[Corrected, 10:16 am. See below] There has been a lot of speculation about what Ann Arbor, MI-based May Mobility has been up to since the company began working in stealth mode early this year. The new company has to do with autonomous vehicles—one of the hottest fields in tech—but until this week its website revealed only a landing page with a “contact us” e-mail address.
That veil of silence has now been partially lifted. May Mobility has emerged from stealth with the announcement that it has just completed a stint in Y Combinator, an elite startup accelerator in Silicon Valley, and is headed back to Michigan to start work on a set of pilot projects. TechCrunch reported on Monday that the company has raised about $3.5 million in venture funding from investors including Maven Ventures (a Silicon Valley seed fund, not to be confused with GM’s mobility arm), Trucks, and Tandem. [Corrected to identify Maven as a seed fund, not affiliated with GM—Eds.]
May Mobility is the brainchild of Ed Olson, a University of Michigan professor, researcher, and director of the university’s APRIL lab, which focuses on robot perception, coordination, and planning. In a phone interview from Palo Alto, CA, he told us a bit more about his startup.
He says May Mobility, with 15 employees, is a “fleet as a service” company offering an “end-to-end mobility solution.” Although the word fleet brings to mind long-haul truckers, that’s not quite what it means in the world of autonomous vehicles.
Big automakers like Ford and tech giants like Google (Waymo) and Uber are racing to own the technology behind self-driving vehicles in what could eventually be a multi-trillion-dollar industry. But rather than targeting consumers directly, some companies and startups are working on aftermarket technologies that are carmaker-agnostic and could be used to implement a vision of shareable autonomous vehicles, in the form of fleets of taxis, shuttles, and the like.
Ted Serbinski, who directs the Techstars Mobility program, says he thinks the definition of fleet will expand to mean simply “more than one car” with the dawn of self-driving vehicles. “If a car can drive itself, it’s a lot easier to manage a fleet,” he notes.
“There are things the robotics community has figured out about autonomous vehicles,” Olson says, describing May Mobility’s innovation in terms that are deliberately vague. “So it’s that, plus intellectual property from [the leadership team’s] previous lives.”
Olson told TechCrunch that May Mobility has created a technology stack that could be an aftermarket installation “to offer autonomy” on existing vehicle platforms designed for fleet service. The company plans to oversee a full fleet operation service, as well.
The average car owner doesn’t drive their vehicle all day long, Serbinski says. They use it to commute, run errands, schlep kids around, and maybe take the occasional road trip. In between those activities, the car stays idle. So in an autonomous world, where shared ownership or leasing models become more prevalent, driverless cars could be deployed by a fleet-as-a-service company like May Mobility instead of being sold to individual consumers.
Olson has identified central business districts, corporate campuses, and other “small, compact areas” with both public and private roads as the company’s primary market.
In 2010, Olson led the winning team in the U.S. Department of Defense’s MAGIC competition after developing a crew of 14 robots that semi-autonomously explored and mapped a 500-by-500-meter arena simulating an urban environment. Named one of Popular Science’s “Brilliant Ten” in 2012, he also spent many hours in university parking lots refining software to help Ford’s small fleet of autonomous research vehicles, a collaboration between U-M and the automaker, see and navigate their surrounding areas.
Olson featured prominently in our coverage of the first demo day at the university’s autonomous vehicle test track, MCity, in 2015. At the time, he was heading up an initiative to create a fleet of 3D-printed autonomous cars in partnership with Local Motors to ferry students around campus. (Olson also worked at the Toyota Research Institute as co-director for autonomous driving development.)
But in January 2017, he started May Mobility and has spent much of the year quietly building his team. Alisyn Malek, who worked in a number of roles at GM, including a stint as an investor with GM Ventures, came on board as co-founder and chief operating officer. Steve Vozar, a fellow U-M alum with a background in robotics research, is the third co-founder and will serve as chief technology officer.
At the “last-minute suggestion” of one of the company’s seed investors, Olson says, May Mobility applied to Y Combinator. So far, the experience has been positive. “My network in Silicon Valley wasn’t good, plus I’m a first-time CEO,” Olson says. “It’s been really useful.”
Despite improved connections in the Valley, Olson says the company has no plans to leave Ann Arbor. “It has the ecosystem, the talent pipeline, and the cost of doing business is a lot less,” he adds.
This fall, May Mobility will start two pilot projects in downtown Detroit and Warren, MI, along with two more in Florida. “We see these pilots as making contact with the ball,” Olson says. “It’s a simple version of our system. Then, in 2018, we’ll begin operations without a safety driver.”
It’s an ambitious plan, but Olson says he thinks it’s ultimately a faster way to market. “The approach being taken by the [major automakers] is not the right path,” he says. “It’s too large and unwieldy.”
Techstars’ Serbinski isn’t sure being first to market should necessarily be the goal for companies like May Mobility.
“I don’t think it matters a whole lot in a highly competitive market,” he says. He uses the example of Facebook dominating its competitors, many of whom pre-dated it, to illustrate his point. “Sometimes first to market can be a negative thing because your competitors can learn from your mistakes. First to scale is probably more important.”
Image courtesy of May Mobility.