Dongyi Tech's Liu Chenhao: The Hierarchical Revolution Behind 40-Rally Humanoid Robot Badminton | IROS 2026
Dongyi Tech has achieved a breakthrough in high-dynamic motion for humanoid robots, realizing 40 consecutive badminton rallies. The core of this achievement lies in a hierarchical cerebrum-cerebellum control architecture. This progress was showcased at IROS 2026.
Key Takeaways
- Key Highlight：Dongyi Tech has achieved a breakthrough in high-dynamic motion for humanoid robots, realizing 40 consecutive badminton rallies. The core of this achievement lies in a hierarchical cerebrum-cerebellum control architecture. This progress was showcased at IROS 2026.
- Innovation & Tech：Highlights advancements in Dongyi, Tech, Liu, demonstrating rapid progress in model capabilities.
- Industry Impact：Reported via 雷峰网 (CN), offering actionable signals for developers and technology leaders.
Badminton, due to its unique aerodynamic characteristics—extremely high initial velocity, rapid deceleration, asymmetrical trajectory, and in-flight self-rotation—has become an ultimate testing scenario for high-dynamic motion control in humanoid robots. The achievement of 40 consecutive rallies demonstrated by Dongyi Tech marks a new industrial-level milestone for humanoid robots in real-time perception, trajectory prediction, and motion planning.
The key to this achievement lies in the so-called "hierarchical revolution," an architectural design with a clear division of labor between the cerebrum and cerebellum. The cerebellum is responsible for low-latency reflexive motion control and body balance, while the cerebrum handles high-level situational awareness, trajectory prediction, and strategic decision-making. This hierarchical architecture enables the robot to complete the full closed-loop process of "seeing the shuttlecock—predicting the landing point—moving into position—swinging the racket to hit" within a millisecond-level time window, while maintaining body stability.
In terms of industry impact, the breakthrough in the badminton rally scenario holds landmark significance. It not only validates the real-time interaction capabilities of humanoid robots in unstructured, highly dynamic environments but also lays the technical foundation for broader embodied intelligence applications in the future—such as collaborative tasks, service interactions, and specialized scenario operations. The maturation of hierarchical control architectures may become one of the key pathways for embodied intelligence to transition from the laboratory to large-scale deployment.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.
Industry Insights & Analysis
As artificial intelligence rapidly evolves, breakthroughs surrounding Dongyi, Tech, Liu, Chenhao are shifting toward scalable, robust real-world implementations.
Driven by both open-source ecosystems and proprietary model architectures, the integration between compute optimization, data engineering, and agentic workflows is accelerating. This development provides a strategic benchmark for upcoming AI tooling and developer workflows.