Long-time New Atlas readers will surely be familiar with the mantis shrimp, the formidable marine crustaceans with powerful spear- or club-like forelimbs used to overwhelm their prey. But their limbs aren't their only remarkable feature. Their eyes are among the most advanced in the natural world, and researchers at the University of Illinois now think that their extreme sensitivity to both light and dark could help self-driving cars to see better in difficult conditions.
There are actually some 400 species of mantis shrimp, but one thing they share is an extraordinary sensitivity to light and dark. Whereas the human eye adapts to either bright or dim environments, the mantis shrimp is able to see detailed nuances in extremes of light and dark at the same time. It needs this to detect potential pray in the bright ocean while hiding in dark holes and crevices.
It's this idea that researchers have applied to cameras, altering their photodiodes to use forward bias mode instead of reverse bias. This means that the electrical current from the photodiodes is no longer proportional to the light input of the camera but instead rises logarithmically, much like the shrimp's sensitivity to brightness. The result is a dynamic range some 10,000 times that of cameras on the market today.
"In a recent crash involving a self-driving car, the car failed to detect a semi-truck because its color and light intensity blended with that of the sky in the background," research team leader Viktor Grueve says in a press release. "Our camera can solve this problem because its high dynamic range makes it easier to detect objects that are similar to the background and the polarization of a truck is different than that of the sky."
The team also took cues from the shrimp's ability to detect light polarization. They did this by adding nanomaterials to the camera's imaging chip which filters light pixel by pixel to detect polizarization.
Above: Everyday objects as seen with a normal camera (left) and the researchers' cameras (middle and right)
The researchers think the breakthrough could help self-driving cars navigate in hazy and foggy weather, and make those difficult transitions between dark tunnels and broad daylight more safely. They also claim the cameras could detect hazards, cars and people three times farther away. New cameras with these capabilities could be mass-produced at US$10 each, they claim – a price that would make them viable for use alongside other cameras.
The team is also exploring potential use in the ocean, where the camera's polarization-detection could be used to analyze sunlight to determine location – potentially valuable because GPS doesn't work under water. The cameras could also be used to record better pictures while under the water.
Other potential applications include detecting cancerous cells (much like mantis shrimp-inspired research out of the University of Queensland) which polarize light differently to normal tissue. They're also working with an airbag manufacturer to see if the technology can help to deploy airbags sooner in the event of a car accident.