It’s not unusual to see a fly effortlessly stick an upside-down landing on a ceiling, but exactly how they pull off the aerial stunt has eluded scientists for decades. Even modern drones cannot compete with the fly’s sophisticated touchdown tricks. Now, a new study offers the most comprehensive exploration of fly landings to date, revealing agile maneuvers that could one day lead to robotic fliers that mimic the insect’s aerobatic abilities.
In a bid to build machines that could imitate insect movements, Bo Cheng, a mechanical engineer at Pennsylvania State University in State College, started by scouring 50 years of scientific literature for studies involving fly landings. He was surprised that such a common occurrence was so underdocumented. Then he realized why: The flies’ lightning-quick moves during landings aren’t easy to observe.
So, Cheng and colleagues used high-speed video to capture and analyze more than 20 bluebottle flies (Calliphora vomitoria), known for their exquisite maneuvers, sticking their inverted landings in a flight chamber. The flies landed in versatile ways. Some stuck the landing by first planting their forelegs on the surface, then swinging their bodies into place, similar to a back flip (see video, above). Other landings looked more like barrel rolls. After taping 18 perfect landings, the team discovered that flies depend primarily on visual cues to make these maneuvers. When a fly sees that it’s about to collide with a ceiling, for example, it must decide in 50 milliseconds how to turn upside down and grab the ceiling with its feet, Cheng and colleagues write today in Science Advances.
But even nimble flies blunder: The study also describes 15 failed landings, which reveal that the insects need to move in a specific range of motion in less than the blink of a human eye to achieve a perfect landing and avoid colliding with the ceiling.
“This is a really interesting new paper,” says Jessica Fox, a biologist studying insect sensory systems using high-speed video at Case Western Reserve University in Cleveland, Ohio, who was not involved in the study. However, the insects’ landings may have been influenced by the experimental setup, she adds. Flies were stimulated with mechanical vibrations to get them to take off—in an area the size of a small box. If they had more room, and if they weren’t frightened into taking off, they may have chosen easier landings, she says. The study shows “what flies can do at their limits, when they are flying fastest and need to make decisions in the shortest amount of time.”
The flies “are just a starting point” in exploring how they and other flying insects—from mosquitoes to bees—control their complicated maneuvers, says co-author Sanjay Sane, an integrative biologist at the National Centre for Biological Sciences in Bengaluru, India. More studies of this kind will help scientists start to pinpoint the most important shared flight maneuvers across species, he says. And once scientists know more about the processes that control fly landings, Cheng adds, they may discover how to create robots that mimic the flies’ barrel rolls and other fly feats. “Just like children mimic their parents, we can let the fly teach a robot.”