Google has provided an update on its Green Light project, which was launched two years ago to combat street-level pollution stemming from vehicles waiting at traffic signals. The project employs machine learning systems to assess map data, determining traffic congestion levels and average wait times at specific intersections. Using this information, AI models are trained to optimize traffic flow at these intersections, reducing idling, braking, and acceleration.
This initiative is part of Google's broader goal to assist its partners in achieving a one gigaton reduction in carbon emissions by 2030. Initially introduced in 2021 and tested at four intersections, the project demonstrated fuel consumption and intersection delay reductions of 10% to 20%. Since then, the pilot program has expanded to numerous partner cities worldwide, including Rio de Janeiro in Brazil, Manchester in England, and Jakarta in Indonesia. Google intends to further expand in 2024, with early estimates suggesting a potential 30% reduction in congestion points.
Google emphasizes the scalability and cost-effectiveness of its project for cities compared to alternative options, encouraging widespread deployment across networks of residential buildings. The AI recommendations seamlessly integrate with existing infrastructure and traffic systems, yielding noticeable results within weeks. For instance, the Manchester trial reported improved emission levels and an 18% increase in air quality.
Additionally, Google highlights the environmental impact of its map routing, claiming to have prevented more than 2.4 million metric tons of carbon emissions—equivalent to removing approximately 500,000 fuel-powered cars from the road for an entire year.
Comments
Post a Comment