GNFactor: Multi-Task Real Robot Learning with Generalizable Neural Feature Fields

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ze, Yanjie, Yan, Ge, Wu, Yueh-Hua, Macaluso, Annabella, Ge, Yuying, Ye, Jianglong, Hansen, Nicklas, Li, Li Erran, Wang, Xiaolong
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913447836385280
author Ze, Yanjie
Yan, Ge
Wu, Yueh-Hua
Macaluso, Annabella
Ge, Yuying
Ye, Jianglong
Hansen, Nicklas
Li, Li Erran
Wang, Xiaolong
author_facet Ze, Yanjie
Yan, Ge
Wu, Yueh-Hua
Macaluso, Annabella
Ge, Yuying
Ye, Jianglong
Hansen, Nicklas
Li, Li Erran
Wang, Xiaolong
contents It is a long-standing problem in robotics to develop agents capable of executing diverse manipulation tasks from visual observations in unstructured real-world environments. To achieve this goal, the robot needs to have a comprehensive understanding of the 3D structure and semantics of the scene. In this work, we present $\textbf{GNFactor}$, a visual behavior cloning agent for multi-task robotic manipulation with $\textbf{G}$eneralizable $\textbf{N}$eural feature $\textbf{F}$ields. GNFactor jointly optimizes a generalizable neural field (GNF) as a reconstruction module and a Perceiver Transformer as a decision-making module, leveraging a shared deep 3D voxel representation. To incorporate semantics in 3D, the reconstruction module utilizes a vision-language foundation model ($\textit{e.g.}$, Stable Diffusion) to distill rich semantic information into the deep 3D voxel. We evaluate GNFactor on 3 real robot tasks and perform detailed ablations on 10 RLBench tasks with a limited number of demonstrations. We observe a substantial improvement of GNFactor over current state-of-the-art methods in seen and unseen tasks, demonstrating the strong generalization ability of GNFactor. Our project website is https://yanjieze.com/GNFactor/ .
format Preprint
id arxiv_https___arxiv_org_abs_2308_16891
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle GNFactor: Multi-Task Real Robot Learning with Generalizable Neural Feature Fields
Ze, Yanjie
Yan, Ge
Wu, Yueh-Hua
Macaluso, Annabella
Ge, Yuying
Ye, Jianglong
Hansen, Nicklas
Li, Li Erran
Wang, Xiaolong
Robotics
Computer Vision and Pattern Recognition
Machine Learning
It is a long-standing problem in robotics to develop agents capable of executing diverse manipulation tasks from visual observations in unstructured real-world environments. To achieve this goal, the robot needs to have a comprehensive understanding of the 3D structure and semantics of the scene. In this work, we present $\textbf{GNFactor}$, a visual behavior cloning agent for multi-task robotic manipulation with $\textbf{G}$eneralizable $\textbf{N}$eural feature $\textbf{F}$ields. GNFactor jointly optimizes a generalizable neural field (GNF) as a reconstruction module and a Perceiver Transformer as a decision-making module, leveraging a shared deep 3D voxel representation. To incorporate semantics in 3D, the reconstruction module utilizes a vision-language foundation model ($\textit{e.g.}$, Stable Diffusion) to distill rich semantic information into the deep 3D voxel. We evaluate GNFactor on 3 real robot tasks and perform detailed ablations on 10 RLBench tasks with a limited number of demonstrations. We observe a substantial improvement of GNFactor over current state-of-the-art methods in seen and unseen tasks, demonstrating the strong generalization ability of GNFactor. Our project website is https://yanjieze.com/GNFactor/ .
title GNFactor: Multi-Task Real Robot Learning with Generalizable Neural Feature Fields
topic Robotics
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2308.16891