Neural networks, AI, and deep learning are being employed by many organizations to help improve your driving experience. By using historical data and monitoring existing users, apps like Google Maps and Waze, are beginning to make traffic prediction something of a science.
But, of course, they are not infallible.
Can Google maps predict traffic?
If you've ever used Google Maps and your smartphone as handy GPS, you've probably noticed its handy ability to predict traffic. It provides estimates of travel time, based on the most recent traffic information the app has on hand.
In its early days, the app only had access to historical statistical data on a particular section of road, but now it has improved and become a lot smarter. So much so that it can give you reasonably accurate arrival time, forecasted to the exact minute.
But how does it do this?
It turns out what Google is doing is monitoring the many millions of users of its app around the world. By using their location and average travel speeds, it can track and collate journey times on many roads.
By feeding this data into a special algorithm, Google Maps can combine it with historical traffic trends to give you a best estimate of your journey time.
They even announced, in 2017, that they were beginning to adopt a neural network to help improve the service.
The more people use the app at any particular place and time, the more accurate the prediction becomes. But, of course, it is not perfect.
It is, after all, challenging, if not impossible, to ever 100% predict future events based on past or current data.
Currently, Google Maps Traffic information is not available in all countries, but they are working on it.
How do I see traffic at different times on Google Maps?
Google Maps has come a long way since its early days. With recent features like traffic prediction, it has become a handy tool for planning journeys.
But did you know you can also check (depending on location) things like fare prediction, crash and speed reports?
One of the coolest features is the real-time traffic prediction feature.
But this "hidden" feature is not well known to most users. You can use it to see information on traffic predictions at certain times of the day.
Here's how you can do it (courtesy of techwiser.com):
If you want to check out this feature, it's pretty simple once you know how.
First of all, with Google Maps open on your smart device, tap the 'Directions' button in the bottom right. From here, enter your starting point and destination to generate a route.
When the route appears, you should be able to see real-time traffic patches in red on the blue route plot. If you then tap on the 'options' button (three vertical dots) on the top right, you can then choose 'set depart & arrive time' in a new pop up.
From here you can then select Time and date for your departure and tap set.
Et voilà, you'll be presented with a prediction for traffic and estimated travel duration at a particular time of day.
How does Waze predict traffic?
Called "Planned Drives" the feature promised to make planned trips smoother as it planned your route to avoid the worst traffic spots, based on your set schedule.
With it you can select when you want to arrive at a destination, and Waze will do the rest.
Being owned by Google, Waze uses similar tech to Google Maps. It takes account of expected traffic conditions using smart algorithms, historical data, and predictive analysis.
It also, like Google Maps, collects and analyses data from the many Waze users in the area. By recording the average speed that cars on a particular section of the road are currently on, or plan to be on, it can churn out a reasonably accurate estimate of your ETA.
Also, like Google Maps, the more users use the app in the location at any one time, the more accurate the predictions.
Waze also offers a means of linking to your calendar and Facebook events so that it can let you know beforehand when it's time to leave and be on the road.
What other ways are Neural Networks being used to predict traffic
But Google and Waze are not the only people working on making traffic prediction cutting edge. Various other organizations are also employing Neural Network tech to help make our commuting experiences more streamlined.
For example, researchers at the Miguel Hernández University (UMH) of Elche have developed a form of AI, using deep learning networks, to predict traffic conditions.
Using data from fixed sensors on loops and connected vehicles, they have been able to predict traffic 15 minutes ahead of time.
Their study, conducted on Spain's A-7 Motorway between Alicante and Murcia, is yielding some interesting results. The team was also granted access to the Levante Traffic Management Centre's last 12-years of data to help them out.
"Their conclusions show that until at least 15 percent of vehicles are autonomous, there won’t be a noticeable benefit regarding the fluidity of traffic and the capabilities of motorways unless solutions are developed to guarantee an efficient coexistence between autonomous and conventional vehicles," reports transport magazine Intelligent Transport.