This robot is named Cody, and he’s from the Georgia Tech’s Healthcare Robotics Lab. I don’t know why the robot is called Cody… It kinda seems like it should stand for something. You know, as in, C.O.D.Y. Anybody got anything? No? Okay then.

Anyway.

Cody is sporting one of those Segway RMP bases with the nifty Mecanum-type omnidirectional wheels. The vertical axis is taken care of by a linear actuator, and Cody has a pair of 7 DOF arms that are able to open things like doors, cabinets, and drawers.

While the capabilities of the arms are certainly impressive, I especially like the intuitive way that Cody can be led around by hand:

It’s way easier than using a gamepad controller, and much more precise than vocal control, minimizing the operator learning curve without sacrificing capabilities. Most of Cody’s code is compatible with ROS, which (ideally) makes it simple for other mobile robot developers to use the clever bits of code that were written to make Cody so intuitive to control.

[ Georgia Tech Healthcare Robotics Lab ] VIA [ Hizook ]

Thanks Travis!

This little helicopter is able to understand you when you tell it what to do. No pushing buttons, no using special commands, you just tell it where you want it to go and (eventually) it goes. Of course, I’m sure it required a bit of work to define where “door” and “elevator” and “window” are, but it’s a much more intuitive way to control a UAV that works when your hands are full, when you’re stressed (think military), or simply when you have no idea now to control a UAV.

I don’t have much in the way of other details on this project, besides the fact that it probably comes from the Robust Robotics Group at MIT, and possibly from someone who lives in this dorm. How do I know? Well, one of the research goals of the RRG is “to build social robots that can quickly learn what people want without being annoying or intrusive,” and this video is on the same YouTube channel. ‘Nuff said.

[ MIT Robust Robotics Group ]

Under development at Willow Garage is a new ROS stack for PR2 that will allow it to completely calibrate itself. According to the Willow Garage blog, “you can start the process before lunch, and by the time you get back, there’s a nicely calibrated robot ready to go.” Robots, for all their charms, tend to rely fairly heavily on being well calibrated… If you’ve ever programmed a humanoid hobby robot, you’re probably familiar with the necessity to “zero” out the servos just about every time you use the thing lest it immediately throw itself to the ground in protest when you try to get it to walk.

It’s not just that calibration is relatively tedious and time consuming, it’s that you need to repeat it over and over again, which is a waste of time in that you’re not actively improving the functionality of the bot. What’s that you say? Something is inefficient and needs to be repeated over and over? That’s just the type of thing that screams “automate me!” to programmers, which is why a calibration stack will be included in a future ROS release. Eventually, the calibration stack will hopefully mature to the point where self calibration can be applied to just about any type of robot running ROS.

[ Willow Garage ]

iCub, who we’ve met before (a few times), was designed to study cognition in children. Thinking, learning, development, stuff like that. As such, iCub was physically modeled on a two year old. Back when iCub was first designed, though, the technology didn’t exist to make functional hands that small, so the robot was equipped with a set of hands equivalent in size to an eight year old. This has just been fixed, with iCub now sporting a pair ‘o mitts appropriate for its age. iCub also got a new, springier pair of legs that should be better able to manage the inevitable faceplants that happen when a child (or a robot child) is trying to teach itself how to walk.

A couple more iCub pics, including a disembodied head, after the jump. (Read more…)

Last August, we posted about ECCE, an “anthropomimetic” robot designed to attempt to duplicate the muscular and skeletal functionality of a human. Kojiro is a robot from the University of Tokyo’s JSK Robotics Laboratory which was built based on the same kinds of ideas. At this stage, Kojiro’s highlight is his spine, which can bend and flex like ours do to allow the robot to move its torso. While it’s true that Kojiro’s spine doesn’t offer a range of motion that’s significantly different from (say) a couple axes worth of conventional servos, the big advantage at this point is that having a structure of artificial muscles, tendons, and bones makes the robot much more forgiving of impacts to both its own structure and whatever it may be trying not to crush. The disadvantage, however, is that such a complex structure is tricky to get to move just exactly the way you want it to, since you’ve got all kinds of different artificial muscles tugging different ways at the same time.

All of this research is focused on building robots that are more human-like, but I’m still wondering just why this is such an appealing thing to roboticists. I totally understand why learning from biological systems is important, but most of the successful applications that I’ve seen are bio-inspired rather than biomemetic, taking elements of biological design and using them to build better robots as opposed to building robots that are modeled on their biological counterparts. As both EECE and Kojiro show, humans are ridiculously complicated inside, and as research towards a potential anthropomimetic endpoint, it’s a good goal. My question, though, is that as a development perspective for practical robotics, are we better off (at this point, anyway) just letting robots be unapologetically robotic?

[ JSK Robotics Lab ] VIA [ IEEE ]

Somehow, the M3-Neony comes from the same lab that inflicted the CB2 robot child on the world. But obviously, they’ve learned from their questionable design decisions of the past and have moved on to research robots for cognitive development and motor learning that don’t look utterly freakish. The M3-Neony is the same size, shape, and weight as a real baby, and just like a real baby (as far as I know, anyway), it has 22 motors, 2 cameras, 90 tactile sensors, and an integrated computer. It’s fully autonomous, and is designed to wander around on its own, after figuring out how to stand and walk based on your programming. Hypothetically, the process of programming M3-Neony to learn to walk will provide insights into how humans figure out the whole process.

