G1 Wbt Inspire Pick Up Drinks: Robot End-Effector and Hand State Data

A dataset of robot state observations, last updated on 2026-05-06. It contains end-effector states computed via forward kinematics and finger states for both hands, with data provided by unitreerobotics. The dataset appears to be related to a pick-up task involving drinks.

Use Cases

• Training reinforcement learning agents for robotic manipulation based on end-effector state observations.
• Developing and validating forward kinematics models based on the provided end-effector pose data.
• Analyzing hand-object interaction dynamics for grasping tasks based on the finger state data.
• Benchmarking robotic control policies for pick-and-place operations based on the described task context.

Strengths

• Includes end-effector states computed via forward kinematics, incorporating waist contribution.
• Provides finger state data for both hands, with dimensionality specified by hand type.
• Last updated timestamp of 2026-05-06 suggests 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.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

unitreerobotics on Hugging Face

Collection Method

Likely collected from robot sensors during task execution.

Time Range

null

Freshness

Last updated 2026-05-06 10:12:06.

Geography

null