This is the second of a three-part series on how bicyclists and autonomous vehicles can cohabitate. Part 1 is here.
The suggestions in this series were originally developed in January at a planning session sponsored by the Eno Center for Transportation that the author designed and facilitated with fellow Mobility Lab contributor Lisa Nisenson. Eno will soon publish a full set of policy, planning, law, and funding strategies for communities looking to a smooth deployment of AVs.
I may have misspoken last week when I wrote that “integrating AVs into environments where we want to encourage bicyclists (and pedestrians) is one of the toughest issues communities face as AV deployment moves forward.”
That makes is sound like we don’t have any good ideas about how to protect bicyclists.
The first part of this series sought to answer why it’s so difficult for autonomous vehicles to see bikes. The problem is that AV sensing technology is not a magic wand akin to x-ray vision. AVs are unable to see around corners or understand what bicyclists are likely to do.
“Bicycles are probably the most difficult detection problem that autonomous vehicle systems face,” said University of California Berkeley research engineer Steven Shladover in IEEE Spectrum.
Now Mobility Mama turns her attention to how we can take what we already know about making roads safe for bicyclists pedaling alongside conventional vehicles and develop it a smidgen further to keep them safe as automated – even driverless – vehicles hit the streets.
There are five main things to keep in mind:
- Communities can make bike and pedestrian safety a governing principle via a resolution on AV usage on public streets. Lisa Nisenson and Andy Boneau have drafted a guide for Writing Effective Resolutions for New Mobility that highlights protecting bicylists (and pedestrians.) While these resolutions are intended as statements of principle, the National Association of City Transportation Officials has successfully defended the role and responsibility of localities in managing the operation of AVs on local streets, including making sure that all modes of transportation, including biking and walking, are safe.
- Cities and counties could separate AVs and bicycle and pedestrian traffic through the use of existing dedicated lanes, creation of additional dedicated lanes, and the management of curb space. The Transportation Research Board is currently studying the use of dedicated lanes for autonomous and connected vehicles as a way to increase roadway capacity. Use of dedicated lanes and curbside management allows for more controlled testing of AVs on local streets while keeping bicycles and pedestrians clearly separated from them.
- Creating temporary or flexible dedicated lanes with movable barriers or time restrictions on lane use. On local streets, it may be more practical to employ tactical urbanism by creating temporary or flexible lanes through movable barriers. Or another approach would be to restrict lane use during peak travel periods. Quick Builds for Better Streets by People for Bikes describes how this might be accomplished.
- Cities, in conjunction with engineering firms or university research programs, could redesign streets and intersections to separate bike, pedestrian, and vehicle traffic, especially at crosswalks. We already know that physically separated bike lanes are the best way to reduce bike fatalities and injuries. This would take what we know and should be doing now, and apply it going forward in anticipation of an increasing number of AVs on the streets.
- As cities host and pilot AVs, they have a chance to integrate bike and pedestrian safety in those tests. This could include testing sensors on bikes that communicate with the infrastructure, as well as implementing road, trail, and intersection designs that separate bicyclists from AV traffic. One way to do this would be to partner with researchers to assess these measures and to better understand how bikes and AVs interact. An example of such a collaboration is The AV Proving Ground Partnership of the Texas Transportation Institute, which integrates bike and pedestrian safety testing into all of its tests, which are located in a variety of environments.
Some early experience with bikes and AVs yielded an interesting and perhaps unexpected observation: people on bikes may be more comfortable sharing the road with an AV than with human-driven vehicles. Why? Because driverless vehicles are more predictable than those with drivers.
That’s the flip side of how AVs still cannot predict what a biker will do. As reported in CleanTechnica, Carlos Ghosn, CEO of Renault and former CEO of Nissan once described cyclists as “one of the biggest problems for driverless cars.” They confuse the vehicles, he said, because at times they behave like pedestrians, at other times like drivers, and “they don’t respect any rules.”
That said, AV-related tech may, in a roundabout way, help improve bike and pedestrian safety for conventional vehicles.
Part 3 will look at how AV-related technology can be used to keep pedestrians and bikers safer well before we see lots of autonomous vehicles on the road.
Photo of LA Mayor Eric Garcetti test driving a semi-autonomous vehicle courtesy of Eric Garcetti/Flickr.