SensorFly is a prototype for what will eventually be an entire swarm of autonomous mini helicopters that can navigate, talk to each other, and carry sensors (including a miniature camera and microphone) to perform tasks like locating survivors after a disaster. Each robot weighs only 29 grams, but they’re remarkably resilient, able to handle collisions with walls as well as swats from vicious racket-swinging grad students.

[ SensorFly ]

If you’re going to have a testbed robot wandering around investigating autonomous operations in office environments, you might as well have it do something useful. Like passing out snacks. Seriously… Greatest. Use. For a robot. Ever. SnackBot works at Carnegie Mellon University, which is way way too far away from me right now, ’cause I’m hungry.

The research will allow the robot to navigate through congested areas in a socially acceptable fashion, detect individual people moving near the robot, recognize when someone that the robot knows approaches it, and autonomously learn to recognize new objects.

Oh, and of course, SnackBot will support “research on snack services drawn from behavioral economics.” I can solve that for you right now: snacks are tasty, people like snacks, and a robot that brings tasty snacks to people will be well liked by all.

[ SnackBot.org ]

If you liked the video of the Carnegie Mellon snake robots we posted last week, here’s a follow-up showing a bunch of different movement tests, complete with a headache-inducing techno soundtrack. And if you still want more, seek help, and then watch the original video on these robots that we posted in March of last year.

[ ModSnake ]

One of the coolest things about snake inspired robots is how they’re in the process of transcending their biological models, as this video from CMU’s biorobotics lab shows. I’m fairly certain that nowhere in nature do snakes leverage their design to, for example, climb up vertical poles with a twisting motion. This may be because real snakes are able to execute a much finer degree of control over their muscles so as to render such a motion unnecessary, but on the other hand, perhaps twisting is actually a primitive emulation of a sidewinder motion. Either way, it’s good to see some wholesome research snakebots again, after the last ones we wrote about, which were decidedly more violent.

[ Modular Snake Robots ]

When Americans finally make it back to the moon, we’re gonna need a base. It’s going to have to have (among other things) laboratories, a powerplant, housing, and a landing pad for resupply craft. CMU’s William “Red” Whittaker (whom you may especially remember from here), explains why robots may have a little bit of work to do before we start taking up residence: “for efficient cargo transfer, the landing site needs to be close to the outpost’s crew quarters and laboratories. Each rocket landing and takeoff, however, will accelerate lunar grit outwards from the pad. With no atmosphere to slow it down, the dry soil would sandblast the outpost.”

That leaves two options: build a big hill to keep dust contained, or pave the landing area to keep dust from getting kicked up in the first place. Either one of these options involves a lot of tedious work… For example, building a protective hill (8.5 feet tall in a 160 foot semi-circle around the landing area) would involve something like 2.6 million pounds of lunar soil. That sounds like a lot, but but two 330 pound robotic rovers the size of riding lawnmowers could get the job done in just six months. Similarly, a couple robots could sift soil in the area around the pad to collect rocks suitable for paving. These particular robots have been developed by Astrobotic Technology, who plan to send robots to the moon and then license the scientific and engineering data that they collect to space agencies and other aerospace companies.

Figuring out which dust reduction method would be best will most likely involve sending a scouting mission (robotic scouting mission, I bet) to the as yet undecided site for the lunar outpost, which is expected to begin operations by 2020.

[ CMU Press Release ]
[ Astrobotic MoonDigger Presentation (PDF) ]

It’s a sobering fact that most insects have more brainpower, or at least more basic life skills, than even the most modern robots. Small flying insects, for example, use a process called “optical flow” to navigate by translating changes in luminance into the relative speed and proximity of objects around them. The little helicopter in the above video is using some functions developed from these principles of insect navigation by researchers at the University of Maryland to find its way down a corridor.

