Fatigue equals car crash

Fatigue, not speed, is the leading cause of car crashes, according to detailed new Mercedes-Benz research.

The German luxury-vehicle giant claims as many as a quarter of all car crashes worldwide are caused by exhausted drivers who make a mistake.

Various scientific studies estimate between 10 and 20 per cent of serious vehicle crashes can be attributed to drowsiness. But, according to an investigation carried out by insurance companies in Germany, fatigue is responsible for one in four fatal highway crashes.

The new research is in direct conflict with the Victoria Police view. Its website insists speed is the biggest killer on the state's roads.

Speed is the main cause of 20 per cent of collisions causing fatal and serious injury in Victoria, it states.

Mercedes-Benz's claims are also backed by the US insurance research body the American National Highway Traffic Safety Administration, which has data showing that in the US, more than 100,000 crashes a year are caused by fatigue. These kill at least 1500 people and injure another 71,000.

Today, Mercedes-Benz takes the issue so seriously it runs a fleet of test drivers on Germany's roads.

To obtain objective indicators of fatigue, the drivers are fitted with electroencephalogram (EEG) skull caps.

“We will learn individual driver behaviour and work from there,” Fruh said.

“A lot of indicators in the car can describe your driving patterns, such as your steering or braking characteristics. If the driver does not move the steering wheel for a prolonged period, this is a sign of fatigue.”

Benz has employed engineers, cyberneticists, mathematicians, computer scientists and psychologists on its investigation team.

It has found that observation of one individual criteria alone does not allow reliable detection of tiredness.

“Fatigue is a highly complex phenomenon that can manifest in many different ways,” Fruh said.

“Accordingly, we will make use of a multitude of factors for fatigue detection, including driving style, trip duration, the time of day and the given traffic situation.

“By comparing this data with a stored model, the system will build an individual driver profile and use probability calculus to determine if the driver is exhibiting the first signs of fatigue.

“The problem is yawning and nodding off are not the first signs. They are advanced signs. Our goal is to detect the transition from being awake to being tired, and to warn drivers before they become over-fatigued.”