Saved in:
Bibliographic Details
Main Authors: Bauza, Maria, Chen, Jose Enrique, Dalibard, Valentin, Gileadi, Nimrod, Hafner, Roland, Martins, Murilo F., Moore, Joss, Pevceviciute, Rugile, Laurens, Antoine, Rao, Dushyant, Zambelli, Martina, Riedmiller, Martin, Scholz, Jon, Bousmalis, Konstantinos, Nori, Francesco, Heess, Nicolas
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.06613
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929499172503552
author Bauza, Maria
Chen, Jose Enrique
Dalibard, Valentin
Gileadi, Nimrod
Hafner, Roland
Martins, Murilo F.
Moore, Joss
Pevceviciute, Rugile
Laurens, Antoine
Rao, Dushyant
Zambelli, Martina
Riedmiller, Martin
Scholz, Jon
Bousmalis, Konstantinos
Nori, Francesco
Heess, Nicolas
author_facet Bauza, Maria
Chen, Jose Enrique
Dalibard, Valentin
Gileadi, Nimrod
Hafner, Roland
Martins, Murilo F.
Moore, Joss
Pevceviciute, Rugile
Laurens, Antoine
Rao, Dushyant
Zambelli, Martina
Riedmiller, Martin
Scholz, Jon
Bousmalis, Konstantinos
Nori, Francesco
Heess, Nicolas
contents We present DemoStart, a novel auto-curriculum reinforcement learning method capable of learning complex manipulation behaviors on an arm equipped with a three-fingered robotic hand, from only a sparse reward and a handful of demonstrations in simulation. Learning from simulation drastically reduces the development cycle of behavior generation, and domain randomization techniques are leveraged to achieve successful zero-shot sim-to-real transfer. Transferred policies are learned directly from raw pixels from multiple cameras and robot proprioception. Our approach outperforms policies learned from demonstrations on the real robot and requires 100 times fewer demonstrations, collected in simulation. More details and videos in https://sites.google.com/view/demostart.
format Preprint
id arxiv_https___arxiv_org_abs_2409_06613
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DemoStart: Demonstration-led auto-curriculum applied to sim-to-real with multi-fingered robots
Bauza, Maria
Chen, Jose Enrique
Dalibard, Valentin
Gileadi, Nimrod
Hafner, Roland
Martins, Murilo F.
Moore, Joss
Pevceviciute, Rugile
Laurens, Antoine
Rao, Dushyant
Zambelli, Martina
Riedmiller, Martin
Scholz, Jon
Bousmalis, Konstantinos
Nori, Francesco
Heess, Nicolas
Robotics
Machine Learning
We present DemoStart, a novel auto-curriculum reinforcement learning method capable of learning complex manipulation behaviors on an arm equipped with a three-fingered robotic hand, from only a sparse reward and a handful of demonstrations in simulation. Learning from simulation drastically reduces the development cycle of behavior generation, and domain randomization techniques are leveraged to achieve successful zero-shot sim-to-real transfer. Transferred policies are learned directly from raw pixels from multiple cameras and robot proprioception. Our approach outperforms policies learned from demonstrations on the real robot and requires 100 times fewer demonstrations, collected in simulation. More details and videos in https://sites.google.com/view/demostart.
title DemoStart: Demonstration-led auto-curriculum applied to sim-to-real with multi-fingered robots
topic Robotics
Machine Learning
url https://arxiv.org/abs/2409.06613