No episode media available here

This is a representative still, not playable footage. Episode video for this dataset lives with its original source.

CALVIN

Name: CALVIN
Creator: Simulation
License: Open Source

Long-horizon language instruction chains

Indexed from an external sourceSimulationUpdated 2022-04-06Open Source

Published by

Freiburg Robot Learning Lab

RobotData indexes this dataset's metadata. Attribution and licensing stay with the original publisher.

Composing Actions from Language and Vision: four simulated environments (A/B/C/D) with 24 hours of teleoperated play and 20,000 language annotations. The standard benchmark for chaining five instructions in a row with zero-shot environment transfer.

Indexed episode structure

Per-episode numbers RobotData has indexed locally (camera count, DoF, control rate, success rate). "Not indexed yet" means we have not recorded that number — it does not mean the publisher's data lacks observations, state or actions. See the source-backed profile above for what the publisher documents.

Observation

2 cameras

State

7 DoF

Action

30 Hz

Result

Not indexed yet

Data modalities

● RGB Video● Depth● Joint Positions● Robot State● Actions○ Force/Torque○ Audio● Language Instructions

Episode explorer

No episode media has been added for this dataset, so there is nothing to play. The explorer appears as soon as real clips exist — either uploaded here or extracted from the source.

Contribute episodes