Driving Vision Zero with Autonomous Vehicles
By Ben Rosenblatt & Michael Flynn
Greenwich London self-driving test vehicle (Photo credit: David Jones)
There has been no shortage of discussion in the urban planning and technology communities over the last few years – and, increasingly, in the mainstream media – on the coming sea change in daily life that will result from autonomous vehicles (AVs). With proclamations that the AV revolution will be in full swing anywhere from 5 to 50 years out – or maybe never at all? – there is both more buzz and more uncertainty today on this topic than any in the transportation sphere in recent memory. If the bullish futurist projections hold, our society is about to undergo a change in its way-of-life that could match the impacts of astounding recent transformations like the proliferation of internet access and growth of advanced smart phone technology.
For all the opinionating and hype, it still isn’t possible to see into the future – how many past predictions about how technologies would imminently transform our cities were way off, or way ahead of their time (flying cars, anyone)? Based on the extraordinary corporate and institutional investments being made by major automakers, large tech companies, and start-ups looking to capitalize on the AV opportunity, it doesn’t take a leap of faith to predict that AVs are coming and disruption in the automotive sector is inevitable. How this “revolution” plays out, however, is still very much up in the air.
We do not seek to predict here exactly how AVs will roll out – the timing (of partial automation, full automation, and mixed traffic), the locations (urban vs. suburban vs. rural), the ownership model (personal ownership vs. on-demand services; single-occupancy vs. shared rides), the fleet types (freight or for-hire), or the technology (internal combustion vs. electric). We do, however, know that with or without AVs, what we want from our transportation systems in 2018 is relatively clear: cities desire safer streets and reduced crashes; cleaner air and lower greenhouse gas emissions; increased access to employment, educational, and recreational opportunities for lower-income communities (i.e. transportation equity); and, overall, a system that supports a high quality-of-life.
AVs can support these goals and may yet contribute toward achieving them, even if the uncertainties abound. In this article, we will focus on the many moving parts at play as we turn the calendar to 2018. Next, we will summarize how these elements have brought on distinctly competing narratives around what the future holds. Finally, we stress that regardless of the direction around AVs, cities can gain by keeping one topic front and center: safety benefits need to be prioritized in official messaging, with action, and through corresponding funding decisions. With their potential to significantly reduce the thousands of fatalities we see on roadways every year, AVs must help finally eliminate the scourge of automobile-related injuries and deaths that have unfortunately become all too endemic to our lives over the last century. And cities can leverage their Vision Zero policies as a primary mechanism to do so.
Moving Parts: Factors that will Impact AV Growth
Much of the uncertainty around what AVs will mean for society, the built environment, and our daily lives has little to do with the progression of the autonomous technology itself. While still important, and included as the first item in the list below, there are other policy decisions that planners and engineers will make that will drive outcomes as AVs proliferate. Meanwhile, these choices will come within the realm of a constantly evolving regulatory framework that features patchwork legislation at federal, state and local levels, heavy influence of special interest groups, and uncertain knowledge about what will work and what won’t.
Technology type: As large companies race to bring to market a fully autonomous vehicle requiring no human control or input, others are betting that a semi-autonomous version might sell. But could a world of semi-autonomy be just as risky – or even more so – than the previous 100+ years of our human-controlled driving experience? With Tesla’s rollback of certain semi-autonomous features, it is increasingly clear that expecting a human to “relax” in the driver’s seat but still be ready to take control of a multi-ton vehicle at a second’s notice might be just too much to ask. With distracted driving already a well-documented problem, shouldn’t we make sure that full autonomy is ready to roll and dispense with the supposed short-term gains from these incremental “semi-autonomous” phases?
