Tech

Computer vision software has the potential to reinvent the way cities move

Harris LummisContributor
Harris Lummis is the co-founder and chief technical officer at Automotus.

In October 2019, The New York Times reported that 1.5 million packages were delivered in New York City every single day. Though convenient for customers and profitable for the Amazons of the world, getting so many boxes from warehouse to customer generates considerable negative externalities for cities.

As the Times put it, “The push for convenience is having a stark impact on gridlock, roadway safety, and pollution in New York City and urban areas around the world.”

Since that article was published, the global pandemic has taken e-commerce to new heights, and experts don’t expect this upward trend to slow down anytime soon. Without strategic intervention, we will find our cities facing increasingly severe traffic problems, safety issues and polluting emissions.

Without strategic intervention, we will find our cities facing increasingly severe traffic problems, safety issues and polluting emissions.

The same frustrations have plagued urban roadways for decades. However, technology is finally catching up, providing new means of addressing the challenges of crowding, pollution and parking enforcement on dense city streets.

As is almost always the case, an effective solution begins by first understanding the detailed circumstances giving rise to the problem. In this case, a simple means of assessing the problem is to observe curbside parking and street traffic using streetlight cameras.

Deploying cameras to monitor public spaces may immediately incite the ire of die-hard privacy advocates (I consider myself among them), which is why companies like mine have taken a privacy-by-design approach to product development. Our technology processes video in real time and addresses further concerns about potential misuse for surveillance purposes by blurring faces and license plates beyond recognition prior to making any kind of image data available either internally or to public officials.

The point of these cameras is not to surveil but rather to leverage concrete data from real-world city streets to generate crucial insights and power automations at the curb. Automotus’ computer vision software is already using this model to help cities manage the aforementioned flood of commercial vehicles on their streets.

This technology can also be used to optimize and incentivize parking turnover. According to one study, drivers in New York City spend an average of 107 hours per year searching for parking spots, at a cost of $2,243 per driver in wasted time, fuel and emissions, which represents $4.3 billion in total costs to the city. Similar wasteful dynamics are unfolding across America and the world. By collecting comprehensive data around the demand for curbside space, cities can design parking policies that ensure proper alignment between the supply of curb space and the way vehicles are actually using it.

In one pilot we ran on the campus of Loyola Marymount University, traffic caused by drivers searching for parking dropped by more than 20% after our data was used to adjust parking policy. Using data to optimize parking results in more efficient turnover, less time spent circling for a spot and reduced traffic delays. Real-time parking availability data can also be used to direct drivers to open parking spots via an application or API.

By arming city planners with accurate, up-to-date information on all forms of curbside activity, we empower them to fully understand the temporal and spatial patterns that rule their curbs. This gives planners the information they need to make informed decisions about curbside policy tailored to their city’s peculiarities.

Suddenly, questions such as “How many ride-hailing drop-offs occur here?” and “Whose delivery trucks are double parking on Tuesday morning?” become trivial to answer. Gone are the days of using vague heuristics to guide policy; this new wealth of information makes possible precise and impactful decisions on the locations of passenger parking, dedicated delivery zones and ride-hailing areas, as well as optimal rates to charge for parking, appropriate penalties for violations and much more.

This tech is also a win for delivery companies. When delivery fleets have data about real-time and predicted parking availability, this can improve route efficiency, saving them money. Instead of paying for curb usage via fines, delivery companies can instead receive an invoice for their time spent at the curb (a tax-deductible expense, I might add).

A study done in Columbus, Ohio, found that designated loading zones decreased double parking violations by 50% and reduced commercial vehicle time at the curb by 28%. Radically increasing the efficiency of delivery translates into savings for companies like FedEx and Amazon, which can then afford to pay fair rates for their curb access and pass on those savings to consumers.

Several interrelated trends make the current moment an especially opportune time to apply new technology to our streets and curbs. Pre-pandemic, many cities already faced declining revenue from parking as citizens shifted toward using ride-shares. Now, thousands of American municipalities are expecting major budget shortfalls in the wake of COVID-19. At the same time, a report from the World Economic Forum predicts that the number of commercial delivery vehicles will increase by 36% in inner cities by the year 2030. Our research suggests that more than 50% of parking violations are unenforced and committed by commercial vehicles.

It’s no coincidence that Columbus was the winner of the 2016 federal Smart City Challenge. When former President Barack Obama pledged over $160 million as part of his “Smart Cities” initiative in 2015, reducing congestion and pollution were among the program’s major goals. Better management of parking and curb space are crucial tools for achieving these aims. Though former President Donald Trump campaigned on a massive infrastructure plan, his delivery on promises in this area were mixed at best. Despite the lack of federal support, there are currently promising initiatives underway in cities such as Santa Monica, which is piloting a zero-emissions delivery zone in the heart of its downtown.

President Joe Biden has outlined a plan to build the infrastructure America needs both to combat climate change and modernize urban transportation. This plan includes a provision for 500,000 public charging stations for electric vehicles; changes to our cities that allow drivers, pedestrians, cyclists and others to safely share the road; and investment in critical clean energy solutions.

Curb management technology is one of a suite of options on the market that federal and local governments can leverage to reduce pollution and improve quality of life in cities. If the incoming administration is willing to champion this novel approach toward solving the problems of urban mobility, America’s infrastructure will not just be modernized but made ready for the future.

I, for one, hope this renewal is realized; our nation’s health, safety and shared prosperity depend on it.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Most Popular

To Top