Kine2Go: Kinematic Motions for the Unitree Go2 Quadruped Robot

MIMUW-Robotics created a kinematic motion dataset for the Unitree Go2 quadruped robot. Forty reference clips from dog, horse, and synthetic robot motions were retargeted to the robot's morphology. The dataset, last updated in June 2026, includes per-clip imitation-learning policies and rendered video rollouts.

Use Cases

• Training legged locomotion policies based on the provided imitation-learning (PPO) policies.
• Regularizing behavioral foundation models using the diverse set of gaits and motions.
• Benchmarking robot control algorithms against the 20 perturbed rollouts per clip.
• Studying motion retargeting from biological and synthetic sources to a specific robot morphology.

Strengths

• Includes 40 distinct reference clips retargeted to the robot.
• Pairs each clip with a trained imitation-learning policy.
• Provides 20 perturbed rollouts per clip for analysis.
• Includes rendered video outputs for visual inspection.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2026-06-13 06:43:24; freshness should be verified.

Provenance

Source

MIMUW-Robotics

Collection Method

Reference clips from dog, horse, and synthetic motions were retargeted to the Unitree Go2 robot.

Freshness

Last updated 2026-06-13 06:43:24.