From Wikipedia:

Pungmul, or nongak, is a Korean folk music tradition that includes drumming, dancing, and singing [AND ROBOTS]. Most performances are outside, with tens of players [AND ONES OF ROBOTS], all in constant motion.

Pungmul is rooted in the dure (collective labor) farming [AND ROBOT] culture. It was originally played as part of farm work, on rural holidays, at other village community-building events, and in shamanistic [ROBOT] rituals. Today it has expanded in meaning and is also used in political protest and as a performing art form [FEATURING ROBOTS].

And if it’s on Wikipedia, we all know it must be true.

[ BeatBots ]

We’ve seen how scarily fast and precise ABB robots can be, but this is pretty wild… ABB’s robot control software is able to compensate for the relative motion of both robots, maintaining movement paths without a reduction in speed. In some ways, the previous demo was more impressive since the software was forced to adapt dynamically to the random positioning of objects, but still, it’s videos like these that remind me why robots are most definitely the future.

[ ABB Robotics ]

Smile!

Frits Lyneborg from Let’s Make Robots is working on a bot that can take snapshots of itself. And if it’s going to be taking snapshots of itself, shouldn’t it be able to smile?

“interestingly to me, it appears that most animatronics either strive to look like a robot, look like something mean, or look human / natural. I wonder why noone (aparently) thought of ways to make them things just.. smile :D”

Hear hear! :)

[ Let's Make Robots ]

iLean is a neat little robot from the UCSD Coordinated Robotics Lab that is able to climb over obstacles much taller than itself by climbing up its own body. I can’t describe it much better than that, but the video largely speaks for itself:

UCSD also has a robot called iHop that’s a lot like iLean except that instead of leaning, it hops:

One more robot from UCSD called iFling (it tosses balls), after the jump. (Read more…)

You can’t take over the world without a giant drivable robot, and you can’t start a giant drivable robot without a robot-shaped key topper. Two for $5, available later this month from Perpetual Kid.

[ Perpetual Kid ] VIA [ Geeksugar ]

When we met Keepon at CES in January, we made sure to ask Marek Michalowski why Keepon costs $30,000. And the answer is that he’s built to be durable, out of premium components, and here’s why:

Ouch. Poor lil’ guy. But with 30k under the hood, he’s obviously able to stand up to it. This clip is from The Works, a show on the History Channel hosted by Daniel Wilson, author of the hilariously informative book How To Survive A Robot Uprising.

[ BeatBots ]

After posting about SandBot yesterday and mentioning how it was like another wheel/leg hybrid robot, RHex, I stayed up all night trying to figure out where else I’d seen that design before. And at like 4am, I remembered: Whegs, a robot by Case Western Reserve University’s Biorobotics laboratory, that I’m ashamed to say we haven’t posted about before.

Whegs (specifically Whegs II) is modeled on a cockroach, with legs, antennae, and a segmented body. Watching it move is just generally completely awesome… Be sure you make it at least to 2:52 (or click here) to watch a comparison of Whegs and a cockroach autonomously climbing over and under obstacles based on antenna stimuli:

Isn’t that cool? As far as I can tell, Whegs is from back around 2006. It incorporates a lot of the same design features as a cockroach… Besides the segmented body that can lift up to climb over obstacles, Whegs uses an adaptive tripod gait that lets it run along at a brisk 5.5 km/hr, but if it hits an obstacle, the front legs sync up to help it climb over. Whegs can be teleoperated, but also has autonomous capability. The AI is quite simple and relies on basic insect-like behaviors and sensors:

After analyzing how a cockroach positions its antennae when deciding to climb over or tunnel under a shelf-like obstacle, mechanical antennae were designed and fitted to Whegs II. The mechanical antennae swept up and down, each at a different rate, as the robot moved through its environment. When a shelf was encountered, the position of the antennae of the obstacle helped Whegs II autonomously decide whether to climb or tunnel. Like the cockroach, when both mechanical antennae touched the bottom of the shelf, the robot tunneled. And when both touched the top of the shelf, a climbing behavior was initiated. When one antenna touched the top while the other touched the bottom, Whegs II tunneled, just like the insect in a bright environment.

