Linda Poon is a staff writer at CityLab covering science and urban technology, including smart cities and climate change. She previously covered global health and development for NPR’s Goats and Soda blog.
Electric bus company Proterra is teaming up with the University of Nevada, Reno, to see how cities can communicate with driverless public transit.
Ride along Virginia Street in Reno, Nevada, you’ll cross through many different parts of the city. There’s the more rural landscape in the north, and as you head south, you’ll hit increasing foot and bike traffic near the University of Nevada, Reno (UNR), before reaching heavy traffic downtown. That variance makes the road a perfect testing ground for autonomous buses.
That’s why, earlier this year, electric bus maker Proterra began quietly using Virginia Street as as a living lab for the next big innovation in public transit. Since January, the company has been working with Reno’s transportation agency and engineers, roboticists, and AI experts at UNR’s Living Lab Coalition to collect data on how to integrate autonomous public transit into cities—starting with the bus. The partnership was announced Tuesday at a green tech conference in New Orleans.
But this isn’t like other self-driving pilots that have sprung up elsewhere. The project isn’t just about teaching the vehicles how to safely navigate the city, but also about figuring out how to prepare a city’s infrastructure for autonomous public transportation. As the buses roll out, researchers are considering what sensors and communication tools are needed, for instance, and what variables the vehicles need to be aware of in the world around them.
The research team is currently outfitting the city’s high-occupancy Proterra buses with a variety of cameras and sensors that, for example, use light to track the distance of nearby objects and heat to detect the presence of pedestrians. As the buses travel along Virginia Street, researchers are gathering data about how well the sensors detect traffic ahead or perceive a potentially dangerous encounter with other vehicles or pedestrians.
Streetlights will also be equipped with radios that can tell the buses what’s happening at an intersection up ahead. That’s the first phase of the project, which will begin in the coming months after the sensors are fully installed. That phase is expected to last at least a year so researchers can gather data during all four seasons. The goal is to test the sensors in as many different road conditions as possible.
“When you have snow on the ground, the way these vehicles and sensors are going to behave is going to be completely different to a rainy day and to extreme heat during the summer,” says Carlos Cardillo, director of the Nevada Center for Applied Research, who leads the Living Lab program. “We also have this ability to change environment, to have traffic conditions from our most deserted highways to the more moderate ones to heavy traffic in the metro area.”
Eventually all that data will be mined and used to develop a set of “robotic” perception algorithms—that’s phase two of the project, which will take place after team feels that adequate data has been collected. “You have to put lots of sensors on the bus, and then you have use algorithm to take all of the data coming from all of those different sensors and fuse them into a single coherent picture of the environment around the vehicle,” says Richard Kelley, a roboticist on the team. Phase three involves licensing and commercializing their end results, hopefully bringing what they learn to cities across the U.S. and maybe even the globe.
It’s perhaps fair to say fully autonomous buses likely won’t be coming en masse to cities anytime soon, something pointed out by researchers and regulators alike. Even Proterra CEO Ryan Popple acknowledged the long road ahead in an interview last year with Bloomberg. Still, chief commercial officer Matt Horton says the collaboration could not have come at a better time.
“All of these sensing and awareness technologies are very relevant today,” he tells CityLab. “These are active safety tools that will augment the drivers' existing capabilities, even if the world doesn't move toward the vision of an all-autonomous future.” The sensors, for example, can improve the bus driver’s ability to detect a car in its blindspot.
When asked whether Proterra plans to jump into the business of developing autonomous buses, Horton will only say that the company will listen to what their customers demand—and right now electrifying fleets and improving safety are two top priorities for transit agencies.
Asked for his thoughts on the future of an all-autonomous public transit, Kelley emphasizes the main focus of their research is how tech can improve safety in a city that’s more likely to see a mix of manned and unmanned vehicles on the road. “Downstream, the question of what it looks like for a city when you have a large number of autonomous vehicles is something that we're going to have to watch and respond to,” he says. “The technology is going to enable such radically different things that if we try to plan out exactly what that future looks like, I think we'll probably get it wrong.”