RM-RL: Role-Model Reinforcement Learning for Precise Robot Manipulation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Chen, Xiangyu, Zhou, Chuhao, Liu, Yuxi, Yang, Jianfei
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914323814678528
author Chen, Xiangyu
Zhou, Chuhao
Liu, Yuxi
Yang, Jianfei
author_facet Chen, Xiangyu
Zhou, Chuhao
Liu, Yuxi
Yang, Jianfei
contents Precise robot manipulation is critical for fine-grained applications such as chemical and biological experiments, where even small errors (e.g., reagent spillage) can invalidate an entire task. Existing approaches often rely on pre-collected expert demonstrations and train policies via imitation learning (IL) or offline reinforcement learning (RL). However, obtaining high-quality demonstrations for precision tasks is difficult and time-consuming, while offline RL commonly suffers from distribution shifts and low data efficiency. We introduce a Role-Model Reinforcement Learning (RM-RL) framework that unifies online and offline training in real-world environments. The key idea is a role-model strategy that automatically generates labels for online training data using approximately optimal actions, eliminating the need for human demonstrations. RM-RL reformulates policy learning as supervised training, reducing instability from distribution mismatch and improving efficiency. A hybrid training scheme further leverages online role-model data for offline reuse, enhancing data efficiency through repeated sampling. Extensive experiments show that RM-RL converges faster and more stably than existing RL methods, yielding significant gains in real-world manipulation: 53% improvement in translation accuracy and 20% in rotation accuracy. Finally, we demonstrate the successful execution of a challenging task, precisely placing a cell plate onto a shelf, highlighting the framework's effectiveness where prior methods fail.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15189
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RM-RL: Role-Model Reinforcement Learning for Precise Robot Manipulation
Chen, Xiangyu
Zhou, Chuhao
Liu, Yuxi
Yang, Jianfei
Robotics
Precise robot manipulation is critical for fine-grained applications such as chemical and biological experiments, where even small errors (e.g., reagent spillage) can invalidate an entire task. Existing approaches often rely on pre-collected expert demonstrations and train policies via imitation learning (IL) or offline reinforcement learning (RL). However, obtaining high-quality demonstrations for precision tasks is difficult and time-consuming, while offline RL commonly suffers from distribution shifts and low data efficiency. We introduce a Role-Model Reinforcement Learning (RM-RL) framework that unifies online and offline training in real-world environments. The key idea is a role-model strategy that automatically generates labels for online training data using approximately optimal actions, eliminating the need for human demonstrations. RM-RL reformulates policy learning as supervised training, reducing instability from distribution mismatch and improving efficiency. A hybrid training scheme further leverages online role-model data for offline reuse, enhancing data efficiency through repeated sampling. Extensive experiments show that RM-RL converges faster and more stably than existing RL methods, yielding significant gains in real-world manipulation: 53% improvement in translation accuracy and 20% in rotation accuracy. Finally, we demonstrate the successful execution of a challenging task, precisely placing a cell plate onto a shelf, highlighting the framework's effectiveness where prior methods fail.
title RM-RL: Role-Model Reinforcement Learning for Precise Robot Manipulation
topic Robotics
url https://arxiv.org/abs/2510.15189