Market segments: Where will we see the quickest uptake in AVs in terms of market share? Many believe that the freight industry is ripe for autonomy. Long-distance trucking checks all the right boxes: the distances are long, mostly on limited-access highways, and an autonomous business model would allow for companies to achieve far higher utilization of their vehicle fleets than today, where driver recovery time for sleep and meals interrupts long-distance trips. Millions of jobs are at stake, however, and in places with limited other freight options via rail or water, goods transport by commercial truck represents a very high employment share. So how about passenger travel? It depends on how one defines it: perhaps AVs will grow fastest on traditional fixed transit routes, as transit agencies seek to eliminate vehicle-operator positions to save heavily on operating costs (labor unions may have some opinions on that prospect). On the other end of the spectrum, the vast share of trips made today by single occupancy passenger vehicles represents a tremendous profit opportunity for companies selling autonomous vehicles (or services, see below). Finally, something in between could represent the sweet spot for AV producers to focus on in the battle for market share. “Microtransit” or pooled trips set up through app-based on-demand ride hailing systems like Chariot, Lyft, Uber, and Via could emerge as the best case for fully autonomous vehicles, acting like on-demand, dynamically routed transit but with a more personalized and “private” experience than today’s fixed-route bus lines. Can a service model like that be both profitable and equitable, however, or would it exacerbate a two-tiered transportation system based on wealth?
Geographies: The idea of bringing a fully autonomous vehicle into a chaotic urban environment with pedestrians, cyclists, curbside activity, potholes, inconsistent lane markings, non-compliance with traffic signals, and bad weather is often cited as a reason why AVs simply won’t work out (anytime soon) in places like Manhattan, New York. Perhaps AVs are a better fit for more controlled settings offered by suburbs with their wide arterials and generally vehicle-oriented transportation networks, coupled with a lack of obstacles like people daring to (legally) walk or bike on them. But this differentiation may lack the subtlety necessary to define smart policy. Sections of New York City, for instance, could be the perfect testing ground for very low-speed (max 15 mph) automated taxis or pooled rides, especially if the area is studied before and after AV pilot implementation to document the impact on safety outcomes like crashes, injuries, and fatalities.
Rotterdam ParkShuttle autonomous minibuses (Photo credit: Paradasos)
Shared or owned: This debate is, in our eyes, the key to successful AV proliferation and by extension mid-21st century transportation policy. As we seek to transform our cities and towns to correct the mistakes of auto-oriented planning and design stemming from the last automobile revolution, how do AVs fit in? Can we use them for good, and still focus on creating walkable, healthy, diverse communities that remain at human-scale? We think so, if the thread of the sharing economy and declining vehicle ownership holds true. Dispensing with the desire to own an asset that sits unused over 95% of the time can be not only a smart economic decision, but translates to a far more effective utilization of precious urban land and street space (opening up new open space and development opportunities in urban cores). If AVs are contained to the realm of fleet-operated shared resources, we dramatically reduce the most common fear of an “owned” AV future – zero-occupancy vehicles cruising and clogging our streets after they have dropped off their owner at a destination. If traditional ownership goes away as AV market share grows, the companies running the services – whether they be today’s traditional automakers, large tech companies, or smaller startups – will all have the incentive to maximize vehicle utilization, and in the process significantly reduce traffic congestion.
Competing Narratives: Utopia to Nightmare
Because of the many uncertainties highlighted above, a wide variety of narratives are still in play, even as we turn to 2018 and AV manufacturers insist that fully autonomous models are ever-so-close to market. The opinions summarized below range from utopian to nightmarish, on a variety of issues.
Utopia vs Nightmare AV Scenarios
The Utopia and Nightmare narratives will be shaped by how places plan and regulate proactively, especially cities with their large populations, highest demands for freight deliveries, and most complex street environments. That said, among the most critical determinants of which pathway we go down in the United States is related to federal action. As Congress attempts to push through all-encompassing legislation that governs how AVs will roll out nationwide in the coming years, it has the potential to essentially preempt almost all local decision-making on AVs. Meanwhile, states themselves are also starting to preempt local municipalities, as in Texas, New York, Colorado, and many others. With limited control over what “narrative” plays out, we suggest that cities focus instead on the levers they do have under their control.
These levers are often based around incentivizing sustainable travel, which has been and remains very important even today, before the AV wave really takes hold. Issues like street design and accompanying right-of-way allocation among various travel modes, road pricing and tolling to influence travel demand, parking pricing and policy to encourage turnover of spaces, and others, are included here. But one policy lever rises to the top, and is already used to justify major investments in transportation improvements: Vision Zero.
Decreased parking demands can allow cities to reallocate roadway space from on-street parking to bicycle lanes, wider sidewalks, loading zones, transit stops, and parklets.
