NVIDIA Generative AI Accelerates Automotive Industry Innovation

The News: NVIDIA is providing the infrastructure and platform that enables automakers to leverage NVIDIA generative AI to reshape the vehicle lifecycle from the design concept to the finished vehicle, including safety engineering to full self-driving. In addition, NVIDIA’s software development kits (SDKs) enable automotive companies to build and deploy custom generative AI solutions. Read the press release on NVIDIA’s website.

NVIDIA Generative AI Accelerates Automotive Industry Innovation

Analyst Take: NVIDIA is providing automakers with the NVIDIA infrastructure and platform enabling automakers to leverage NVIDIA generative AI to reshape the entire vehicle lifecycle process from design concept to finished vehicle which includes safety engineering and full self-driving. NVIDIA’s CEO, Jensen Huang, associates the current AI-driven transformation with the evolution of mobile applications such as GPS and other sensors in smartphones. He suggests that foundational AI models, such as ChatGPT for text and Stable Diffusion for images, serve as a base on which innovators can build the next wave of applications. “Generative AI, large language models, and recommender systems are the digital engines of the modern economy,” says Huang.

Toyota Transforms Design Workflow

Toyota is one automotive company that is incorporating NVIDIA generative AI into its initial sketches and engineering limitations into the design workflow. The Toyota Research Institute (TRI) unveiled this innovative approach, seeking to enhance its designers’ creative input, while at the same time optimizing critical elements such as aerodynamics from the beginning. TRI has published their research report, “Drag-guided diffusion models for vehicle image generation” in which the TRI specialists detail the methodology through which engineering limitations are integrated into their design process. The authors explain that they have included factors such as drag, fuel economy, and important chassis measurements like the cabin’s dimensions and ride height, which have important implications for vehicle handling, comfort, and safety. According to Avinash Balachandran, director of TRI’s Human Interactive Driving Division, this methodology approach combines Toyota’s engineering expertise with leading-edge NVIDIA generative AI capabilities, resulting in designs that satisfy both aesthetic and functional requirements from the start.

NVIDIA Automotive Platform and Software Development Kits

• NVIDIA TAO: Automakers are able to leverage this technology to develop robust models using techniques like transfer learning on automotive datasets.
• NVIDIA Picasso: This provides automakers with a cloud-based platform for creating custom generative models.
• NVIDIA Omniverse: This gives automakers a portal to access leading generative engines that automate tedious workflows in the design process. By utilizing NVIDIA’s Omniverse, 2D sketches are instantly transformed into NURBS models, helping designers visualize in 3D. In addition, it helps automakers build digital replicas of their manufacturing units facilitating smoother real-world operations and saving the company financially.
• NVIDIA’s NeMo: Automakers are able to develop chatbots for improving customer interactions with language and image applications. For example, Mercedes is pioneering an AI voice assistant using natural language generation (NLG).

In addition, NVIDIA researchers are playing an active role in the development of autonomous driving employing neural radiance fields (NeRFs) to turn AV sensor data into fully interactive 3D environments for scaled testing and validation. NVIDIA generative AI is playing an important role in helping the automotive industry to transform the vehicle lifecycle workflow process. This is good, not only for automakers, but also for consumers, as it will result in not only aesthetically, modern cars, but also better engineering with humans as “co-pilots with generative AI.

Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.

Other insights from The Futurum Group:

NVIDIA AI Workbench Could Simplify Generative AI Builds

NVIDIA Q1 Earnings

5G Factor Video Research Note: NVIDIA Softbank Collaborate on Gen AI and 5G/6G Apps