Artificial Intelligence and the City
Closed-circuit television (CCTV), also known as video surveillance, is a part of our public lives, for better or worse. There are an estimated 1 billion CCTV cameras operating around the world in 2021. And while the most common application is to provide footage to investigators after criminal acts, the technology has also started to inform how spaces are used and how our cities can be developed.
“Data analytics can tell you why people or vehicles are behaving the way they are. Who is there? Where do they go? How many of them are there? What’s their behavior?” said Karen Burns, co-founder of AI video analytics firm Fyma, located in Tallinn, Estonia. “The AI is able to understand what happens in the picture.”
Defining the Way Spaces are Used
Applying AI to analyze the footage from these cameras opens up a realm of new possibilities in understanding the way people interact with their environment and how this data can be used to inform projects from urban development to advertising.
“Public space design is in many places still stuck in probably the 1950s,” said Burns. “You could use all of these cameras, turn them into sensors and understand how people are using a particular space and at what times. Is something tripping them up in in the same location? You can take that data and turn it into a benefit for the people using that space.”
Fyma has been working with retail centers to understand traffic footfall. “They're using it to do all sorts of social analytics to understand how long people are spending in front of a billboard,” she said. “We're testing this out with various pilots in commercial real estate and airports.”
“For example, in airports, they want to understand the ‘dwell time’ in areas inside the terminal and outside. How long are people spending there? Where is it busiest? What is the traffic density? Because then they can give value-added services for those passengers.”
A shopping center in Tallinn, Estonia, has been using the technology to identify bottlenecks. In the United Arab Emirates, Fyma was chosen for a government project with the Dubai Road and Transport Authority to help understand pedestrian and driver behavior, so potential safety issues could be identified.
The technology can also identify if the weather is impacting the way people or drivers behave. Since COVID-19 began, AI has been used to examine areas where people are finding it difficult to social distance.
Parking operators have also found the technology useful. “We can direct a car not just to a car park, but to an available space, rather than drive around for half an hour,” said Burns. “It enhances the customer experience and reduces CO2 emissions.”
However, unlike surveillance cameras, Fyma claims to protect individual privacy by turning the video quality down so there is no identifying information.
Burns sees urban space changing through data analytics. “We can turn entire city centers into these really flexible work and travel spaces that cater to people's needs depending on the time of day or year,” she said.
Streets could be converted to a pedestrian area on weekends when there is a high density of people, and traffic could be diverted by understanding the data in a neighborhood.
“AI should really be like a little helper in your pocket that should enhance whatever you do and help your decision-making to be better and have more quality,” she said.
This will also change job roles in project management. “A lot of people will end up having to understand an element of machine learning AI as part of their daily job,” she said.
Project managers will need to look at having people on their team who can manage the data and build algorithm systems.
“We don't want to replace people with AI,” she said. “AI will still need people to help it learn the right kinds of things to avoid biases that we already have everywhere in urban planning.”
Advertisement
Advertisement