THE message couldn鈥檛 be clearer: don鈥檛 drive under the influence of alcohol or drugs. Yet people still put themselves and others in danger by getting behind the wheel while intoxicated. Despite decades of education programmes and law enforcement, road collisions attributable to intoxicated driving remain unacceptably high. We need a better approach and we are in the middle of a technological boom that may just deliver one.
Seat belts and airbags have dramatically reduced the number of lives lost on roads. However, they do little to prevent the cause of crashes. Someone dies every 25 seconds globally due to road traffic injuries, and one in five of the drivers involved will test positive for alcohol following the collision.
Advertisement
Some cars already have intelligent driver monitoring systems that help to reduce deaths by human error, such as inattention. These use cameras to monitor drivers鈥� alertness. When there is a deviation from the behaviour it has been trained to recognise, such a system can either warn the driver or take complete control of the vehicle to stop a collision from occurring.
These systems may . Already, they must be installed in all new European vehicles. In 2024, they must be included in all new US vehicles. Australia and New Zealand will also follow suit.
Now, a second wave of intelligent driver monitoring technology is on its way. Internal cameras can be combined with other biometric sensors, such as heart rate monitoring or skin conductibility, to determine a driver鈥檚 internal state in real time. In other words, a car could determine when its driver is impaired by drugs or alcohol.
These technologies can already spot certain behaviours that are hallmark features of intoxication. For example, alcohol significantly impairs alertness and attentiveness, which such systems can detect, and the more someone drinks, the more pronounced these effects are.
Therefore, it might be possible to retrain such systems to identify and monitor the individual 鈥渋mpairment signature鈥� for a range of other substances that are known to negatively affect driving. Ideally, this information would be used to automatically block the person from driving.
My colleagues and I are . The first stage is to capture the exact signature of different drug types and characterise their effects on eye movements and driver behaviour. Then we can train the technology to efficiently identify these patterns.
Right now, my research group is determining the unique impairment signature for drivers intoxicated by , or alcohol.
Creating individual profiles will help to differentiate between a driver who has taken legal prescription drugs needed to treat a medical condition, such as benzodiazepines for anxiety, and someone who has consumed illegal drugs, such as methamphetamine.
Some people may ask: 鈥淏ut won鈥檛 cars be driving themselves, anyway?鈥�
It is true that semi-automated vehicles are on track to be legalised in the UK as early as next year, for instance. But although these systems can control a car鈥檚 movements, they require an alert and capable driver to take over if needed. Fully driverless technology is still a long way from being realised.
In the meantime, driver-state monitoring systems will help stop crashes by ensuring that a driver is fit to be behind the wheel.
Over the next decade, these technologies could vastly reduce the number of traffic collisions. If we embrace them, we may finally be able to stop deaths linked to intoxicated driving.
