Volvo uses AI and virtual worlds to train driver assistance systems
Volvo Cars has announced the use of artificial intelligence and virtual simulations to improve the safety of its cars. ADAS (driver assistance systems) can now learn not only from real accidents, but also from thousands of simulated scenarios that Volvo creates using advanced 3D technology.
Here's What We Know
Volvo analyses data on emergency braking, sudden manoeuvres and driver interventions collected by the car's sensors. This is done using a hybrid way of representing 3D data, also known as "Gaussian splatting". This technology allows the creation of realistic 3D scenes based on real images. In these virtual environments, it is possible to modify the behaviour of other road users by testing the system under different conditions.
Depicting possible accident scenarios using the "Gaussian splatting" technology. Illustration: Volvo Cars
According to the developers, this significantly speeds up autopilot training: whereas it used to take months to collect data in rare and complex situations, it can now be tested in a matter of days.
Volvo combines real-world tests and virtual modelling to create the safest cars possible. The project is being carried out together with Zenseact (Volvo's AI subsidiary) and supported by the Wallenberg AI, Autonomous Systems and Software Programme (WASP). The research is also part of a PhD programme at leading universities in Sweden.
In the future, such technologies could make safety systems even more adaptive and efficient - especially in unpredictable situations.
Source: Volvo Cars
