Google has recently launched the Project Green Light, an initiative using AI to model traffic patterns and adapt traffic lights to optimise flow across cities and reduce emissions at intersections.
The initiative is currently active in 70 intersections in 12 cities around the world, but will be rolled out to more locations next year. From the testing phase, conducted over the course of 2022 and 2023 in Abu Dhabi, Bali, Bangalore, Budapest, Haifa, Hamburg, Hyderabad, Jakarta, Kolkata, Manchester, Rio de Janeiro and Seattle, Google concluded that, with the help of Green Light, emissions were reduced by up to 10%.
How it works
Studies have shown that pollution at intersections in cities can be 29 times higher than on open roads. Besides the wait time when some vehicles keep their engines running until the light turns green, stopping and starting contribute to half of the emissions at intersections, as it requires more fuel than simply driving.
Green Light, a Google Research initiative, uses AI and Google Maps driving trends to model traffic patterns and make recommendations for optimising the existing traffic light plans. City engineers can implement these changes “in as little as five minutes”, using existing infrastructure. By optimising not just one intersection, but coordinating across several adjacent intersections to create waves of green lights, cities can improve traffic flow and further reduce stop-and-go emissions.
Green Light identified opportunities where we previously had no visibility and directed engineers to where there were potential benefits in changing signal timings. This provided valuable insights for our city with 2,400 traffic signals.
David Atkin, Transport for Greater Manchester
While many cities already have synchronized lights on main roads, Green Light can provide more reliable and more widespread data. ” Our city partners tell us that prior to Green Light, they would try to optimise traffic lights using expensive sensors or time-consuming manual vehicle counts — and these solutions do not provide complete information on key parameters they need”, said Yossi Matias, Google Vice President for Engineering and Research.
Green Light is capable of analysing thousands of intersections simultaneously, improving the flow through multiple intersections in the city. It builds an AI-based model of each intersection, including its structure, traffic patterns (such as patterns of starting and stopping), light scheduling and how traffic and the light schedules interact, as well as a model of the interaction between traffic lights.
“Based on this model, we develop AI-based optimisations and then provide recommendations to city engineers via the Green Light interface. As an example, we might identify an opportunity to coordinate between intersections that are not yet synced and provide a recommendation around the timing of the traffic lights so that traffic flows more effectively along a stretch of road”, explained Matias.
Measuring traffic flow before and after cities implemented the Green Light recommendations, Google determined that the number of times vehicles stopped at an intersection was reduced by up to 30%, which can potentially reduce emissions by up to 10%.
