Saved in:
Bibliographic Details
Main Authors: Walke, Homer, Black, Kevin, Lee, Abraham, Kim, Moo Jin, Du, Max, Zheng, Chongyi, Zhao, Tony, Hansen-Estruch, Philippe, Vuong, Quan, He, Andre, Myers, Vivek, Fang, Kuan, Finn, Chelsea, Levine, Sergey
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2308.12952
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913199296610304
author Walke, Homer
Black, Kevin
Lee, Abraham
Kim, Moo Jin
Du, Max
Zheng, Chongyi
Zhao, Tony
Hansen-Estruch, Philippe
Vuong, Quan
He, Andre
Myers, Vivek
Fang, Kuan
Finn, Chelsea
Levine, Sergey
author_facet Walke, Homer
Black, Kevin
Lee, Abraham
Kim, Moo Jin
Du, Max
Zheng, Chongyi
Zhao, Tony
Hansen-Estruch, Philippe
Vuong, Quan
He, Andre
Myers, Vivek
Fang, Kuan
Finn, Chelsea
Levine, Sergey
contents We introduce BridgeData V2, a large and diverse dataset of robotic manipulation behaviors designed to facilitate research on scalable robot learning. BridgeData V2 contains 60,096 trajectories collected across 24 environments on a publicly available low-cost robot. BridgeData V2 provides extensive task and environment variability, leading to skills that can generalize across environments, domains, and institutions, making the dataset a useful resource for a broad range of researchers. Additionally, the dataset is compatible with a wide variety of open-vocabulary, multi-task learning methods conditioned on goal images or natural language instructions. In our experiments, we train 6 state-of-the-art imitation learning and offline reinforcement learning methods on our dataset, and find that they succeed on a suite of tasks requiring varying amounts of generalization. We also demonstrate that the performance of these methods improves with more data and higher capacity models, and that training on a greater variety of skills leads to improved generalization. By publicly sharing BridgeData V2 and our pre-trained models, we aim to accelerate research in scalable robot learning methods. Project page at https://rail-berkeley.github.io/bridgedata
format Preprint
id arxiv_https___arxiv_org_abs_2308_12952
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle BridgeData V2: A Dataset for Robot Learning at Scale
Walke, Homer
Black, Kevin
Lee, Abraham
Kim, Moo Jin
Du, Max
Zheng, Chongyi
Zhao, Tony
Hansen-Estruch, Philippe
Vuong, Quan
He, Andre
Myers, Vivek
Fang, Kuan
Finn, Chelsea
Levine, Sergey
Robotics
Machine Learning
We introduce BridgeData V2, a large and diverse dataset of robotic manipulation behaviors designed to facilitate research on scalable robot learning. BridgeData V2 contains 60,096 trajectories collected across 24 environments on a publicly available low-cost robot. BridgeData V2 provides extensive task and environment variability, leading to skills that can generalize across environments, domains, and institutions, making the dataset a useful resource for a broad range of researchers. Additionally, the dataset is compatible with a wide variety of open-vocabulary, multi-task learning methods conditioned on goal images or natural language instructions. In our experiments, we train 6 state-of-the-art imitation learning and offline reinforcement learning methods on our dataset, and find that they succeed on a suite of tasks requiring varying amounts of generalization. We also demonstrate that the performance of these methods improves with more data and higher capacity models, and that training on a greater variety of skills leads to improved generalization. By publicly sharing BridgeData V2 and our pre-trained models, we aim to accelerate research in scalable robot learning methods. Project page at https://rail-berkeley.github.io/bridgedata
title BridgeData V2: A Dataset for Robot Learning at Scale
topic Robotics
Machine Learning
url https://arxiv.org/abs/2308.12952