8th April 2025
Volvo Cars is harnessing the power of AI-generated, lifelike virtual environments to accelerate the development of its safety technologies, including advanced driver-assistance systems (ADAS), with the ultimate goal of making driving even safer.
By transforming incident data, such as emergency braking, abrupt steering, or manual driver intervention, captured by sensors in their latest vehicles, we can now simulate, reconstruct, and analyze real-world events in entirely new ways. This helps them gain deeper insights into how accidents can be prevented.
At the heart of this innovation is a cutting-edge computational method known as Gaussian splatting. This technique enables the creation of detailed, realistic 3D environments from actual visual data. Within these digital worlds, we can modify variables like road users or traffic behavior to explore how different scenarios might play out.
This breakthrough allows our safety systems to be tested against a wide variety of traffic conditions, including rare and complex edge cases, at a scale and pace previously unattainable. What once took months to achieve can now be done in just days, helping us refine and improve safety features faster than ever before.
“We already have millions of data points of moments that never happened that we use to develop our software,” says Alwin Bakkenes, Head of Global Software Engineering at Volvo Cars. “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.”