How Google Maps used AI to help predict traffic during the new crown pandemic.

Google Maps may have had 13 years of practice in correctly mastering traffic patterns, but the new crown pandemic in 2020 has caused problems for its work, and driving during this period is clearly a very different issue,media Slash Gear reported. Google has revealed that it has been forced to adapt its prized algorithms to cope with the traffic uncertainty associated with COVID-19, including discarding old data that has traditionally been crucial to better predicting road traffic.

How Google Maps used AI to help predict traffic during the new crown pandemic.

Typically, Google Maps relies on a wealth of historical information to calculate how traffic will behave on different roads. “For example, a pattern might show that vehicles on Northern California’s 280 Freeway typically travel at 65 mph between 6 a.m. and 7 a.m., but only 15-20 mph in the late afternoon,” explains Johann Lau, Google Maps manager. We then combine this database of historical traffic patterns with real-time traffic conditions, using machine learning to generate predictions based on both sets of data. “

The granularity of this information is important because what happens on the road may not be what people actually expect when they are driving. Google uses these localized data and combines it with artificial intelligence-driven historical trends to predict upcoming traffic. As a result, Lau says, travel length prediction accuracy is already above 97 percent. Although Google says it has seen global traffic halve as early as early 2020 – it is a change in local opening hours. Some areas are limited, some areas are not restricted, and the old model is not.

How Google Maps used AI to help predict traffic during the new crown pandemic.

“To explain this sudden change, we recently updated our model to make it more agile — automatically prioritize historical traffic patterns over the last two to four weeks and prioritize patterns from any previous time.” Lau explains.

Google combines this information with “authoritative information” provided by local governments, including road closures, repair and construction sites, the possibility of COVID-related slowdowns, and accident reports from drivers.

。 A recent update to Google Maps drew on Waze’s honed technology to make the reporting process easier to travel.

How Google Maps used AI to help predict traffic during the new crown pandemic.

Perhaps the most interesting thing to prove is how all this continues to evolve over time, as platforms like the Android car operating system become more common. Google’s connected car software was originally launched on the Poleser 2 electric car and is expected to be available on cars such as General Motors, Volvo and Audi. Running Google Maps as the default navigation system and having access to the car’s underlying hardware will be a richer pipeline of traffic and road data that Google will be able to aggregate anonymously.

In fact, while the aesthetics of digital dashboard design will vary from car manufacturer to car manufacturer, the underlying data fed back to Google will likely benefit every car company that uses the Android operating system. Google, for example, may not rely on active user reporting of snow and ice traffic conditions, but will be able to identify them based on feedback from each vehicle’s traction control and stability system. The data can then be shared across all vehicles running Android-based platforms, allowing drivers to get advance warning of potentially dangerous conditions ahead or simply bypass them.