It has also conducted tests for pilots to test whether its software could improve the efficiency of traffic lights. Inefficiently-timed traffic lights aren’t just a waste of time. As Google’s chief sustainability officers Kate Brandt pointed out at an event for journalists last night, traffic lights are also harmful to the environment as well as public health. The company unveiled a slew of sustainability-centric products and updates today that aims to help users make more informed, environmentally friendly decisions. It’s also conducting research on a project which could switch the AI to traffic light signals to help reduce pollution
If your car stops at an intersection idle time can result in the use of fuel, and “more pollution from the street,” Brandt said. Google’s latest project will use AI to analyze and determine the conditions of traffic and timing at the city’s intersections and time the intersections more effectively. Brandt stated that one of Google’s AI research teams has been capable of accurately calculating the data, and then train a model to improve inefficient intersections.
Google has been running pilots in 4 locations across Israel so far in collaboration with municipalities like Haifa, Beer-Sheva, and the Israel National Roads Company. Google claims to have seen the results of a “10 -20 percent cut in the amount of fuel used and an improvement in intersection delay duration” up to now. Google hasn’t provided any specifics regarding the average usage of these intersections however a video from this event shows a busy intersection. Google also didn’t provide any details on the way in which the AI could work with current technology and the lighting around particular intersections.
“It’s still early days,” Brandt said, “but due to this promising result, we’re currently launching with new trials with Rio de Janeiro and speaking with other cities across the world.” Although we’re probably some time away from the day we can have AI control all of our intersections with traffic This is just one step towards the vision of completely intelligent cities that the sector has been collectively researching for years.
The concept of letting AI determine when vehicles move or stop may be a bit unsettling and even risky at worst, however, the aim of improving the efficiency of traffic lights is an effective one. If we can get through testing and safety precautions in place, we might actually be able to cut down the amount of polluting emissions and fuels.