Sign in to view source links and access this dataset
Description
16 episodes of robot teleoperation data for a 'Move To Tube Rack' task, created using LeRobot. The dataset contains 4,905 video frames recorded at 30 frames per second, totaling approximately 500 MB of video files and 100 MB of structured data files. It was authored by gistailab and last updated on July 6, 2026.
Use Cases
Training imitation learning models based on the 16 teleoperated demonstration episodes.
Developing visual servoing policies based on the 4,905 frames of video data.
Benchmarking robot manipulation performance on a 'Move To Tube Rack' task.
Analyzing robot state-action trajectories likely contained in the structured data files.
Strengths
Contains 4,905 video frames providing dense visual context for the task.
Includes 16 distinct demonstration episodes for learning from multiple examples.
Data is structured into manageable chunks of 1,000 frames each.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count for the structured data is unknown, which may limit suitability assessment.
The dataset is limited to a single task, which may restrict generalizability.
Provenance
Source
gistailab via Hugging Face.
Collection Method
Created using LeRobot, likely involving human teleoperation of a mobileai_robot.
Freshness
Last updated 2026-07-06 12:34:12; freshness should be verified.
License is unknown; terms of use must be verified before application.