Volvo reconstructs crashes with AI in virtual worlds to make safer cars
Briefly

Volvo's ADAS developer, Zenseact, has shifted to developing software in-house, significantly improving innovation speed and testing capabilities. By leveraging advanced simulation technologies like Gaussian splitting, they can instantly create multiple scenarios for more thorough testing. This approach enables daily software iteration, facilitating rapid responses to any issues. Volvo's substantial investment in data centers further enhances their ability to simulate various driving scenarios, ultimately aimed at increasing safety and reducing accident rates in autonomous vehicles. The transition to software-defined vehicles (SDVs) signals a major evolution in how automotive technology evolves.
"If there's something, we solve it in a day," Coelingh said. "It's so much faster. So we iterate much faster... innovation speed is fundamentally different than before."
"Gaussian splitting is a technology where we can take one point, one traffic scenario, and explode it into thousands or tens of thousands of scenarios from this real-world data... and then we can enclose the simulation and test our software against this."
Read at Ars Technica
[
|
]