Google Maps using DeepMind AI to predict your arrival time

San Francisco, Sep 4 (IANS) As people traverse over 1 billion kms with help from Google Maps in more than 220 countries, the company is using artificial intelligence (AI) machine learning (ML) models to predict whether the traffic along your route is heavy or light, an estimated travel time, and an estimated time of arrival (ETA).

San Francisco, Sep 4 (IANS) As people traverse over 1 billion kms with help from Google Maps in more than 220 countries, the company is using artificial intelligence (AI) machine learning (ML) models to predict whether the traffic along your route is heavy or light, an estimated travel time, and an estimated time of arrival (ETA).

Google has partnered with DeepMind, an Alphabet AI research lab, to improve the accuracy of its traffic prediction capabilities.

“Our ETA predictions already have a very high accuracy bar – in fact, we see that our predictions have been consistently accurate for over 97 per cent of trips,” said Johann Lau, Product Manager, Google Maps.

By partnering with DeepMind, Google has been able to cut the percentage of inaccurate ETAs even further by using an ML architecture known as Graph Neural Networks.

“This technique is what enables Google Maps to better predict whether or not you’ll be affected by a slowdown that may not have even started yet,” Lau said in a statement on Thursday.

To predict what traffic will look like in the near future, Google Maps analyzes historical traffic patterns for roads over time.

“We then combine this database of historical traffic patterns with live traffic conditions, using machine learning to generate predictions based on both sets of data,” Lau said.

Since the start of the Covid-19 pandemic, traffic patterns around the globe have shifted dramatically.

“We saw up to a 50 per cent decrease in worldwide traffic when lockdowns started in early 2020,” Lau informed.

To account for this sudden change, Google has updated its models to become more agile — automatically prioritising historical traffic patterns from the last two to four weeks and deprioritising patterns from any time before that.

The predictive traffic models are also a key part of how Google Maps determines driving routes.

“If we predict that traffic is likely to become heavy in one direction, we’ll automatically find you a lower-traffic alternative. We also look at a number of other factors, like road quality,” Google said.

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.

Authoritative data lets Google Maps know about speed limits, tolls, or if certain roads are restricted due to things like construction or Covid-19.

“And incident reports from drivers let Google Maps 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,” Google added.

–IANS

na/

ANA NEWS WIRE Disclaimer:
The African News Agency (ANA) is a news wire service and therefore subscribes to the highest standards of journalism as it relates to accuracy, fairness and impartiality.
ANA strives to provide accurate, well sourced and reliable information across Text, Images and Video. Where errors do appear, ANA will seek to correct these timeously and transparently.
The ANA platform also contains news and information from third party sources. ANA has sought to procure reliable content from trusted news sources but cannot be held responsible for the accuracy and opinions provided by such sources on the ANA platform or linked sites.
The content provided for on the ANA News Wire platform, both through the ANA news operation and via its third party sources, are for the sole use of authorised subscribers and partners. Unauthorised access to and usage of ANA content will be subject to legal steps. ANA reserves its rights in this regard.
ANA makes every effort to ensure that the website is up and running smoothly at all times, however ANA does not take responsibility for, and will not be held liable for times when the website is temporarily unavailable due to technical issues that are beyond our control.