AV researchers release winter-driving training tool

Monday, February 3, 2020

Autonomous vehicle (AV) developers now have a new winter-weather tool at their disposal thanks to engineering researchers.

Teams at Waterloo Engineering and the University of Toronto collaborated on a free, open-source dataset to test and train AV perception algorithms in nasty conditions.

“We want to engage the research community to generate new ideas and enable innovations,” said Krzysztof Czarnecki, the engineering professor who leads the Waterloo team.

Krzysztof Czarnecki

Krzysztof Czarnecki leads the Autonomoose research team at Waterloo Engineering.

Released today, the dataset is a collection of over 100 annotated video clips recorded by Autonomoose -the highly automated Lincoln MKZ hybrid that has been under development at Waterloo for more than three years - in a variety of Waterloo Region winter driving conditions.

Click here for the full story.