DeepMind Announces Plans to Utilize AI Models in Physical Robotics
Google’s New AI Models for Robotics
Introduction to Gemini 2.0
On Wednesday, Google DeepMind unveiled two innovative AI models aimed at enhancing robotics. These models operate on the advanced Gemini 2.0 infrastructure, which Google touts as its most sophisticated AI to date.
Partnership with Apptronik
Google has joined forces with Apptronik, a robotics firm based in Texas, to create what they refer to as the "next generation of humanoid robots." Apptronik is recognized for its collaborations with other industry leaders like Nvidia and NASA. Recently, Google also took part in a substantial $350 million funding round for Apptronik, indicating a significant interest in advancing humanoid robotics technology.
Demonstration of Capabilities
In a series of demonstration videos, Google showcased the capabilities of Apptronik robots powered by the new AI models. The robots executed a variety of tasks, including:
- Plugging devices into power strips
- Filling a lunchbox
- Managing plastic vegetables
- Zipping up bags
These actions were all performed in response to simple vocal commands, highlighting the robots’ responsiveness and dexterity.
Key Features of Gemini Robotics
Google’s blog elaborated on three essential characteristics that AI models for robotics should possess:
- General: The ability to adapt to varying situations.
- Interactive: Quick understanding and response to instructions or environmental changes.
- Dexterous: Capable of performing tasks that humans typically manage with their hands, like carefully manipulating objects.
Gemini Robotics-ER
The second model, Gemini Robotics-ER (Extended Reasoning), is designed specifically for developers and researchers in the field of robotics. It serves as a foundational tool for these experts to train their own models. Trustworthy organizations like Agile Robots, Agility Robotics, Boston Dynamics, and Enchanted Tools also have access to this advanced platform.
The Broader Robotics Landscape
Google’s initiative represents just one aspect of the growing interest in AI applications for robotics. In the previous months, OpenAI made headlines by investing in a startup called Physical Intelligence, which aims to integrate general-purpose AI into physical robots. This startup focuses on creating large-scale AI models designed to enhance robotic functionalities.
In addition to investments, OpenAI also appointed a new leader for its robotics efforts, coming from Meta, indicating a strategic shift in enhancing AI capabilities in physical tasks. Meanwhile, Tesla has entered the humanoid robotics domain with its Optimus robot, signifying rapidly growing competition within the industry.
Forward-Looking Statements
Google’s CEO, Sundar Pichai, emphasized the importance of robotics as a testing ground for applying AI advancements in real-world situations. He noted that the robots powered by Gemini 2.0 are engineered to be adaptable, responding dynamically to their surroundings and altering their operations as needed.
In summary, the collaboration between Google and Apptronik highlights the continued evolution of robotics technology, emphasizing AI’s role in enabling highly capable and flexible machines. As companies like Google, OpenAI, and Tesla invest in this sector, the robotics landscape is set to undergo significant transformations, paving the way for new applications and innovations.
