Robo-DM: Data Management For Large Robot Datasets

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Kaiyuan, Fu, Letian, Huang, David, Zhang, Yanxiang, Chen, Lawrence Yunliang, Huang, Huang, Hari, Kush, Balakrishna, Ashwin, Xiao, Ted, Sanketi, Pannag R, Kubiatowicz, John, Goldberg, Ken
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909618756648960
author Chen, Kaiyuan
Fu, Letian
Huang, David
Zhang, Yanxiang
Chen, Lawrence Yunliang
Huang, Huang
Hari, Kush
Balakrishna, Ashwin
Xiao, Ted
Sanketi, Pannag R
Kubiatowicz, John
Goldberg, Ken
author_facet Chen, Kaiyuan
Fu, Letian
Huang, David
Zhang, Yanxiang
Chen, Lawrence Yunliang
Huang, Huang
Hari, Kush
Balakrishna, Ashwin
Xiao, Ted
Sanketi, Pannag R
Kubiatowicz, John
Goldberg, Ken
contents Recent results suggest that very large datasets of teleoperated robot demonstrations can be used to train transformer-based models that have the potential to generalize to new scenes, robots, and tasks. However, curating, distributing, and loading large datasets of robot trajectories, which typically consist of video, textual, and numerical modalities - including streams from multiple cameras - remains challenging. We propose Robo-DM, an efficient open-source cloud-based data management toolkit for collecting, sharing, and learning with robot data. With Robo-DM, robot datasets are stored in a self-contained format with Extensible Binary Meta Language (EBML). Robo-DM can significantly reduce the size of robot trajectory data, transfer costs, and data load time during training. Compared to the RLDS format used in OXE datasets, Robo-DM's compression saves space by up to 70x (lossy) and 3.5x (lossless). Robo-DM also accelerates data retrieval by load-balancing video decoding with memory-mapped decoding caches. Compared to LeRobot, a framework that also uses lossy video compression, Robo-DM is up to 50x faster when decoding sequentially. We physically evaluate a model trained by Robo-DM with lossy compression, a pick-and-place task, and In-Context Robot Transformer. Robo-DM uses 75x compression of the original dataset and does not suffer reduction in downstream task accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15558
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Robo-DM: Data Management For Large Robot Datasets
Chen, Kaiyuan
Fu, Letian
Huang, David
Zhang, Yanxiang
Chen, Lawrence Yunliang
Huang, Huang
Hari, Kush
Balakrishna, Ashwin
Xiao, Ted
Sanketi, Pannag R
Kubiatowicz, John
Goldberg, Ken
Robotics
Artificial Intelligence
Databases
Machine Learning
Recent results suggest that very large datasets of teleoperated robot demonstrations can be used to train transformer-based models that have the potential to generalize to new scenes, robots, and tasks. However, curating, distributing, and loading large datasets of robot trajectories, which typically consist of video, textual, and numerical modalities - including streams from multiple cameras - remains challenging. We propose Robo-DM, an efficient open-source cloud-based data management toolkit for collecting, sharing, and learning with robot data. With Robo-DM, robot datasets are stored in a self-contained format with Extensible Binary Meta Language (EBML). Robo-DM can significantly reduce the size of robot trajectory data, transfer costs, and data load time during training. Compared to the RLDS format used in OXE datasets, Robo-DM's compression saves space by up to 70x (lossy) and 3.5x (lossless). Robo-DM also accelerates data retrieval by load-balancing video decoding with memory-mapped decoding caches. Compared to LeRobot, a framework that also uses lossy video compression, Robo-DM is up to 50x faster when decoding sequentially. We physically evaluate a model trained by Robo-DM with lossy compression, a pick-and-place task, and In-Context Robot Transformer. Robo-DM uses 75x compression of the original dataset and does not suffer reduction in downstream task accuracy.
title Robo-DM: Data Management For Large Robot Datasets
topic Robotics
Artificial Intelligence
Databases
Machine Learning
url https://arxiv.org/abs/2505.15558