“`html
Science & Tech
‘We possess a method for guiding a fly as you would a vehicle’

Kenichi Iwasaki collaborating with fruit flies.
Photos by Niles Singer/Harvard Staff Photographer
Geneticists discover a technique to transform minuscule insects into living automatons
Fruit flies, among the most extensively examined species in science, are agile little beings that can adeptly maneuver their surroundings. They were also among the initial creatures to be altered by genetic engineers.
A recent study, published last month in PNAS, leveraged these traits by converting fruit flies into what the co-authors refer to as “living micro-robots,” whose actions can be influenced by sensory signals. The paper was completed at Harvard’s Rowland Institute, committed to experimental inquiry across various engineering and scientific fields.
“Generally, people envision robots as contraptions made with plastic, metal, wiring, and software,” stated co-author Aleksandr “Sasha” Rayshubskiy, a fellow at the Rowland Institute. But since fruit flies — scientifically known as Drosophilae melanogaster — are comprehensively understood, they can be perceived as living machines.
“You can prompt them to perform tasks that you desire,” Rayshubskiy commented.

The research group employed two techniques to manipulate the flies’ movements, the first involving illumination and a wheel. “We can rotate the light and wheel either clockwise or counterclockwise, prompting the flies to turn,” detailed Charles Neuhauser, a postbaccalaureate fellow involved in the study.
The second method utilized odor, again to steer the flies toward one side. “We make them perceive that they are detecting a scent either to the left or right of their bodies,” Neuhauser explained.
The paper illustrates how these methods were implemented to guide flies through a maze and block out phrases like “Hello, world,” a well-known phrase among computer programmers. “We can modify their genetics to activate actions,” remarked co-author Kenichi Iwasaki, a research associate at the Rowland Institute. “We can manipulate them in the same manner as one would influence a traditional robotic system. We can guide a fly as you would operate a car.”
But do these flies genuinely qualify as robots? “Our research challenges that conceptual frontier,” Rayshubskiy noted. “I define a robot as a device that can perform useful tasks for you.”
The project emerges from “two domains that typically don’t engage,” he added. “You have the realm of engineering where individuals construct robots. They don’t yet regard the fly as a participant in that sphere. Then, there are those who are deeply passionate about fruit flies but rarely explore the robotic realm.
“I’ve positioned myself as a connector between these two domains,” Rayshubskiy explained.
Even though the fruit fly’s 1 mg, 2.5 mm body is minuscule, Rayshubskiy envisions numerous opportunities for future implementations. “They can carry their own weight,” he noted. “Not only could we direct them to specific locations, but we could also trigger behaviors within them that enable them to accomplish additional tasks.”
He provided a practical example. “Cleaning buildings is remarkably challenging,” Rayshubskiy stated. “They constantly accumulate debris that builds up over time. Imagine directing a swarm of fruit flies, thousands of them, onto a building’s surface, where they would remove all the grime accumulating over time. You could employ swarms of flies as cleaning tools.”
A subsequent paper, currently nearing completion by the same research team, examines how fruit flies can be made to engage with external objects, effectively investigating their potential as diminutive pack animals.
“Our findings lay a foundation for deploying swarms of natural ‘robots’ that, with further advancements, could revolutionize applications such as environmental monitoring and disaster management,” stated Rosy Hosking, director of research affairs at the Rowland Institute.
The upcoming research also sheds light on the functioning of the fruit fly brain. “There’s a conflict between our guidance, which is indifferent to the fly’s feelings, and the fly’s own perceptions of safety,” Rayshubskiy remarked. By examining that conflict, and the fruit fly’s reactions, he said, “We could unravel the concealed factors regarding what the fly values by compelling it to perform specific actions.”
“`