Better Road Safety For Cars? Look To The Insects, Say Scientists

Self-driving cars seem to be the road some automakers are going down, but they are not without their own set of flaws. Concerns over autonomous cars being able to make judgement calls have made them hard to integrate onto the roads with humans.
 
In order to avoid accidents when driving, autonomous vehicles employ a collision avoidance sensor using a LiDAR system. However, the technology has its challenges as well. For one, the actual LiDAR modules can only be miniaturized to a certain extent, and most require large amounts of energy to be powered up.
 
Scientists at Pennsylvania State University are now looking to nature instead as a guiding light for self-driving cars. 
 
The team studied the neural circuits which keep insects such as locusts from bumping into other objects while flying. This resulted in the developing of an optoelectronic sensor that uses eight photosensitive ‘memtransistors’.
 
The processor measures just 40 square micrometers and takes up only a few hundred picojoules of energy. By comparison, that’s tens of thousands of times less than a conventional LiDAR system.
 
The device helps estimate the distance of automobiles by measuring changes in the intensity of the headlights—meaning, the brighter the lights, the closer the car. When tested in actual vehicles, the system could predict two-vehicle accidents two to three seconds before they even occurred. 
 
This might not seem long, but even a couple of seconds can give the vehicle enough time to swerve out of the way and save its passengers. According to the paper released in ACS Nano, only 25% of driving happens at night, yet most fatal accidents occur when driving in the dark. And with how things are going in terms of self-driving cars being on the rise, it’s clear that we need a solution fast.
 
 
 
[via New Atlas and Pennsylvania State University/ACS Nano, Photo 10445178 © Maksim Toome | Dreamstime.com]