This is MTRAN2, a modular robot who’s a few years old. I’m posting this because the video is neat, but also because I wanted to introduce you to MTRAN2 before you meet his younger sibling, MTRAN3. Old or not, MTRAN2 is a pretty clever little guy. Each one of his modules includes a battery pack, two motors, connectors, a tiny computer, a wireless receiver, and some sensors. The above video shows MTRAN2 in a walking configuration, encountering a slope that might make him unstable. MTRAN2 then reconfigures himself into a snake, which is a slower but far more stable way to move. The behavior is entirely autonomous, and MTRAN2 is able to reconfigure into a variety of different forms, including asymmetrical walkers and even a wheel. It’s a completely different take on flexible locomotion from, say, the RHex, and although it trades off speed and simplicity, there’s theoretically nothing the MTRAN2 couldn’t reconfigure itself to do.

[ MTRAN2 ]

This video of a portrait drawing robot comes from Sylvain Calinon, who also helped design the active learning system I posted about yesterday.

“When the robot recognizes a face in its field of view, it first grabs a snapshot of this face and extracts the relevant features and contours characterizing it. The robot then grabs a pen and start drawing the user’s portrait, starting from the rough contours and adding details iteratively.”

If you could set one of these things up at some kind of fair, you’d make a killing. I mean, who wants a silly caricature when you could get your portrait drawn in robotic detail?

[ Sylvain Calinon’s Homepage ]

Robots are notoriously lousy at coping with unfamiliar situations. One of the major obstacles to having robots assist us at home is that they’re going to need to be flexible, and home users aren’t going to want to sit down and write a new servo control subroutine every time their robot needs to perform a task for which it has not been preprogrammed. The robot in the video above is able to learn via observation and replication, which is how we humans figure things out. The HOAP-3 robot, made by Fujitsu, has been programmed by Sylvain Calinon and other researchers at the Swiss Federal Institute of Technology in Lausanne to watch how a human performs an action, and then duplicate that action. If the robot doesn’t get it right, it can learn on the fly as a human corrects its motions, which (if you think about it) is pretty much how you were taught to write, throw a baseball, ride a bike, and so forth. I’d much rather have a dumb robot with the ability to learn than a smart robot that I can’t teach new things to, especially if the teaching process is no more complicated than a game of Simon says. Er, make that Evan says.

[ Sylvain Calinon’s Homepage ] VIA [ New Scientist ]

Yesterday’s robot video was kinda creepy, but this one is creepy enough that I’m going to have to label it as not safe for people who are medically squeamish, because it’s graphic and totally weird. Please, take the warning seriously (think, horror movie and potential nightmares), but it helps to keep in mind that it is 100% fiction, albeit very well done.

[ Microbia ] VIA [ Suicide Bots ]

No, this is not an actual documentary. But it could be, man, it could be.

[ Microbia ] VIA [ Suicide Bots ]

Kiva Systems designs robots that work in distribution centers, moving and sorting materials and components. Looks like they’re not bad dancers, either…

Happy holidays to you and your robot masters friends from all of us here at BotJunkie.

[ Kiva Systems ] VIA [ Gizmodo ]

At a medical devices conference in Boston last month, iRobot apparently had on display a prototype of an “elder care” robot named CiCi. Although iRobot has refused to comment on CiCi, they’ve at least confirmed that it exists as a concept. The unit on display in Boston “stood about the size of a desktop computer and was the shape of a human upper body without arms.” Sources at MIT say that “the robot has audio features that can be used to check on elderly patients at home and would be linked to a network to alert clinicians elsewhere of a medical problem.” Other than that, not much information is available… It sounds like a much more sophisticated and interactive version of the iRobot ConnectR. Although CiCi could be just a proof of concept design, it’s also possible that iRobot is waiting to see what the reception is like for ConnectR before they commit to producing an elder care robot. It’s also worth noting that elder care robots are undergoing heavy development in Korea, and elder care was forecast to be a major part of the growing robotics market at RoboDevelopment.

[ Mass High Tech ]

Well, I’m disappointed. It looks like a drone dogfight, which would be the only thing more robotic than the acting in Top Gun, is not going to be feasible for at least the next 25 years, according to the Pentagon… Dyke Weatherington, deputy director for the Pentagon’s Unmanned Aircraft Systems (UAS) Task Force, states that “there’s really no way that a system that’s remotely controlled can effectively operate in an offensive or defensive air combat requirement. The requirement of that is a fully autonomous system, we don’t have that level of autonomy yet and frankly in the roadmap that will take many years to get to.” He’s got a point there, sort of, but I also think that the speed of development of aerial combat drones is being vastly underestimated, especially when it comes to the AI. I’m not saying I think it’ll happen in the next year, but 25 years is a very, very long time when it comes to computers. I mean, 25 years ago was what, 1982? Think about the amount of change that has happened between then and now, and project it forward. We’re rapidly approaching the point where hardware is not going to be the limiting factor, and where computers will be able to operate (and interoperate) fast enough to make faster and arguably better decisions in an active combat environment than humans are capable of. The hardest part, in my opinion, is not going to be developing the autonomy: it’s going to be getting past all of those prickly ethical issues that come up when you have robots deciding whether or not to shoot at people.

BTW, the drone matchup above is the Boeing X-45 versus the MiG UCAV.

VIA [ Danger Room ]

Building your own robot doesn’t have to be complicated, as long as you’re not building a complicated robot. Sure, you might have to learn how to solder or something, but building a bot is a great excuse to get comfortable with the skill. Vibrobots are about as simple as it gets: these robots use unbalanced motors (the things that make your cellphone vibrate) to power them along. Evil Mad Scientist Laboratories has a neat tutorial on how to make your own vibrobot with little more than a motor, a battery, a toothbrush, and some tape. Before you scoff at this little bot, at least take the time to watch the video below (or skip to the end of it, anyway) and see just how well it works… I guarantee you will be impressed:

If you like the idea but would rather fork over $60 for a pre-assembled, slightly fancier vibrobot, you can buy one at Etsy that runs on solar power rather than a battery. Check it out, after the jump. (Read more…)

One of the cool things about working in an emerging high tech industry (which is what robotics undoubtedly is) is that you don’t have a be a big corporation with a correspondingly big R&D budget in order to create a useful product. iRobot PackBots are the standard in heavy duty, reconfigurable small field robots, but they cost tens of thousands of dollars or more, which is usually out of the price range of anyone except the military. Javier Rodriguez Molina, a graduate student at UCSD, has designed a robot called Gizmo that offers a lot of high-end field usability and flexibility in a package that costs less than $1000.

Gizmo’s primary feature is the ability to create a large (200m diameter) wireless network bubble, sort of like a LANdroid. Each Gizmo can link with other Gizmos to create an exponentially larger network. This allows Gizmos to communicate with each other and with controllers via computers, cellphones, remote controls, or whatever you want. The Gizmo platform itself is scalable and modular, and allows for mounting and control of cameras, sensors, arms, or (again) whatever you feel like sticking on there. While Gizmo currently runs on wheels, he can be fitted with treads or even, eventually, wings. Says Rodriguez, “people see Gizmo and immediately think of a new idea for what it can do. I’m sure it has important uses that we haven’t even thought of yet.” And that’s the whole point: making Gizmo flexible and inexpensive allows users to spend their money and brain power where it matters… Getting the robot to do exactly what they need it to do.

[ Calit2 ] VIA [ LiveScience ]