Feeling drowsy at the wheel? You’re not alone. New research from the Virginia Tech Transportation Institute indicates that driver fatigue is at play in more crashes than previously estimated. In a 100-car naturalistic driving study conducted by the institute, results showed that fatigue was a cause of 20% of crashes, compared to the 2-3% previously estimated based on surveys, simulator studies, and test tracks.
This is not the only study which has produced some alarming results. The National Highway Traffic Safety Administration (NHTSA) recently reported that drowsy driving causes more than 100,000 crashes a year, resulting in an estimated 40,000 injuries and 1,550 deaths.
Embedded eye-tracking technology that could monitor a driver’s alertness and drowsiness levels is already in advanced stages of development by some automotive OEMs and suppliers, and the successful development of such a system could enable a car to automatically assess and respond to movements and certain actions by the driver.
Avatars and eye-tracking
Many OEMs have been using eye-tracking and virtual reality to improve safety in cars, and earlier this year, Jaguar Land Rover developed an innovative 3D instrument cluster which uses the latest head- and eye-tracking technology to create a natural-looking 3D image on the instrument panel. Cameras positioned in the instrument binnacle or steering column area track the position of the user’s head and eyes. Software then adjusts the image projection in order to create a 3D effect by feeding each eye two slightly differing angles of a particular image. This creates the perception of depth which allows the driver to judge distance.
As well as detecting drowsiness, there are numerous other uses for eye-tracking technology which could benefit the future of in-car safety, such as allowing OEMs an insight into where drivers spend most of their time looking. With this data, developments could be made to incorporate design appropriate decisions on where to place certain features, or which to leave out altogether. The technology could also be used to control additional features in the car more easily than using haptic technology.
Visteon’s Christian Feltgen told Megatrends that eye-tracking technology is gaining in importance as a driver distraction solution. Feltgen is Visteon’s Global Director of Cockpit Electronics, and Vice President of the Tier 1 supplier’s Technology Office. “I think we were the first company to display this technology within the automotive industry and promote it to our customers, not only for the purpose of monitoring distraction but also for the purpose of additional ease of control for in-car features.”
Feltgen believes that eye-tracking will play a role in the future of in-car HMI, not only in the fight against distraction but also as a feature in the multimodal HMI system, and explained that if it is possible to know what the driver is looking at, it is also possible to help guide his interaction with the car. “It is a technology which can not only keep on the look-out for what the driver shouldn’t be doing, but make it easier by enabling a pre-selection of features depending on where the driver is looking.”
The next big thing – face-tracking?
Mercedes-Benz has been researching the analysis of brainwave patterns to identify drowsiness and inattentiveness, but it still has limitations, not least of which is the need to attach electrodes to the driver’s head. Similar technology has come from Toyota Europe; since 2006, the OEM has been developing 3D Face Tracker, designed to analyse the driver’s ‘state’ or mood, by detecting their facial expressions. Jonas Ambeck-Madsen, Senior Manager, Advanced Technology, Research and Development at Toyota Europe, told Automotive World, “The approach is to focus on a 3D model rather than the 2D model used by most face tracking systems because it is more robust.” Using images of real faces to create a 3D model, the system monitors 238 points on a driver’s face to analyse his or her expression.
Even though eye-tracking technology is just gaining relevance in the industry, Toyota’s Ambeck-Madsen has argued that it has limited value when used on its own. Integrating a gaze detector would enable the system to know when the driver is looking for too long at an audio system, for example, helping to identify ‘gazing’, the act of “looking but failing to see”, in Ambeck-Madsen’s words. This is particularly likely to occur when a driver is tired or bored, he said, for example after driving on a highway for several hours. While Toyota has given no indication of plans to commercialise 3D Face Tracker, it underlines the potential for technology to tackle driver distraction.
From mining to automotive
Along with many other technologies now gaining momentum in automotive applications, developments in eye-tracking technology have filtered in from another industry sector. An Australian company called Seeing Machines recently developed eye-tracking technology to tackle driver fatigue, and began a key trial for new safety technology for the commercial transport industry in 2013. The Driver Safety System (DSS) technology began its life in the mining industry, but it has the potential to save lives and dramatically reduce fleet costs in the passenger transport sector.
DSS is designed to detect driver fatigue (“micro-sleep”) and distraction. It relies on eye-tracking algorithms to monitor the driver’s face, sending alerts if it detects either distraction or micro-sleep. The system kicks in at speeds above 70kph (43mph), as long-haul motorway routes are deemed the most dangerous in terms of fatigue and distraction for drivers.
Ken Kroeger, Chief Executive of Seeing Machines, provided some background on the distraction and micro-sleep detection improvements in the mining industry. While the system of alerts quickly taught the drivers to drive in a more alert manner, it couldn’t address fatigue in the same manner. “The brain can be trained to not be distracted, but you can’t teach somebody to not be tired,” Kroeger told Megatrends.
Many OEMs are already exploring the use of eye-tracking technology within their cars and as the concern for driver safety increases, so too will the technology implemented behind the wheel.
In reality, the number of people driving vehicles whilst drowsy is probably much higher than statistics show; knowing how drowsy someone was prior to a crash, unlike drink driving, is impossible to determine. However, if technology such as this makes further tracks in the automotive industry, it is hoped that even if it can’t be solved, driver fatigue at the wheel can at least be tackled.