How Volvo is Using AI to Predict Car Accidents Before They Happen

3 min read
Co-founder of JustAINews
Share
Key Points
  • Volvo Cars adopts Gaussian splatting to create realistic virtual worlds for safer and faster safety software development.
  • AI-generated environments allow Volvo to simulate rare and complex traffic incidents, significantly speeding up software testing.
  • Collaboration with NVIDIA, Zenseact, and Wallenberg AI helps Volvo leverage powerful AI technologies to predict and prevent accidents.
Refreshed Volvo XC60 exterior in a road in the mountains
Credits: Volvo Cars

Volvo Cars is now using AI to generate lifelike virtual worlds to improve the development of safety software, particularly driver-assistance systems (ADAS). These virtual scenarios are based on real-world sensor data gathered from Volvo’s latest cars, capturing incidents like emergency braking or sharp steering.

By turning this data into detailed virtual worlds using a technique called Gaussian splatting, Volvo can create realistic 3D environments. Engineers can manipulate these virtual environments, adjusting road users, traffic behavior, and obstacles to analyze numerous potential scenarios.

Illustration that showcases how does Volvo uses AI to prevent car accidents
Credits: Volvo Newsroom

Predicting the unpredictable with Volvo's AI

The power of AI lies in its ability to simulate countless rare yet dangerous traffic situations, known as "edge cases." Previously, testing these cases could take months. Now, Volvo can evaluate thousands of these scenarios in just days, significantly speeding up the software development process.

“We already have millions of data points of moments that never happened that we use to develop our software. Thanks to Gaussian splatting, we can select one of the rare edge cases and explode it into thousands of new variations of the scenario to train and validate our models against. This has the potential to unlock a scale that we’ve never had before and even to catch edge cases before they happen in the real world.”

Alwin Bakkenes, Head of Global Software Engineering at Volvo Cars

Collaboration and innovation

Volvo collaborates closely with AI and software company Zenseact to develop these virtual environments. Additionally, Volvo participates in research supported by Wallenberg AI, Autonomous Systems and Software Program (WASP), exploring neural rendering technologies for future safety enhancements.

This initiative builds on Volvo’s longstanding tradition of using data-driven approaches to improve vehicle safety. Historically, data collected by Volvo’s Safety Research Team led to the creation of influential safety innovations, including the Whiplash Injury Protection System and Side Impact Protection System.

Accelerating development with NVIDIA

Volvo’s partnership with NVIDIA further accelerates its AI capabilities. Volvo’s latest electric vehicles, built on NVIDIA accelerated computing technology, gather detailed data through advanced sensors.

NVIDIA's AI supercomputing platform processes this information rapidly, helping Volvo refine its safety software and speed up AI development.

This platform is part of a major investment by Volvo Cars and Zenseact to create one of the largest data centers in Northern Europe, significantly boosting Volvo’s ability to leverage AI for safer roads.

Additionally, Volvo’s groundbreaking work on Gaussian splatting and generative AI will be presented at the NVIDIA GTC conference. The presentation, accesible via this link, is available live and on-demand, providing further insights into the technology behind Volvo’s safety advancements.

Tagged: , ,

Subscribe to JustAINews!

Get the industry's biggest AI news straight to your inbox.
Subscribe

Related posts

© Copyright 2025 - Just AI News - All Rights Reserved
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram