Problem / Objective

Driver drowsiness, fatigue and distraction at the wheel are often the cause of serious accidents worldwide. According to a study done by the Central Road Research Institute, drowsy drivers are responsible for 40% of the total road accidents. Automotive and fleet management organisations are challenged with addressing the humungous safety issue arising due to driver drowsiness and distraction. PathPartner’s driver monitoring solution, using a single low-power in-vehicle camera and advanced AI and deep learning technologies, provides reliable solution for driver drowsiness and inattention detection. It alerts the driver to reduce the chances of serious accidents and thus providing a safer and comfortable drive.

Solution / Approach

PathPartner Technology is a embedded systems company, headquartered in Bangalore. It has developed a Driver Monitoring System (DMS) with complete hardware and software solution for driver identification and alertness detection. It uses advanced computer vision algorithms to enable reliable driver monitoring on the edge. It's in-car technology uses a single low-power camera inside the vehicle for advanced analysis: driver Identification; drowsiness detection and prediction; driver distraction alert; driver action classification; driver emotion estimation; and in-cabin occupancy. It is capable of precise monitoring such as eye tracking and facial analysis even under unfavourable operating conditions such as poor illumination, dark glasses and different camera positions.

It gives reliable performance at an affordable cost by utilising standard, readily available automotive grade platforms and cameras. It triggers audio and visual alerts if the driver is found to be drowsy or distracted from driving. It also monitors the position of occupants for optimized airbag deployment and seat belt reminders to augment their on-road safety.

It is suitable to be integrated into a wide range of Advanced Driver Assistance Systems (ADAS) and Autonomous Driving (AD) applications. PathPartner DMS is platform agnostic and works on hybrid architectures including ARM, DSP, GPU, FPGA & vision accelerators.

Impact / Implementation

PathPartner Driver Monitoring Solution has achieved 95% accuracy in detecting scenarios of driver distraction. The alerts and warnings aid the driver to either stay focused or take a small break from driving, in order to mitigate the risk of a mishap. The solution is extremely modular and customizable, applicable to various use cases: commercial fleet owners can effectively monitor and enhance safety of cabin crew during the entire journey and lower the fleet operational costs; automotive vendors can enhance NCAP ratings to satisfactorily meet regulatory demands; and automotive Original Equipment Manufactures (OEMs) can add additional safety features in their futuristic AI Cockpit designs.

Image from PathPartner