How does Google Maps calculate your ETA

Google Maps: Estimated arrival times become more accurate

Just a few days after Google switched to displaying the much more detailed maps in Germany, the free navigation app is now available with further improvements under the hood, which should make the estimated time of arrival (ETA) significantly more accurate during route planning.

As Google reported in the house blog, they have teamed up with the wholly-owned subsidiary DeepMind and will from now on use machine-learning algorithms to estimate the so-called ETA. ETA stands for Estimated time of arrival.

Our ETA predictions already have a very high level of accuracy. In fact, we see that our predictions were consistently accurate for over 97% of trips. By partnering with DeepMind, we have been able to reduce the percentage of inaccurate ETAs [...] even further - with significant improvements in places like Berlin, Jakarta, São Paulo, Sydney, Tokyo and Washington DC. This technique allows Google Maps to better predict if you are experiencing a slowdown that may not even have started.

[...] Our predictive traffic models are also an essential part of determining driving routes using Google Maps. If we predict that traffic is likely to get heavy in one direction, we will automatically find an alternative with less traffic.

At the end of August, Google rolled out an extensive update for its online maps, which have now been available for 15 years, which not least refreshed the display in the Google Maps app.

The company generates a new, significantly more detailed graphic map display from high-resolution satellite images. As a result, different vegetation should be better recognizable, but Google also promises more details in the map display in cities. The new maps should be noticed immediately when you open the app, regardless of whether you are looking at the display in urban areas or in rural areas.

Google Maps - Transit & Essen
Developer: Google LLC