It looks like Volvo has found a way to utilise artificial intelligence in a way we can all get behind. Using data gathered from its fleet of cars, the Swedish car maker is synthesising various situations and incidents to test and study its cars' safety features.
This is all done through a new technique that is known as Gaussian splatting, and it uses 3D scenes and subjects from real world visuals to create a virtual world that can then be manipulated to Volvo's desire. From here, they can create a slate of road situations and behaviours to see how their cars would fare in different outcomes.
The hope of this idea is that it will enable the company to test situations that would otherwise be too complex or rare to create physically, and thus allow them to begin creating counters and safety features that could further protect the driver and passenger from harm.
Volvo's head of global software engineering, Alwin Bakkenes, stated: "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 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."
Gaussian splatting is powered by Nvidia technology, and even uses an AI supercomputing platform to study the data and provide insights so that Volvo can continue to lead the automotive sector as a safety pioneer.