India is a dangerous place to drive. According to a 2017 report published by the Telegraph, the fatality rate on Indian roads at that time was 19.5 per 100,000 residents. The U.S. was 11.5. Much of the danger lies in India’s notoriously bad roads. Yet drivers still bear some responsibility as well.
Unfortunately, it does not appear that India’s roads are getting any safer. Enter Harvard’s John A. Paulson School of Engineering and Applied Sciences and their challenge to engineering students to come up with solutions. One such solution involves monitoring driver behavior so as to prevent accidents caused by drivers falling asleep at the wheel.
Keep Them Awake, Keep Them Safe
The prospect of falling asleep behind the wheel is something many of us have faced. Sometimes you just don’t realize how tired you are until you’ve been on the road for quite some time. Without warning, you can start to nod off thanks to a combination of fatigue and monotony. And as we all know, nodding off while driving creates a potentially deadly scenario.
Understanding this, Harvard students went to work on creating a system that keeps drivers safe by keeping them awake. They developed a device capable of detecting drowsiness by monitoring a driver’s vital signs in real time. If those vital signs suggest a driver is in danger of falling asleep, a warning alarm sounds.
The Harvard project is not the first to rely on monitoring drivers in danger of falling asleep. What makes their project unique is the fact that the students zeroed in on heart rate rather than simply deploying lasers and cameras to watch a driver’s eyes – as some existing technologies already do.
Earlier Detection of Drowsiness
A system of cameras and lasers is designed to detect when a driver’s eyes close. Such a system is better than nothing, but it could ultimately be too little too late. It only takes a split second to lose one’s concentration on the road. So if a driver is drowsy enough that his eyes close, even momentarily, an alarm still might not be quick enough to prevent an accident.
The Harvard project monitors heart rate variability over a given period in time. It is capable of detecting variations within a defined threshold, variations that can better predict when falling asleep is imminent. This allows the device to sound the alarm before the driver closes his eyes.
Harvard students say their biggest challenge was dealing with all of the data their detection device produced. They had a raw signal that needed to be processed in order to make it useful. They relied on the advanced signal processing technologies that companies like Rock West Solutions have developed in recent years.
Signal processing is all about taking a raw signal and eliminating the noise, Rock West engineers say. When you are monitoring something like heart rate using a radar module, you’re dealing with a tremendous amount of noise that is of no value to the task at hand. That noise has to be filtered out through signal processing.
Only Useful Information, Please
If you want to come up with a system that prevents drivers from falling asleep based on monitoring their heart rate, you cannot afford useless information that gets in the way. You have to be able to zero in on heart rate signals themselves. It is a tall task but one that Harvard students seem to have made good progress on.
With any success, their work will make roads safer in India. That should ultimately mean safer roads around the world.
