Google’s latest AI model, Gemini 2.0, is making waves not only in smartphones but also in the realm of robotics. Building upon the foundation of Gemini 2.0, which has already shown its prowess in tasks like processing PDFs and answering complex queries, Google DeepMind has introduced two new AI models specifically designed for robotic applications. These advancements promise to usher in a “new generation of helpful robots,” according to the company’s blog.
Gemini Robotics: Bridging the Gap Between Virtual and Physical
The first model, Gemini Robotics, is a vision-language-action (VLA) model. It takes the core capabilities of Gemini 2.0 and extends them to include physical actions as output responses. While on a Pixel phone, Gemini might answer a question or execute a digital task, in a robot, the same command translates into a physical action in the real world.
Gemini Robots-ER: Embodied Reasoning for Enhanced Spatial Awareness
The second model, Gemini Robots-ER, is a vision-language (VLM) model equipped with “advanced spatial understanding.” This embodied reasoning empowers the AI to navigate and interact with its environment, even as it changes dynamically. Demonstrations showcased robots successfully differentiating between objects of varying colors and textures, even correctly identifying and sorting real and artificial fruits. Another example showed the robot understanding the context of a packed lunch, identifying a Tupperware container of granola and placing it in a lunch bag.
DeepMind’s Contribution and Partnerships
Google emphasizes DeepMind’s crucial role in developing Gemini into a powerful “brain” for robots. It’s remarkable to consider that the same AI powering our smartphones could soon drive humanoid robots. Google is actively collaborating with companies like Apptronik to develop the next generation of humanoid robots, and the Gemini Robots-ER model will be available for testing with partners like Agile Robots, Agility Robots, Boston Dynamics, and Enchanted Tools. While the arrival of these advanced robots is imminent, no specific timeline has been announced.
Addressing Safety Concerns
Google is aware of the potential safety concerns surrounding AI-powered robots and is proactively addressing them. They are incorporating safeguards to prevent robots from causing harm, using frameworks like the ASIMOV dataset to assess the safety implications of robotic actions in real-world scenarios. Google is also collaborating with experts in the field to ensure responsible development and deployment of AI applications.
Conclusion: A Future Shaped by AI-Powered Robotics
Google’s introduction of Gemini 2.0 for robotics marks a significant step forward in the field of AI and robotics. While the exact timeline remains undisclosed, these advancements promise to revolutionize various industries and aspects of our daily lives. The focus on safety and collaboration with leading robotics companies suggests a commitment to responsible development and deployment, paving the way for a future where humans and robots can coexist and collaborate effectively.
