NVIDIA introduces Project GR00T, a foundation model for humanoid robots, and significant upgrades to the Isaac Robotics Platform, empowering developers with new tools and AI capabilities for robotics innovation
NVIDIA Project GR00T
NVIDIA, a pioneer in artificial intelligence (AI) and robotics, continues to push the boundaries of robotics and embodied AI with the announcement of Project GR00T and major updates to its Isaac Robotics Platform. Project GR00T introduces a general-purpose foundation model tailored for humanoid robots, while the Isaac platform enhancements provide developers with advanced tools and AI capabilities to drive innovation in robotics.
At the heart of Project GR00T is the ambition to develop humanoid robots capable of understanding natural language and emulating human movements through observation. Powered by NVIDIA’s cutting-edge technologies, these humanoid robots equipped with GR00T will quickly learn coordination, dexterity, and various skills crucial for seamless navigation and interaction with the real world. NVIDIA CEO Jensen Huang showcased several GR00T-powered robots demonstrating diverse tasks during the GTC keynote.
At the heart of this innovation lies Jetson Thor, a new computer specifically tailored for humanoid robots. Equipped with the NVIDIA Thor system-on-a-chip (SoC), Jetson Thor offers a modular architecture optimized for performance, power efficiency, and size. With features like a next-generation GPU based on the Blackwell architecture and a transformer engine, Jetson Thor delivers significant AI performance, enabling the execution of complex multimodal generative AI models such as GR00T.
NVIDIA’s commitment to advancing humanoid robotics extends beyond technology development. The company is Collaborating with industry pioneers like Agility Robotics and Sanctuary AI, NVIDIA aims to accelerate the development of human-centric robots that revolutionize labour and daily life.
Also Read: NVIDIA GTC AI Conference 2024: Topics, Schedule, Panel and Speakers Details
The Isaac Robotics Platform receives major updates to further support the development of embodied AI models for diverse robot embodiments. Isaac Lab, a GPU-accelerated application built on Isaac Sim, facilitates reinforcement learning through parallel simulations, while OSMO streamlines data generation, model training, and hardware-in-the-loop workflows across distributed environments.
Additionally, NVIDIA introduces the Isaac Manipulator and Isaac Perceptor, offering state-of-the-art dexterity and multi-camera 3D surround-vision capabilities, respectively. These advancements empower developers to automate new robotic tasks efficiently and enhance autonomy in material handling operations.
With the anticipated availability of these new Isaac platform capabilities in the next quarter, NVIDIA reaffirms its commitment to driving breakthroughs in robotics and AI, fostering collaboration, and enabling innovations beyond imagination. Learn more about Project GR00T and the Isaac Robotics Platform to embark on the journey of robotics transformation.
Must Read: NVIDIA GTC 2024 Keynote by CEO Jensen Huang, Check Key Highlights and Announcements
This post was last modified on March 19, 2024 8:56 am
Rish Gupta is an Indian entrepreneur who serves as the chief executive officer (CEO) of…
Are you looking to advance your engineering career in the field of robotics? Check out…
Artificial intelligence is a topic that has recently made internet users all over the world…
Boost your learning journey with the power of AI communities. The article below highlights the…
Demystify the world of Artificial Intelligence with our comprehensive AI Glossary and Terminologies Cheat Sheet.…
Scott Wu is the co-founder and Chief Executive Officer of Cognition Labs, an artificial intelligence…