If you have an Android mobile with Google services, Google Maps will be installed by default as a navigation application. In addition to having a huge database with maps and routes, Maps shows us real-time traffic information.
In a new blog post, Google wanted to explain how can they predict the traffic and establish the best routes. Machine learning, user data collection and more: this is how Maps can analyze traffic.
Here’s how Google Maps determines traffic conditions
When we open Google Maps and set a route, the app It tells us the estimated travel time, the path to take, whether there will be traffic or not (and in which sections) and, ultimately, all the necessary information relating to said itinerary. Google Maps uses information from the location data you get from users, in order to be able to report traffic forecasts in real time.
Google not only uses real-time user data, it has a history of road behavior to make predictions about traffic conditions.
However, from Google, they indicate that this information is not sufficient to predict what will happen after 10, 20 or 50 minutes of travel, and that is where “technology really comes in.” To predict traffic behavior over certain time periods, Google uses machine learning. It does this by combining the historical traffic patterns of roads over time. In other words, Google knows that on X roads, between certain times, X vehicles usually drive at a certain speed.
Google makes it clear that when selecting the route, also takes into account the type of road. In addition to knowing if the route is paved or if it is dirt, they know the size of the road and how often it stops more on one or the other, which also makes it possible to predict the routes.
They also use authorized data from local governments to find out the speed limits, tolls or to find out if certain roads are restricted to traffic. This information is combined with that of the users, which really allows us to know if there is construction, closed or collapsed lane, objects on the road and the like.
Google also uses DeepMind technology to improve prediction capabilities
Likewise, Google is using technology from DeepMind, an AI lab, to improve prediction capabilities. According to the company, currently Google can predict traffic conditions with a 97% prediction.
They also say that since the start of the COVID-19 pandemic, they have noticed up to 50% reduction in global traffic and that machine learning models have been updated with the latest data to maintain forecast accuracy.