Unitree
excl. VAT
Couldn't load pickup availability
HangZhou YuShu TECHNOLOGY CO., LTD
https://www.unitree.com/
sales_zty@unitree.com
Building 1, No.88 DongLiu Road 88
Room 306, Hangzhou
China
Pricing Models
Compare options clearly, quickly, and without the fine print. Our recommendation is highlighted.
Flexible financing
Inclusive
Lowest price
Inclusive
Flexible terms
Inclusive
ㅤ
Switch between the tabs to view content.
Platform for hands-on robotics, AI, and mechatronics education: from introductory labs to master's and research projects involving humanoid locomotion and manipulation.
Testbed for new algorithms in perception, reinforcement learning, motion planning, and human-robot interaction – including data acquisition and benchmarking in real-world scenarios.
Modern training environment for prospective mechatronics engineers, programmers, and automation technicians, ranging from secure basic operation to custom applications and add-ons.
Ideal for teams that want to use humanoid robotics in challenges: rapid prototyping, scenarios such as grasping, balancing, navigation, or collaboration with other robots.
An eye-catcher for technology centers, museums, and events: the G1 EDU demonstrates modern AI and robotics functions live, making complex technologies tangible for visitors.
Development and testing of service robotics applications in controlled environments – such as simple transport, gripping, or interaction tasks in laboratories, on campuses, or in test areas.
The Unitree G1 EDU, available in 10 modular versions, is the professional solution for hardware-level capture, testing, and implementation of complex AI and motion algorithms in research and industry. With its integrated NVIDIA Orin NX computing unit and sophisticated sensor system, this humanoid robot is perfectly suited for demanding laboratory environments and industrial test fields where standard systems reach their performance limits. As an indispensable link between digital AI simulation and physical interaction, the EDU platform noticeably optimizes your research speed and the real-time control of complex manipulation tasks.
| Model / Variant | Degrees of Freedom (DOF) | Hand Type & Kinematics | Tactile Sensing | Features & Core Specification |
|---|---|---|---|---|
| G1 EDU U1 | 23 DOF | Standard End-effectors | None | Standard Version. NVIDIA Orin NX (100 TOPS), 120 Nm knee torque, 3 kg arm payload, programmable. |
| G1 EDU U2 | 29 DOF | Standard End-effectors | None | Advanced Version. Extended joint structure: Waist increased to 3 DOF, arms increased to 7 DOF each. |
| G1 EDU U3 | 43 DOF | 2× Dex3-1 (3-Finger) | None | Based on U2 (Advanced). Equipped with force-controlled 3-finger hands for precise object manipulation. |
| G1 EDU U4 | 43 DOF | 2× Dex3-1 (3-Finger) | Integrated Sensing | Based on U2 (Advanced). Force-controlled 3-finger hands combined with tactile surface feedback. |
| G1 EDU U5 | 41 DOF | 2× RH56DFQ-2 (5-Finger) | None | Based on U2 (Advanced). Upgrade to anthropomorphic 5-finger hands (model RH56DFQ-2R/L) for human grip scenarios. |
| G1 EDU U6 | 41 DOF | 2× RH56DFTP-2 (5-Finger) | 17 sensors per hand | Based on U2 (Advanced). High-end 5-finger dexterity hands with 17 highly sensitive tactile pressure points per hand. |
| G1 EDU U7 | 41 DOF | 2× Revo 2 Basic (5-Finger) | None | Based on U2 (Advanced). Integration of two five-fingered Revo 2 Basic robotic hands for optimized laboratory processes. |
| G1 EDU U8 | 37 DOF | 2× Dex3-1 (3-Finger) | None | Based on U1 (Standard). Combines basic joint structure with two force-controlled 3-finger hands (without tactile sensing). |
| G1 EDU U9 | 37 DOF | 2× Dex3-1 (3-Finger) | 33 sensors per hand | Based on U1 (Standard). Force-controlled 3-finger hands with maximum tactile matrix sensing (33 sensors per hand). |
| G1 EDU U10 | 35 DOF | 2× Revo 2 Basic (5-Finger) | None | Based on U1 (Standard). Combines standard joint kinematics with two five-fingered Revo 2 Basic hand models. |
| Physical Property | Details and Lab Measurements |
|---|---|
| Dimensions (Standing) | 1270 × 450 × 200 mm |
| Dimensions (Folded) | 690 × 450 × 300 mm |
| System Weight | approx. 35 kg |
| Joint Range of Motion | Hip (Pitch): ±154° | Knee: 0 to 165° | Waist (Z-axis): ±155° |
| Integrated Environmental Sensing | Intel RealSense depth camera + 3D LiDAR system |
| Connectivity & Wireless | WiFi 6 (802.11ax) & Bluetooth 5.2 |
Advantage for professionals: The Unitree G1 EDU, in its modular configurations, offers robotics engineers, AI developers, and academic institutions an unbeatable, standardized open-source platform for researching humanoid kinematics without performance bottlenecks. The integrated NVIDIA Orin NX board drastically shortens pure development and compilation time, as compute-intensive AI models are executed directly "on-the-edge" – minimizing latencies and significantly reducing development costs per project. Thanks to the flexible selection of sophisticated gripping systems and tactile sensors, the system meets complex laboratory manipulation requirements with zero loss, while the robust, foldable design effectively prevents costly transport damage.
