Founded in 2004, Engineered Arts’s flagship product is a full humanoid called RoboThespian.  RoboThespian’s head features a moving jaw, internal colored lights, and a small LCD screen for each eye, which allow for substantial expression.  The robot has over 30 degrees of freedom, powered by unique hybrid pneumatic-electric actuators.  The movements (which you can see in the video below) are surprisingly fluid — I find them reminiscent of the animatronics at Disney, but unlike those, RoboThespian also contains sensors and can be programmed to interact with people in a variety of ways, including speech.

A more recently introduced product is SociBot, a half-humanoid (from the waist up) that dispenses with the pneumatics and relies on fully electric actuators.  Less costly and easier to install and maintain than RoboThespian, SociBot also features a fully animated face, using a picoprojector inside the head.  This allows an even wider range of expressions, though I do imagine it places some constraints on the ambient lighting.  SociBot features speech recognition, speech synthesis in over twenty languages, visual person tracking and gesture recognition, and even an RFID reader to allow it to recognize tagged people or objects as they approach.  SociBot can serve something like a (half-)humanoid kiosk, answering questions and greeting visitors.

Working even further down the cost scale, a SociBot-Mini model is available.  This is actually a full-sized robotic bust, made miniature by omitting limbs and half the torso, but with the same facial animation and head/neck articulation as its full-sized cousin.  This robot is aimed at individual researchers or hobbyists who want an affordable starter platform.

So what’s next for Engineered Arts?  Even RoboThespian, which actually does have legs, cannot currently walk.  However, the company has embarked on a 2-year research program to develop legs and dynamic balance to make robots that can walk in a humanlike manner.

Whether they will succeed in that goal, only time will tell.  But already it’s clear that founder Will Jackson and his team have built some remarkably agile and social robots.  There are plenty of great ideas here that hobbyists could learn from to make our own little humanoids seem more human-like, too.

For a look around the Engineered Arts factory, see this BBC News video, and see here for more videos of RoboThespian in action.

Entropica is a powerful new kind of artificial intelligence that can reproduce complex human behaviors, including the ability to autonomously set and implement its own goals.  In this video, we will see how Entropica can walk upright, use tools, cooperate, play games, make useful social introductions, globally deploy a fleet, and even earn money trading stocks, all without being told to do so.

Here’s the full pitch video:

Sounds great, doesn’t it?  But unfortunately, there’s a lot here that causes my hype detector to peg the red zone.  Let’s take the first claim: “Entropica can walk upright.”  As humanoid robot hobbyists, we know what an impressive feat that is; you have to coordinate a large number of actuators, maintain balance while moving forward, shift weight from side to side, deal with possible slopes or small obstacles, and so on.  Entropica can do all that?  Well, no.  What’s actually shown is Entropica solving the classic pole-balancing problem.  That’s a control problem that’s easy to solve with any of a dozen different techniques, and has almost no relationship to walking at all.

OK, how about making useful social introductions?  A robot that can say hello, introduce itself, and exchange a bit of personal information would indeed be pretty neat, especially if it did so without being specifically programmed to do that.  But again, no, that’s not what’s shown here.  Instead they show Entropica adding links in an acyclic graph, while other links are randomly removed.  Again, there are other fairly trivial solutions to that same problem.

Now, not to be a complete curmudgeon, it is normal and expected for new approaches to first cut their teeth on toy problems.  (It’s not all that normal to overhype the results of such toy-problem tests this much, at least on this side of the pond, however.)  It’s entirely possible that the authors (A. D. Wissner-Gross of Harvard and MIT, and C. E. Freer of Hawaii) really have found a useful new approach to solving problems.  The basic idea — that an agent should make choices that maximize its future options — makes some sense as a general principle.  Whether it scales up to nontrivial problems remains to be seen.

The full paper is behind a paywall at Physical Review Letters; and for more curmudgeonly analysis, see this New Yorker article.

Michael Overstreet has achieved prominence in the U.S. humanoid robotics scene via his Bioloid-based robot “Boomer,” and is now taking it up a notch by trying to 3D print his own DARwIn-OP.  The Darwin is a fairly high-end research/education robot based on twenty Robotis MX-28 servos.  Those servos are quite pricey, so the robot as a whole is still going to cost him about $6k — but that’s about half what an off-the-shelf Darwin costs.

