autoevolution
 

This System Will Know If You Are Driving Distracted

Driver assistance systems have come a long way since the first generation of ABS. Nowadays, you can buy a car that helps you keep to your lane, keep a distance from the vehicle in front of you, and stop if necessary. But what should the car do if you are not paying attention? A new system is here to help.
What sensors "see" when a driver uses a mobile phone without a handsfree device 20 photos
Photo: Fraunhofer institute
Tesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and backTesla Autopilot FSD trip from San Francisco to Los Angeles and back
Europe's largest application-oriented research organization, the Fraunhofer Institute, has developed a new system that is meant to gauge driver attentiveness.

To be specific, we are referring to the Fraunhofer Institute for Optronics, System Technologies, and Image Exploitation IOSB's latest work, which is claimed to be the world's first system of this kind that can use image data and concluding from that data alone.

What is the goal of the innovative technology, and what is its current use?

The goal is to ensure safer roads in the context of more vehicles equipped with SAE Level 3 driver assistance systems, along with future SAE Level 4 assistance systems. With those two standards, the former being the only one currently available, drivers might be tempted to stray their attention from the road and towards a smartphone or something else.

Suppose a driver is not paying attention to the road while using an assistance system that combines an automated braking function, active lane assist, and cruise control. In that case, the risk of an accident can be increased because of the time needed for the driver to step in and react when the vehicle cannot handle the situation further.

For example, if a vehicle with this type of system is used on the highway and the driver falls asleep or is found to be inattentive or otherwise incapable of promptly taking control of the vehicle, the assistance systems on board should be able to detect that situation and adjust accordingly.

Tesla Autopilot FSD trip from San Francisco to Los Angeles and back
Photo: Whole Mars Catalog / YouTube
While most if not all vehicles with these systems have several fail-safes, they can still be tricked by people with an imagination. Doing so will make the vehicle driver itself, but the system will operate as if the driver is just getting a small nudge of help from technology. Once the system's technology can no longer operate in optimal conditions, the system is shut off to prevent any risk of injury.

Well, if the driver were paying attention to their smartphone the whole time, they would not be ready to simply step, hold the wheel, keep within the lines of the lane, and make sure you were going in the correct direction.

Limitations imposed by current technology try to keep these vehicles safe, but it is not enough. The goal is to get zero deaths in car accidents across the world as soon as possible. Reaching that objective is impossible if current driver assistance technology is not improved as soon as possible.

What can the new system do for cars?

Tesla Autopilot FSD trip from San Francisco to Los Angeles and back
Photo: Whole Mars Catalog / YouTube
The new system, yet unnamed (as it appears), can track eye and hand movements to distinguish if the driver is paying attention to driving, the road ahead, both, or none. The worst situation is the latter, but people who cannot be bothered to pay attention to driving should not be driving at all.

Once the system is implemented, those who used driver assistance systems as means to drive distracted by their smartphones or their brunches on wheels will not be able to continue driving while distracted.

The system will be able to gauge if a driver is sleeping, feeling sick, or just not paying attention, along with taking further actions to stop the vehicle from being involved in an accident.

These systems will not be flawless, and they will probably be tricked somehow. We hope that humans will not try to trick the more advanced systems to experience distracted driving using these driver assistance systems.

In the future, when autonomous cars roam the streets, a system like this one will have to be employed to ensure that passengers are correctly wearing their seatbelts.

How does the new system work, and do other humans get to see video footage of the driver and passenger?

The cameras are fitted inside the interior and pointed at the driver and the other passengers. The occupant monitoring system can detect several activities, as well as associated objects. The passenger analysis feature allows the system to detect if the driver is playing with the children in the backseat. Doing so is distracted driving.

The system can also figure out if someone is reading a book or a magazine or falling asleep. From the start, automakers can fit these cameras anywhere inside the vehicle, and the cameras consist of traditional video cameras, infrared cameras, and 3D cameras.

By joining those images together and analyzing the data, the system can estimate how fast the driver can take back control of the wheel If an audio notification requests it.

When data privacy is concerned, the system does not send any images from the vehicle anywhere else. Privacy is an important aspect of the system, and the developers were careful not to jeopardize the aspect.
If you liked the article, please follow us:  Google News icon Google News Youtube Instagram X (Twitter)

Editor's note: Photo gallery shows driver using Tesla's Autopilot technology for extended driver assistance on a longer route, and those images are used for illustration purposes only. There is no link between the system and Tesla Motors.

 Download: Activity Detection Inside the Vehicle (PDF)

About the author: Sebastian Toma
Sebastian Toma profile photo

Sebastian's love for cars began at a young age. Little did he know that a career would emerge from this passion (and that it would not, sadly, involve being a professional racecar driver). In over fourteen years, he got behind the wheel of several hundred vehicles and in the offices of the most important car publications in his homeland.
Full profile

 

Would you like AUTOEVOLUTION to send you notifications?

You will only receive our top stories