A humanoid robot from "Walker" series of UBTECH conducts sorting work at the Beijng World of Robots in Beijing, capital of China, Feb. 24, 2025. (Xinhua/Ju Huanzong)SHENZHEN, March 22 (Xinhua) -- In Geely's Zeekr 5G smart factory in Ningbo, east China's Zhejiang Province, dozens of advanced humanoid robots are seamlessly collaborating on factory floors -- sorting materials, transporting boxes and assembling delicate car parts -- marking a new era of human-like teamwork in manufacturing.Developed by Shenzhen-based tech company UBTECH, the Walker S1 robots in the auto factory mark a significant milestone, as it's the first time multiple humanoid robots have been trained to collaborate across different tasks and complex zones in a real-world factory environment.The robots operate using UBTECH's "brain network" system, which transforms them from individual workers into a synchronized, collaborative team.According to Jiao Jichao, vice president of UBTECH, the system combines a "super brain" for decision-making and "mini-brains" for real-time actions.The "super brain" uses AI such as DeepSeek to analyze data and assign tasks, like a manager overseeing workflows while the "mini-brains" of each robot allow them to perceive surroundings, adjust grip strength for fragile parts, and share live updates with teammates, Jiao explained.In action, the robots were seen tracking components across different zones with cameras, mapping workspaces, and delicately handling flexible materials like film without causing damage.To improve their skills, the robots gather data while performing tasks. Jiao said this process has already generated "hundreds of millions" of real-world industrial data points, which are stored in a knowledge database. Robots can quickly learn new skills for different assembly line roles with this resource.UBTECH has already partnered with automakers like BYD and BAIC and logistics giant SF Express to advance applications of the Walker S series."We are planning to partner with more factories, comprehensively promote the implementation of multi-robot collaborative training, and scale up the use of humanoid robots in industrial settings," Jiao said.