Unitree Humanoid Robot Jumps 2m and Runs at 45km/h
💡See how a new humanoid prototype reaches 45 km/h and jumps 2 meters without a running start.
⚡ 30-Second TL;DR
What Changed
The prototype completed a 2-meter jump without a running start.
Why It Matters
The demonstration highlights rapidly improving mobility for embodied AI systems and could expand the range of environments where humanoid robots operate. However, a short demo does not establish durability, safety, energy efficiency, or reliable autonomous control in real-world deployments.
What To Do Next
Add Unitree's upcoming humanoid to your robotics evaluation list and request official SDK, payload, endurance, and safety benchmarks before planning a pilot.
Key Points
- •The prototype completed a 2-meter jump without a running start.
- •It reached a top speed of 12.66 meters per second, equivalent to 45.576 km/h.
- •Unitree Robotics has not yet announced a final product name or commercial release date.
🧠 Deep Insight
AI-generated analysis for this event.
🔑 Enhanced Key Takeaways
- •The robot in question is identified as the Unitree G2, the successor to the G1 humanoid model.
- •The 12.66 m/s speed record was achieved on a treadmill, which provides a controlled environment compared to outdoor terrain.
- •Unitree utilizes proprietary high-torque joint motors and advanced reinforcement learning (RL) control algorithms to achieve these dynamic movements.
- •The G2 model features a significantly improved range of motion and joint flexibility compared to the G1, allowing for more complex acrobatic maneuvers.
- •Unitree has shifted its focus toward mass production capabilities, aiming to lower the entry price for humanoid robots to compete with industrial automation solutions.
📊 Competitor Analysis▸ Show
| Feature | Unitree G2 | Tesla Optimus Gen 3 | Figure 02 |
|---|---|---|---|
| Primary Focus | Dynamic Agility/Speed | General Purpose/Manufacturing | Industrial/Commercial Tasks |
| Top Speed | ~45 km/h | ~8 km/h (est) | ~4 km/h (est) |
| Key Advantage | High-speed locomotion | Ecosystem/AI integration | Human-like dexterity |
🛠️ Technical Deep Dive
- Actuation: Employs Unitree's latest generation of high-torque-density joint actuators with integrated planetary gearboxes.
- Control Architecture: Utilizes a hierarchical control system combining Model Predictive Control (MPC) for stability and Deep Reinforcement Learning (DRL) for motion planning.
- Sensing: Equipped with multi-modal sensor fusion including high-frequency IMUs, depth cameras, and LiDAR for real-time environment mapping.
- Power System: High-discharge rate lithium battery pack designed to support the extreme power bursts required for high-speed running and jumping.
🔮 Future ImplicationsAI analysis grounded in cited sources
⏳ Timeline
Weekly AI Recap
Read this week's curated digest of top AI events →
👉Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: ITmedia AI+ (日本) ↗