Covering Scientific & Technical AI | Tuesday, October 8, 2024

ISC 2022: Digital Twins Keynote Preview 

In this contributed Q&A, ISC’s Nages Sieslack interviews Rev Lebaredian, vice president of Omniverse and simulation technology at Nvidia, about the use of digital twin technology for science and industry. Lebaredian will be delivering the opening keynote at ISC 2022 in partnership with Michele Melchiorre, senior vice president of production system, technical planning, tool shop at BMW Group.

How is the computational model that Nvidia has embraced different from traditional digital simulations. What role will supercomputing play in Digital Twin technology?

Rev Lebaredian: All things that are designed and built by humans are typically first built in a virtual world. Bicycles, cars, bridges, factories … are all designed and simulated with various CAD and CAE tools, well before they are built in the real world. Physically accurate and extremely fast simulation is key to designing the best and most efficient products, processes, and environments. We can quickly test many iterations of a design in the virtual world at a fraction of the cost of what it would take to build them in the real world.

Once the digital version of the product is complete, it’s transformed into its real world dual—one built from atoms instead of electrons. In most cases today, that’s the end of the road for the digital version.

But, if we link the two manifestations—digital and real—they can evolve with each other. We can capture data from the real world through IoT sensors and devices and feed it into the digital model, keeping the “twins” in sync.

Applying accurate physical simulation to the digital twin gives us incredible superpowers. We can teleport to any part of the digital twin just like we can in a video game and inspect any aspect of it reflected from the real world. We can also run simulations to predict the near future or test many possible futures for us to pick the most optimal one.

Digital twins give us the ability to apply powerful design practices common in modern software engineering–continuous integration, continuous delivery–to the design and engineering of all things. They allow us to make layout, process, operational, or systems changes in the digital twin before deploying to the physical world where there is great risk of downtime and safety.

Digital twins encompass the monitoring, operating, and optimizing of autonomous systems through living simulations. All of the agents and systems must operate in the same space-time–requiring a new class of computing architecture at data center scale. The real-time nature of this new computing architecture demands precise and synchronous timing (precision timing protocol) to orchestrate physical and virtual agents operating in the same time domain. Both the software and the supercomputer need to be scalable, low latency, and support precise time.

We’ve built the Omniverse platform and new technologies like NVIDIA Modulus for AI-physics simulation, and OVX infrastructure, to unlock the full potential of Digital Twins—from design and creation to physically accurate simulation on supercomputers.

In light of your upcoming keynote at ISC 2022, can you detail the cooperation Nvidia has with other car manufacturers?

Rev Lebaredian: We are working with hundreds of automakers, trucking companies, robotaxis and shuttles to deliver a range of semi-autonomous to fully autonomous vehicles. Our auto ecosystem spans OEMs, Tier 1 suppliers, sensor makers, software companies and research institutions. Volvo Cars, Zoox and SAIC are among the growing ranks of leading transportation companies using the newest NVIDIA DRIVE solutions to power their next-generation AI-based autonomous vehicles. We have been partnering with Mercedes Benz to develop their fleet of autonomous cars, build intelligent cockpits, and redesigning car architectures for an AI driver.

All of the top car manufacturers have expressed interest in Omniverse digital twin projects and many are actively engaged in Omniverse projects, exploring how Omniverse can serve as an open and extensible platform to connect every part of their development process - from styling and concept design to engineering simulation to Factory Digital Twins and AI for logistics.

How does digital twin serve science and how can the scientific community benefit from this technology?

Lebaredian: For the scientific community, digital twins bring enormous potential and opportunity.  With digital twins developed and operated in platforms like NVIDIA Omniverse and based on open standards like Universal Scene Description (USD), the scientific community can combine outputs from traditional simulations, AI models, and observed data into a single, aggregated, living model of a complex system. What was previously impossible with traditional modeling techniques, are not only expensive and time-consuming to execute, but also only provide a ‘point in time’ view of the system. With true digital twin simulations, the scientific community can interactively run infinite what-if scenarios which will allow researchers to explore a broader solution space for use cases such as drug discovery, climate research, and renewable energy. Lastly, with visual simulations of digital twins, the scientific community has an enormous opportunity to convey the impact of their research findings to broader non-technical audiences, influencing authoritative bodies like policy makers, and the general public.

What are the other areas we can see digital twin being applied in the future? What is the timeline for this?

Lebaredian: We will see digital twins affecting every industry, if not already - financial markets, industrial manufacturing, transportation, infrastructure and city planning, retail, supply chains, energy, climate and environment, healthcare. Digital twins will also usher in the next wave of robotics.

Digital twins are already here. Nvidia has been working with simulated virtual worlds for over a decade. We built a digital twin of our autonomous vehicle fleet, and the worlds they are born and raised in–Omniverse. We built Isaac Sim to create and simulate digital twin environments for training and testing robots. We are building digital twins of our data centers and supercomputers to enable true software automation. Our partners and customers are using Omniverse to build digital twins of automotive factories, warehouses, cities, 5G networks, self-driving car fleets, climate and weather.

Finally, what can the ISC attendees expect from this keynote address?

Lebaredian: ISC attendees will get to see how a new class of digital twin simulations and supercomputers are transforming industries–ushering in a new era of AI. They will get a behind the scenes look at the invention of hyperscale software and hardware, required to power development and operation of large scale, physically accurate and super-realtime digital twin simulations. Last but not least, they’ll get an intimate look at the vision the BMW Group has set for their factories of the future, their current progress, and what’s coming next!

The joint keynote, Supercomputing: The Key to Unlocking the Next Level of Digital Twins, takes place on Monday, May 30, at ISC 2022.

About the author: Tiffany Trader

With over a decade’s experience covering the HPC space, Tiffany Trader is one of the preeminent voices reporting on advanced scale computing today.

AIwire