Robot observation data for a bed-making task, published by unitreerobotics on Hugging Face. The dataset includes end-effector states computed via forward kinematics and finger states for both hands. It was last updated on March 27, 2026.
Use Cases
- Training robot control policies based on end-effector state observations.
- Analyzing robot hand kinematics and finger actuation patterns.
- Benchmarking robot performance on domestic manipulation tasks like bed-making.
- Developing forward kinematics models from robot root to end-effectors.
Strengths
- Includes end-effector states computed via forward kinematics, incorporating waist contribution.
- Contains hand state data with dimensionality dependent on hand type, such as the Inspire Hand.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count, file formats, and license are unknown, which may limit suitability assessment.
Provenance
- Source
- unitreerobotics on Hugging Face
- Collection Method
- Likely collected from robot sensor observations during task execution.
- Time Range
- null
- Freshness
- Last updated 2026-03-27 12:05:42; freshness should be verified.
- Geography
- null