Mapping the future of Toronto's autonomous vehicles

Fancy Maps from Ecopia AI and autonomous shuttles are coming to life in Toronto as part of the city's ambitious AV plans

Sami Haj-Assaad

Oct 14, 2021 • October 14, 2021 • 5 minute read • Join the conversation

Ecopia AI Kitchener-Waterloo test map Photo by Ecopia AI

Back in 2019, Toronto laid out a plan that would get the city ready for autonomous vehicles as soon as 2022. It was the first plan of its kind in North America, and helped to give Ontario’s capital a shiny high-tech badge to show off.

Story continues below

This advertisement has not loaded yet, but your article continues below.

But making a plan and seeing it through are two different things. While things around the world became sidelined thanks to the COVID-19 pandemic, things are picking back up again in Toronto’s quest to become autonomous-vehicle friendly.

Self-driving vehicles need data to navigate the streets — and lots of it. This data usually comes in the form of detailed maps that can be interpreted by the vehicle’s sensors and onboard computers. These maps are the rosetta stone for autonomous vehicles, and are one of the most important aspects of the whole equation.

Ecopia AI: Toronto’s AV Cartographer

The city’s recent announcement of a mapping partner, Ecopia AI, is a significant step in realizing the AV roadmap. However, Toronto is a busy, bustling city with lots of obstacles to overcome when it comes to accurate mapping. From an overabundance of signs and rules to streetcar and bike lanes, it seems the streets of the big city can confuse even seasoned drivers — especially with major construction projects popping up all the time.

Story continues below

This advertisement has not loaded yet, but your article continues below.

Fortunately, Ecopia AI makes use of several different sources to create its maps. “Ecopia’s HD map creation process involves imagery sourced remotely from aerial and street-view imagery providers,” said Jon Lipinski, co-founder and President of Ecopia AI in an email interview. “Ecopia’s proprietary AI systems are able to leverage the most accurate and up-to-date geospatial imagery and generate HD vector maps with Geographic Information Specialist (GIS)-professional level accuracy at scale.”

You’ll note that Ecopia AI seems to be staying away from leveraging LiDar from its mapping processes, and seems content on the accuracy and detail in its HD mapping sources. Those unconvinced are encouraged to see the company’s latest results of mapping the Waterloo region to enable the deployment of autonomous vehicles, found at Open HD Maps.

Story continues below

This advertisement has not loaded yet, but your article continues below.

Looking through a demo map of Kitchener, Waterloo and Cambridge, you can see it is loaded with information. “The maps created by Ecopia include the necessary land cover features such as every building, curb, driveway, parking spot, road, road centerline, white line, yellow lane, bike signs, and turns that serve the need for accurate foundational data for safe deployment of AVs,” explained Lipinski.

There are some missing points of data that may also be crucial for AV deployment, including height and depth (for example, knowing whether a bridge may be an obstacle for a taller vehicle) and perhaps the location of traffic lights and signs, though some of these details can likely be added in through AI or manually.

Story continues below

This advertisement has not loaded yet, but your article continues below.

The aerial mapping process may be a concern where roads are on top of each other, like how the Gardiner covers Lakeshore in Downtown Toronto, but overall the maps seem detailed enough to provide a starting point for Toronto’s AV ambitions.

“The development of a high definition map of Toronto is a key milestone in our Autonomous Vehicle Tactical Plan, and Ecopia’s efforts will help drive us towards that future. This initiative is a great example of Toronto’s robust innovation ecosystem and our ability to foster public and private partnerships that help us achieve common goals and a vision for a healthy, equitable, liveable and sustainable city,” said John Tory, Mayor of the City of Toronto about the partnership with Ecopia AI.

Story continues below

This advertisement has not loaded yet, but your article continues below.

Self Driving Trials Arrive With Shuttles in Toronto, but is it a first?

Another part of Toronto’s self-driving plan is coming to life in the form of a trial of an autonomous shuttle that’s being deployed out of the Rouge Hill GO Station which provides transportation through the West Rouge neighbourhood. This trial program commences in October and requires passengers to book a trip as needed, rather than just hopping on and off.

So far, there’s no plan to make this a permanent mobility solution. The current project is a learning exercise that will show how AV can be deployed in the city, how it will perform in fall and winter weather, and how passengers and other motorists will react to self-driving shuttles.

While Toronto was the first North American city to develop an AV roadmap, Whitby became one of the first cities to deploy an autonomous shuttle into an already established transit route, DRT Route 300.

Story continues below

This advertisement has not loaded yet, but your article continues below.

The shuttle makes a six-kilometre journey from the Whitby GO Transit station, looping through the Port Whitby area. At the time of its trial, it was the longest AV shuttle route on the continent. Like the Toronto shuttle, this is just a trial to see how shuttles can integrate into the broader transit plan in the future.

The vehicle used in these trials is a Local Motors Olli 2.0, which has a capacity for eight passengers, as well as an AV attendant in case there’s something on the route that the vehicle can’t handle on its own.

The Force Behind It All: Autonomous Vehicle Innovation Network

Toronto’s AV roadmap as well as other projects around the province are a byproduct of Ontario’s Autonomous Vehicle Innovation Network (AVIN) and the Driving Prosperity initiative which helps to grow the auto sector in the province to push more companies to invest. The network is an 85-million dollar program that will work with small and medium-sized enterprises working in the world of high-tech mobility including driverless cars, vehicle cybersecurity, and real-time fleet routing.

Story continues below

This advertisement has not loaded yet, but your article continues below.

AVIN is behind many of the province’s AV pushes, providing support and resources. For example, Ecopia AI received $994,706 from the Government of Ontario to match an industry contribution of $2,039,561 to develop the mapping technology related to Toronto’s AV project.

“Through pilot programs for testing automated vehicles on Ontario roads, and investments into innovative projects like this initiative by Ecopia, Ontario has established itself as a global leader in the connected and automated vehicle industry,” said Caroline Mulroney, Minister of Transportation. “Autonomous vehicles have the potential to make transportation more efficient, safe, and environmentally friendly. Our government is proud to support Ontario-based businesses working to make a more connected and convenient transportation system a reality.”

By providing the means to connect the province’s post-secondary institutions, R&D centres, innovation hubs, and connected and autonomous (C/AV) testing facilities, AVIN is playing a significant role in the development and hopeful rollout of next-generation mobility technology. With the recent progress in Toronto and elsewhere in the province, it seems that the AV future is getting ever closer.