The validation test took place in the Group’s driving simulator, which comprises an Audi TT car in an immersive curved screen. Twenty drivers spent up to an hour undertaking a monotonous driving task - following a car on a motorway - and were asked at five-minute intervals to assess their tiredness using the Karolinska Sleepiness Scale.
Drivers’ assessments of tiredness were compared to assessments of tiredness generated by Blueskeye’s machine learning algorithms in line with the official EU testing protocol. The algorithms use Near Infra-Red cameras mounted in the car’s left and right pillars to capture and analyse eye, head and facial muscle movement multiple times per second and identify early signs of fatigue.
The machine assessments of tiredness are said to have passed the EU’s threshold of 40 per cent sensitivity, indicating the ability of the system to identify instances when the driver is drowsy and achieving the standard required by the EU for inclusion in new cars.
In a statement, Blueskeye’s Chief Scientific Officer, Professor Michel Valstar, said: “Our ultimate goal is to help automotive manufacturers develop a vehicle which can respond to occupants’ emotions as measured by our AI face and voice analysis using vehicles’ pre-existing cameras and microphones.
“Validating our technology for driver drowsiness and attention demonstrates that our underpinning technology works. It also means we are well placed to help our clients meet the anticipated, more stringent, legislative demands of the EU’s Euro NCAP Vision 2030."
Dr David R Large, a Senior Research Fellow in the Human Factors Research Group, said: “This is a perfect example of how collaborating and sharing knowledge with commercial companies, like Blueskeye, is a great opportunity to apply our academic expertise and world-class facilities to turn research into reality and solve real-world challenges.
“The automotive industry is constantly evolving and crying out for new technology to make travel safer and more sustainable - our participants were generally very positive about their car monitoring them for signs of drowsiness and welcomed the idea of their car warning them or intervening if it detected that they were sleepy."
AI is a gamble we cannot afford without cybersecurity
I am reminded of a quote and example from palaeontology. Herbivorous dinosaurs like stegosaurs/triceratops originally weere bi-pedal with large...