Instead of timed lights, this company is using AI to keep traffic moving

Automobiles have benefited from remarkable advances in technology over the past 100 years – just compare a Ford Model T with today’s Mustang Mach-E. Yet the system of timed lights that controls traffic flow remains stuck in the past.

Traffic signals haven’t changed much since the first electric traffic light was installed in Cleveland, Ohio, in 1914. The earliest ones employed only red and green lamps; a yellow light was added in 1920 to give motorists time to stop safely.

Mechanical timers were soon added to the signals, relieving traffic officers from the drudgery of changing the lights manually. Thanks to the egg-timer-inspired innovation, New York City was able to reassign most of its 6,000 traffic cops, saving the city more than $12 million in the 1920s.

The technology didn’t advance much for decades after that, even while traffic volumes grew exponentially.

Toronto became the first city in the world to introduce computer-controlled traffic lights in 1964. The floor-to-ceiling Sperry UNIVAC 1107 computer gradually took over all traffic signals in the Toronto area by 1970, which significantly improved vehicular flow without the expense of widening streets and intersections.

Computerized traffic signalization has been a benefit for cities managing heavy traffic volumes, but motorists recognize its limitations. Anyone who has waited at a red light at 1 a.m. with no other cars in sight knows the frustration of dealing with obsolete computer programming.

All of that is poised to change.

NoTraffic is an Israeli start-up that intends to revolutionize how traffic can move more efficiently. The company’s platform utilizes machine vision, cloud and edge computing, vehicle-to-everything (V2X) connectivity – technology that allows your car to “talk” to things around it – and the Internet of Things (IoT), which allows everyday objects to exchange data and information. It was named one of “Time” magazine’s 100 most influential companies of 2022.

“We have IoT in our refrigerators, so why not in our streets where there are life-and-death situations?” said Tal Kreisler, co-founder and CEO of NoTraffic. “The urban environment has become more complex and dynamic. Congestion is the by-product of an unmanaged network. We can do better.”

Already adopted by several cities in California, Arizona and Texas, NoTraffic uses sensors and artificial intelligence (AI) to monitor traffic and adjust traffic lights in real time to bring about more efficient movement of people and vehicles.

In one case study, conducted in Redlands, Calif., the platform was shown to reduce traffic congestion in an intersection by up to 50 per cent, cutting 33 tons in carbon dioxide emissions and saving 2,700 hours of delays during a one-year period.



NoTraffic’s platform relies on IoT sensors installed at eye level at each intersection using machine vision and radar to classify all road users, including bicycles, scooters, pedestrians, cars, heavy trucks, transit and emergency vehicles. “Connected” vehicles that already use autonomous technology can integrate directly with the platform.

“IoT devices are processing all the real-time data using AI algorithms with the end result of increasing safety and efficiency of movement,” said Kreisler. “It’s like a huge chess game.”

Founded in 2017, NoTraffic recently announced its expansion into Canada, teaming up with Rogers Communications and its wireless 5G network.

Its first Canadian pilot project is in Vancouver. The technology is being installed at multiple intersections and roundabouts and the company is working with the city, Rogers and the University of British Columbia to improve road coordination, safety and environmental sustainability. The partnership will contribute to Vancouver’s goal of cutting the city’s carbon footprint in half by 2030.

“With its history of embracing innovation, Canada is a priority market for NoTraffic and the ideal location for the next step of our expansion,” Kreisler said.

While optical devices are used, Kreisler said that privacy is maintained, since the technology does not collect identifying information or perform facial recognition or licence plate reading. Installed in existing traffic light cabinets, the system calculates variables and adjusts the signalling to match the traffic in real time. This data is also passed along to other nearby intersections to help it understand the flow of traffic and adjust according to needs.

“The IoT technology is making real-time decisions without human intervention,” said Kreisler. He said traffic planners can still, for example, set priorities for the system, such as giving pedestrians or transit vehicles higher priority.

“We are transforming low-tech infrastructure to support modern demands,” said Kreisler. NoTraffic can retrofit an intersection with its “plug and play” technology in a matter of hours.

Since the Vancouver pilot got underway, Kreisler said NoTraffic and Rogers have received strong interest from several Canadian cities, with more projects launching soon – just not in Canada’s largest city.

Toronto’s Transportation Services division recently made the transition to the Sydney Coordinated Adaptive Traffic System (SCATS), an AI-based solution developed by Australia’s New South Wales government. It was installed as part of the Smart Signals program identified in Toronto’s MoveTO action plan.

“SCATS is a prominent AI traffic signal solution used by many jurisdictions in North America and the world,” Hakeem Muhammad, a transportation spokesperson for the city, said in an email.

Rather than classify and assess all road users, SCATS gives priority to transit and emergency vehicles to make Toronto’s streets safer. According to the SCATS website, the system relies on modified timing plans and traditional pedestrian pushbuttons and road-embedded sensors and not the optical and radar systems used by NoTraffic.

“Following a thorough investigation of a number of AI traffic signal solutions, SCATS was a front-runner as it’s able to manage the city’s Vision Zero safety requirements better than other solutions,” said Muhammad.