Safety Benefits: Vision Zero Cities and the AV Revolution
If cities are serious about making sure AVs work for and not against them, they can and should use their Vision Zero commitments as springboards towards immediate action. Regardless of the various uncertainties and which AV narrative(s) we end up seeing, there is no doubt that leveraging the strong computing power and technological innovation behind AV investment can help us achieve broader safety goals for society at large. This starts with our commitments to ultimately reduce fatalities caused by vehicles down to zero.
Not only should safety benefits accrue to society through the AV revolution; they must. In 2018, we still face a worldwide public health crisis around traffic safety. It is a crisis that began almost as soon as motor vehicles became the dominant form of personal transport 100 years ago, and has continued through today, worsening as automobiles have proliferated in growing consumer markets like China and India, even as improved medical treatment and increased use of seatbelts and autonomous braking features make individual crashes less deadly to vehicle passengers.
Every year, almost 1.3 million people worldwide die in road crashes, or over 3,200 per day on average. This is the equivalent of about eight Boeing 747s crashing and killing everyone onboard, every single day, for 365 days, each and every day of the year. Similarly, somewhere between 20 and 50 million people are injured or disabled due to road crashes annually – the midpoint of this estimate of 35 million would mean that an equivalent number of people to the entire population of Canada is annually impacted by traffic violence.
To eliminate even 10% of these needless deaths and injuries would be welcome, but if AVs can achieve serious crash reduction of up to 90%, there is a significant human health and societal welfare benefit staring right at us. More than any roadway redesign program, AV technology provides the greatest potential to actually achieve the “Zero” in a Vision Zero goal. So what can cities with Vision Zero commitments – or those looking to officially make such commitments – do?
Cities can and should use their Vision Zero programs and associated funding to embark on some or all of the following (as Boston, for one, has begun to do):
Employ pilot projects for fully autonomous vehicle technologies at gradually larger geographic scales, and measure the before and after safety and network efficiency outcomes. Restrict or ban human-controlled driving from certain districts as necessary to control the environment further and illustrate potential benefits of a completely driverless city. Urban islands, pedestrian-dominated districts (such as Lower Manhattan), and campuses (academic, corporate, or otherwise) are logical places to start.
Apply camera and sensor technology to automate enforcement, collect collision and “near miss” information, and perform before and after studies of any pilot projects of AV technology. If legislative approval is required to do so, discerningly tie the automated data collection program in with pressing and much needed Vision Zero safety goals.
Use data to make the case for continued AV rollout, in the context of Vision Zero. In the same way that a street redesign project can be shown to improve safety outcomes, do the same detailed before and after measurement for early-stage autonomous vehicle pilot projects. Tell the story to clearly illustrate any safety benefits, and demonstrate how AVs can help reach Vision Zero.
Reference Vision Zero consistently on all AV policy development and strategic decisions. This will aid in public understanding of an abstract issue, in gaining political capital to test things out, and potentially in the ability to capture federal or other grant funding for AV projects.
Adopt a short- and long-term AV strategy as part of an updated or new Vision Zero Action Plan. Each city is unique and will face specific implementation challenges on AVs that require a proactive approach to grow local AV market share. Include specific goals, strategies, and benchmarks that lay out rationale, policy levers, and measurement tools that will be applied to execute an AV strategy in the context of Vision Zero.
Moving Forward in 2018: What Cities Will Do
It isn’t possible to know exactly how autonomous vehicles will be adopted – where, when, by whom, and in which form(s). On top of that, many cities find their ability to influence the outcomes limited by Federal and State pre-empting legislation. However, many of the tools that cities have traditionally used to encourage more sustainable forms of transportation are just as applicable to AVs, and perhaps even more so given how much is at stake and the likely short time to act. These tools include designing and pricing streets to prioritize more efficient and sustainable modes of transport, using parking policy to discourage personal vehicle ownership and single-occupancy trips, and leveraging technology to improve passenger and freight efficiency.
However, Vision Zero traffic safety initiatives may be the strongest “lever” of all that cities have to steer AV adoption and usage towards safer, cleaner, and more equitable transportation systems. Connecting the potentially massive safety benefits of AVs with Vision Zero programs that are well established at municipal, state, and federal levels will increase cities’ abilities to proactively plan for and fund early-stage pilot projects on a technology that stands ready to transform the way we move, permanently.
