterative distillation method that transfers high-fidelity simulations into a simplified, GPU-accelerated environment.
  arXiv  Detail & Related papers  (2025-05-13T04:42:30Z)
• TUGS: Physics-based Compact Representation of Underwater Scenes by Tensorized Gaussian [6.819210285113731]
  Underwater Gaussian Splatting (TUGS) can effectively solve the modeling challenges of the complex interactions between object and water media.<n>Compared to other NeRF-based and GS-based methods designed for underwater, TUGS is able to render high-quality underwater images with faster rendering speeds and less memory usage.
  arXiv  Detail & Related papers  (2025-05-12T07:09:35Z)
• 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. 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. 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)
• A General Infrastructure and Workflow for Quadrotor Deep Reinforcement Learning and Reality Deployment [48.90852123901697]
  We propose a platform that enables seamless transfer of end-to-end deep reinforcement learning (DRL) policies to quadrotors. Our platform provides rich types of environments including hovering, dynamic obstacle avoidance, trajectory tracking, balloon hitting, and planning in unknown environments.
  arXiv  Detail & Related papers  (2025-04-21T14:25:23Z)
• OceanSim: A GPU-Accelerated Underwater Robot Perception Simulation Framework [9.848290383558572]
  We propose OceanSim, a high-fidelity GPU-accelerated underwater simulator.<n>We propose advanced physics-based rendering techniques to reduce the sim-to-real gap for underwater image simulation.
  arXiv  Detail & Related papers  (2025-03-03T00:32:09Z)
• Stonefish: Supporting Machine Learning Research in Marine Robotics [5.021710505685786]
  This paper highlights recent enhancements to the Stonefish simulator, an open-source platform supporting development and testing of marine robotics solutions.<n>Key updates include a suite of additional sensors, as well as, visual light communication, support for tethered operations, improved thruster modelling, more flexible hydrodynamics, and enhanced sonar accuracy.
  arXiv  Detail & Related papers  (2025-02-17T15:13:41Z)
• UGSim: Autonomous Buoyancy-Driven Underwater Glider Simulator with LQR Control Strategy and Recursive Guidance System [3.8632181427836945]
  This paper presents the UGSim, a simulator for buoyancy-driven gliders.<n>It is designed to address unique challenges that come from the complex hydrodynamic and hydrostatic impacts on buoyancy-driven gliders.<n>The simulator is provided to accelerate the development and the evaluation of algorithms that would otherwise require expensive and time-consuming operations at sea.
  arXiv  Detail & Related papers  (2025-01-29T18:50:41Z)
• Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networks [93.38375271826202]
  We present a method to improve generalization and robustness to distribution shifts in sim-to-real visual quadrotor navigation tasks. We first build a simulator by integrating Gaussian splatting with quadrotor flight dynamics, and then, train robust navigation policies using Liquid neural networks. In this way, we obtain a full-stack imitation learning protocol that combines advances in 3D Gaussian splatting radiance field rendering, programming of expert demonstration training data, and the task understanding capabilities of Liquid networks.
  arXiv  Detail & Related papers  (2024-06-21T13:48:37Z)
• Aquatic Navigation: A Challenging Benchmark for Deep Reinforcement Learning [53.3760591018817]
  We propose a new benchmarking environment for aquatic navigation using recent advances in the integration between game engines and Deep Reinforcement Learning. Specifically, we focus on PPO, one of the most widely accepted algorithms, and we propose advanced training techniques. Our empirical evaluation shows that a well-designed combination of these ingredients can achieve promising results.
  arXiv  Detail & Related papers  (2024-05-30T23:20:23Z)
• An Efficient Detection and Control System for Underwater Docking using Machine Learning and Realistic Simulation: A Comprehensive Approach [5.039813366558306]
  This work compares different deep-learning architectures to perform underwater docking detection and classification. A Generative Adversarial Network (GAN) is used to do image-to-image translation, converting the Gazebo simulation image into an underwater-looking image. Results show an improvement of 20% in the high turbidity scenarios regardless of the underwater currents.
  arXiv  Detail & Related papers  (2023-11-02T18:10:20Z)
• DRIFT: Deep Reinforcement Learning for Intelligent Floating Platforms Trajectories [18.420795137038677]
  Floating platforms serve as versatile test-beds to emulate micro-gravity environments on Earth. Our suite achieves robustness, adaptability, and good transferability from simulation to reality.
  arXiv  Detail & Related papers  (2023-10-06T14:11:35Z)
• OSCAR: Data-Driven Operational Space Control for Adaptive and Robust Robot Manipulation [50.59541802645156]
  Operational Space Control (OSC) has been used as an effective task-space controller for manipulation. We propose OSC for Adaptation and Robustness (OSCAR), a data-driven variant of OSC that compensates for modeling errors. We evaluate our method on a variety of simulated manipulation problems, and find substantial improvements over an array of controller baselines.
  arXiv  Detail & Related papers  (2021-10-02T01:21:38Z)
• PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics [89.81550748680245]
  We introduce a new differentiable physics benchmark called PasticineLab. In each task, the agent uses manipulators to deform the plasticine into the desired configuration. We evaluate several existing reinforcement learning (RL) methods and gradient-based methods on this benchmark.
  arXiv  Detail & Related papers  (2021-04-07T17:59:23Z)
• ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulation [99.2543521972137]
  ReLMoGen is a framework that combines a learned policy to predict subgoals and a motion generator to plan and execute the motion needed to reach these subgoals. Our method is benchmarked on a diverse set of seven robotics tasks in photo-realistic simulation environments. ReLMoGen shows outstanding transferability between different motion generators at test time, indicating a great potential to transfer to real robots.
  arXiv  Detail & Related papers  (2020-08-18T08:05:15Z)

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.