Four autonomous May Mobility shuttles are operating in Grand Rapids, Mich., part of a year-long pilot project to study their viability.Peter Nowak/The Globe and Mail
With its glass roof and bench-like seating for four, May Mobility’s autonomous shuttle feels like a cross between a Tesla and a golf cart. The electric engine powering the vehicle and its 40-kilometre-an-hour top speed only deepen that impression on both fronts.
The particular shuttle in which our group of journalists is riding is among four operating since July here in Grand Rapids, the second-largest city in Michigan. City officials consider the year-long pilot a test that has relevance for every major municipality – including a number of Canadian cities currently engaged in or considering similar projects.
After all, if self-driving buses can make it in Michigan – where car culture is in the blood – they can likely succeed anywhere.
“This is way more a sociology experiment than it is a technology experiment. We’re trying to figure out, how do you get people to use this?” says Mike Morin, director of Start Garden, a startup accelerator in Grand Rapids thatʼs attached to the project.
“You’re really not going to learn what this is until you put it on the streets and see how the community deals with it. We’re Michigan – if you can own a car, you do own a car.”
So far, officials are pleased with the results. The shuttles, which patrol a set five-kilometre route spanning 20 stops in the downtown core, tallied 26,000 rides in their first three months of service. They run Tuesday to Saturday and are free to use.
Each vehicle is equipped with lidar (laser-based sensors that can detect distances to an object), radar and cameras, and communicates wirelessly with sensors built into light posts and other street infrastructure along the route. The ultimate goal is to create a fast and easy-to-use public transit option that’s more efficient and affordable than human-driven buses.
A human operator sits at the controls for now, ready to take over. During our short, 10-minute ride, it happens frequently. Our operator steps in to guide the shuttle to a stop and again to ease it through an intersection.
That’s on purpose, explains Edwin Olson, chief executive officer of Ann Arbor, Mich.-based May Mobility Inc. Some autonomous test vehicles in other jurisdictions have angered human drivers by being too cautious. In Arizona, for example, motorists have thrown rocks at Google’s Waymo vehicles for idling at intersections and stop signs for too long, waiting for traffic to clear.
In Grand Rapids, having a low threshold for operators to step in is tacit acceptance that the technology is still a work in progress – a way to gather data and figure out what works in real-world conditions.
“We can help alleviate congestion, but if we’re being a pest to other vehicles, we’re not accomplishing that,” Olson says.
The next steps for May Mobility will involve increasing the frequency of shuttles and decreasing the amount of human intervention.
Similar autonomous shuttle buses are in various stages of testing in a number of Canadian cities. The town of Beaumont, Alta., south of Edmonton, began a six-month pilot in May, while Toronto has announced it intends to begin testing next year.
Montreal also concluded a pilot recently, which saw self-driving shuttles operating at a maximum speed of 15 km/h on the streets between Olympic Stadium and the Maisonneuve Market. City officials are now analyzing the data gathered before deciding on next steps.
The technology worked well, according to Transdev Canada Inc., the Brossard, Que.-based company running the project. Passengers generally enjoyed their rides, with the main complaint being that the shuttles drove too slowly.
“When people get in an autonomous shuttle, even in heavy traffic, they rapidly forget they’re in an autonomous shuttle,” says Camille Boulier, vice-president of business strategy and communication for Transdev. “The main lesson we got from Montreal is that people are ready.”
The biggest obstacle to further testing, she adds, is funding. Pilot projects can require hundreds of thousands or millions of dollars. Convincing governments to help fund them can be challenging.
Amir Khajepour, the Canada Research Chair in mechatronic vehicle systems and a professor at the University of Waterloo, would like to see more public-funded testing as well. (Mechatronics is a multidisciplinary branch of engineering.)
Much of the work being done in autonomous driving in the United States is being bankrolled by large companies such as Google and Uber. Canada doesn’t have the same level of privately funded activity, which means the country is behind.
Khajepour has tested a drive-by-wire shuttle – one that is steered by electronic systems rather than mechanical – on the University of Waterloo campus and plans to launch a full self-driving version next year. His goal is to make the generated data open source and free to use for non-commercial researchers, which would ultimately lower the cost of such tests.
“One of the difficulties in any such research is the availability of data,” he says. “Obviously, those commercial companies are not sharing their data.”
Shopping for a new car? Check out the new Globe Drive Build and Price Tool to see the latest discounts, rebates and rates on new cars, trucks and SUVs. Click here to get your price.
Stay on top of all our Drive stories. We have a Drive newsletter covering car reviews, innovative new cars and the ups and downs of everyday driving. Sign up for theweekly Drive newsletter, delivered to your inbox for free. Follow us on Instagram,@globedrive.