EN
Right now, roads all over the world are filled with 1.2 billion vehicles. And it’s a number that is constantly growing. Every year, these 1.2 billion vehicles are involved in 50 million road accidents. And out of these 50 million accidents, 1.3 million have a fatal outcome.
While there is no one cause to blame for these tragic accidents, we have been able to identify three of the most common culprits: distraction, fatigue, and intoxication.
For more than 20 years now, we’ve been delivering our driver monitoring technology to the automotive industry, with the goal of making sure that every new car put on the market is better equipped to help the driver overcome challenging and dangerous situations.
It won’t take long before we start seeing the positive effects of Driver Monitoring Systems (DMS). Slowly but surely, the number of accidents will start decreasing. But only a small fraction of the global vehicle fleet is updated each year.
For every new car put on the market, there are many older, less safe, cars still in use – the majority of which will remain on the roads for years to come. If we’re serious about improving global road safety – and saving lives – we can’t just ignore these millions of vehicles.
This is an issue that has kept us occupied for years. And in 2018, our best engineers and developers started working on a solution.
We recently spoke with Magnus Brunzell, VP of Applied AI Systems at Smart Eye. With a Masters in Engineering, Magnus has years of experience working with automotive Tier-1 suppliers such as Delphi/Aptiv. He has had roles in project management, engineering, sales & marketing and has been a managing director. I invited him to talk today about Smart Eye’s latest product, AIS, from the Applied AI Systems business unit.
“AIS is actually more of a line of products, with several variants to meet different needs. It is a Driver Monitoring System that detects the driver’s face, eyes and gaze while driving and can give warnings for behaviors such as distraction and drowsiness. It’s a standalone system that consists of our proprietary hardware and software, and fills a void in the market for namely commercial vehicles and the aftermarket.”
“One common customer is the fleet operator. They have a fleet of existing vehicles and care about the safety of their drivers as well as the commercial value of the vehicles. The fleets are typically trucks, buses, delivery vans, taxis etc. Other customers can be insurance companies and car rental companies, who want to avoid accidents for obvious reasons.
All of the above can deploy AIS in their vehicles, but we also work with truck, bus and EV manufacturers for original installation, so fleet operators can use our off-the-shelf solution instead of spending large amounts on product development.”
“To make the products scalable and to fit our different types of customers, we have split the components into two units. One electronic control unit (ECU) and one camera. The ECU can be installed pretty much anywhere in the vehicle compartment and the camera should be installed where the driver’s face is visible.
We have developed our own hardware in-house and it is of course according to automotive specifications. The system captures images of the driver and analyzes the driver’s face and eyes especially; so for example, using our software, we can determine the eye gaze direction of the driver. We use the driver images to feed our AI-powered software, so that the system is able to detect early signs of drowsiness or distraction and prevent dangerous situations. It’s important to note that no images are stored from this operation–it all runs locally and nothing is sent to the cloud.
Our solution uses infrared technology at a frequency that is invisible to the human eye and we provide our own light sources, which are infrared LEDs. This means that the AIS works very well both in very sunny and very dark environments.
AIS is based on the same great eye tracking and driver monitoring software that Smart Eye uses in all interior sensing systems. This means that we have a lot of “mileage” even before we launch, and that we are confident in high performance for AIS.
The system is also very easy to use. It is plug-and-play out of the box and should be up and running in just a few minutes. The user doesn’t need to calibrate the system since we have developed a self-calibrating function. For settings and installation support we use a smartphone app developed by Smart Eye, and to monitor and control your fleet of systems there is a cloud solution.”
“AIS has detection of distraction, drowsiness and dangerous behavior. We define distraction as when the driver takes away the eyes from the road for a certain period of time, which means that we also detect if you use your mobile phone to text. Drowsiness is detected by studying the driver over time and finding the early signs of sleepiness. Dangerous behavior, in this context, is to detect the use of mobile phones, and secondary activities such as eating and drinking.”
“To strengthen the team, we hand-picked some key people with extensive experience in the automotive industry, both from car manufacturers and suppliers. This team has led the development of both hardware and software that were developed in parallel. We have also put a lot of effort into choosing the right manufacturing partner, who can contribute with deep domain expertise. Of course we are guided by the legislators globally, and we did not forget to listen to our customers to really get all the necessary functionality.”
In the end, it’s about saving lives, and with AIS we will reach many new customers to do so. We believe that driver monitoring systems will be as natural in cars as seat belts and airbags are, but the big difference is that driver monitoring systems will avoid accidents rather than reduce their effects. Despite some early challenges with COVID-19 and supply chain delays, we have secured components for the first batch of products being produced as we speak. At the time of this publication, we are doing extra tests for this first batch and will ship them very soon.
