How Google Is Reducing Emissions at City Intersections with AI

As road transportation continues to contribute significant greenhouse gas emissions globally, an initiative by Google Research called Green Light is using artificial intelligence (AI) to manage traffic lights and reduce emissions, specifically at city intersections.

Google has identified intersections that are pollution hot spots, with emissions that can be 29 times higher than open roads. Interestingly, about half of these emissions come from vehicles stopping and starting. While some stop-and-go traffic is inevitable, Google believes it can prevent a part of it by improving the timing of traffic lights.

Traditionally, cities had two options for managing traffic lights - installing expensive sensor hardware or manually counting vehicles, both of which provide incomplete information.

Google Research has come up with a third option— using AI and data from Google Maps driving trends to model traffic patterns and recommend new timing configurations. The project has been deployed across 70 intersections in 12 cities globally, including Bangalore, Hyderabad and Kolkata.

How does Project Green Light work?

Green Light builds a simulation of each intersection, including structure, traffic patterns, light scheduling and their interactions. The AI then identifies opportunities to coordinate lights better and reduce stops. For example, it may recommend syncing currently uncoordinated lights to create "green waves" along a stretch of road.

Recommendations are also provided to city engineers through an online dashboard. With existing infrastructure, Google says these changes can be implemented in as little as 5 minutes. The AI continually monitors to measure impact, reporting with updated recommendations as needed.

Early results show a potential 30% reduction in stops and a 10% cut in greenhouse gas emissions. The Green Light project is now live across 70 intersections in 12 cities worldwide, saving fuel and lowering emissions for up to 30 million monthly car rides.

Vineet Kumar Goyal Commissioner of Police, Kolkata, India said, "Green Light has become an essential component of Kolkata Traffic Police. It serves several valuable purposes which contribute to safer, more efficient, and organized traffic flow and has helped us to reduce gridlock at busy intersections. Since November 2022, we have implemented suggestions at 13 intersections. The outcome is excellent as per the feedback from commuters and traffic personnel."

With its ability to analyze thousands of intersections simultaneously using real-time data, the Green Light project has proven successful from Seattle to Bengaluru. However, there's a greater potential in cities needing access to advanced traffic systems.

Google plans to expand the Green Light project to more cities soon. This collaboration between Google and local government bodies is an excellent example of how AI can address climate change and reduce carbon emissions globally, giving way to cleaner and greener city environments.