Ubtech Robotics has successfully conducted the world's first multi-unit, multi-scenario, and multi-task collaborative training for humanoid robots at Zeekr's 5G smart factory. This training aims to explore and establish a universal solution for group operations of humanoid robots in multi-task industrial environments, promoting the evolution of humanoid robot technology from single-unit autonomy to collective intelligence.
In this training, Ubtech applied swarm intelligence technology to humanoid robots for the first time, enabling multi-robot collaboration. To achieve this, the company developed BrainNet, a software architecture for humanoid robot swarm intelligence, and designed the Internet of Humanoids (IoH). This innovation provides a feasible approach for implementing both hardware and software aspects of swarm intelligence.
At the core of the software architecture, Ubtech adopted a design that combines the "Super Brain" with the "Smart Cerebellum." The "Super Brain," based on multimodal embodied reasoning large models, focuses on intelligent hybrid decision-making technology, capable of handling high-dimensional decisions in complex production line tasks. Meanwhile, the "Smart Cerebellum," built on Transformer models, develops cross-domain fusion perception and multi-robot collaborative control technologies, supporting parallel distributed learning among multiple units and accelerating skill generation and transfer.
The Walker S1 robots involved in the training were deployed across various complex scenarios in Zeekr's 5G smart factory, including the final assembly workshop, SPS dashboard area, quality inspection zone, and door assembly section. In these areas, the robots successfully performed multi-task collaborative operations such as sorting, transportation, and precision assembly, demonstrating their practical application potential in industrial settings.
This training not only verified the multi-task and multi-scenario collaborative capabilities of Ubtech's humanoid robots but also laid a solid foundation for the widespread adoption of humanoid robots in the industrial sector.
