Rothenberg says Google has prioritized supporting large cities that employ traffic engineers and can remotely control traffic lights, while expanding globally to prove the technology works well in a variety of conditions , suggesting that it could, if widely adopted, have a significant impact on the world. emissions.
Using Maps data, Google can infer timings and signal coordination at thousands of intersections per city. An AI model developed by the company’s scientists can then analyze traffic patterns over recent weeks and determine which lights might be worth adjusting, primarily in urban areas. It then suggests parameter changes to reduce stop-and-go traffic. The system’s filters attempt to block certain careless suggestions, such as those that might be unfriendly to pedestrians.
Some of Google’s recommendations are as simple as adding two extra seconds at specific times between the start of a green light and the time the next one on the road turns green, allowing more vehicles to pass through both intersections without stopping. More complex suggestions may involve two steps, adjusting both the duration of a particular light and the offset between that light and an adjacent one.
City engineers log into an online Google dashboard to view recommendations, which they can copy into their lighting control programs and apply in minutes remotely, or for non-networked lights , by stopping in person at the control box of an intersection. In both cases, the fact that Google calculates all of this using its own data saves cities from having to collect their own – whether automatically via sensors or manually through laborious counts – and also from having to calculate or monitor their own adjustments.
In some cities, the parameters of an intersection may remain unchanged for years. Rothenberg says the project has in some cases brought attention to intersections in areas typically overlooked by city leaders. Google’s system allows for changes every few weeks based on changing traffic patterns, but for now it doesn’t have the ability to adjust in real time, something many cities don’t have at all. way not the necessary infrastructure. Rothenberg says Google collaborated with the traffic engineering faculty at Israel’s Technion University and UC Berkeley on Green Light, whose users also include Haifa, Budapest, Abu Dhabi and Bali.
To validate how Google’s suggestions work, cities can use traffic counting from video footage or other sensors. Applying computer vision algorithms to city video feeds could potentially help Google and users understand other effects that are difficult to detect in conventional traffic data. For example, when Google engineers observed the green light going into effect in Budapest in person, they noticed fewer people passing the red light because drivers no longer had to wait for multiple red-to-green light cycles to cross the intersection. .
Green Light is ahead of some competing options. Mark Burfeind, a spokesman for transportation analytics provider Inrix, says the company’s dataset covers 250,000 of the estimated 300,000 signals in the United States and helps about 25 government agencies study parameter changes. synchronization. But he doesn’t actively suggest adjustments, leaving traffic engineers to calculate their own. Inrix’s estimates underscore the far-reaching climate consequences of small changes: Every second of waiting for the average signal in King County, Washington, where Seattle is located, burns 19 barrels of oil per year.
Google has a “significant” team working on Green Light, Rothenberg says. Future plans include exploring how to proactively optimize lights for pedestrian needs and whether to notify Maps users that they are crossing an intersection set to the green light. When asked if Google will eventually charge for the service, she said there are no plans to do so, but that the project is still in its early stages. His journey has yet to pass a red light.
Updated 10/10/2023, 5:15 p.m. EDT: This article has been updated with comments from Transport for Greater Manchester.