Why is this exciting? Well, autonomous robot navigation in unfamiliar environments has always been a fairly intensive thing. I mean, have a look at all the hardware Boss has bolted on just to find its way around a 2-D environment. High resolution cameras, lasers, sonar, radar, lidar, GPS… Not to mention a SUVload of computers. This is not especially practical for micro-UAVs. But flying insects can navigate complex environments using this optical flow technique based on little more than the equivalent of low-res (I’d think?) cameras, which is pretty much exactly what you want in a micro-UAV nav system. And, you get the bonus of a freakishly efficient ability to dodge angry hand swipes.

[ UM Autonomous Vehicle Lab ]

If there’s one thing that’s hard to find here on Earth, it’s the moon. I guess it’s pretty tough to find a good analog for the lunar surface, ’cause NASA has decided to test out a potential lunar rover called Scarab in Hawaii, of all places. Yeah, the volcanic landscape there is kinda like the lunar surface, but the test site 9,000 feet up Mauna Kea also has snow, fog, wind, rain, and 40 degree daytime temperatures, most of which are decidedly un-moony. But you take what you can get, I suppose.

Scarab has been designed by Carnegie Mellon to drill into the lunar surface in search of useful stuff like hydrogen, oxygen, and a convenient mixture of the two called water. It’s able to navigate autonomously around the dark side of the moon, without relying on constant contact with Earth or constant power from the sun. The radioactive isotope power generator on Scarab is good for ten years (that’s ten years), the trade-off being that you can’t get the power out of the generator very fast… It’s like the energizer bunny on downers. The generator outputs 100 watts of power (i.e. a weak lightbulb’s worth), which has to keep the rover’s systems going while it’s either moving around, or drilling. Consequently, the rover just does stuff very, very slowly. But that’s cool, there’s no rush. It’s got ten years, remember?

Video after the jump. (Read more…)

As big of a fan as I am of robot festivals, it generally takes booze, sex, or violence to motivate me to travel to one. And Pittsburgh is not exactly at the top of my list. But Robot 250, sponsored by Carnegie Mellon University and the City of Pittsburgh, actually looks pretty cool. The festival takes place all around the Pittsburgh and includes lots of robotic art installations, many of which are interactive. The picture above is of Mower, a robotic sheep that walks around (on, um, six legs) and trims grass with a small blade inside its mouth. Mower’s head contains obstacle avoidance sensors and a GPS locator for fully autonomous grass mitigation operations. It was designed by Osman Khan, a visiting professor at CMU.

Okay, so it may not be quite as efficient as one of these (especially since it seems to be trailing an extension cord?), but it’s way cuter and it goes “baaaaa!” Yay! Check out the Robot 250 website for pics and movies of some of the other installations.

[ Robot 250 ] VIA [ Crave ]

Most of the time, it’s a bit frustrating to write about nanobots. We have to talk about them in the context of teeny tiny little scales that you can’t really identify with… 250 microns long? What does that even MEAN? CMU’s nanobots are barely, just barely, large enough to identify with. They’re about the size of a grain of sand, and you can see one in action alongside a penny:

So yeah, that’s still pretty nano, but I at least feel like I have some conception of the bot’s actual size rather than having to rely on an abstract measurement. These little guys are controlled via an external magnetic field, and by rocking back and forth very quickly, they can reach a top speed of 13 mm (60 body lengths) per second. They’re capable of movement on surfaces that are generally smooth and non-stick, and will work equally well underwater. Although most people would call these things micro-robots, all they really are are solid little magnets being pushed around by other magnets… But that’s okay, we’ll let it slide, ’cause they’re little and cute and all.

[ CMU Nanorobotics Lab ] VIA [ Communist Robot ]

Since 1994, Carnegie Mellon has been running the Mobot Slalom competition, where home built autonomous robots follow a white line through a sequence of gates as fast as they can. To complicate matters, the lines diverge and converge at several points, requiring the bots to have some sort of built-in reasoning. They also have to deal with uneven terrain, inclement weather, and have to fit through the 18 inch square gates. There’s this rule about the design of the bots:

“Animals (except primates) may be used to assist with vehicle control as long as such use is humane (does not harm the animal in any way) and conforms with applicable University regulations.”

Intriguing… Too bad the website doesn’t elaborate.

[ Mobot Slalom ] VIA [ Communist Robot ]