Google says predicting traffic and determining routes is incredibly complex, and it will continue looking for ways to keep users out of gridlock and on the safest, most efficient routes possible. "We also look at a number of other factors, like road quality", Google said. To predict future traffic, Google said it examinations historical traffic patterns for streets after some time.
"We saw up to a 50% decrease in worldwide traffic when lockdowns started in early 2020", said Johann Lau, Google Maps' product manager. But, as the search giant explains in a blog post today, its features only work thanks machine learning tools from DeepMind, the London-based AI lab owned by Google's parent company Alphabet.More news: Stephen A. Smith Doubles Down on Steve Nash 'White Privilege' Comment
Google added that it made some changes in the data it uses to make the predictions to follow the subsequent shift in road usage, caused by the COVID-19 outbreak.
Google says to be using a Machine Learning model named Graphic Neural Networks to make improved traffic predictions.More news: Only 'iPhone 12 Pro Max' will feature fastest mmWave 5G, report claims
The feature showing traffic lights at intersections had previously been available on Google Maps in Japan for years, according to the company.
Two other sources of information are important to making sure Google recommends the best routes - authoritative data from local governments and real-time feedback from users. It began prioritizing historical traffic patterns from the last two to four weeks and deprioritizing older patterns. And incident reports from drivers allow Google Maps to quickly show if a road or lane is closed, if there's construction nearby, or if there's a disabled vehicle or an object on the road. Morning and afternoon traffic patterns on which Google has years of data are suddenly irrelevant with so many commuters now working from home.More news: Kate Bishop slings into Marvel's Avengers game as new playable character