Anderson and his colleagues took it a step further and programmed the Smellicopter to chase away odors like a moth would. If you are able to smell an odor, the source is very likely to be near you. The same goes for insects like moths, which do something called crosswind casting, in which they lock onto a presumably upwind source and fly towards it, then shift their bodies to the left or the right as needed to stay focused on the scent. Anderson’s team trained the Smellicopter to do the same. “If the wind shifts or if you get off course a bit, you’ll lose the scent,” Anderson says. “And so you throw a crosswind to try to get back on that track. And in this way, the Smellicopter gets closer and closer to the source of the smell.
The researchers call this a “cast-and-surge” algorithm: the drone heads for a scent – in the lab they used a mixture of flower compounds – and turns left or right if it loses the scent. , then advance once it locks. again. The drone is also equipped with laser sensors that allow it to detect and avoid obstacles while it is sniffing.
And, boy, it works well: researchers have found that the Smellicopter gets to the source of an odor 100% of the time. This is due in large part to the extreme sensitivity of a moth’s antenna, which can detect minute odors not on a scale of parts per million, or billion, but thousand billion. A moth increases its efficiency even further with physics: by flapping its wings, it circulates air above its antennae, helping to pick up more scent. Here, too, the researchers took inspiration from nature, using the quadrotor’s rotating blades to move more air over their borrowed antenna.
Of course, at the moment, humanity may not have much use for a flower-sniffing moth drone, so researchers are now exploring ways to use gene editing to create butterflies with antennae that detect odors like those associated with bombs. But could these Frankenmoths be as sensitive to the scents of man-made materials as ordinary moths are to pheromones from potential mates and the scent of flowers? In other words, can researchers re-tune a sense of smell that evolution has perfected for the moth over hundreds of millions of years of evolution?
“Theoretically, you could get more sensitivity, ”says Anderson,“ because the moth antenna can sense a variety of different chemicals, much like how we can smell a variety of different things. The idea for his lab would be to genetically engineer a moth antenna to be filled with the particular protein involved in detecting a desired chemical. It would concentrate the powers of the antenna on one scent, not several.