DeepMind to Utilize AI Models for Enhancing Physical Robots
Google DeepMind Unveils Advanced AI Models for Robotics
Introduction of Gemini 2.0
On a recent Wednesday, Google DeepMind launched two innovative artificial intelligence models for robotics, named Gemini Robotics and Gemini Robotics-ER (Extended Reasoning). These models operate on Gemini 2.0, which Google claims to be its most advanced AI technology to date. Unlike previous AI models that primarily generated text and images, Gemini Robotics focuses on enabling robots to perform physical tasks.
Partnership with Apptronik
Google has announced a collaboration with Apptronik, a robotics development company based in Texas. This partnership aims to create the next generation of humanoid robots using the capabilities of Gemini 2.0. Apptronik has previously collaborated with companies like Nvidia and NASA, showcasing their expertise in robotics. Interestingly, last month, Google participated in Apptronik’s recent funding round, where they raised $350 million.
Robot Demonstrations
In a series of demonstration videos, Google’s robots, enhanced by Gemini’s AI models, showcased their current abilities. These robots successfully executed tasks such as:
- Plugging devices into power outlets
- Filling a lunchbox with items
- Sorting and moving plastic vegetables
- Zipping up bags
These tasks were performed in response to voice commands, highlighting the robots’ capability to integrate verbal instructions and physical actions. However, Google has yet to announce when this technology will become available to the public.
Key Features of Gemini Robotics
In its announcement, Google identified three essential qualities that AI models for robotics must possess:
- General Adaptability: Robots must be able to adjust to various situations and environments.
- Interactive Response: They should understand and react promptly to new instructions or environmental changes.
- Dexterity: This involves performing tasks that require fine motor skills, such as handling different objects similar to how humans use their hands.
The Role of Gemini Robotics-ER
Gemini Robotics-ER serves as a foundational tool designed specifically for roboticists. It provides a framework for developing and training their own AI models. Currently, Apptronik and a select group of "trusted testers," which includes companies like Agile Robots and Boston Dynamics, have access to Gemini Robotics-ER.
The Competitive AI Robotics Landscape
Google’s efforts in robotics are part of a broader trend in the tech industry. Other notable companies are also investing in AI for robotics. For example, OpenAI recently invested in a startup named Physical Intelligence, which focuses on integrating AI into the physical realm. This startup aims to develop large-scale AI models and algorithms that can power robotic applications.
OpenAI, in the same announcement, hired a former leader from Meta’s Augmented Reality project to head its robotics and consumer hardware division. Similarly, Tesla is making strides in humanoid robotics with its Optimus robot, which aims to tap into the growing market of robotic applications.
Google’s Vision for Robotics
Sundar Pichai, CEO of Google, communicated via a post on X (formerly known as Twitter) that the company views robotics as a practical testing ground. He emphasizes the potential of using advanced AI models to enable robots to adapt in real time to their surroundings.
Through innovation and collaboration, Google is paving the way for significant advancements in robotics, merging the digital capabilities of AI with the physical abilities of machines. The transformation of AI in robotics promises exciting trends and practical applications in various sectors.
