BotJunkie is merging with Automaton to form the best robotics blog on the Net! Please continue following our stories at our new home and update your RSS reader with our new feed. See you there!
Writing by Evan Ackerman on Wednesday, 4 of November , 2009 at 4:17 am
“David Hanson’s robot faces look and act like yours: They recognize and respond to emotion, and make expressions of their own. Here, an “emotional” live demo of the Einstein robot offers a peek at a future where robots truly mimic humans.”
Writing by Evan Ackerman on Wednesday, 4 of November , 2009 at 4:03 am
Nobody seems to know what exactly this is or if anything ever came of it, although the consensus is that it’s probably a dead concept. The two noses are kinda neat, though. I’m not sure what the aerodynamic implications are, but they certainly add a bunch of payload space. I also like the idea of an optionally-manned UCAV… For that matter, why do we need dedicated UAVs at all? Why not just retrofit existing combat aircraft with optional unmanned systems for added flexibility and, for that matter, safety? The answer, of course, is that you can’t make an optionally manned UCAV that looks anywhere near as badass as this.
Writing by Evan Ackerman on Tuesday, 3 of November , 2009 at 1:02 am
Late nineties is fairly prehistoric when it comes to robots with fast reaction times. I mean, I was in high school in the late nineties! But while I was busy being a miserably outcast geek, Toshiba was demonstrating a robot that could (sort of) play volleyball. According to a Toshiba press release from back then, the robot was nicknamed “the beach-volleyball-playing robot” (seriously) and had multiple talents, including voice recognition and individual facial identification.
Toshiba’s press release also included a little bit of development background, where they discussed the present and future of robotics from a perspective that is now 12 years old:
“The 21st century will see robots not only programmed to act in certain ways, but able to respond to their environment, including working with and around people. Such robots will perform sophisticated tasks, including taking care of patients or executing complicated manufacturing processes, on the basis of simple instructions.
Currently, robots, and the objects they work on, must be precisely positioned, and robots require detailed programming to control their movements and actions. All this severely limits repertoire and usefulness.
Toshiba’s new robot demonstrates solution to these problems that will bring greater flexibility and capabilities to robots.”
Robots do still require fairly detailed programming to control their movements and actions, even as they’re getting much better at learning how to learn and make decisions for themselves. They’re also much better at responding to their environment, but working with and around people still hasn’t really come to pass. Toshiba seems to be working as hard as anyone to make all of this stuff work, though, with their series of (still mostly demonstrative) household assistance robots.
Writing by Evan Ackerman on Tuesday, 3 of November , 2009 at 12:22 am
I love the inhuman speed of pick and place robots. We’ve seen a lot of incredible examples from ABB, but Adept Technology has just broken the “300 cycle per minute barrier” with their Quattro robot. The secret, they say, is that using four arms instead of three enables higher speeds and faster accelerations. If you want to see what 300 cycles per second looks like, pay attention at about 1:30 in the video and remind yourself that you’re not watching it on super fast forward.
Writing by Evan Ackerman on Tuesday, 3 of November , 2009 at 12:20 am
Robots are slowly learning to take advantage of nonverbal cues, which are an integral part of natural human communication. Things like eyebrow movement can be used to convey information about emotional states, but subtle nonverbal cues can also provide more pointed information… Think about how much can be said with a quick glance. Professor Bilge Mutlu from the University of Wisconsin is trying to teach robots to communicate information with glances, by seeing how well effectively they can “leak” the answer to an object guessing game played with a human.
As you can see from the video, two robots took place in this experiment: Robovie R-2, and the creepy as f**k Geminoid. Interestingly, participants in the experiment responded better to cues leaked by Geminoid, while at the same time, the didn’t realize that they were doing so, implying that however deep in the Uncanny Valley Geminoid may reside, in some cases, having a robot with distinctly humanoid features really does make a difference.
There is one itsy bitsy teeny tiny little caveat at the end of the paper: “We found that the leakage cue affected the performance of only pet owners and not others, which might suggest that pet owners become, through their interaction with their pets, more sensitive to nonverbal behavior.” Er, okay, well, I guess it was good for them to check for that kind of thing, but that kinda makes the effects of shifty-eyed robots a lot less impressive.
Writing by Evan Ackerman on Monday, 2 of November , 2009 at 1:57 am
The video that includes Boston Dynamics’ BigDog getting kicked and slipping on ice is nothing new, and it’s still one of the best robot videos I’ve ever seen. The vid above includes that stuff, but also some footage that I don’t think we’ve seen before of BigDog descending slopes and getting its foot caught in a concrete block. It’s all very impressive, but even more importantly, it’s damn funny stuff.
Writing by Evan Ackerman on Monday, 2 of November , 2009 at 1:45 am
Some robots are, granted, awesome dancers, but most are just too robot-y. If you think about it, it’s a lot like early computer animation… It’s hard to convincingly mimic all of the subtle movements that humans use when dancing (or doing pretty much anything else). This is exactly why the motion capture suit was invented, and it’s a wonder that nobody’s used them to program a robot before. Or maybe they have, but this is the first video I’ve seen of it in action. And it really does seem to make a difference to the robots’ moves… I’d venture to call them significantly more lifelike.
The research team from South Korea that’s developing this technique is focusing specifically on hip movements, because as you know, your hips are what keep the rest of your body from falling over when you’re bustin’ those crazy dance moves, yo. It still appears to be a fairly restrained system, but if it does for robots what motion capture did for CGI, Dancing With The Stars could get a lot more interesting.
That’s a show, right? I wouldn’t know, ’cause I of course wouldn’t watch anything of the sort.