The large-scale introduction of self-driving vehicles will have major impacts on our urban centres. There is no debate over whether autonomous vehicles will be transformational. However, what their impact will be on congestion remains to be seen.

One argument is that cars that can drive themselves will ultimately mean more vehicles on our streets, causing increased congestion and more traffic that has to be managed. The opposing argument is that the technology will help decrease the number of vehicles on our roads, enabling cities to redesign spaces that cars no longer need for new purposes.

The experts who expect the arrival of self-driving vehicles to ultimately reduce the number of cars on our streets see it as an opportunity to transform our urban landscapes. Self-driving cars could improve the “efficiency” of our existing roads through the use of real-time communication and platooning, enabling autonomous vehicles to travel closer together, in narrower lanes.

The result would be that existing streets could handle greater volumes of cars, in many cases reducing the need to build new roads.

Redesigning Urban Spaces

The focus in the past has all been on the automobile, but now infrastructure expansion and improvement is becoming more inclusive. The global reality is that reducing the number of cars on our streets isn’t an option yet. Innovations, like the deployment of autonomous vehicles, must form part of the solution.

“We know that we must reduce the number of cars in our cities, but the question is how we do that,” says Timothy Papandreou, from Google X, formerly Director, Office of Innovation at San Francisco Municipal Transportation Agency (SFMTA). He thinks that cities must be willing to establish flexible frameworks that will allow private sector innovators to bring solutions to the table. Autonomous vehicles will be one eventual option, but he insists that other solutions must also be considered and that cities must adequately prepare.

“Providers of new services, like Uber or Zipcar, can all contribute to resolving some of our transportation issues, but cities must stop focusing on building the infrastructure or operating the networks themselves,” he explains. “They must simply provide the framework for new providers to offer their services.”

One of the most significant impacts that self-driving vehicles will have on city planning is the case of parking spaces. “On holidays, we generally see a 15 to 20% drop in traffic, streets and roads are much less congested,” elaborates Timothy Papandreou. “Imagine a similar drop happening with the introduction of automated vehicles.”

Large parking lots could be redeveloped as parks or garden spaces, and eliminating the need for street parking spaces might lead to larger sidewalks, and more public or retail space. A carefully planned deployment of shared and self-driving vehicles can increase the developable area by up to 20%, allowing innovation to redesign space that previously needed to be reserved for parking.

“What can we do with that new space?” asks Rachel Skinner, Development Director for WSP in London. She produced, in collaboration with Farrells, a thought-provoking white paper entitled Making Better Places. “With automated cars, you don’t need to get to the car; the AV comes to you, and that has a profound impact on our parking needs.”

The 6.8 million parking spaces in the City of London covers 8,000 hectares of space that city planners would be able to reallocate. The area, per capita, is fairly consistent across cities like New York, Paris, and Hong Kong.

These former parking spots could have a new life as urban gardens, and help cities to be more resilient. Shared and autonomous vehicles can offer cities the opportunity to transform paved surfaces into green spaces that can naturally absorb excess water, which would have a direct impact on the long-term capacity of municipal drainage systems.

Preparations in the Private Sector

Autonomous cars will also change the landscape for private developers, since many municipalities currently have development regulations that include mandatory parking space requirements: new residential buildings can require 1.5 parking spaces for every two-bedroom dwelling, while retail centers can require 3 parking spaces per thousand square feet. Then there are shopping malls, where developers have traditionally needed to include multi-level parking lots.

“Is it still best practice to invest in giant parking lots that may not be needed in the future?” asks Douglas Webber, Associate Vice-President of Buildings Sustainability for WSP in Canada. “Developers could end up investing in parking space that might end up empty if people are no longer using personal vehicles, or they could misjudge the market and end up with a building that’s short on parking space, which will frustrate consumers and could drive them to their competitors.”

It is an issue he has researched extensively, examining a variety of opportunities, like building parking lots that could be easily transformed for other uses. However, because of the regulated requirements for parking space - heights, widths, ventilation systems, etc. - his conclusion for the moment is that repurposing would be cost deterrent. For the moment, building a structure that could be dismantled and rebuilt elsewhere is what makes the most sense.

Our Advice for Developers

Ahead of this shifting tide, developers can temper their approach through phased development.

In the end, it all comes down to one central question: what kind of cities do we want to bequeath to the future? If we don’t want urban centres that double as giant parking lots, then the time has come to do something about it. Preparing the landscape for autonomous vehicles is one important step forward.

“Just twelve months ago, people were asking what this autonomous issue was, and many felt like they were on the outside looking in,” says Scott Benjamin, Principal ITS Engineer at WSP in Australia. “Now, people are starting to realize that they can have a say in what the outcome of this redevelopment will be.”

More on this subject