With many companies developing autonomous vehicles around the world, the University of Toronto believes that the winter driving environment needs to be given more attention and has formed a partnership to tackle this very issue.
“Winter conditions aggravate the remaining challenges in autonomous driving,” said Steven Waslander, an associate professor at the University of Toronto Institute for Aerospace Studies (UTIAS) in the Faculty of Applied Science and Engineering. “Reduced visibility limits perception performance, and slippery road surfaces are a big challenge for vehicle control.”
For autonomous vehicles to drive safely and efficiently in all conditions, especially winter, vehicles need to fully observe their surroundings despite limits to their sensor range to get advanced warning of challenging situations and to react quickly to changing conditions, Waslander explained.
As leader of the new partnership, he is bringing together a range of University of Toronto Engineering researchers and members of the university’s Robotics Institute. These include Timothy Barfoot, Jonathan Kelly and Angela Schoellig, alongside other notable academia, and industry minds.
The project, named WinTOR: Autonomous Driving in Adverse Conditions, seeks to transform Toronto into a global hub for research and development related to autonomous driving in winter. The project includes multiple partners from the autonomous vehicles sector, including General Motors Canada, LG Electronics, Applanix and Algolux.
A fund of C$$12m (US$9.5m) supports the new partnership, provided by an array of sources including the Ontario Research Fund, the Natural Sciences and Engineering Research Council of Canada, and donations from project partners. The team will be made up of 20 individuals including graduate and undergraduate students, professors and engineers.
“With this investment, the Ontario Research Fund is supporting important work of U of T researchers that will benefit all Canadians,” explained Christine Allen, associate vice-president and vice-provost, strategic initiatives. “And it’s exciting to see that two of the initiatives to receive funding, focused on advanced robotics and innovative mobility, are among U of T’s key areas of strategic, interdisciplinary focus.”
“Algolux’s mission is to solve the issue of computer vision robustness in harsh driving conditions, a fundamental problem not effectively addressed by current approaches,” said Felix Heide, co-founder and chief technology officer, Algolux. “As a Canadian company, we are thrilled to bring our expertise to this project and continue to advance the state-of-the-art in perception technologies.”
The WinTOR project is divided into three main areas:
- Sensor filtering for object detection: new ways of analyzing the data from sensors such as visual cameras, radar and lidar will help to separate the signals that represent real objects from the noise caused by falling or blowing snow. Strategies will include both pre-processing techniques and improved artificial intelligence algorithms trained to be aware of the limits of their own performance.
- Sensor fusion, localization and tracking: While today’s self-driving cars can reliably determine where they are in relation to their surroundings, the techniques they use begin to break down under adverse driving conditions. The team will leverage new algorithmic strategies in vision and lidar registration, as well as new sensing options, such as ground-penetrating and automotive radar, to make localization algorithms more resilient in adverse conditions.
- Prediction, planning and control: self-driving cars of the future will need to change the way they drive in response to winter hazards. For example, they might take a slightly different path to avoid a snowdrift or slow down when driving over a section of road that their sensors have perceived as particularly slippery. They will learn the implications of adverse weather on the vehicles around them and be able to assess the increased uncertainty of outcomes, enabling them to plan actions that can be executed reliably in winter conditions.
“We continue to seek additional faculty, partners and funding to grow the effort. We have many more ideas to work on, from multi-hypothesis prediction and interaction planning to attentive perception and explainable, efficient AI for autonomous driving,” Waslander said.