Terra Robotics
https://terra-robotics.de/
mail@terra-robotics.de
Zeißstraße 10,
32584 Löhne
Germany
Dimensions: 1270 × 480 × 380 mm
Weight: approx. 35 kg
Number of axes (DOF): 23
Max. running speed: 1.8 m/s
Max. jump height: approx. 0.65 m
Battery: 2.4 kWh
Runtime: 2.5 – 3.5 h (depending on load and operation)
Charging time: approx. 1.5 h
Sensors: 3D LiDAR, depth camera and IMU
Manipulation: Two 3-finger force-position hands
Computing platform: Onboard 8-core CPU (with option for AI expansion, e.g., Jetson Orin NX)
Protection rating: IP54 (standard)
Temperature range: –10 °C to +45 °C
The Unitree G1 EDU utilizes advanced AI control for realistic movement sequences, learning processes, and interaction. Its 23 finely tuned joints work synchronously in real time, enabling fluid gestures, precise balance, and rapid response behavior. With its adaptive motion control, the G1 EDU independently adjusts to new environments and tasks – ideal for robotics research, AI training, and complex experiments.
Equipped with 3D LiDAR, a depth camera, and IMU sensors, the Unitree G1 EDU detects its surroundings in real time. It creates detailed spatial maps, analyzes movements, and reacts precisely to changes – whether for navigation, interaction, or learning tasks. The advanced sensor fusion enables adaptive responses and situational learning, allowing the G1 EDU to operate safely even in dynamic research environments.
The G1 EDU combines high-resolution sensor technology, adaptive motion algorithms, and learning-capable control into a perfectly coordinated unit. Its integrated cameras and sensors capture the environment and position in real time – for precise movements, natural gestures, and fluid interaction. Whether for research, robotics training, or industrial development: the G1 EDU impresses with consistent performance, adaptability, and reliable autonomy at the state-of-the-art level.
The Unitree G1 EDU combines precise sensor technology with adaptive control, optimizing movements in real time.
Thanks to 3D camera sensors, it precisely detects objects and people, reacts adaptively, and learns with foresight.
Ideal for research, education, and AI development – with the highest level of accuracy and efficiency in humanoid robotics training.
The Unitree G1 EDU is designed for academic and industrial research. Thanks to its ROS integration, SDK, and modular architecture, the G1 EDU can be flexibly adapted to projects—from motion control to autonomous interactions. Ideal for universities, laboratories, and development centers researching humanoid robotics at the highest level.
The Unitree G1 EDU humanoid robot is suitable for colleges, universities, research labs, AI labs, vocational training, technical schools, robotics competitions, showrooms, trade fairs, and controlled service or assistance scenarios. Terra Robotics positions the Unitree G1 EDU specifically for teaching, robotics research, AI development, humanoid locomotion, manipulation, and human-robot interaction.
The Unitree G1 EDU is a humanoid robot from Unitree specifically designed for education, academic research, and applied development. The focus is not only on human-like movements but also on the ability to test algorithms, perception, motion control, navigation, and interaction in a reproducible manner within real-world environments.
The Unitree G1 EDU is more geared toward research, teaching, and development than a mere basic or demonstration variant. Particularly relevant are the open platform orientation, ROS integration, SDK support, modular architecture, and the ability to implement custom research projects, motion sequences, AI applications, and interaction scenarios on a real humanoid robot platform.
According to Terra Robotics, the Unitree G1 EDU features dimensions of 1270 × 480 × 380 mm, a weight of approx. 35 kg, 23 degrees of freedom, a maximum running speed of 1.8 m/s, a maximum jump height of approx. 0.65 m, a runtime of 2.5 to 3.5 hours, a charging time of approx. 1.5 hours, 3D LiDAR, a depth camera, IMU, and two 3-finger force-position hands.
The Unitree G1 EDU is of interest to colleges and universities because it enables practical teaching in robotics, artificial intelligence, mechatronics, sensor fusion, autonomous navigation, manipulation, and human-robot interaction. Students and research teams can conduct real-world experiments with the humanoid robot instead of working only with simulations or theoretical models.
3D LiDAR, depth camera, and IMU help the Unitree G1 EDU to detect the environment, distances, movements, and its own body position. This sensor technology is particularly relevant for research into autonomous navigation, mapping, obstacle avoidance, sensor fusion, state estimation, object recognition, and stable humanoid movement in real-world test environments.
Yes, according to Terra Robotics, the Unitree G1 EDU is equipped with two 3-finger force-position hands, making it suitable for practical gripping, positioning, and manipulation scenarios. Research teams can use it to investigate topics such as grasp planning, visual servoing, contact models, object interaction, and mobile manipulation under real-world conditions.
