Dexterous Grasping with Real-World Robotic Reinforcement Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Huang, Dongchi, Zhang, Tianle, Li, Yihang, Zhao, Ling, Li, Jiayi, Fang, Zhirui, Xia, Chunhe, He, Xiaodong
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918145252392960
author Huang, Dongchi
Zhang, Tianle
Li, Yihang
Zhao, Ling
Li, Jiayi
Fang, Zhirui
Xia, Chunhe
He, Xiaodong
author_facet Huang, Dongchi
Zhang, Tianle
Li, Yihang
Zhao, Ling
Li, Jiayi
Fang, Zhirui
Xia, Chunhe
He, Xiaodong
contents Dexterous grasping in the real world presents a fundamental and significant challenge for robot learning. The ability to employ affordance-aware poses to grasp objects with diverse geometries and properties in arbitrary scenarios is essential for general-purpose robots. However, existing research predominantly addresses dexterous grasping problems within simulators, which encounter difficulties when applied in real-world environments due to the domain gap between reality and simulation. This limitation hinders their generalizability and practicality in real-world applications. In this paper, we present DexGraspRL, a reinforcement learning (RL) framework that directly trains robots in real-world environments to acquire dexterous grasping skills. Specifically, DexGraspRL consists of two stages: (i) a pretraining stage that pretrains the policy using imitation learning (IL) with a limited set of expert demonstrations; (ii) a fine-tuning stage that refines the policy through direct RL in real-world scenarios. To mitigate the catastrophic forgetting phenomenon arising from the distribution shift between demonstrations and real-world environments, we design a regularization term that balances the exploitation of RL with the preservation of the pretrained policy. Our experiments with real-world tasks demonstrate that DexGraspRL successfully accomplishes diverse dexterous grasping tasks, achieving an average success rate of nearly 92%. Furthermore, by fine-tuning with RL, our method uncovers novel policies, surpassing the IL policy with a 23% reduction in average cycle time.
format Preprint
id arxiv_https___arxiv_org_abs_2503_04014
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dexterous Grasping with Real-World Robotic Reinforcement Learning
Huang, Dongchi
Zhang, Tianle
Li, Yihang
Zhao, Ling
Li, Jiayi
Fang, Zhirui
Xia, Chunhe
He, Xiaodong
Robotics
Dexterous grasping in the real world presents a fundamental and significant challenge for robot learning. The ability to employ affordance-aware poses to grasp objects with diverse geometries and properties in arbitrary scenarios is essential for general-purpose robots. However, existing research predominantly addresses dexterous grasping problems within simulators, which encounter difficulties when applied in real-world environments due to the domain gap between reality and simulation. This limitation hinders their generalizability and practicality in real-world applications. In this paper, we present DexGraspRL, a reinforcement learning (RL) framework that directly trains robots in real-world environments to acquire dexterous grasping skills. Specifically, DexGraspRL consists of two stages: (i) a pretraining stage that pretrains the policy using imitation learning (IL) with a limited set of expert demonstrations; (ii) a fine-tuning stage that refines the policy through direct RL in real-world scenarios. To mitigate the catastrophic forgetting phenomenon arising from the distribution shift between demonstrations and real-world environments, we design a regularization term that balances the exploitation of RL with the preservation of the pretrained policy. Our experiments with real-world tasks demonstrate that DexGraspRL successfully accomplishes diverse dexterous grasping tasks, achieving an average success rate of nearly 92%. Furthermore, by fine-tuning with RL, our method uncovers novel policies, surpassing the IL policy with a 23% reduction in average cycle time.
title Dexterous Grasping with Real-World Robotic Reinforcement Learning
topic Robotics
url https://arxiv.org/abs/2503.04014