By Chris Skeldon 
Google has shared an update on its AI-driven mission to reduce wait times, fuel use and polluting emissions at intersections — and early indications suggest that stops could be reduced by about 30%, with emissions falling by about 10%.
Read also: The unusually fast and bright radio burst from deep space cannot be explained by current theories
Multiply that by all the city intersections in the world and we have a potentially huge impact. Research shows that pollution is about 29 times higher at these intersections than on the open road.
All this optimization is based on a large amount of data collected from Google Maps. The app is able to monitor vehicle paths and speeds – which is why it can warn you about traffic ahead – and create AI models of how vehicles and lights interact with each other.
Currently, urban planners rely on expensive sensors or even manual vehicle meters to know what is happening at intersections. Google Maps and AI technology means that more data can be processed much faster, and what’s more, no updates to existing infrastructure are required.
“Green Light is able to analyze thousands of intersections simultaneously, optimizing flow across multiple intersections in a city,” wrote Yossi Matias, vice president of engineering and research at Google.
“Our AI-based recommendations work with existing infrastructure and traffic systems, and city engineers can monitor the impact and see results within weeks.”
Some of the ideas that Google’s AI could offer include how to change traffic light times or how traffic lights at different intersections could be linked together in order to keep vehicles moving more smoothly during the day and night.
The Green Light Project, as the scheme is called, is now operating in 12 cities and four continents around the world: Seattle, Rio de Janeiro, Manchester, Hamburg, Budapest, Haifa, Abu Dhabi, Hyderabad, Bangalore, Kolkata, Jakarta and Bali. Support for more cities is in the works and Google is encouraging new signups.
Google engineers aren’t the only ones looking to make traffic lights more efficient to reduce greenhouse gas production. Earlier this year we saw a study that showed how self-driving cars could work in partnership with traffic signals and human drivers to improve the efficiency of intersections.
Other researchers have looked into clever ways to prioritize buses and bicycles, making greener transportation options more attractive. And as these charts and studies show, even small improvements can make big differences when you consider how many people take road trips every day.
“Green Light has identified opportunities where we previously had no clear view and pointed engineers to where there are potential benefits from changing signal times,” says David Atkin, Director of Analytics and Reporting at Transport for Greater Manchester in the UK.
“This has provided valuable information for our city of 2,400 traffic lights.”
Translated by Matthews Lineker from Science Alert