Volvo Cars is leveraging artificial intelligence (AI) and advanced virtual environments to enhance the development of its safety software, including driver assistance systems (ADAS), with the goal of creating even safer vehicles.
By synthesising incident data collected from the advanced sensors in their new cars – such as emergency braking events, sharp steering manoeuvres, or manual interventions – Volvo engineers can reconstruct and analyse these scenarios in detail. This process enables a deeper understanding of how incidents occur and how they can be prevented in the future.
This innovative approach is made possible through a computational technique known as Gaussian splatting. This method allows for the creation of a vast array of realistic, high-fidelity 3D scenes and subjects derived from real-world visuals. Within these virtual environments, engineers can manipulate various factors – such as adding or removing road users, altering traffic behaviours, or introducing different obstacles – to explore how these changes impact safety outcomes.
“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 corner 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.”
The use of such advanced virtual testing environments allows Volvo to conduct extensive testing and development without the logistical constraints and potential risks associated with real-world testing. This approach not only accelerates the development process but also contributes to Volvo’s commitment to a ‘better tomorrow’ by prioritising safety and innovation.
