Can self-driving cars handle a Canadian winter?

Newly released data from a collaboration between U of T Engineering, the University of Waterloo and Scale AI will help train future self-driving cars to handle the challenges of winter driving.

This week, Professors Steven Waslander (UTIAS) and Krzysztof Czarnecki (University of Waterloo) and their teams unveiled the Canadian Adverse Driving Conditions (CADC) dataset. Based on scans of real Canadian roads, the dataset acts as a virtual training course for the computer algorithms that enable cars to drive themselves.

“There are lots of great training datasets out there already, but they were collected on sunny, summer days,” says Waslander. “If you take algorithms trained on those datasets and try to use them in adverse conditions, they tend to get confused. They can misclassify objects — such as pedestrians and other vehicles — or even miss them entirely, all because of the changes in sensor data caused by snowfall.”

More information about winter driving datasets