← Sinan Lab › All datasets › BridgeData V2
Dataset · UC Berkeley RAIL

BridgeData V2

Landing page →

Paper abstract: 60,096 trajectories across 24 environments on a low-cost robot. Download size and depth-camera model not stated on fetched page.

Facts

Organisation
UC Berkeley RAIL
Form tags
single arm
Embodiments (as stated)
WidowX 250 6DOF robot arm
Episodes
60,096
Hours
unverified
Frames
unverified
Tasks
13 skills
Scenes
24
Modalities
rgb · depth · language
Format
raw (JPEG images, zip downloads)
Licence
Creative Commons Attribution 4.0 International License
Commercial use
allowed (stated)
Released
2023-08
Size
unverified
Sources · 13 sources
name https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'WidowX 250 6DOF robot arm'
org https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'WidowX 250 6DOF robot arm'
url https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'WidowX 250 6DOF robot arm'
embodiments https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'WidowX 250 6DOF robot arm'
form_tags https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'WidowX 250 6DOF robot arm'
episodes https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet '60,096 trajectories' = '50,365 teleoperated demonstrations' + '9,731 rollouts from a scripted pick-and-place policy'
tasks https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet '13 skills' across '24 environments'
scenes https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet '13 skills' across '24 environments'
modalities https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet Fixed over-the-shoulder RGBD camera, two randomized RGB cameras, wrist RGB camera; 'Each trajectory is labeled with a natural langauge instruction'
format https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'stored as JPEGS'; teleop and scripted data provided as separate zip files
license https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'All data is provided under the Creative Commons Attribution 4.0 International License'
commercial_ok https://rail-berkeley.github.io/bridgedata/ Official page · Fetched 2026-09-18 · No snapshot stored yet 'All data is provided under the Creative Commons Attribution 4.0 International License'
release_date https://arxiv.org/abs/2308.12952 Official page · Fetched 2026-09-18 · No snapshot stored yet arXiv v1 2023-08-24 (v3 2024-01-17)

Related bodies in the index

Matched by stated robot names or form; a match means the form is covered, not that a model was trained on it.

SinanLab · Sinan LabClear direction, grounded judgement.