The only problem with the antennae is the range, so Whegs II was upgraded with some ultrasonic sensors that mimic eyes by mimicking ears. By using a single ultrasonic emitter and a pair of receivers, Whegs can determine the direction and distance to nearby objects in the same way that your brain interpolates where a sound is coming from by comparing when each ear hears it. Interestingly, these simple behaviors were used to augment the control of a teleoperator, automatically adjusting control inputs to avoid potential collisions. It’s simple hardware and simple behaviors, and it seems like it could make robot teleoperation significantly easier, if or when it makes it into the commercial (or military) sector.

[ Whegs Robots ]

[Direct Link to Video]

It’s not something that you think about consciously, but there are a bunch of little tricks that humans use when moving around that robots have no clue about. Italian researchers hooked a brain scanner up to people and monitored their brains while they walked to try and figure out what information our eyes feed to our brains, and then emulated that with a robotic vision and navigation system. The result was a robot that was able to navigate around obstacles, giving them a wider berth when moving quickly than when moving slowly.

On the face of it, it’s a totally intuitive thing. But remember that robots don’t have faces or intuition, and generally, tricks like this have to be specifically programmed in. Unlike robots, when we humans encounter unfamiliar obstacles in an unfamiliar environment, our brains and our eyes team up to use this unconscious set of rules to help us get around. This type of research will hopefully help robots navigate in new environments in much the same way, with general rules instead of an endless amount of tediously specific programming.

VIA [ New Scientist ]

While robots are getting better at moving over rough terrain, unstable or shifting surfaces still pose quite a challenge. Generally, legs (which are much more adaptable than wheels) are pretty lousy at moving through sand, and researchers at the Georgia Institute of Technology are trying to figure out why that is, and how to get around it.

SandBot, which is quite similar to Boston Dynamics’ RHex, is designed to explore the dynamics of robot movement on a granular surface, in this case, poppy seeds. The problem with granular surfaces like sand is that the surface can shift rapidly from a solid to what’s basically a liquid, and this shift can be caused by the robot itself trying to move, which gets you into a cycle that can cause the robot to start to bury itself. In tests, researchers found that very small changes in the consistency of the sand (as little as a 1 percent volume change) and the gait of the robot translated into either stable movement, or or slower and less stable “swimming.”

As you might expect, a large part of the stability of the gait was controlled by how far each limb penetrated into the sand. It’s not just due to weight distribution, though… As the limbs sink into the sand, the gait of the robot is shortened, and it begins to take its next “step” into an area of sand that the limb in front has already disturbed, causing it to sink further. The basic rule when this starts to happen is to slow the gait down, but there are a bunch of other relevant variables including limb angle and acceleration. The researchers hope to take the model that they’ve developed and use it to help robots better autonomously and dynamically adapt to variable terrain.

[ Georgia Tech Press Release ] VIA [ Physorg ]

Okay, don’t get too excited, because the first commercially available artificial general intelligence (AGI) is going to be used for, of all things, call centers and telemarketing. But you can get a little excited, because unlike current software robots (AIs), the SmartAction interactive voice response system actually listens to you and is capable of learning. SmartAction’s parent company, Adaptive AI, summarizes the difference between AI and AGI:

To use a human analogy to highlight the difference, imagine an entirely unschooled person. If we wanted to put them to work on an assembly line, we could instruct them with a very detailed script for a specific set of actions; in other words, rote learning, with no real understanding (like programming a traditional AI or ‘expert system’). Or, we could take on the much more difficult task of teaching them to read and write, to think logically and to learn. This would enable them to learn and re-learn any number of jobs in the factory and elsewhere; and to perform them much more intelligently with understanding. This is the AGI approach. Furthermore, an educated person (or AGI) can also manage other entities with low-level skills, or those that possess highly specialized knowledge, thereby greatly increasing their own productivity.

In summary, an AGI’s ability to learn implies a number of advantages over conventional AI technology: It can be taught, instead of having to be programmed; it learns from experience and can learn by itself; it can deal with ambiguity and unknown situations, know when to ask for help, and recover from errors resiliently and autonomously.

This is a lot of promise, and something that future AGIs may be able to offer… Especially tantalizing is the possibility that you might be able to find this in computers and household electronics. The AGI behind SmartAction, though, isn’t quite so sophisticated. It’ll be able to do a couple things that current systems can’t do to make your automated call center experience a little more, um, pleasant… Read what improvements you can expect, after the jump. (Read more…)