Video: To Enable the Robo-Insects of the Future, Researchers Capture Butterfly Flight at 3,000 FPS

Neither bio-mimicking robots nor insect-analog micro aerial vehicles (MAVs) are new concepts. But where super high-speed video capture, competitive figure skating, and lepidopterology collide, there PopSci shall be. Today, that means turning our attention to Johns Hopkins University, where engineering undergrad Tiras Lin is potentially shaking up insect-like aerial robot design.

For a proper visual explanation of what Lin and colleagues are up to, the video below is thorough. But briefly: DARPA and other defense- and public safety-related research entities in both the public and private sectors have been exploring the idea of tiny, sensor-capable drones the size of aircraft for years now (regular readers have read about many of them on this site). But actually recreating mechanically the kind of flight achieved by insects is notoriously difficult.

Users want MAVs they can pilot through complex urban environments, where the variables–obstacles, tight spaces, variable air pressure and wind speeds–make it difficult to fly. Wishing to tap real insects’ tricks, mechanical engineering junior Lin crossed over into entomology, using a high-speed camera array to capture butterfly flight–wing flapping, body deformation, and anything else that contributes to mass distribution as a butterfly moves through the air.

His high-speed rig allowed him to capture 3,000 one-megapixel images per second (compare that to 24 frames per second for standard video), allowing him to dissect the forces at play as the butterflies flapped their wings roughly 25 times per second. Using three cameras, he was able to capture three dimensional data and analyze the way butterflies’ bodies and wings move in sync to provide them with their maneuverability.

His findings? Butterflies appear to be very much like figure skaters, using angular momentum as they flap their wings to modify their moments of inertia (this is akin to figure skaters tucking their arms to increase the speed of their spins and outstretching them to slow their rotation–essentially manipulating their rotation by redistributing mass). This refutes earlier assumptions that a butterfly’s wings don’t have enough mass relative to their bodies to be a factor in maneuverability. And it just might change the way roboticists approach robo-insect design going forward.

Much more via the video below.

[JHU]