🔥36氪•Freshcollected in 1m
Hangmo Tech raises 100M RMB for intelligent variable-stiffness joints
💡Learn how variable-stiffness hardware combined with AI models is solving energy and safety bottlenecks in robotics.
⚡ 30-Second TL;DR
What Changed
Secured nearly 100 million RMB in angel funding from investors including Zhongtou Wanfang.
Why It Matters
The integration of variable-stiffness hardware with AI-driven perception models marks a significant step toward safer, more energy-efficient human-robot interaction in industrial and rehabilitation settings.
What To Do Next
Evaluate the use of variable-stiffness actuators in your next robotic design to improve safety and energy efficiency in human-collaborative environments.
Who should care:Developers & AI Engineers
🧠 Deep Insight
AI-generated analysis for this event.
🔑 Enhanced Key Takeaways
- •Hangmo Tech's founding team originates from the Beihang University Robotics Institute, specifically leveraging years of research in biomechatronics and soft robotics.
- •The company's variable-stiffness technology utilizes a proprietary mechanical design that allows for real-time adjustment of joint compliance without requiring complex sensor feedback loops.
- •The IRMO M1 exoskeleton targets specific industrial scenarios such as logistics and construction, focusing on reducing musculoskeletal strain for workers performing repetitive heavy lifting.
- •The EPA large model architecture integrates multimodal sensor fusion, allowing the exoskeleton to predict user intent by analyzing both environmental visual data and internal proprioceptive joint feedback.
- •The angel round funding is earmarked for scaling production capacity and establishing a dedicated R&D center in Beijing to accelerate the iteration of the FlexmoJoint series.
📊 Competitor Analysis▸ Show
| Feature | Hangmo Tech (FlexmoJoint) | Fourier Intelligence | ULS Robotics |
|---|---|---|---|
| Core Tech | Variable-Stiffness Joints | Force-Feedback Actuators | Industrial Exoskeletons |
| Target Market | Industrial/Consumer | Medical/Rehab | Industrial/Logistics |
| AI Integration | EPA Large Model | Clinical Data Models | Task-Specific Algorithms |
| Pricing | Competitive (Startup) | Premium (Medical) | Mid-Range (Industrial) |
🛠️ Technical Deep Dive
- FlexmoJoint Architecture: Employs a dual-motor antagonistic configuration to achieve variable stiffness, enabling the joint to switch between rigid and compliant states in under 50ms.
- Energy Recovery Mechanism: Utilizes regenerative braking during the deceleration phase of the gait cycle, feeding energy back into the battery system to achieve the reported 31.2% efficiency gain.
- EPA Model Implementation: The Exteroception layer processes RGB-D camera data for terrain mapping, while the Proprioception layer monitors joint torque and angle at 1kHz frequency to ensure stability.
- Material Science: Incorporates carbon-fiber reinforced polymers in the exoskeleton frame to maintain a high strength-to-weight ratio, keeping the total unit weight under 8kg.
🔮 Future ImplicationsAI analysis grounded in cited sources
Hangmo Tech will pivot toward human-robot collaboration (HRC) in manufacturing.
The focus on impact resistance and adaptive stiffness is a prerequisite for safe, un-caged interaction between robots and human workers in shared spaces.
The company will likely license its FlexmoJoint technology to third-party humanoid robot manufacturers.
Variable-stiffness actuators are a critical bottleneck for humanoid robots requiring natural, human-like movement, creating a high-value B2B market opportunity.
⏳ Timeline
2024-05
Hangmo Tech officially incorporated in Beijing by Beihang University research team.
2025-02
Successful prototype testing of the first-generation FlexmoJoint module.
2026-03
Completion of the IRMO M1 exoskeleton pilot testing in industrial logistics environments.
2026-06
Announcement of nearly 100 million RMB angel funding round led by Zhongtou Wanfang.
📰
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: 36氪 ↗