4 ms·
Harvard researchers create insect-sized robot that can both fly and swim
- tomcam 11y agoThis version seems tethered to a power source and not self-contained with a battery.
- dang 11y agoUrl changed from http://bgr.com/2015/10/29/harvard-robobee-fly-swim-insect/ http://bgr.com/2015/10/29/harvard-robobee-fly-swim-insect/, which points to this.
- hising 11y agoDid not read, only watched the video. It seems that it can go from flying to swimming, but can it go from swimming to flying?
- seangrant 11y ago"While this RoboBee can move seamlessly from air to water, it cannot yet transition from water to air because it can’t generate enough lift without snapping one of its wings. Solving that design challenge is the next phase of the research, according to Chen."
- jawarner 11y agoI'm impressed by the simplicity of this design. How much control can you get out of the robot over that power cable? Is timing and computation done on the robot itself, or on the other side of the wire?
- Nzen 11y ago2012 2min vid about inspiration, construction, & control : https://www.youtube.com/watch?v=K6PpCTMEAVM https://www.youtube.com/watch?v=K6PpCTMEAVM . There is a processor onboard that controls the flapping.
- petersouth 11y agosweet, we need to use these things to pollinate when the bees die out.
- msravi 11y agoWhy are all the robotic insects built with flapping wings, as opposed to propellers or rotors that drones, airplanes, boats, and helicopters are built with? Is there something inherently different at that scale that requires flapping wings?
- nether 11y agoYes, the tendency to shed vortices is much greater at small scales. A rapid flapping motion is basically a vortex generator and more efficient than a wing (rotating or fixed) which relies more on steady state, laminar flow. See https://en.wikipedia.org/wiki/Reynolds_number https://en.wikipedia.org/wiki/Reynolds_number for more explanation.
- mtrpcic 11y agoAdditionally, the energy required (relevant to total mass) to generate lift is a lot smaller at this scale thanks to the Square-cube law[0], and it's a lot more efficient to flap a wing on an as-needed basis than to run a rotor at a continual and steady rotation. [0] https://en.wikipedia.org/wiki/Square-cube_law https://en.wikipedia.org/wiki/Square-cube_law
- bordercases 11y agoI don't really like the "flying submarine" angle, as historically interest-piquing it might be. A lot of the success seems like it might come from the scale that the machine works at, i.e. there are non-linear cost functions involved when it comes to changing the scale at which the aerodynamics are treated at. But yeah, super cool.