Recently revealed at the GTC 2024 event, NVIDIA's Project GROOT marks a major advancement in the field of humanoid robots and artificial intelligence (AI). Its ambitious goal is to integrate AI into the heart of robots, so that they can learn and interact more naturally and effectively with the real world.
Objectives and Features
GROOT, which stands for Generalist Robot 00 Technology, is a versatile base model for humanoid robots. It acts as their brain, allowing them to understand natural language and replicate human movements by observing human interactions. This rapid learning ability is essential for them to navigate, adapt and interact effectively with their environment.
Technologies and Innovations
NVIDIA is also introducing Jetson Thor, a new computing platform for humanoid robots, based on the NVIDIA Thor system-on-chip (SoC). This modular architecture is optimized for performance, power and compactness, with a next-generation GPU based on the NVIDIA Blackwell architecture. Jetson Thor dramatically simplifies design and integration efforts, with an integrated security processor, high-performance CPU cluster, and 100 GB Ethernet bandwidth.
Partnerships and Applications
NVIDIA is partnering with leading humanoid robot companies, such as 1X Technologies, Agility Robotics, Apptronik, Boston Dynamics, and others, to develop a comprehensive AI platform. These partnerships aim to accelerate the development and enable the integration of robots into various aspects of daily life, from assisting people to optimizing industrial operations.
Major updates to the Isaac platform
The GROOT project uses tools from NVIDIA's Isaac platform to create new base models for different robot shapes in various environments. Among these tools, Isaac Lab for reinforcement learning and OSMO, an IT orchestration service, particularly stand out. These embodied AI models require massive amounts of real and synthetic data for their development.
Conclusion
NVIDIA's Project GROOT represents a significant advance toward more intelligent and capable humanoid robots. By combining cutting-edge hardware and software technologies, NVIDIA aims to transform humanoid robot learning in simulation and the real world. This project not only advances research in robotics and AI, but also paves the way for innovative applications in various sectors, promising to significantly improve our interaction with robotic technology.
