Google Maps is improving its ability to deliver more accurate arrival times by predicting when and where traffic jams will appear…
Every day, Google Maps helps people around the world navigate an astounding 621,371,192 or 6.2+ million miles. While that’s an amazing amount of data, even more impressive is that Google says that it’s accuracy rate is over 97 percent. But, the tech giant believes that it can still improve. And, is doing so by partnering with DeepMind to use machine learning in order to make its ETAs or estimated arrival times more accurate than ever.
Google Maps Improving Commute ETAs with DeepMind AI
Prior to its teaming up with DeepMind (which is an Alphabet research lab), Google Maps relied on historical traffic patterns and live traffic conditions to understand overall traffic conditions. Now, the two are working together, developing a graphic neural network, which takes a number of factors into account, including time of year, road quality, speed limits, collisions, and closures.
Armed with its new technology and data, Google Maps has experienced an improvement of 50 percent for real-time ETAs in cities such as Berlin, Jakarta, São Paulo, Sydney, Tokyo, and Washington D.C. Now, Google Maps is actually able to warn drivers about traffic jams before they form.