Motorcyclists are exposed and vulnerable in traffic compared to other road users. Annually, about 250 drivers are grievously injured in Sweden, according to statistics from the Swedish Transport Agency. The trend is growing – in 2018, 47 motorcyclists were killed, which is the highest rating in a decade.
The researchers Stefan Candefjord and Bengt Arne Sjöqvist have developed the algorithm that makes it possible to automatically trigger an alarm to SOS Alarm, via an app in the mobile phone, in the event of a motorcycle accident. The sensor alarm technology could also be used for other unprotected road users such as horse riders and all-terrain vehicle drivers. (Image credit: Detecht and Yvonne Jonsson)
“If a single vehicle accident occurs and the driver ends up unconscious in the ditch, it may take a long time before anyone notices what has happened”, says Chalmers scientist Stefan Candefjord, being a biker himself, and also one of the inventors of the algorithm that is currently used by SOS Alarm in a pilot project.
Fast and safe help for unprotected road users
Stefan Candefjord and Bengt Arne Sjöqvist, who both conduct research at the Department of Electrical Engineering, are very involved in the field of prehospital e-health/Digital Health. The research is about formulating smart IT solutions as support for decisions, aiming to offer the right care and quick treatment, even before the patient reaches the hospital. In this case, it is critical to quickly detect that an accident has taken place to alert the appropriate rescue resources and thus reducing injuries and deaths.
“Our starting point was to develop a function that is similar to eCall, which modern cars of today are equipped with, though instead directed to unprotected road users”, says Bengt Arne Sjöqvist. “The advantage of a mobile app is that it is considerably easier to distribute to the users than a specially designed hardware would be – since a smartphone already is present in almost every person's pocket.”
Originally, the planned target group was single practitioners such as horse riders, cyclists, and all-terrain vehicle drivers. In 2017, the scientists reached out to a group of master’s students via Chalmers Ventures, who signed up for the idea. The company Detecht was established with an emphasis on motorcycle drivers – a homogeneous target group with related interests and a high safety awareness, which the app can fulfill.
“The two of us and Detecht are complementing each other. They are knowledgeable in the field and are driving the business model, while we can concentrate on the actual research part of the project,” says Stefan, who also sits in the company board and will continue to add to the progress of the functions of the algorithm.
For his part, Bengt Arne works within the agenda of the "Prehospital ICT Arena", with a related research project referred to as TEAPaN (Traffic Event Assessment, Prioritizing and Notification). The purpose is to form an IT structure that, in a coordinated way, is able to link several eCall solutions – the motorcycle app being one of many examples – with the society's rescue resources so that they are ranked correctly, and the effort is enhanced based on what really has happened.
Besides the safety aspect, the app also offers the users a few social features such as sharing routes, statistics, and pictures with other motorcyclists. This is something that hopefully makes the app even more motivating to use. In just Sweden, there are over 300,000 registered motorcycles, and globally the interest is also very extensive – the market is said to be substantial.
Being tested by SOS Alarm and 15,000 bikers
For the last three months, including May with a likelihood of an extension, the app is tested in a pilot project at SOS Alarm.
“SOS Alarm is usually being restrictive in integrating new functions into their system, so it is gratifying that they are welcoming this technology and want to evaluate how sensor-controlled alarms can aid in their work”, Bengt Arne says.
Nor has it been hard to recruit motorcycle drivers who are keen to participate in the test.
The test results for April show that the about 15,000 motorcyclists involved in the project altogether have driven 120,000 kilometers, corresponding to three laps around the globe. One single minor accident did occur. The crash was correctly detected, the alarm was triggered as planned, and the operator at SOS Alarm sent intended rescue units to the position of the accident. The number of false alarms has been low, only once was the alarm released unjustified. One possible reason is that the driver probably had the phone lying loosely during the ride, which may have affected the data that the app registered and therefore incorrectly was interpreted as if the motorcycle overturned at speed. We will analyze all such events to decide what further improvements in the algorithm that can be made.
Stefan Candefjord, Researcher, Department of Electrical Engineering, Chalmers
In the future, functions grounded on artificial intelligence are planned to be incorporated, which will improve the motorcyclists’ experiences. For instance, the app can then suggest roads, fika stops and driving routes based on the driver’s earlier choices and preferences.
From a societal viewpoint, it is the opportunity to save lives that is the vital factor. The time that passes from the occurrence of an accident until the victims get help can be directly vital for the outcome. Besides providing better safety for the driver, the app is also providing reassurance for family and friends, who speedily can get an alert regarding an accident.
This is how the app works
The mobile phone's integrated sensors record g-force, speed, and rotation. The information from the run is examined using an algorithm that detects driving behavior that shows that the driver is not in control of the motorcycle. If the alarm is triggered, the driver has 60 seconds to switch it off; otherwise, an alarm message is automatically activated, contacting SOS Alarm with information about the position. Then, the alarm operator initially tries to make contact with the driver over the phone. If the driver requires assistance, or is not contactable, rescue units are notified to the current location.
Essentially any smartphone can be used, the only prerequisite is that the driver downloads and activates the app "Detecht – your motorcycle app". Anyone interested can take part as a test driver.