This other robot is called M3-Synchy, and it’s kinda like the bot we wrote about yesterday in form, although rather than being a telepresence robot it’s designed to study how humans communicate with multiple robots, specifically using non-verbal means. Seems like it would be pretty cool if these two bots teamed up. Or, like, fought each other. Either way.

[ Osaka University (Translated) ] VIA [ GetRobo ]

Most telepresence robots (with a few exceptions) aren’t especially presence-y, in that you can see people, and people can see you, but you’re pretty much just a head on a screen on a robotic stick with wheels. MeBot, a project from the Personal Robotics Group at MIT, adds a little bit of personality to telepresence by providing ways for users to communicate non-verbally, through things like head movement, arm movement, and posture:

The clever bit is that you, as the user, don’t need to tell the robot to do any of the expressive stuff that it does with its screen. It watches what you’re doing with your head, and duplicates those socially expressive movements with the robot. Is it effective? You bet:

We conducted an experiment that evaluated how people perceived a robot-mediated operator differently when they used a static telerobot versus a physically embodied and expressive telerobot. Results showed that people felt more psychologically involved and more engaged in the interaction with their remote partners when they were embodied in a socially expressive way. People also reported much higher levels of cooperation both on their own part and their partners as well as a higher score for enjoyment in the interaction.

Even though it has those little 3 DoF arms, MeBot isn’t designed to do anything in particular with its additional axes of motion. You currently control them sympathetically using a second set of arms, the positions and movements of which are duplicated by the arms on the robot. Conceivably, you could add some grippers to the robot and a more comprehensive control system on the other end, but that would defeat a large part of the purpose (and the beauty) of MeBot: it’s designed to be purely expressive, implying a natural simplicity that requires no extra effort or skill. It just does its thing while you do yours, which is how all the best systems (hardware and software alike) tend to function.

Another vid with a few more details, after the jump. (Read more…)

The NY Times posted an article online on Tuesday about food serving robots. We’ve already covered pretty much all of the bots in the article here on BotJunkie (like Snackbot, Motoman, Chief Cook, ramen bots, sushi hand, and that octopus balls robot)

But, there was this interesting quote:

Dr. Heather Knight, a roboticist at the NASA Jet Propulsion Laboratory, said that the industry is trying to change “the perception of robots.”

“The Japanese have always been more comfortable with it, but particularly in the West, there’s this whole Frankenstein thing that if we try to make something in the image of man, to make a new creature, we’re stealing the role of God, and it’s going to turn out wrong because that’s not our role,” she said.

Dr. Knight goes on to suggest that having robots serve people food will get people to like robots, ’cause people like food. Maybe that’s true. I’m not entirely sure, though, about the premise that service robots in general (or perhaps humanoid service robots specifically) make people uncomfortable because of this Frankenstein effect involving humans, uh, overstepping our bounds (I guess?) when it comes to creating things. I’m inclined to believe that it’s something more straightforward: robots are foreign to our general experience, and that makes people uncomfortable. You could argue, I suppose, that this is just semantics, and that robots are foreign to us because we’re not ‘meant’ to be creating them, but I wouldn’t want to suggest that because something makes us uncomfortable there’s something inherently wrong about it. Really, it’s just a matter of education (and cultural education), and you can cite any number of historical precedents to this. Anyway, it’s an interesting thing to think about.

Dr. Knight gave at talk at Ignite LA last fall on social robotics, here’s the vid:

This brings up another interesting issue, that of trust… I feel like it might be possible to trust the current generation of robots in the same way that you trust a business partner, but social trust is another matter entirely. It’s going to be hard to make that happen, since social trust is such a nebulous thing, and it’ll take a lot of hard work by social roboticists to make it happen.

[ NY Times ]
[ Dr. Heather Knight ]

You can buy a shirt with Franken-Bot on it here

Thanks Stark!

When we first saw PR2 Alpha plugging itself into a standard wall socket last year, it was a sort of hit and miss behavior where the robot would stab at the outlet a bunch of times until it actually made a connection. As they say in the video, this is a “sub-optimal” behavior. I commented at the time that:

This is one of those situations where developing a robot that’s able to plug into a socket 20% of the time (but try over and over) is (I imagine) much more efficient than developing a robot that’s able to plug into a socket 100% of the time.

Now, I’m not sure whether I’m exactly wrong about my assertion… There are many significant hardware differences between PR2 Alpha and the PR2 in the above video, and undoubtedly a lot of programming was necessary for the one-shot plug-in. This stuff is a big investment of time and money. As a programmer myself, I totally understand how when something is sub-optimal, you want to optimize, and the improved behavior is certainly beneficial to PR2. However, my question is (still) whether the amount of resources invested in getting PR2 to hit that outlet 100% of the time is really worth it when a much lower percentage of success is still perfectly adequate when it comes to getting the robot to charge itself.

Either way, optimal is optimal, and the same library of behaviors that allow PR2 to plug itself in which such accuracy and precision will certainly come in handy when it tries to do other stuff, like tying my shoelaces for me ’cause I’m too lazy to bend over.

[ Willow Garage ]

If you didn’t manage to watch the Flyfire video in the 10 seconds or so between when we posted it and when MIT yanked both the video and the project website, good news: everything’s back up now. Our post from a few days ago with more info is here.

[ Flyfire ]