The automotive industry operates on a backdrop of rapid innovation driven by advances in technology, evolving safety regulations, and the need to lower environmental impact. What’s remarkable today, is the pace of innovation. In the next ten years, we are going to see more changes in automotive than we’ve seen over the past 40. I say this because the coming wave of modernization is characterized by a level of maturity in multiple technologies and influences. Mobile operators will be able to deliver ultra-low-latency vehicle-to-vehicle and vehicle-to-cloud communication. Safety regulators around the world will ensure we get closer to no fatalities or severe injuries on roads (Vision Zero). Driver-assist features are becoming ever more prevalent as we gradually shift to driverless vehicles. And electrification is moving toward zero emissions, with governments worldwide putting plans into action to phase out fossil fuels.
You could simplify the coming wave of innovation into safer for people, easier for people, and better for the planet. What we do at Tobii can contribute to all these outcomes. But in this post, I am going to focus on how we address the ‘safer for people’ element with our Tobii Driver Monitoring System (DMS).
As a Swede, I am quite proud that Vision Zero began as a national road safety project in Sweden. Introduced in 1995 and ratified by the government two years later, the number of road fatalities has been steadily dropping in Sweden ever since — even though the amount of traffic on our roads has risen by nearly 30 percent.
And the concept has spread. The EU, for example, has set targets and interim milestones with the goal of reaching zero fatalities and severe injuries by 2050. Many major cities in the US have also committed to the idea, putting measures into place that not only address the technical solution, but also the cultural aspects and attitudes toward traffic safety.
The road to zero fatalities is paved with many initiatives. Reduced speed limits, the addition of cycle lanes, speed cameras, and driver-warning systems that use geofencing technologies are just some of them. The most significant risk, however, is in the vehicle — the driver.
According to the Euro NCAP 2025 Roadmap, more than ninety percent of road accidents are caused by human mistakes. Speeding, driving under the influence, drowsiness, fatigue, and distraction are among the main contributory factors — and sudden incapacitation is a growing issue as populations age.
To help prevent these kinds of accidents, vehicle manufacturers are rapidly developing prevention systems that monitor driver attention and warn people when they are in the danger zone. In some parts of the world, these driver monitoring systems (DMSs) will shortly be mandatory on all new vehicles, and I’d say within a decade or so they will be compulsory everywhere — a bit like the way seat belts became mandatory.
A DMS monitors attention. Attention monitoring is not a new subject, scientists and psychologists have been doing research in this field for well over a century, and its application is broad. Attention can be used to assess a person’s reading abilities or health issues like neurodegenerative diseases and brain trauma.
Today, the essential component is an infrared camera, placed unobtrusively at a short distance from the subject. In a vehicle, it can be embedded in the dash, the A-pillar, or in the rear-view mirror. As the camera takes pictures, advanced image processing and machine learning extract physical attributes of the driver, such as how often they blink, what they are looking at, and which way their head is facing, combining them into a data stream.
A driver monitoring system uses the data stream to build up a picture of driver awareness. A sporadic gaze pattern, for example, may indicate subconscious driving — also known as highway hypnosis or white-line fever. The duration and frequency of blinking are indicators of sleep deprivation. The more frequent and prolonged they are, the more acute the problem. Even fixation duration, the amount of time we spend dwelling on a certain point, can be used to assess drowsiness. And changes in pupil size are useful for assessing cognitive load or impairment.
Tobii DMS can generate alert signals when a driver is inattentive or showing signs of drowsiness or cognitive overload. When combined with data from other sensors, such as the speedometer, the vehicle can determine the best course of action — issue a warning or hand over to the autopilot.
Because technologies like attention computing generate tons of data, it’s possible to leverage that data in additional safety and usability features. You could, for example, monitor how often a driver checks their rear-view mirrors, or enable gaze-control for interaction with systems like in-car entertainment.
Like a lot of good ideas, creating a reasonably good prototype may not necessarily be difficult. The challenge comes when you need to create a solution that works for the widest possible population in types of conditions. For vehicles, the solution needs to work in varying lighting conditions — bright sunlight, darkness, and everything in between. It needs to cater for people wearing sunglasses, prescription glasses, face masks, hats, and makeup. Highly regulated industries, like the automotive sector, require solutions based on standards to ensure safety and long-term compatibility. And if that’s not difficult enough, small form factor, minimal compute load, and low power consumption are all vital design considerations — assuring that the creation of a reliable DMS is a significant challenge.
Fulfilling any one of these design considerations is a challenge but being able to deliver all of them is where I believe Tobii makes a difference. I’d say it’s down to our collaboration and development strategies — how we constantly work toward our vision of a world where technology works in harmony with people — which have enabled us to become a global business partner. As any subsystem or component supplier knows, working closely with a range of partners, including OEMs and Tier-1 suppliers, and operating in different industries, especially highly regulated sectors like automotive and healthcare, is one of the best ways to deliver robust technology.
Throughout our history, we have collaborated with major chipset makers and consumer electronics manufacturers to design our technology for ease-of-integration and our supply chain for volume production — ensuring scalability. Our work with niche players in the gaming industry and innovators in healthcare ensures our technology can deliver highly granular and robust data. Our long relationship with renowned universities and research institutes has positioned us as the preferred supplier for human behavior insights. This kind of multipronged approach — working with partners with widely varied requirements — means that we have solved most of the technical challenges that DMSs present.
In addition to solving the technical challenges, I’d say that any automotive solution needs to meet tough cost requirements. Cost is complex. The total bill of materials and development and production expenditure for the complete system need to be factored in — and this is where Tobii’s approach to innovation and our full-stack capabilities come into play.
We’ve built all our solutions on the same tech stack — a platform developed and refined through collaboration with industry leaders in XR and PC, for example, as well as healthcare and automotive. We do this to ensure that every one of our customers reaps the benefits of innovation — no matter where it takes place, whatever product we are developing, and irrespective of the intended market.
Today, Tobii works with OEMs and Tier-1 suppliers in the automotive industry throughout the vehicle lifecycle, supplying expertise and technology, such as objective consumer insights on new design features, as well as showroom enhancement and solutions for driver monitoring systems. In line with our automotive strategy, today we announced the acquisition of Phasya (see Press release) — a medical research and AI software company focusing on cognitive state. Of specific interest is Phasya’s range of validated algorithms that can establish driver state based on fatigue, distraction, and cognitive-load signals generated by systems such as Tobii DMS. In addition to this acquisition, we have announced partnerships with camera solutions provider Sunny SmartLead and NVISO — the world leader in human behavior AI for Occupant Monitoring Systems (OMS).
If you’re curious to find out more about Tobii DMS, why not check out our
product page where you can download the product sheet, which details the data points we can provide, camera requirements, and placement options. You can also get in touch by filling out our contact form.
The DMS, how it works, and how it will help create a future without accident-related fatalities — Tobii's Peter Tiberg explains.Learn more
Subscribe to our stories about how people are using eye tracking and attention computing.