Nvidia unveils Omniverse Blueprint for digital twins

NVIDIA has announced the launch of the NVIDIA Omniverse Blueprint, which aims to enable industry software developers to assist their computer-aided engineering (CAE) customers in creating digital twins with real-time interactivity across various industries, such as aerospace, automotive, manufacturing, and energy.

This blueprint is designed to aid software developers, including firms like Altair, Ansys, Cadence, and Siemens, in streamlining costs and reducing energy usage, thus facilitating a faster time-to-market for their customers. The Omniverse Blueprint serves as a reference workflow which incorporates NVIDIA acceleration libraries, physics-AI frameworks, and interactive physically-based rendering to achieve significantly faster simulations and real-time visualisation.

Jensen Huang, the founder and CEO of NVIDIA, commented on this development: "We built Omniverse so that everything can have a digital twin. Omniverse Blueprints are reference pipelines that connect NVIDIA Omniverse with AI technologies, enabling leading CAE software developers to build groundbreaking digital twin workflows that will transform industrial digitalisation, from design and manufacturing to operations, for the world's largest industries."

A key initial application of this blueprint is in computational fluid dynamics (CFD) simulations, which are crucial for the virtual testing and optimisation of designs in sectors like automotive and aerospace. Traditional engineering workflows encompass physics simulation, visualisation, and design optimisation, often requiring weeks to months. NVIDIA's collaboration with Luminary Cloud, showcased at SC24, facilitates a virtual wind tunnel for simulating and visualising fluid dynamics with real-time interaction, even amidst vehicle model changes.

The Omniverse Blueprint utilises three primary NVIDIA technologies to offer real-time physics digital twin capabilities: the NVIDIA CUDA-X libraries for solver acceleration, the NVIDIA Modulus physics-AI framework for model training, and the NVIDIA Omniverse APIs for wide-ranging data interoperability and real-time visualisation.

Ansys is among the early adopters, integrating the Omniverse Blueprint with its Fluent fluid simulation software, resulting in accelerated CFD simulations. Ajei Gopal, Ansys' President and CEO, stated, "By integrating NVIDIA Omniverse Blueprint with Ansys software, we're enabling our customers to tackle increasingly complex and detailed simulations more quickly and accurately. Our collaboration is pushing the boundaries of engineering and design across multiple industries."

Luminary Cloud is also leveraging the blueprint, utilising its Modulus-based simulation AI model to run simulations significantly faster than traditional solvers, enabling real-time visualisation of aerodynamic flow using Omniverse APIs. Other companies such as Altair, Beyond Math, Cadence, Hexagon, Neural Concept, Siemens, SimScale, and Trane Technologies are exploring similar integrations.

The Omniverse Blueprint is compatible with major cloud platforms, including Amazon Web Services, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, and the NVIDIA DGX Cloud.

Additionally, Rescale, a cloud platform that assists organisations in scientific and engineering advancements, employs the NVIDIA Omniverse Blueprint to facilitate the training and deployment of custom AI models across cloud service providers. This enables organisations to generate training data, train AI models, and run inferences easily.