Robotics and humanoid robotics & civilian drones discussion

humanoid robot picking tea leaves. This is quite difficult stuff at this stage still.

An intelligent traffic management robot is on duty at a crossroad in the Hubin business area of Hangzhou, east China's Zhejiang Province, May 11, 2026. Fifteen intelligent traffic management robots have been deployed at key intersections of the West Lake scenic area, Hubin business area, and main urban roads in Hangzhou, undertaking the tasks such as showing the way for tourists, dissuading riders of non-motorized vehicles and pedestrians from violating traffic rules, and directing the traffic. (All photos by Xinhua/Huang Zongzhi)

Figure robot live streaming working shift.

A heavy-duty industrial robot developed by Shanghai based Chaifu Robot has set a new Guinness World Record of 5,000.36 kg for the highest payload capacity, marking the first time a Chinese industrial robot has topped the global load capacity ranking. The previous record, held by Japan's Fanuc Corporation since 2016, stood at 2,300 kg.

We have at home humanoid robot doing service roles now.

Chinese firm pushes humanoid robot intelligence forward with 300 FPS control speed​

Real-time 300FPS edge inference enables fast, low-latency robot control without cloud dependency.
By
May 20, 2026 08:13 AM EST


Chinese firm Horizon Robotics has released an open-sourced AI model, named HoloMotion-1, designed for whole-body humanoid robot control.

The company says the 4-billion-parameter robot cerebellum model represents a major leap in robot motion intelligence, pushing cerebellum models beyond the million- and ten-million-parameter scales commonly used previously.
HoloMotion-1 can perform real-time inference at 300 frames per second on edge devices, enabling faster, more responsive humanoid robot movements.

According to Horizon Robotics, the new model marks a significant advance in scalable humanoid robot control and edge AI deployment.

Zero-shot motion learning​

HoloMotion-1 is a humanoid motion foundation model designed to improve real-time whole-body robot control through large-scale motion learning, reports .

HoloMotion-1 is a system designed to help humanoid robots copy and perform human-like movements more reliably in real time. Instead of relying only on small motion capture (MoCap) datasets, which are recordings of human movement made in controlled environments, it uses a much larger and more varied collection of motion data.

This includes curated MoCap data, motion data created inside the company, and movements reconstructed from real-world videos taken “in the wild.” This mix gives the robot a much wider range of examples, helping it handle new or unseen movements and situations where its sensors may not work perfectly.
Real-world zero-shot transfer of the HoloMotion policy.
To manage this complex data, HoloMotion-1 uses a Transformer-based neural network, a type of deep learning model that is especially good at understanding sequences over time, such as motion steps. This is better than older MLP (Multi-Layer Perceptron) policies, which are simpler neural networks that struggle with long and complex motion patterns, according to the firm’s .

For efficient real-time use on , the system uses a Mixture-of-Experts (MoE) Transformer. This means only a few parts (“experts”) of the model are activated at each step, saving computing power. It also uses KV-cache (Key-Value cache), a technique that speeds up repeated calculations. Together, these allow the system to run at about 300 frames per second on edge devices.

Finally, the system uses a sequence-level PPO (Proximal Policy Optimization) training method. PPO is a reinforcement learning technique, and here it is applied to whole motion segments instead of single time steps, making training more efficient and stable when learning from large, mixed datasets.

Agile humanoid tracking​

To test how well the system works in the real world, was directly installed on a Unitree G1 . Importantly, it was used without any extra training on real-world data. All the computing needed for the robot’s movement was done on its own built-in computer system.

The system used the MoE Transformer along with a technique called KV-cache, which helps the model reuse past calculations efficiently. According to Horizon Robotics, this combination allows the robot to make very fast decisions, running at about 200–300 cycles per second on its onboard hardware. At the same time, the robot’s movement system itself runs at 50 cycles per second to keep the motion smooth and stable.

The results show that the robot could successfully transfer what it learned in simulation to the real world without extra adjustment. It was able to perform many different movements it had never been directly trained on in real hardware, including dancing, crawling, sitting, and martial arts-style kicks.

The system was also tested with live human control using devices like motion capture suits and VR-based controllers. In these tests, the robot followed human movements closely and responded smoothly, showing stable and reliable real-time tracking of user actions.
Researchers highlight that HoloMotion follows a 4-step plan for humanoid robot control: Imitate Any Pose, Follow Any Command, Move on Any Terrain, and Control Any Robot Type. HoloMotion-1 completes the first step by letting robots copy many human movements from videos or live input. It also acts as a base for future improvements.

Please, Log in or Register to view URLs content!

This photo taken on May 20, 2026 shows the exterior view of the Paxini Super EID Factory in 764 Aerospace, north China's Tianjin Municipality. Located on the site of a former state-owned broadcasting equipment factory, the 764 Aerospace, a digital industrial park in Hexi District of Tianjin, is home to more than 10 high-tech enterprises. Featuring embodied intelligence industry, the park focuses on advancing embodied intelligence data collection, robot debugging, algorithm R&D and scenario implementation, and has built a full-chain industrial cluster covering the whole process from data collection to scenario deployment. (Xinhua/Sun Fanyue)

Click to expand...

A staff member wearing sensor devices carries out data collection under home scenarios at Paxini Super EID Factory in 764 Aerospace, north China's Tianjin Municipality, May 20, 2026. Located on the site of a former state-owned broadcasting equipment factory, the 764 Aerospace, a digital industrial park in Hexi District of Tianjin, is home to more than 10 high-tech enterprises. Featuring embodied intelligence industry, the park focuses on advancing embodied intelligence data collection, robot debugging, algorithm R&D and scenario implementation, and has built a full-chain industrial cluster covering the whole process from data collection to scenario deployment. (All photos by Xinhua/Sun Fanyue)



A staff member wearing sensor devices carries out data collection under industrial scenarios at Paxini Super EID Factory in 764 Aerospace, north China's Tianjin Municipality, May 20, 2026.



Engineers adjust robots at tech firm SYNROX in 764 Aerospace, north China's Tianjin Municipality, May 20, 2026.



Engineers adjust a humanoid robot at tech firm SYNROX in 764 Aerospace, north China's Tianjin Municipality, May 20, 2026.



This photo taken on May 20, 2026 shows the exterior view of the Paxini Super EID Factory in 764 Aerospace, north China's Tianjin Municipality.

report here of smart robots in China entering into different part of lives in China.