Unitree Open-Sources 500-Hour Whole-Body Household Manipulation Dataset for G1 Humanoid
Covers 28 everyday household tasks—including opening doors, wiping tables, pouring liquids, and bimanual tote carrying—with synchronized wrist tactile and joint torque telemetry.
⚡ Key Engineering & Supply-Chain Takeaways
- Unitree Robotics open-sourced G1-Home-500H, a 500-hour whole-body loco-manipulation dataset recorded on 43-DoF Unitree G1 EDU humanoids.
- Includes synchronized 50Hz joint positions/torques, 94-taxel Dex3-1 tactile arrays, Livox Mid-360 LiDAR point clouds, and stereo wrist RGB-D streams.
- Formatted natively in Hugging Face LeRobot Parquet v2 and RLDS schemas for immediate VLA pre-training.
🔬 Technical & Supply-Chain Specification Matrix
| Dataset Volume | 500 Hours / 72,000 Trajectories Across 28 Household & Logistics Tasks |
|---|---|
| Robot Platform | Unitree G1 EDU (43-DoF) + Dual Dex3-1 Tactile 3-Finger Hands |
| Sensor Modalities | Stereo RGB-D + Dual Wrist Cameras + 94 Force Taxels/Hand + Joint Torque |
| License & Format | CC-BY-4.0 · Native LeRobot v0.4 Parquet & Open-X RLDS Compatible |
Bridging the Whole-Body Loco-Manipulation Data Gap
While tabletop robotic arm datasets have multiplied over the past two years, open-source datasets that couple bipedal lower-body balance and squatting locomotion with bimanual upper-body dexterous manipulation have remained scarce.
Unitree Robotics has released G1-Home-500H, a 500-hour multi-sensor dataset collected across 12 physical apartment and warehouse mockups in Hangzhou using fleets of 43-DoF Unitree G1 EDU humanoids equipped with Dex3-1 tactile hands.
28 Contact-Rich Tasks With Synchronized Tactile & Torque Telemetry
What makes G1-Home-500H particularly valuable for Vision-Language-Action (VLA) foundation model training is its high-frequency force-feedback recording:
- Whole-Body Coordination: Tasks include squatting to lift 12kg storage totes from the floor, opening spring-loaded refrigerator doors while holding a tray, wiping kitchen counters, and pouring liquids between containers.
- Tactile + Proprioceptive Alignment: Every trajectory logs 500Hz motor phase current torque estimates alongside 100Hz 3-axis normal/shear force readings from all 94 tactile taxels per hand, enabling policies to learn slip prevention and compliant force control.
🔗 Related AI Hardware in RECIWO Database
All 38 Devices →Unitree G1 Humanoid Robot (2026 Developer Edition)
Features 23 to 43 high-torque joint degrees of freedom, 3D LiDAR + depth vision perception, and end-to-end reinforcement learning & VLA manipulation policies.
Booster T1 Soccer & Manipulation Humanoid (Dev Kit)
1.2-meter ultra-rugged research humanoid built for high-impact dynamic locomotion, RoboCup autonomous soccer, and whole-body sim-to-real reinforcement learning.
Unitree Go2-W Wheeled-Legged Embodied AI Dog
16-DoF hybrid quadruped equipped with pneumatic hub-motor wheels on all four knees, climbing stairs like a legged dog and cruising flat pavement at 2.5 m/s.
📚 Related Deep-Dive Engineering Teardowns
Browse All 83 Articles →Booster T1 ($29,900) vs. Unitree G1 ($16,000) Humanoid Developer Shootout: Fall Survival, AGX Orin Compute & URDF SDKs
RECIWO Labs compared the two most popular mid-sized research humanoids in university and corporate robotics labs: the 29.5kg Booster T1 and the 35kg Unitree G1 Developer Edition.
Quantizing Vision-Language-Action (VLA) Models on NVIDIA Jetson Orin NX Super: FP8 vs. INT4 Control Loop Jitter
RECIWO Labs benchmarked OpenVLA-7B, Pi-0-Fast, and ACT policies on the 157 TOPS Jetson Orin NX Super, measuring action-chunk inference latency and grasp precision across FP16, FP8, and AWQ INT4.
Sim-to-Real Reinforcement Learning on Unitree G1: Domain Randomization & Actuator Net Calibration in Isaac Lab
RECIWO Labs trained 4,096 parallel Unitree G1 humanoid agents in NVIDIA Isaac Lab 2.2, detailing the exact domain randomization parameters required for zero-shot outdoor stairs and gravel locomotion.
