Who needs a sensor from the manufacturer? Researchers from the University of Washington have equipped their drone with one of nature's finest detectors: a moth antenna.
“Nature really blows our human-made odor sensors out of the water,” said UW doctoral student Melanie Anderson , lead researcher of the aerial vehicle known as the "Smellicopter."
"By using an actual moth antenna with Smellicopter, we’re able to get the best of both worlds: the sensitivity of a biological organism on a robotic platform where we can control its motion.”
The live antenna responds to chemical signals, allowing the flying vehicle to navigate toward specific odors. The team added the antenna sensor to an open-source hand-held commercially available quadcopter platform, along with two rear plastic fins to create drag and keep the drone oriented upwind.
Anderson and fellow researchers turned to Manduca sexta hawkmoths for their flying robotic platform, placing the moths into a refrigerator to anesthetize them before removing an antenna. The antenna, once taken, still remains active for up to four hours, and that range of time can be extended if stored for longer in the refrigerator, according to the UW inventors.
After adding tiny wires into either end of the antenna, the researchers connected the appendage to an electrical circuit and measured the average signal from all of the antenna's cells. The Smellicopter antenna picked up floral scents and ethanol more quickly — and took less time to recover— than a human-made sensor tested out by the Smellicopter engineers.
The team published their results on Oct. 1 in the journal IOP Bioinspiration & Biomimetics .
The Smellicopter uses a protocol known as "cast and surge" to mimic how moths find specific odors. The UW drone starts at the left for a specific distance. If a smell is not detected, the copter moves to the right for the same distance. Once the Smellicopter detects an odor, it changes its flying pattern and moves toward the object of interest.
A camera and four infrared sensors guide the UAV, measuring nearby obstacles at a rate of ten times per second. When an object is detected within 8 inches of the device, the drone changes direction by going to the next "cast-and-surge" stage.
“So if Smellicopter was casting left and now there’s an obstacle on the left, it’ll switch to casting right,” Anderson said. “And if Smellicopter smells an odor but there’s an obstacle in front of it, it’s going to continue casting left or right until it’s able to surge forward when there’s not an obstacle in its path.”
During tests in the UW research lab, Smellicopter was naturally tuned to fly toward smells that moths find interesting, such as floral scents. But researchers hope that future work could have the moth antenna sense other smells, such as the exhaling of carbon dioxide from someone trapped under rubble or the chemical signature of an unexploded device.
In a short Q&A below, Anderson explains more about how the Smellicopter works, and where it works best.
Tech Briefs: What inspired the choice to use a live antenna? I imagine that this was kind of “out-of-the-box” idea compared to the usual adding of a sensor.
Melanie Anderson: Measuring the electrical signal from a moth antenna (an electroantennogram) has actually been done in research before. We are just the first ones to combine that with a small flying robotic platform! The moth antenna is many times more sensitive than any portable commercial chemical sensor. Additionally, it is light-weight and low-powered — perfect for a small drone platform.
Tech Briefs: Can you take us through the day you tested it for the first time? You mentioned in the UW video [shown above in this article] that it was very exciting and that you weren’t sure that it was going to work. What were you most concerned about, and how did you feel it performed that day?
Melanie Anderson: The antenna also produces signals in response to movement and touch (mechanical stimulus) as well as odor. There were concerns that the vibrations from the rotors and the extra flow of air that the rotors provide over the antenna would make it difficult to separate out odor detections. But the response to odor is much stronger than the other signals, and the antenna works extremely well in our setup on the drone.
We were all very excited that it worked that day! It felt very validating that our efforts to create the circuit that reads the signals from the antenna and combine that with the drone platform panned out to produce such an exciting result!
Tech Briefs: The antenna senses a chemical signal – how does that then inform where the drone goes? Does it just follow a strong smell? Can it somehow differentiate between smells?
Melanie Anderson: Like a heart monitor shows the electrical signal from the heart as it beats, the antenna also produces these pulse-like electrical signals when it smells odor. We can insert small wires into the antenna to measure this signal and see the pulses in response to odor. These pulses inform the drone of when the antenna smells something. For now, the antenna responds most strongly to floral scent and moth pheromone, but we are working on genetically engineering the moth so that it will be sensitive to other scents and can be used in scenarios such as locating bombs or finding trapped survivors in a disaster.
Many different animals who search out odors do so by relying on the wind direction. You can safely assume that if you are smelling odor, then the source of that odor will be upwind from you, since the odor is carried by the wind. In this way, when you smell odor, then you travel upwind, and when you lose that odor, then you travel crosswind until you pick up the trail again. The drone does a simplified version of this where it passively orients upwind using the fins on the back of the drone like weather vanes, and then surges forward when it encounters odor. When it doesn't smell that odor anymore, then it casts left and right until it smells the odor again.
Tech Briefs: What’s next for you and your team regarding this research?
Melanie Anderson: We are very excited to work on genetically engineering the moth antennae to make the Smellicopter useful in a variety of scenarios. For my doctoral research, I will also be exploring simulation-based search methods so that the Smellicopter can fly efficiently in different spaces such as those with obstacles.
Tech Briefs: What application is most exciting to you?
Melanie Anderson: Search-and-rescue is very exciting. It is an application where we currently use nature in the form of sniffer dogs rather than man-made odor sensors to find trapped persons. If we were able to use a tiny drone, or a swarm of tiny drones, to find these trapped persons instead, we would be able to locate them faster and keep the search dogs and rescue workers out of harm's way.
Read the UW team's research in the journal IOP Bioinspiration & Biomimetics .
What do you think about the Smellicopter? Share your questions and comments below.