Small Fruit Flies, Big Heroes: The Microrobots of the Future

When people think of microrobots, what usually comes to mind? Perhaps the swarming magnetic robots in the movie Big Hero 6, or the nanomachines portrayed in science-fiction films. In most cases, these imagined machines are composed of inorganic components—cold, precisely engineered mechanical devices.

However, a concept proposed by Kenichi Iwasaki and colleagues in the Rayshubskiy laboratory at Harvard University suggests a very different vision. Instead of manufacturing tiny mechanical robots, the team explores a biohybrid strategy that integrates living organisms with engineering control systems. Their idea is to guide biological organisms using external stimuli, effectively turning them into controllable microscopic agents. For this purpose, the researchers chose the fruit fly Drosophila melanogaster, a classic model organism widely used in biological research.

To achieve reliable control over the flies' behavior, the researchers developed two different guidance strategies. The first relies on visual stimulation, exploiting a well-known reflex called the optomotor response. This innate behavior causes flies to adjust their movement according to patterns moving across their visual field. When a visual pattern shifts in a particular direction, the fly tends to turn in the same direction to stabilize its path. For example, if a pattern moves from left to right, the fly typically turns right to align its movement with the perceived motion.

Figure 1. Visual guidance experiments in fruit flies and related data（Image source：Kenichi Iwasaki， CC BY 4.0 ）

Using this principle, the team designed a rotating pinwheel pattern composed of alternating blue and black stripes. The pattern was projected around a freely walking fruit fly within an experimental arena. When the pinwheel rotated clockwise, the fly tended to turn to the right; when it rotated counterclockwise, the fly turned to the left. By continuously adjusting the rotation direction, the researchers could steer the flies along specific trajectories. Remarkably, using this visual guidance system the flies could be directed to trace complex paths, even forming letters that spelled “Hello World.” The accuracy of following the intended trajectory reached approximately 94 percent.

Despite its effectiveness, continuous projection around the fly is not always practical for real-world scenarios. To overcome this limitation, the researchers developed a second control strategy based on optogenetics—more specifically, optogenetic manipulation of the fly's olfactory system.

Figure 2. Optogenetic guidance experiments in fruit flies（Image source：Kenichi Iwasaki， CC BY 4.0 ）

In this approach, the team genetically modified fruit flies so that their olfactory receptor neurons could respond to specific wavelengths of light. Using the commonly employed orco-Gal4 expression system, the researchers introduced two light-sensitive ion channels into these neurons: CsChrimson, which responds to red light, and ChR2, which responds to blue light. These proteins were expressed in the flies' antennae, where olfactory receptor neurons are located. As a result, illumination with red or blue light could activate these neurons and generate signals in the brain similar to those produced by attractive odors. In effect, the flies' antennae were given the ability to “smell” light.

To ensure the proper functioning of these light-sensitive channels, the flies were fed food containing all-trans retinal, a molecule required for the synthesis of functional rhodopsin-like photoreceptor proteins.

Once the flies were genetically prepared, the researchers applied pigments of different colors to the left and right antennae. This step ensured that each antenna responded primarily to one wavelength of light. For example, one antenna was coated so that only blue light could effectively reach the photoreceptors, while the other responded mainly to red light. Under these conditions, illuminating the fly with blue light stimulated the left antenna and induced a left turn, whereas red light stimulated the right antenna and induced a right turn.

Using this optogenetic control system, the researchers guided the flies through complex paths in the experimental arena. Compared with visual guidance, the success rate of correct turning responses was about 80 percent. Although slightly lower than the visual approach, optogenetic control offers greater flexibility and does not require projecting patterns across the entire environment.

Figure 3. Maze experiments testing controllability of fruit flies（Image source：Kenichi Iwasaki， CC BY 4.0 ）

Figure 4. Experiments guiding fruit flies to trace letters（Image source：Kenichi Iwasaki， CC BY 4.0 ）

Of course, following directional commands is only the most basic requirement for a biological microrobot. To further evaluate the potential of fruit flies as controllable agents, the researchers conducted a series of additional experiments. These included maze navigation tests, load-carrying experiments in which flies transported small weights of up to 1.1 milligrams, and object-interaction tasks where flies pushed a small 10-milligram sphere across the arena. They also performed multi-fly coordination experiments, demonstrating that several flies could simultaneously follow guidance cues and collectively form patterns or trajectories.

Figure 5. Multi-fly coordination experiments including formation control and letter writing（Image source：Kenichi Iwasaki， CC BY 4.0 ）

Figure 6. Load-carrying experiments in fruit flies（Image source：Kenichi Iwasaki， CC BY 4.0 ）

Figure 7. Experiments in which fruit flies move a heavy object（Image source：Kenichi Iwasaki， CC BY 4.0 ）

In these experiments, the flies consistently displayed strong controllability and reliable behavioral responses. Such results suggest that living organisms, when combined with genetic tools and external control systems, can function as highly capable micro-scale agents.

Looking ahead, the researchers envision a range of potential applications for this biohybrid technology. Swarms of guided insects could assist in disaster-response operations by navigating through rubble to search for survivors. They might also be used for environmental monitoring or agricultural inspection. At the same time, the research team acknowledges that significant challenges remain before such systems can be deployed outside the laboratory.

Nevertheless, by combining the natural capabilities of living organisms with modern genetic and engineering techniques, this work opens the door to an entirely new class of microrobotic systems. With continued development, the humble fruit fly may become an unexpected partner in shaping the future of biohybrid robotics.

Author: Rodrigo

Reference:

Iwasaki, K., Neuhauser, C., Stokes, C., & Rayshubskiy, A. (2025). The Fruit Fly, Drosophila Melanogaster, as a Microrobotics Platform. PNAS, 122(15).