Google DeepMind Revamps Manufacturing Automation with Advanced Robotics Models
Google DeepMind's latest advancements in robotics are transforming manufacturing automation across factories and production lines. Led by CEO Demis Hassabis, the company has developed models that can handle video understanding, task orchestration, and multi-robot collaboration. The Gemini Robotics 2 model introduces whole-body motion control for humanoid robots, allowing them to perform multiple tasks.
The Gemini Robotics ER2 model functions as a vision-language-action model that acts as an agent for robots. It enables machines to communicate with factory workers, understand the physical environment, and plan tasks lasting several minutes. This model also allows for multiple robots to work together as a coordinated team.
Google DeepMind's goal is to 'enable AI in the physical world' that can work alongside humans to solve complex challenges. The company's focus on safety is evident in its introduction of ASIMOV-Agentic, a new benchmark for agentic safety orchestration and uncertainty resolution.
The manufacturing applications of these models remain to be proven in production environments. Factories will need to evaluate whether the models deliver sufficient reliability and cost savings to justify deployment at scale.