China’s humanoid robotics sector is expanding its data infrastructure, with the Beijing Humanoid Robotics Technology Innovation Center collecting more than 3 million data entries within four months of operation.
According to CCTV News, more than 300,000 of these entries are open source. Center staff said the model could reduce data collection costs by more than 50% for small and medium-sized robotics companies and improve access to training data.
This facility focuses on simulating repetitive tasks for training large-scale models. One example is washing fruit. An operator controls a robotic arm to mimic human movements. This process involves picking up the fruit, turning on the faucet, cleaning the surfaces, and putting the fruit back into the basket.
Each robotic arm has six joints, tailored to the movements needed to perform a task. This sequence is repeated 500 times to generate 500 data entries that allow the system to interpret the fruit washing task.
Additional training scenarios include changing diapers, organizing equipment, and transferring products such as bottled water. These processes position the center as a data generation hub, generating training datasets across multiple applications.
Centralized data generation improves cost efficiency. Reducing data collection costs will enable small and medium-sized enterprises to participate in robot development and expand their operations more efficiently.
Similar facilities operate in other regions. At Shanghai’s Zhangjiang Robot Valley, a data center supports the simultaneous operation of robots across simulated environments such as homes, supermarkets, and retail stores, allowing the system to adapt to a variety of use cases.
Data services are also being developed as part of the artificial intelligence value chain. In Chengdu, companies focus on data collection, labeling, and training to support the widespread implementation of AI systems.
The development of data production centers is contributing to the expansion of robotics applications with structured datasets that support model training and system performance across tasks such as fruit handling and other business processes.
Source: Global Times
