It’s been two years since Google first launched Project Green Light as a new way to address street-level pollution caused by cars stopped at traffic lights. At the Sustainability ’23 event on Tuesday, some of this program’s early findings were discussed and a new wave of expansion for the program was announced.
Green Light uses machine learning systems to scan Map data to calculate the amount of traffic congestion present at a given light and the average wait times for vehicles stopped there. This information is then used to train artificial intelligence models that can autonomously optimize traffic timing at that intersection, reducing unnecessary waiting times and the amount of braking and acceleration vehicles must do at those points. It’s all part of Google’s goal to help its partners reduce carbon emissions by one gigaton by 2030.
When the program was first announced in 2021, it was pilot tested at only four intersections in Israel in partnership with the Israel National Highways Company, and Google reported that it observed “a 10 to 20 percent reduction in fuel and intersection delay time” during these tests. The pilot program subsequently grew to a dozen partner cities around the world, including Rio de Janeiro in Brazil, Manchester in the UK and Jakarta in Indonesia.
“We’re happy to share this today… We plan to scale to more cities in 2024,” Google’s VP of Geo Sustainability, Yael Maguire, told reporters. “The early numbers show we have the potential to achieve a 30 percent reduction in stops.” shows that it is. We believe green light is unique because it is more scalable and cost-effective for cities than alternative options. “Ideally, cities would place models along grids of neighborhood blocks to create radiating ‘waves’ of green light.”
“Our AI recommendations work with existing infrastructure and traffic systems,” Maguire continued, adding: “City engineers can monitor the impact and see results within a few weeks.”
Maguire also stated that improvements in emission levels were seen in the Manchester test and air quality improved by up to 18 percent. The company also touted the effectiveness of Maps routing in reducing emissions, with Maguire noting that it helped avoid more than 2.4 million metric tons of carbon emissions, the equivalent of taking nearly 500,000 fuel-powered cars off the roads for a year.