Winning over an uncertain public is one of the biggest challenges facing the realisation of autonomous vehicles (AVs). In the last few years, a wave of high-profile accidents has caused public excitement for driver assistance and self-driving technology to turn into discomfort.
One notorious incident involved a Tesla Model S using a beta version of the automaker’s Full Self-Driving feature: the car came to an abrupt halt while crossing the San Francisco Bay Bridge, resulting in an eight-car pile up. Nine people were injured, and traffic was held up for more than an hour. SAE Level 4 developers are faring no better. In November 2023, General Motors’ Cruise had its licence to operate on Californian roads revoked following an incident in San Francisco the month prior.
Consumer sentiment towards automation technology is nearing rock bottom in the US. 93% of respondents in a January 2024 Forbes Advisor poll indicated they had concerns about some aspect of self-driving cars, with safety and technology malfunctions the most widespread. 61% indicated they would not trust an AV to transport a loved one. These concerns are backed by National Highway Traffic Safety Administration data: 9.1 crashes occur per million driverless miles, compared to only 4.2 in human-operated vehicles.
Growing distrust means that companies operating in the autonomous driving space must make significant efforts to demonstrate the safety and reliability of their technology. For Michigan-based AV firm May Mobility, this means two things. First, developing self-driving software should not be bound to a rigid and therefore potentially dangerous rules-based system. An entirely rules-based system not coded for every possible scenario will inevitably lead to unsafe behaviour. Second, self-driving technology must be deployed in situations where everyday members of the public—not a self-selecting audience of robotaxi users—can actually experience it.
Safety is everything
Unlike many firms in the autonomous driving space, May Mobility has kept a low profile. This is partly due to a lack of controversy: in the nearly seven years that May Mobility has been in operation, there have been minimal reported crashes, injuries, or traffic jams. Despite this, the company has continued to raise significant amounts of money while others have seen their funding dry up. In November 2023, it announced a US$105m Series D investment round led by Japanese telecommunications company NTT.
The secret to this success is May’s prioritisation and strong track record of vehicle safety. “We develop AVs, but we’re particularly focused on validating this amazing technology that just works and is safe to use,” remarks Sarah Gryniewicz, Strategic Commercial Development Lead at May Mobility. Her dream for autonomous driving is simple: it must not stand out to the vehicle occupants. Rather, they should almost forget they are experiencing an AV. “It should be boring because it is driving completely safely—in fact, boring is a really fantastic goal to work towards.”
Creating a “boring” autonomous driving experience means driving in an entirely predictable and human-like manner, enabled by May Mobility’s patented self-driving software algorithm, Multi-Policy Decision Making (MPDM) technology. Instead of an AV adhering to a rigid and rules-based system, which can lead to serious issues when encountering a situation for which it has not been prepared, MPDM is developed to allow for emergent behaviour and enables the car to navigate through almost any situation.
MPDM works by continuously running thousands of simulations on vehicle hardware in real-time to anticipate any possible change or development in the driving situation. Using this information, the vehicle is able to decide for itself what action is the safest. “What makes our software special is that it’s learning even while it’s on the road,” adds Karsten Kutterer, Senior Communications Manager at May Mobility. “When you have a team of engineers, they can only imagine so many rules to put into the code.”
Such intensive simulation may give the impression that May Mobility’s self-driving software is particularly demanding and can only run on the most advanced hardware, but Gryniewicz dismisses this idea. “MPDM is designed to home in on scalability—it’s extremely energy and data-efficient.” She explains that the particular method of decision-making uses less compute power than relying on a large base of existing data. This isn’t to say the system entirely disregards the benefits of foundational knowledge: before a vehicle is brought into operation in a town or municipality, it is equipped with and trained on a map of the local area.
Bringing AVs into the mainstream
“We think AVs powered by our software can make a difference everywhere, not just in major cities and robotaxis,” explains Gryniewicz. “There’s a ton of people out there for whom there is a mobility gap, so having something that’s scalable makes it more useful across all of these applications.” She states that May Mobility’s technology is designed to be platform agnostic, able to be tuned and adapted across a variety of different vehicle types, ranging from passenger cars to buses.
Gryniewicz emphasises that the company’s system enables a degree of affordability that other solutions may not be able to reach. Crucially, May Mobility achieved this without sacrificing other pivotal redundancy features, such as backup power and sensor communication systems and a fallback vehicle safety system.
We want to enact change within the community that makes up our customers and partners
In combination, these factors are intended to make May Mobility’s autonomous driving software, alongside the actual vehicles it produces, the ideal solution for its target audience: transit agencies, municipalities and businesses. “We’re very bullish about these segments. There are transit companies all over the US that need to fill gaps and help people get around. I think there are really good use cases for autonomy in that,” she explains. “These cities and private companies have budgets, and we all know they are facing driver shortages. They need sustainable economics, and you can use AVs to fulfil that need.”
The company’s strategy is to start at a small scale within a given area and then expand as trust grows. This will be achieved not only by demonstrating to locals that the technology is safe to use but also by prioritising accessibility and expanding mobility access in areas that were previously lacking. “We want to enact change within the community that makes up our customers and partners,” says Gryniewicz. Accessibility—such as for those in wheelchairs—is also a key AV design priority.
This approach forms not only May Mobility’s strategy for commercial success but also what it believes is the best way to sway consumer sentiment towards the AVs. Rather than presenting it as a threat to human jobs—a common critique of robotaxi firms—the technology is instead cast as a tool for safely connecting communities and individuals with limited mobility options. “AVs need to serve all people, and we design them for that purpose. That’s something we’re proud of, but it’s also something that’s especially important to our markets,” concludes Gryniewicz.