According to Terra Robotics, the Unitree G1 EDU U6 is the premium version of the series and is equipped with Tactile Dex5-1 hands. This variant is particularly suitable for demanding manipulation tasks, fine motor skills studies, precise gripping of delicate objects, and complex human-robot interactions in research and development.
Before deploying the Unitree G1 EDU, the target application, safety concept, test area, floor condition, network environment, ROS or SDK workflow, development resources, personnel qualifications, training requirements, maintenance, charging location, potential AI enhancements, and desired manipulation tasks should be assessed. Terra Robotics can provide support in planning the humanoid robot appropriately for education, research, or development.
The Unitree G1 EDU can be purchased, leased, or requested as a customized project solution from Terra Robotics. Purchasing is suitable for universities, research institutions, and companies that wish to use the humanoid robot on a permanent basis. Leasing offers predictable monthly costs, while a project request is particularly useful for training, installation, ROS or SDK integration, AI enhancement, research projects, and complex robotics pilot projects.
The Unitree G1 EDU is a humanoid robot specifically designed for education, academic research, and applied development. The focus is not just on demonstrating human-like movements, but on the ability to reproducibly test algorithms, perception, and control. This makes the G1 EDU a suitable robotics development platform for universities, institutes, schools, innovation labs, and teams looking to validate Embodied AI, autonomous navigation, or human-robot interaction in real-world environments.
Those looking to purchase a humanoid robot to build robotics software in a practical way will especially benefit from the EDU orientation: a focus on open interfaces, ROS integration, and a setup that can be used for curricula, project work, and structured series of experiments. This turns humanoid robotics into a measurable learning and research tool – instead of a mere show object.
For teaching and research, it is important that dimensions, performance values, and sensors can be clearly planned. The Unitree G1 EDU combines a compact design with practice-oriented equipment:
Humanoid robotics is particularly exciting where environments are built for humans: narrow paths, different floor surfaces, work surfaces, changing obstacles, and dynamic situations. The Unitree G1 EDU uses AI-based motion control to enable realistic, fluid movement sequences. The 23 joint axes (DOF) are crucial for finely tuning body posture, gait, and balance – an important factor for gait optimization, whole-body control, and resilient benchmarking.
For teaching and research, repeatability is key: only when identical test runs with comparable results are possible can improvements be clearly measured. The G1 EDU is therefore suitable for experiments involving stability, disturbance robustness, and adaptive behavior – for example, when changing speed, starting and stopping, or changing direction. With speeds of up to 1.8 m/s and dynamic movement options, you can also map realistic scenarios without having to rely exclusively on simulation.
Modern AI robotics stands or falls with perception. The Unitree G1 EDU combines 3D LiDAR, a depth camera, and IMU sensors to capture the environment and self-motion in real time. This is particularly relevant for autonomous navigation, obstacle avoidance, mapping, and robust localization. In learning and research projects, this allows for the practical handling of typical pipeline topics: sensor fusion, state estimation, object detection, distance measurement, and model stability under changing light and environmental conditions.
In dynamic research environments, the truly interesting problems often arise: moving people, variable objects, partially obscured structures, or changing scenes. This is exactly where a humanoid robot with reliable environment detection is valuable, because the gap between "works in the lab" and "works in everyday life" can be made visible early on. This increases the quality of datasets, improves the significance of experiments, and accelerates iterative optimization.
Many projects evolve from navigation to mobile manipulation. The Unitree G1 EDU features two 3-finger force-position hands, which make grasping and handling scenarios significantly more realistic than simple grippers. This allows tasks such as picking up, putting down, repositioning, and positioning to be investigated under real contact conditions. This is relevant for research and teaching because grasping is rarely perfect: objects vary, friction differs, and small deviations must be compensated for by control, perception, and planning.
In this way, the G1 EDU can serve as a humanoid platform for grasp planning, visual servoing, contact models, or "contact-rich" manipulation. This is also an advantage for practical training content: students not only learn the principle but see directly how force, position, timing, and posture are related.
The central EDU added value lies in its development capability. The Unitree G1 EDU is positioned as an open platform for research and education and supports ROS integration as well as SDK-based development. This allows the robot to be integrated into existing robotics workflows: from data acquisition and logging to component architecture and tests in a sim-to-real setup. For teaching projects, this means that groups can work on different focus areas (e.g., perception, navigation, manipulation) without the system becoming a dead end.
With the onboard 8-core CPU and the option for AI expansion (e.g., Jetson Orin NX), modern AI requirements can also be better addressed: local inference, edge compute scenarios, or latency-critical control.
In everyday life, predictability and availability count. With 2.5 to 3.5 hours of runtime (depending on load) and about 1.5 hours of charging time, the Unitree G1 EDU can be easily integrated into typical lab days, seminars, and block courses. The IP54 protection class (standard) supports use in practical environments such as workshops, halls, or test areas, while the temperature range of −10 °C to +45 °C offers additional scope for controlled outdoor tests or varying climate conditions. These are important criteria for everyone who wants to not just test humanoid robotics in the short term, but anchor it sustainably in training or research.
