Google's AI Stoplight Program Calming Traffic in Dozens of Cities

It's been two years since Google first debuted Project Green Light, a novel means of addressing the street-level pollution caused by vehicles idling at stop lights. At its Sustainability ‘23 event on Tuesday, the company discussed some of the early findings from that program and announced another wave of expansions for it.

Green Light uses machine learning systems to comb through Maps data to calculate the amount of traffic congestion present at a given light, as well as the average wait times of vehicles stopped there. That information is then used to train AI models that can autonomously optimize the traffic timing at that intersection, reducing idle times as well as the amount of braking and accelerating vehicles have to do there. It's all part of Google's goal to help its partners collectively reduce their carbon emissions by a gigaton by 2030. (...)

“Today we're happy to share that… we plan to scale to more cities in 2024,” Yael Maguire, Google VP of Geo Sustainability, told reporters during a pre-brief event last week. “Early numbers indicate a potential for us to see a 30 percent reduction in stops. We believe green light is unique because it is more scalable and cost effective for cities than alternative options.” Ideally, cities would deploy the models across grids of neighborhood blocks to create “waves” of propagating green lights.

“Our AI recommendations work with existing infrastructure and traffic systems,” Maguire continued. “City engineers are able to monitor the impact and see results within weeks.” Maguire also noted that the Manchester test reportedly saw improvements to emission levels and air quality rise by as much as 18 percent. The company also touted the efficacy of its Maps routing in reducing emissions, with Maguire pointing out at it had "helped prevent more than 2.4 million metric tons of carbon emissions — the equivalent of taking about 500,000 fuel-based cars off the road for an entire year."

by Andrew Tarantola, Endgadget | Read more:
Image: Google

[ed. Good news! (but maybe 'vertical' integration will be another element to add (see below).]