Waymo is set to venture into Toronto, navigating its way through the city’s frequently snow-covered roads, a challenge that will require time and extensive training data to conquer, according to experts. The U.S. firm recently announced plans to commence testing in Canada’s largest metropolitan area, initially with human drivers at the wheel in the upcoming days, focusing on mapping and safety validation tasks. The ultimate objective, as Waymo revealed, is to have driverless taxis available for summoning by 2028 within the city.
To ensure readiness for the diverse driving conditions in Toronto, Waymo intends to conduct round-the-clock tests throughout the week. The company, however, did not elaborate on the specific challenges posed by these conditions. Nevertheless, Steven Waslander, the director of the Robotics Institute at the University of Toronto, emphasized that the pivotal factor distinguishing Toronto is its winter climate, a feature not yet addressed commercially by Waymo in its existing operations in warmer cities like Los Angeles and Phoenix.
Autonomous vehicles heavily rely on various sensors such as cameras, radar, and lidar systems to interpret their surroundings and make decisions. The presence of snow and ice can impede the effectiveness of these sensors, complicating object detection and navigation. Overcoming these challenges demands continuous testing and refinement in adverse weather conditions, a process that Waymo has been actively engaging in, including operations in snowy regions like upstate New York and Michigan.
Despite the progress made, Waymo acknowledges the need for more winter driving data to enhance its systems further. The company employs innovative solutions like heaters and mini windshield wipers to maintain clear sensors and collaborates with other vehicles to gather real-time weather data. As Waymo gears up for its Toronto debut in 2028, the company aims to address and resolve any winter-related issues through ongoing technological advancements and data-driven strategies.
Experts suggest a phased approach to autonomous driving deployment, with systems initially operating in favorable conditions and gradually expanding capabilities to adapt to more challenging environments. This evolutionary process, they agree, will unfold over several years, reflecting the intricate nature of integrating self-driving technology into diverse and dynamic urban landscapes.