The Covid 19 has rocked the foundations of the world. Nobody was prepared for a pandemic of this proportion: The supply chains got disrupted, factories closed down, and we are still teetering at the edge, 15 months into the breakout. But now that we know how a pandemic can turn our whole world upside down, are we better prepared to take on another act of God?
Not to worry, NVIDIA’s Ominverse offers a silver lining.
At GTC Spring 2021, NVIDIA announced Omniverse, a powerful multi-GPU real-time simulation and collaboration platform for 3D production pipelines. It is built on Pixar’s Universal Scene Description (USD), an open-source 3D scene description and file format.
Omniverse has five parts: Nucleus, Connect, Kit, Simulation and RTX renderer. These components, along with connected third-party digital content creation (DCC) tools and additional connected Omniverse microservices complete the full Omniverse ecosystem.
According to NVIDIA, Omniverse can play a pivotal role in designing, optimising and operating factories of the future. For instance, BMW is leveraging NVIDIA Omniverse and NVIDIA AI to usher in a new era of digitalisation in automobile production.
Credits: NVIDIA
Factory planning requires reams and reams of data collected from various applications. But the data might not be up to date and can have many compatibility issues. The Omniverse platform will enable live data to be collected and collated from all the relevant databases to create a joint simulation without reimporting data. “The capability to operate in a perfect simulation revolutionises BMW’s planning process,” said Milan Nedeljković, the Board Member for Production at BMW.
With an RTX renderer, the Omniverse can deliver photorealistic quality while allowing employees to collaborate from different time zones or geographical location. They are also given freehand to re-adjust and optimise a process or production system whenever it is needed.
Omniverse's simulation has digital humans. These virtual representations of workers are built from data gathered from various associates who work at the plant. The digital humans can be used to test new workflows for work economics and efficiency.
NVIDIA also has an Isaac Robotic platform to help with logistics to improve the material flow in all areas of production. Quite useful, considering BMW produces 2.5 million vehicles per year. “Synthetic data generation and domain randomisation available in Isaac are key to bootstrapping machine learning. Isaac sim generates millions of synthetic images and varies the environment to teach the robots,” explained Jensen Huang, at the Summit.
NVIDIA implements domain randomisation, which can generate infinite permutation of photorealistic objects textures, orientations and lighting conditions. Using Nvidia Fleet command within the Omniverse, workers can orchestrate robots and other devices in the factory. They can also monitor manufacturing cells, update software, launch robot missions and tele-operate when a robot needs a helping hand.
Omniverse will reduce the planning times, improve flexibility and precision and make planning processes 30% efficient. “This is redefining collaboration,” said Milan Nedeljković, emphasising how Omniverse has helped in maintaining the scale and complexity of BMW’s production.
- OpenAI Says AGI is Here. But Not Everyone is Buying It
- Why Oracle is Bringing Post-Quantum Security to Java
- What Really Happens When You Turn Your Selfie Into a 1980s AI Pic?
- AI Agents Are Making Data Security A Bigger Enterprise Problem
- OpenAI is Getting Nervous About Reinforcement Learning
- Meta’s Bold Open-Weight Bet is Meant to Muse Developers