News | Distracted Driving

DriveWell: gamification and AI to reduce distraction when driving | Video []

July 5, 2019

A US-made app promises to raise awareness among drivers to reduce distraction when driving through a system that uses artificial intelligence, behavioral analysis and sensors to recommend a smarter and safer driving style. The app is called DriveWell and we at HDmotori had the opportunity not only to try it out in Milan’s traffic, but also to have the company explain it to us.

Every year, 50 million accidents occur worldwide. In Europe, 52% of accidents are caused by distraction. 35% of car journeys are made with the smartphone in hand, responsible for 75.6% of all car distractions. We could go on like this for hours, stoning off numbers that are difficult to remember. Let’s try lowering this data into real life instead.

For example, did you know that driving by looking at your smartphone or any other device is more dangerous than driving after drinking? The reaction time of a sober motorist when it comes to emergency braking is about 0.54 seconds. After a couple of drinks, a distance of 1.5 metres should be added to this reaction time, while using a smartphone this distance can be extended to 21 metres.

Regulations are becoming increasingly strict with people using smartphones while driving, but until you get into people’s heads and make them understand how dangerous distraction from smartphones is for themselves but also for other road users, we will not go very far. The question lies in making drivers aware of a change in mentality and behavior.

We can’t always travel with the “mommy” sitting next to us telling us not to use our smartphones, and we can’t rely exclusively on the protective saint on the dashboard. That’s why in the United States, more precisely at Cambridge Mobile Telematics (CMT), they have thought of a system capable of monitoring, through a combination of mobile sensors and artificial intelligence, the driving behaviour of drivers.

The objective of the CMT – born in 2010 within the Computer Science and Artificial Intelligence Lab of MIT and developed over the years in the field of “behavior-based-insurance” – is of course to make the roads safer through a solution that, through an app for smartphones and a wireless tag, analyzes the way the driver drives to make them more aware of the risks and implications of a distracted and unsafe driving.

It all starts with an app called DriveWell and, as the name suggests, helps drivers drive with their heads, providing them with interesting insights and statistics about the way they drive. In particular, the app uses the smartphone to analyze our behavior, counting how many times we unlock the phone display, how many times we take it out from the compartment, how many times we open the apps of emails and messages.

DriveWell uses the data collected by these rather sensitive and precise sensors to track our route and detect, for example, if we brake too abruptly – an indicator of potential distraction when driving. A well-done app, full of statistics and with accurate graphics that go hand in hand with a perfect gamification system, which allows drivers to challenge themselves for a more responsible and safe driving.

Through artificial intelligence and machine learning, the system slowly learns our driving style, enriching kilometer after kilometer a database in which the final score is an average of how we brake, how we accelerate, how we make turns, how often we exaggerate with speed and especially how and how much we use the phone.

The concept of gamification is not only found in the infographic and in the final score, but also in a “social” section that allows drivers to compare their driving style with that of other users using the app, choosing an area that can be for example a neighborhood or a city.

To get an even clearer picture of the situation and make the most of artificial intelligence, the DriveWell app is combined with a wireless tag that we can place on the windshield using a sticker. A tag connected via Bluetooth to your phone. First of all, the tag can immediately send a call to the rescue in case of an accident, and already here the game is worth the candle.

But its presence is also important because, being installed on a single car, it helps the artificial intelligence system (thanks to behavioral analysis) to understand if we are driving or are passengers of a vehicle, for example a taxi. Also thanks to the tag, the app recognizes if we are driving or not, reducing energy consumption, which under normal conditions is around 1-3% of the daily budget on an iPhone.

Both the tag and the app can work independently with each other, but it is obviously in their combination that it is possible to have the maximum potential of this system. For example, you have the “find my vehicle” function, collision warning and much more.

A system that is already in use in 6 continents, with a database of billions of km already covered, and that caters a little to all given the low costs. In reality, the app is downloaded free of charge, and its cost and that of the tag are absorbed by insurance companies that decide to install it to allow customers to have a sort of “black box”, also useful to reconstruct the dynamics of accidents or to determine whether the customer is a more or less virtuous driver.

As Ryan McMahon, VP of Insurance of Cambridge Mobile Telematics, explained to us during the test in Milan, the app is often launched free of charge in contests aimed at determining the most virtuous driver. But its application can also be extended to fleetband rideshares (such as Uber or taxi drivers) who want to ensure first of all an optimization of their activities and that their drivers drive in compliance with the Highway Code and safety regulations.

Full article and video