NVIDIA supports humanoid robotics
Summary
The collection, which was announced by NVIDIA Founder and CEO Jensen Huang at CES today, features the NVIDIA Isaac GR00T Blueprint for synthetic motion generation helps deveopers to generate exponentially large synthetic motion data to train humanoid robots using imitation learning. Imitation learning allows humanoids to acquire new skills through observing and mimicking expert human demonstrations. The models were trained on 18 quadrillion tokens, including 2 million hours of autonomous driving, robotics, drone footage and synthetic data. Combining Omniverse — a developer platform of application programming interfaces and microservices for building 3D applications and services — with Cosmos is essential, because it helps minimise potential hallucinations commonly associated with world models by providing crucial safeguards through its highly controllable, physically accurate simulations. Together, NVIDIA Isaac GR00T, Omniverse and Cosmos are supporting physical AI and humanoid innovation progress significantly.
Classifications
AskAI Classifications
LabelsLinked Companies
