Saved in:
Bibliographic Details
Main Authors: Chu, Shuguang, Huang, Zebin, Li, Yutong, Lin, Mingwei, Carlucho, Ignacio, Petillot, Yvan R., Yang, Canjun
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.09203
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913732607606784
author Chu, Shuguang
Huang, Zebin
Li, Yutong
Lin, Mingwei
Carlucho, Ignacio
Petillot, Yvan R.
Yang, Canjun
author_facet Chu, Shuguang
Huang, Zebin
Li, Yutong
Lin, Mingwei
Carlucho, Ignacio
Petillot, Yvan R.
Yang, Canjun
contents 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/.
format Preprint
id arxiv_https___arxiv_org_abs_2503_09203
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MarineGym: A High-Performance Reinforcement Learning Platform for Underwater Robotics
Chu, Shuguang
Huang, Zebin
Li, Yutong
Lin, Mingwei
Carlucho, Ignacio
Petillot, Yvan R.
Yang, Canjun
Robotics
Machine Learning
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/.
title MarineGym: A High-Performance Reinforcement Learning Platform for Underwater Robotics
topic Robotics
Machine Learning
url https://arxiv.org/abs/2503.09203