Here at Xconomy we usually focus on technologies already hitting the marketplace rather than laboratory-stage investigations. But last week we got wind of a project that’s so cool we just had to write about it: an effort to build tiny robot planes with flexible structures and built-in reflexes that would allow them to ricochet off walls or objects unharmed and recover their flight paths, the same way house flies bounce off windows.
They’re called “biomimetic micro air vehicles” or MAVs, and they’re the subject of a study at Harvard University and Andover, MA-based Physical Sciences Inc. (PSI) that just won funding from the Air Force Office of Scientific Research. PSI does contract R&D work on aerospace, energy, environmental, manufacturing, and medical technologies, and the eventual goal of the MAV project would be to create new kinds of indoor reconnaissance or surveillance craft, carrying tiny cameras, chemical sensors, and the like.
I got the lowdown on the project last week from Tom Vaneck, Physical Sciences’ vice president of space technologies and manager of disruptive technologies—of which the fly-like MAVs would certainly be one. Last time I talked with Vaneck, he was the head of Aurora Flight Sciences‘ Cambridge, MA-based R&D lab; he says he left the aerospace contractor for PSI earlier this year because “I am from a technology sense a little bit ADD,” and that at PSI, “I’m able to have my fingers in many different technology pies.”
Vaneck says there are two fundamental things to think about when a flying object hits a non-moving object. “One, how do you design a structure that can withstand the impact—because if the structure breaks or you are no longer able to generate lift or thrust, you’re done. Two, how do you recover without having to do a lot of environmental sensing or sophisticated computation—you need a method that’s almost instinctual, that automatically reorients the vehicle so that it can fly again.”
Well, those are both problems that evolution—“which has had a long, long time and an infinite budget,” in Vaneck’s words—has already solved. “When a fly hits a window it doesn’t fall down; it goes on to do it a hundred more times,” Vaneck notes.
So PSI is putting the $100,000, Phase 1 Air Force grant into a joint study with Robert Wood, a builder of biologically inspired robots at the Harvard Microrobotics Laboratory; he’s the creator of the world’s first artificial insect wing with enough lift to get itself off the ground. Together, researchers from PSI and Wood’s lab will study how houseflies and dragonflies recover from collisions, and think about materials such as carbon-fiber composites and a control system that could be used to duplicate the behavior.
The control system may be the harder problem to solve, since it will actually require the engineers to abandon most of the traditional principles of controlled flight. “If you think about a fly, its wing-beating motion if almost a resonant condition,” says Vaneck. “The fly is not continually thinking about moving its wings up and down. Its nervous system just creates a stimulus such that the wings flap, and through their design they generate lift. Now, after a collision, maybe one wing is generating more lift than the other; the control simply needs to go from one resonant condition to another. We think we can manage that without a computer. You just need a mechanism with several ‘set points’ that it can switch between.”
Vaneck hopes the 9-month, Phase 1 grant will give the researchers enough time to build a simple prototype and “understand enough of how nature does this to map this over to a man-made system.” Then PSI will apply for a larger, longer Phase 2 grant that would lead to the construction of a working, remote-controlled MAV. “If we can make this work, it will fundamentally change the way people operate small unmanned aircraft,” he says.
I couldn’t resist asking Vaneck whether he ever worries that his work might result in the kinds of creepy insectoid probes often shown in movies like The Matrix or Minority Report. “You can’t help but think about that,” he answers. “Any technology can be morphed into something that is unintended. And there is this visceral reaction—if a movie gadget has to be evil and nasty, it is probably going to look like an insect. But the flip side of that is that insects are very robust systems.”
Vaneck also points out that a robot plane that looked and behaved like an insect might have the advantage of stealth. “If it’s truly bouncing around like an insect,” he says, “maybe it gets overlooked, because it kind of looks like something from the natural world.”
Of course, there’s a flip side to that as well: One good swing of the flyswatter could destroy a very expensive gadget.
