G1 WBT Brainco: Robot End-Effector and Hand State Data for Dishwasher Loading

Unitreerobotics provides a dataset of robot state observations for a dishwasher loading task. The data includes end-effector states computed via forward kinematics and hand/finger states for both robot hands. The dataset was last updated on March 27, 2026.

Use Cases

• Train reinforcement learning agents for bimanual manipulation based on end-effector and hand state observations.
• Benchmark robot state estimation methods using the forward kinematics-derived end-effector poses.
• Analyze grasp dynamics and hand-object interaction based on the finger state data.
• Develop sim-to-real transfer models for dishwasher loading tasks using the recorded state sequences.

Strengths

• Includes 12-dimensional end-effector state vectors for both arms.
• Contains hand/finger state data with dimensionality specified by hand type.
• Last updated on 2026-03-27, suggesting recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and sample data are unknown, which may limit suitability assessment.
• Data may reflect temporal or task-specific bias inherent to the single demonstration scenario.

Provenance

Source

unitreerobotics on Hugging Face

Collection Method

Likely recorded from a physical or simulated robot performing a specific task.

Time Range

null

Freshness

Last updated 2026-03-27 12:04:41.

Geography

null