24/7 Emergencies | +352 691 260 215
SMS, WhatsApp, Telegram, Call
Emergency response and post-incident inspection. We operate throughout Luxembourg and within a 50 km radius beyond our borders.
The Unitree G1-D is an end-to-end platform for humanoid robot data and training. It combines a high-performance humanoid robot platform with streamlined data acquisition tools and comprehensive model training and inference tools. The platform supports workflows for data acquisition, processing, labeling, review, data asset management, distributed training, custom model development, and deployment. It is available in G1-D Standard and G1-D Flagship configurations.
| Spec | Value |
| Models - Available configurations | G1-D Standard; G1-D Flagship |
| Dimensions - Overall dimensions at minimum column height, G1-D Standard | 1260 x 500 x 500 mm |
| Dimensions - Overall dimensions at minimum column height, G1-D Flagship | 1260 x 525 x 570 mm |
| Dimensions - Overall dimensions at maximum column height, G1-D Standard | 1680 x 500 x 500 mm |
| Dimensions - Overall dimensions at maximum column height, G1-D Flagship | 1680 x 525 x 570 mm |
| Mechanical - Total weight including battery, G1-D Standard | Approx. 50 kg |
| Mechanical - Total weight including battery, G1-D Flagship | Approx. 80 kg |
| Degrees of Freedom - Total DOF excluding end effector, G1-D Standard | 17 |
| Degrees of Freedom - Total DOF excluding end effector, G1-D Flagship | 19 |
| Degrees of Freedom - Single arm DOF excluding end effector | 7 |
| Degrees of Freedom - Arm degrees of freedom | 7 x 2 |
| Degrees of Freedom - Waist degrees of freedom | 2 |
| Degrees of Freedom - Column degrees of freedom | 1 |
| Degrees of Freedom - Base degrees of freedom | 2 |
| Arm - Maximum single arm payload | Approx. 3 kg |
| Arm - Maximum single arm payload note | The maximum load of the arm varies greatly under different arm extension postures. |
| End Effector - End effector options | Optional 2-finger gripper; 3-finger dexterous hand without tactile; 3-finger dexterous hand with tactile; 5-finger dexterous hand |
| Waist - Waist joint range of motion | Z-axis: ±155°; Y-axis: -2.5° to +135° |
| Column - Column lifting speed | Approx. 60 mm/s |
| Workspace - Vertical workspace | 0 - 2 m |
| Mobile Operation - Mobile lifting design | Combines wheels and lifting mechanisms |
| Baseplate - Maximum mobility speed, G1-D Flagship | 1.5 m/s |
| Baseplate - Chassis drive type, G1-D Flagship | Differential drive, supports 360° in-place rotation |
| Baseplate - Chassis sensors, G1-D Flagship | LiDAR x1; depth camera x2; physical collision sensor x2; low-obstacle detection sensor x2 |
| Computing - Basic computing power | 8-core high-performance CPU |
| Computing - High computing power module | NVIDIA Jetson Orin NX 16GB (100 TOPS) |
| Sensors - Perception sensors | Head HD binocular camera x1; wrist HD camera x2 |
| Connectivity - Wi-Fi 6 | Yes |
| Connectivity - Bluetooth 5.2 | Yes |
| Battery - Battery, G1-D Standard | Upper body battery, quick-release: 9 Ah |
| Battery - Battery, G1-D Flagship | Chassis battery, built-in: 30 Ah |
| Battery Life - Battery life, G1-D Standard | Approx. 2 hours |
| Battery Life - Battery life, G1-D Flagship | Approx. 6 hours |
| Control - Lifting accuracy | ±0.5 mm |
| Control - End-effector gripper accuracy | ±0.1 mm |
| Control - Accuracy note | Accuracy varies with different end-effector configurations. |
| Control - System teleoperation latency | < 100 ms |
| Control - Sampling rate | 60 Hz |
| Accessories - Manual controller | Yes |
| Accessories - Visualization computer | Yes |
| Software - Upgraded intelligent OTA | Yes |
| Software - Secondary development | Yes |
| Data Acquisition Pipeline - Pipeline steps | Creating new collection tasks; task editing and assignment; data acquisition and annotation; data upload and review; data storage; data export |
| Training and Inference Pipeline - Pipeline steps | Model architecture selection; training configuration; real-time monitoring; parameter editing; simulation testing; model export and deployment |
| Training - Distributed training GPU utilization | Up to 90% GPU utilization |