Article Source

• Scaling AV Data With Omniverse and Cosmos

Scaling AV Data With Omniverse and Cosmos

Improve AV performance by amplifying thousands of driving scenes into billions.

The AV data factory consists of fleet data, geo-spatially accurate 4D reconstruction, AI to generate scene and traffic variations, training, and closed-loop evaluation.

OmniMap fuses map and GEO satellite imagery to construct drivable 3D environments. Neural Reconstruction Engine (NRE) creates high-fidelity 4D simulation environments from AV sensor logs. Using text or image prompts, Edify 3DS can search for existing scenes, or generate 3D objects, transforming scenes in NVIDIA Omniverse™ that are 4D grounded. Lastly, Cosmos can generate near infinite variations of driving scenarios using text prompts.

NVIDIA built Omniverse, Cosmos, and supporting AI models to generate synthetic scenarios and amplify training data to turn thousands of human demonstrations into billions of virtually driven miles.

NVIDIA Cosmos: A World Foundation Model Platform for Physical AI

The next frontier of AI is physical AI. NVIDIA Cosmos™—a platform of state-of-the-art generative world foundation models, advanced tokenizers, guardrails, and an accelerated data processing and curation pipeline—accelerates the development of physical-AI-embodied systems such as robots and autonomous vehicles.

The integration of Cosmos with NVIDIA Omniverse™ creates a sim-to-real synthetic data and multiverse simulation engine for physical AI. This enables the generation of every possible outcome for a given use case, providing physical AI systems with the foresight needed to make more accurate and informed decisions.