Swarm robotics is a growing field of engineering that focuses on getting many small robots to work together. Much of the simple reasoning each robot uses and the rules for group coordination are modeled after superorganisms like bee colonies or flocks of birds. These animals are able to combine the efforts of hundreds of individuals towards a single task, without getting in anyone’s way or using telepathy. And although robotic technicians keep assuring me that they have noble goals for their work that will help humanity, I can’t help but keep picturing a supervillain with an army of tiny autonomous robots at his or her disposal... Some swarm-y villains include Marvel’s Gah Lak Tus. The original world-devourer Galactus was a giant individual, but this new generation version is a collection of robots that invades planets. There’s another Marvel villain known as The Swarm, but he’s made up of actual bees, not bee-like robots. I had a surprisingly difficult time finding any more comic examples of such villainy with a cursory Google search, but there are lots of robot swarms to be found elsewhere in pop culture. One famous example is Michael Crichton’s novel Prey about self-replicating nanorobots that take over the world in a sinister mass of "grey goo". Many of the reasons robot swarms would be helpful to us are also tools that could be used for evil. Small autonomous robots are invaluable in being able to enter unstable or radioactive buildings to gather information, find trapped people, or bring back samples. On the other hand, the same idea could be used for a nefarious purposes to map out the defenses and terrain of the next place to be invaded. There is an episode of The X-Files about robotic cockroaches of alien origin looking for sources of methane they can harvest should their creators choose to visit. The Cybermen from Doctor Who run a similar gambit in ”Closing Time” when they release silver rat-like Cybermats to scout out power sources and siphon electricity to the ship they’re trying to repair. The Cybermen themselves bear some resemblances to eusocial insects in that they all work together and follow one supreme CyberKing or CyberQueen for the communal success of their species. Cybermen come into being by assimilating real living people into robot suits and altering their brains to be more obedient and without emotions. Another famous race of robot hybrids that enjoy assimilating their foes (as famous as sci-fi TV shows get anyway) are the Borg from Star Trek: The Borg also demonstrates a collective conscious know in sci-fi as a hive mind, and the term, of course, originates from the seemingly telepathic qualities of bee hives. If there are any real nerds reading this blog, they're starting to question my logic. However, we're straying a little off topic since the Borg is a group of human-machine cyborgs, not purely robotic swarms. But one of the tools they use to assimilate others is a swarm of tiny robots! They inject these nanites into the bloodstreams of the person they are converting. The nanites rebuild the person from the inside out with new mechanical parts. This is actually one of the goals of swarm robotics: to create nanorobots that could diagnose and medically treat a patient from the inside. I'm hoping by then though that our technology allows us to create much friendlier-looking robots like the nanites from Mystery Science Theater 3000. We use rules that we know about swarm behavior to design robot protocols. Trying to program 20 robots to follow centralized commands would take too much time and computer power, but you can get the same group results by just telling each robot how close it should be to its neighbors and who to follow. This TED talk goes in depth in showing how they got tiny robocopters to play nicely with one another. It’s long, but make sure you at least watch the great clips at the end showing cool flight formations and a music video! Those robots were quadcopters, the only bioinspiration involved was in creating the cooperative behavior. Many scientists and engineers are working on more biomimetic, flapping flying robots, but insects' wing maneuvers are so complicated that it's not going to happen overnight. Here's some of the progress that has been made on a microaerial vehicle (MAV) at Harvard: Swarm robotics is a new enough field that it is a two-way street between the science and the actual engineering. We’re still figuring out how individual ants or bees make decisions, and one way to explore that is by building robot models that mimic insect behavior. In some recent research, investigators were trying to determine if ants used complicated angular math to figure out what the most efficient route is to take, but it turns out they just followed each other and didn’t need the math at all. Here’s avideo of little robots following the light trails that other robots create (similar to ants laying a chemical trail for its nestmates). By mapping out the efficiency of the communication among superorganisms, engineers can design better traffic patterns to avoid traffic jams and collisions. It’s also an idea that will get us one step closer to self-driving cars, which I personally look forward to. After all this you might start feeling a little concerned about your safety and privacy in an increasingly robot-filled world. Drones fly through the air spying on you, there are robotic flies on the wall watching your every move, and soon even your car may turn on you! Don't worry though, I'm sure the water is still safe. ...Right?
0 Comments
Leave a Reply. |
Alanna DurkinExploring the realm of biologically inspired design one superhero example at a time, with some other natural sciences mixed in. Archives
September 2016
Categories
All
|