The automotive industry is on the verge of the largest technology revolution it has experienced since the first car was invented over a century ago. Consumers are seeing higher reliability, increases in efficiency and more luxury in vehicles than ever before, but these advances are meaningless without robust safety-focused features to support them.

Seeing Machines Seeing Machines

According to the World Health Organisation, over 1.3 million people die in road accidents every year, with 94% of these fatal accidents caused by human error.

Seeing Machines has spent over 17 years focusing on the development of highly reliable, real-time human monitoring technology that saves lives.

Underpinned by a team of technical and scientific experts, as well as automotive industry veterans, Seeing Machines with its FOVIO driver monitoring platform technology is leading the charge globally to help mitigate injuries and deaths caused by driver distraction and drowsiness.

From monitoring basic driver attention for intelligent driver-aware Advanced Driver Assistance Systems (ADAS), to safely enabling “Co-Pilot” technologies for [semi-] autonomous driving of the future through understanding of the driver’s state, Seeing Machines continues to lead the implementation of state-of-the-art Driver Monitoring System (DMS) technology through solid data-driven Human Factors and optical path research, as well as algorithm development and validation geared toward real-world driving environments.

Every day in the US, 11 US teens die from texting while driving.

Seeing Machines’ technology can help to mitigate this problem today. Our FOVIO driver monitoring platform detects and alerts when drivers are not paying attention to the driving task. Seeing Machines’ driver monitoring systems are already deployed in over 200 commercial trucking and transport fleets worldwide to detect and mitigate distraction-related driving events and get truck drivers home safely to their families.

Now, with the emergence of intelligent vehicles across multiple transport sectors, Seeing Machines’ technology will be integral in safely enabling automotive manufacturers (OEMs) through this transition as semi-autonomous and autonomous vehicles evolve.

Safety – a critical challenge for autonomous vehicle manufacturers. 

 

General Motors, a household brand in the USA, launched the world’s first semi-autonomous car in 2017, the Cadillac CT6 with Super Cruise. Seeing Machines’ FOVIO driver monitoring technology debuted in the Cadillac as an integral part of the vehicle’s safety architecture.

Seeing Machines’ DMS uses a camera and sophisticated algorithms to check if the driver is paying attention to the road, and outputs data to the vehicle systems to determine if the driver is continuously paying enough attention to the driving task to allow for complete removal of their hands from the steering wheel while in automated highway driving mode.

The DMS is a critical new market solution for overcoming the safety challenges associated with the safe roll-out of SAE Level-2 through Level-4 autonomous driving technologies – that of keeping drivers engaged and ready to retake control of the car whenever required.

Seeing Machines

The Importance of Reliable DMS

Worldwide regulators are responding to the technology advancements in the automotive industry with recommendations around driver monitoring technology as well as impacts to vehicle safety ratings, recognising the adoption of this technology as vital in the push for a safer motoring public.

In September 2017, Euro NCAP (European body responsible for vehicle safety ratings and testing) unveiled its “Pursuit of Vision Zero” Roadmap 2025, with the goal of zero automotive accidents. It identified Driver Monitoring as a primary safety feature, required by 2020 for ANY new on-road vehicles to achieve a 5-star safety rating.

Responding to a fatal accident involving a semi-autonomous vehicle with “Autopilot” mode engaged, the US National Transportation Safety Board (NTSB) published its investigation report which concluded that overreliance on the feature and prolonged driver disengagement from the driving task contributed to the accident. Several specific safety recommendations were issued in the NTSB’s report for design of semi-autonomous vehicles, including the adoption of more effective monitoring of driver attention, commensurate with the capability level of the automated driving system.

More about Driver Monitoring Technology

The DMS consists of a small infrared camera and near infrared lighting pods mounted on the driver’s side vehicle cockpit and is focused toward the driver’s face. The raw video from the camera is passed to main processing element with Seeing Machines’ FOVIO Driver Monitoring Engine (DME) where it is processed real-time and measures attention by continuously searching for and tracking the features of the driver’s face and eyes. Simultaneously tracking the multiple points on the face, eyelids and pupils in a large head box over large and sudden head rotations and eye gaze angles, operating robustly in all foreseeable driving conditions including direct sunshine and total darkness, regardless of race or ethnicity, and even when the driver is wearing sunglasses is key to securing a highly reliable real-time core signal set for our classifiers for driver attention and driver state to work from.

Based on thresholds set by the OEM or Tier One electronics integrator, if the driver looks away from the road or closes their eyes for more than a certain period (for example, can be dependent on the speed of the vehicle), the system will begin to escalate a series of warnings to encourage the driver to re-engage their attention to the road. Notably, in most circumstances, as soon as the driver looks back to the road, the warnings to the driver may be switched off. If the driver does not or cannot re-engage their attention to the driving task after the full series of escalating warnings, the DMS will output this information to the vehicle systems to take action (for example, automatically slowing and stopping the vehicle). Integrated with the various vehicle ADAS and Human Machines Interface (HMI) systems, the DMS can acts as a friendly co-pilot, reminding the driver to pay attention, but with an absolute minimum of effort and reducing the number of unnecessary warnings.

SAE Levels of Autonomous Driving

The Society of Advanced Engineers (SAE) defined six levels of autonomous driving, outlined below for easy reference:

Monitors EnvironmentLevelDriver Involvement
HUMAN

0

No automation

Human driver is in complete control. Maybe the car shifts its own gears, if you , you know, hate fun.

1

Driver assistance

Driver must be ready to take over at any time. Features like adaptive cruise control, parking assist with automatic steering, lane keeping assist, in any combination.

2

Partial automation

Automated system is doing most of the driving, including steering, braking, and accelerating, but human must be alert and ready to take over at any time.

VEHICLE

3

Conditional automation

In specific situations and environments, like highway driving, the car is in complete control. Human driver can read or text or whatever if they want.

4

High automation

The car can drive itself independently in most environments, with some exceptions for weather or unusual environments. Human may still need to take over at times.

5

Full automation

All robot, all the time. Humans just along for the ride, and, ideally, to tell the car where to go. No restrictions where the car can drive.

Partners

Seeing Machines works directly with Tier 1 suppliers and automotive OEMs to deliver bespoke human monitoring solutions across the Automotive industry. Close alliances ensure that our customers are able to tap into the wealth of expertise in the automotive and engineering teams at Seeing Machines and understand the depth of solution available to them in this increasingly competitive and fast-moving industry.