DMS: I'm watching you, distracted driver

A recent report from the National Highway Traffic Safety Administration in the United States shows that in 2018, a total of 400,000 people were injured in car accidents caused by driver distraction; in the UK, one-third of fatal traffic accidents every day are caused by distracted driving; in China, many people are killed or injured in car accidents caused by distracted driving every year. In order to improve the safety of vehicles and roads, Euro NCAP has decided to make the driver monitoring system (DMS) a necessary safety feature indicator in the five-star rating from 2020. Therefore, more and more companies have begun to study or deploy driver monitoring systems (DMS) to monitor the driver's status and reduce the probability of accidents.

Shanghai Baolong Technology and Ambarella to develop advanced AI driver monitoring platform

(Image source: Ambarella official website)

Ambarella and Shanghai Baolong Automotive Technology Co., Ltd. announced a collaboration to develop a high-performance driver monitoring system (DMS) platform with advanced artificial intelligence capabilities to assess driver fatigue. The DMS platform will be built on Ambarella's CV22AQ CVflow® computer vision chip and Baolong's visual perception software, which can capture the driver's facial information, including the features of the eyes, head and mouth, through image sensors. Then, Ambarella's computer vision system-on-chip (SoC) performs image processing and artificial intelligence analysis to determine the driver's fatigue level in real time, ultimately improving vehicle safety.

Eyesight partners with multiple companies on driver monitoring systems

(Image source: Eyesight)

Eyesight Technologies, an AI computer vision solutions company, has launched the driver monitoring system DriverSense, and is constantly adding new features to detect driver distraction caused by mobile phone use, smoking, distraction and drowsiness to prevent accidents. The DriverSense system analyzes the driver's facial features, including head posture, gaze direction, blinking frequency and eye openness to detect signs of fatigue and distraction. It not only recognizes the driver's behavior through physiological characteristics, but also detects smoking and mobile phone interference, allowing automakers to intelligently remind drivers based on the type of distraction detected. For example, the driver's use of a mobile phone will trigger a related alarm, while signs of fatigue will trigger a more urgent alarm.

(Image source: Eyesight)

With its advanced technology, Eyesight has also successfully established cooperation with many companies. For example, the company has signed a strategic cooperation agreement with Hefei Zhixin Automotive Technology Co., Ltd. to jointly develop a driver monitoring system; it also cooperates with SEAT's Xplora team to study the driver's eye and head movements to detect whether they are asleep or distracted. It can also identify drivers who have driven the vehicle before and adjust seats, mirrors, heating settings and other cockpit functions according to their personal preferences. In addition, it will be able to identify pedestrians and analyze whether the driver has also discovered pedestrians; it has also established a partnership with Samsung Electronics Co., Ltd. in South Korea to integrate Eyesight's advanced artificial intelligence computer vision driver monitoring software into Samsung's in-vehicle camera solution software to create the most advanced fully integrated driver monitoring system for direct installation by automakers.

Eyeware and Melexis partner on driver monitoring system

(Image source: Eyeware official website)

Swiss 3D eye tracking technology company Eyeware and Belgian microelectronic semiconductor solution provider Melexis have teamed up to create an advanced driver monitoring system that uses Melexis' latest MLX75027 3D ToF sensor and Eyeware's attention monitoring technology to accurately monitor the driver's gaze and head movement even in sunlight. The system uses the distance data provided by the 3D ToF sensor to further enhance the reliability of head and gaze tracking, so it can monitor head movements over a wider range. The Melexis MLX75027 3D ToF sensor has VGA (video graphics array) resolution and uses a high modulation frequency (20-100MHz) to drive infrared illumination. Thanks to the patented pixel design, it is almost completely unaffected by light and can operate robustly and reliably even when the light is constantly changing. In addition, the sensor is compact and can be easily integrated into the rearview mirror.

