Back in July, we wrote about how UPenn’s GRASP Lab had taught their quadrotors to work together to grasp and move things. The next step, it seems, is teaching the quadrotors to work together to grasp and move things and actually build buildings. The video above shows a team of quadrotors cooperating to construct the framework of a (rather small) building. The building’s structure is held together with magnets, and the quadrotors are able to verify that the alignment is correct by attempting to wiggle the structural components around, which is pretty cool.

It’s fun to speculate about how this technology might grow out of the lab into the real world… To build actual buldings, you’d either need much bigger quadrotors (which is possible), lots of small quadrotors cooperating in big pieces (also possible), or buildings built out of much smaller components (which might be the way to go). The quadrotors probably wouldn’t be able to do all the work, but they have the potential to make construction projects significantly more efficient.

[ GRASP ]

Wow. Unlike that blind robot juggler, this quadrotor is juggling one ball the hard way, with a tennis racket duct-taped to itself. Someone just needs to toss a couple more balls in there and see what happens…

[ ETH Zurich Flying Arena ]

Kinect’s 3D sensor is so cheap and effective that it’s getting bolted onto any robot that moves, and quadrotors are just the latest victims. UC Berkeley’s quadrotor is using the Kinect for autonomous flight and dynamic obstacle avoidance, and as long as you don’t come at it from behind, it works great. The nice thing about using Kinect like this is that it translates into a SLAM system, where the robot can fly around and make a 3D map of a space using the same data that it’s relying on to keep from crashing in to stuff.

[ UC Berkeley Hybrid Systems Lab ] VIA [ Trossen ]

While you were stuffing your face on turkey (or forest loaf, whatever that is) and vegging out, the guys over at Utah Aerials (remember their robot ghost?)were busy constructing a flying turkey robot that drops pumpkin pie bombs. In other words, my fantasy animal.

Somehow, adding a turkey costume to a quadrotor doesn’t seem to effect its ability to achieve stable flight. Adding a pumpkin pie definitely does, but the bot manages to stay airborne long enough to achieve its mission. Kinda. Anyway, a pie make it from the sky to the ground, and that’s the important thing.

PIE!

[ Utah Aerials ]

ETH Zürich’s dancing quadrotors are back (with a friend), dancing to a new arrangement entitled ‘Rise Up.’ You know, it would be pretty cool if they could somehow allow people to submit music and choreography for the robots to act out… I have no idea how that would work, I’m just saying, it would be cool.

Pretty soon, that won’t even be necessary, though… The robots will do all their own choreography, creating a dance on the fly for whatever music you decide to throw at them. And then, the world! Bwa ha ha!

[ Music in Motion ]

Those crazily precise autonomous quadrotors from the GRASP Lab at UPenn have learned a few new tricks, presented in this latest video from their YouTube channel. First, the robots recover from being thrown into the air, which is a skill that you definitely want to have in a small surveillance robot. Most impressively, however, they’re able to fly themselves through hoops tossed into the air by a human. This type of adaptability should be at least somewhat familiar from other high-speed systems we’ve posted about, and while it seems pretty incredible to us (and it IS incredible) that robots can make decisions and take actions in mere fractions of a second, to paraphrase Commander Data in one of those lousy Star Trek movies, to a robot, fractions of a second are nearly an eternity.

[ GRASP @ UPenn ]

If you were Google, what would you do with an autonomous, camera-equipped quadrotor? Google, who is Google, must have a pretty good idea (or ideas), since they’ve bought their own Microdrone, which (last time we checked) costs somewhere in the neighborhood of $60k. According to the German publication Wirtschaftswoche, Microdrones GmbH says that their UAVs “are superbly suited to deliver more up-to-date recordings for mapping service Google Earth,” although that may just be a potential use that they’re suggesting as opposed to what Google is actually planning. Still, it’s kinda fun to think about… Like, it might be possible for Google to deliver a live version of Google Earth, at least over a small area, using a Microdrone (or a network of microdrones!) and a georeferenced live video network link.

Or they could be doing something entirely different.

The point is, I guess, that the Microdrone offers a way to collect data that is pretty much limited only by the imagination of the user, and Google has proven to be pretty creative when it comes to collecting and deploying imagery. So now that they have one, the sky’s the limit. Or rather, it isn’t.

[ Microdrones GmbH ] VIA [ Blogoscoped ]

As soon as they figure out how to get quadrotors to operate outside of a motion capture environment as well (or nearly as well) as they do inside, surveillance could very well be their first major application, the appeal being that in addition to being highly mobile, quadrotors can set themselves down and provide static surveillance without wasting energy staying aloft. This isn’t a capability that most currently fielded UAV systems possess, despite some extremely cool research into fixed-wing perching.

Although it may not be the greatest thing, from a moral standpoint, that militaryish applications are driving robot development, there’s a pronounced trickle-down effect as eventually all that military hardware (or components thereof) becomes available to normal people who just want to do cool stuff with robots.

[ GRASP @ UPenn ]

Quadrotors: they can do everything you can do, only better. Way better. Markus Hehn and Raffaello D’Andrea at the Institute of Dynamic System and Control at ETH Zurich have gone from dancing quadrotors to a quadrotor that can actually balance an inverted pendulum in flight. Impressive, yes. Useful? Maybe… If you have quadrotors that are trying to lift something unbalanced, for example, being able to compensate for that could certainly come in handy. Of course, all of this comes with the caveat that you need a pretty swanky external vision system to get it all working properly, but maybe if you mount that on a couple other quadrotors, you’d be good to go.

[ IDSC FMA ]

We’ve posted a lot around here about how swarm robotics is potentially really, really awesome, but besides kidnapping children we haven’t seen that many relevant examples of practical swarm robotics. The above video, from the GRASP Lab at UPenn, shows a group of autonomous quadrotors (these bad boys) teaming up to lift heavy and off-balance loads.

Each quadrotor weighs 500 grams and can deliver some 1250 grams of thrust, making their individual payload capacity somewhere around half a kilogram. This means that a couple together could lift a kilogram, and you can do the math on from there, but there are lots of reasons why you might want a bunch of extra robots cooperating on the lift, which gets back to why swarm robotics has so much potential in the first place. For example, having extra bots protects against mechanical failure of an individual bot. It also protects against complications like wind. Or maybe whatever you’re lifting has a long distance to go or needs to be in the air for a while, and the bots can switch off to go recharge themselves.

It’s interesting to compare these cooperative quadrotors with that distributed flight array from ETH Zurich that we wrote about last month. It’s a different approach, certainly, but the premise is similar, and it’ll be lots of fun to see how each of these projects evolves.

[ GRASP ]