NXP and Momenta collaborate on driver monitoring system

(Image source: NXP official website)

NXP Semiconductors (NXP) announced that it is collaborating with Momenta to develop automotive-grade driver monitoring solutions (DMS). Such solutions will form the basis of systems that monitor driver attention and increase road safety. The purpose of the collaboration is to enable automakers to deploy DMS applications on mass-produced vehicles. The solution will combine the high-performance energy-saving architecture of NXP's Open Vision Platform (S32V2) with Momenta's deep learning software and expertise to optimize, compress and accelerate deep neural networks so that they can run efficiently on automotive-grade DMS embedded platforms. It can not only monitor whether the driver is distracted, but also ensure the high performance of other visual processing tasks in the car, reducing deployment costs for customers.

Seeing Machines launches Guardian backup driver monitoring system

(Image source: Seeing Machines)

Australian company Seeing Machines has released a retrofit driver monitoring system specifically for use in autonomous vehicles to promote the adoption of semi-autonomous and autonomous vehicles. The Guardian backup driver monitoring system is designed to ensure that the backup driver in an autonomous vehicle remains alert, awake and ready to perform driving tasks when necessary. Using Seeing Machines automotive-grade FOVIO driver monitoring technology, it can be easily used in the retrofitted driver monitoring system of SAE L3 (eyes off the road) to L5 (no driver required) autonomous driving test fleets.

The main function of the camera-based Guardian backup driver monitoring system is to track the face and eyes of the driver of autonomous and semi-autonomous vehicles being tested on the road, report driver status information (such as whether the driver is looking at the road or not), and identify increasingly serious driving distraction events (such as the driver's lack of attention to the road scene).

(Image source: Seeing Machines)

In addition, the company has also worked with Qualcomm to deploy a DMS solution for a global high-end car manufacturer. Because the solution uses the efficient heterogeneous computing modules of the Qualcomm Snapdragon platform, it can be expanded to different grades of cockpits and can be integrated with the infotainment system to ensure that the driver can focus on the road.

Subaru introduces driver monitoring system to detect driver distraction/fatigue

Subaru first introduced the DriverFocus driver monitoring system on the revamped 2019 Subaru Forester's top-of-the-line Touring version, and plans to expand it to the 2020 Subaru Legacy model, becoming standard equipment on the Legacy XT model, Limited and Touring version, and becoming an optional system on the Limited version of the Legacy mid-size sedan.

Developed by Japan's Mitsubishi Electric, DriverFocus uses a small infrared camera embedded in the car's infotainment system and uses facial recognition software to identify signs of driver distraction and fatigue. Once the camera scans the driver's face, the system tracks the driver's eyes. If the driver's eyes are off the road for 2 to 3 seconds, the system will issue an audible and visual alert to remind the driver to concentrate. In addition, both Subaru and Mitsubishi Electric said that the technology does not record data and violate privacy.

Edgetensor develops AI driver monitoring system to monitor whether autonomous driving drivers are distracted

(Image source: Edgetensor)

Edgetensor, a US startup, has developed an AI-based driver monitoring system for autonomous vehicles. The system is designed to use cameras to run powerful AI processing software on low-cost commercial hardware, using an AI open source inference engine optimized for two low-power devices (cameras) to monitor passengers in real time with high precision and cost-effectiveness. It also provides facial tracking, head posture detection, eye and mouth tracking, and gaze and iris tracking to ensure that the driver remains focused. In addition, the system can even recognize a person's mood based on facial posture. If the system's software determines that the driver is not paying attention, the system will immediately send an alert.

Edgetensor also provides in-car monitoring solutions and facial recognition capabilities to authenticate the driver and make corresponding adjustments in the car. In addition to being used in autonomous vehicles, this technology can also be used simply to monitor driver behavior. For example, a UBI provider can use this technology to assess risk to determine whether the driver is driving recklessly or distracted.

Although the fundamental way to eliminate driver distraction is to improve the safety awareness of all drivers and enhance their sense of driving responsibility, when it is impossible to ensure that all drivers have the same high level of driving safety awareness, utilizing the power of the above-mentioned driver monitoring system is also a good way to avoid traffic accidents caused by distracted driving.