Michael’s Darwin clone, named “Robby,” isn’t quite able to walk yet, but it’s getting close.  Building a humanoid robot can be a long, difficult task, even when you have a well-developed open platform to work from.  Read more about it at Make, and follow the progress at Michael’s blog.

This event isn’t entirely new to the RoboGames venue; it and related events were first introduced in 2007.  But they haven’t been very consistent about it, either — the couple of times I attended RoboGames in the last few years, we were lucky to get someone to actually run the standard Robo-One style competitions, and never mind anything extra.

But apparently more effort was put into the humanoid events this year, because robotics reporter Lem Fugitt caught some cool videos of the weight-lifting competition.  First, here’s an interesting-looking humanoid from Korea called RnD_Eska:

And then here’s a NAO called Looney doing the same event:

The official rules are hosted at the RoboGames site, but are cribbed from the 2007 HuroCup event, run by FIRA.

I’m a fan of anything that pushes the advancement of humanoid robotics, and especially if it can fire up the crowd and maybe encourage a few of the onlookers to go home and build something.  Not everybody cares for robot fighting competitions, so things like music, dance, and gymnastics are important to reach a wider audience.  Does weight-lifting have a useful place in this spectrum?  Weigh in with your thoughts below.

Buzzword or not, deep-learning techniques have produced some impressive results in recent years.  For example, the latest version of Android OS uses these techniques to dramatically improve the speech recognition on its phones, to a level comparable to Apple’s Siri — except that while Siri sends your voice over the network to be interpreted by big servers in the cloud, Android can do it locally on the device (at least in the case of “voice typing”).

That brings me to the reason why all this is relevant to robotics.  While it seems like a lot of the progress in AI lately is being done with supercomputers (like IBM’s Watson) or truly giant networks of servers (Google), that’s only the beginning.  Throwing computing power at an AI problem certainly helps, but once the techniques are worked out, most of them can be scaled back down to the level where they will run on smaller, robot-sized computers.

Of course, there are also those who think “cloud robotics” is the future — that is, robots with very little intelligence onboard, which instead rely on server-side processing.

In either case, when Kurzweil’s drive to create human-level AI is combined with Google’s deep pockets, army of computer scientists, and massive computing infrastructure, I predict some exciting advances in the next few years.  Stay tuned!

[Via MIT Technology Review]

Enter the new TakkTile touch sensor.  This amazing gadget, developed at the Harvard School of Engineering and Applied Sciences, takes advantage of yet another sensor that cell phones have made tiny, reliable, and cheap: the MEMS barometer.  These little chips are designed for sensing air pressure, and are intended as altimeters (and weather sensors, I suppose) in cell phones.  The Harvard team has found that if you pour rubber into the metal case surrounding the actual sensor, then they no longer notice air pressure, but instead sense pressure applied to the case — and do so just as sensitively as they previously sensed altitude.  They’ve combined this with a bit of circuitry and firmware to make a self-contained touch sensor that speaks to a host controller over the common I2C protocol.

In the demonstration video below, the researchers show how the sensor handles a 25-lb weight and being struck with a hammer just fine, yet remains sensitive enough to clearly detect a 5g weight.  For the metric-impaired, that’s the weight of a few paperclips.

The team is encouraging “community development” by making everything open, from the CAD drawings of the sensor board, to the source code, to instructions on how to pour your own rubber.  That last step is a bit onerous for most of us, though, because air bubbles tend to get trapped between the rubber and the sensor, greatly reducing sensitivity.  To avoid that, you need to use a vacuum chamber to suck the air out.

Fortunately, those of us who lack the time, skill, or vacuum chamber for that procedure can order pre-made sensor boards from TakkTile.  A strip of five ready-to-use sensors costs $149, or about $30/sensor.  That’s not pennies, but for a touch sensor this sensitive, it’s pretty much unheard of until now.

I can already think of some handy applications for these things.  If you put several on the bottom of each foot on a humanoid robot, it would be able to sense how its weight is distributed on the feet, and adjust to maintain its balance.  This could supplement IMU readings (which are already commonly used) to produce much more robust biped standing and walking.

What applications for these new sensors can you think of?  Weigh in with your thoughts below.

[Read more: Gizmag, Harvard press release, TakkTile home page.]

COMAN is one of the new breed of robots making good use of compliance, that is, joints that have a bit of give.  Ordinarily, servomotors hold their position as rigidly as they can.  That can be tuned in any modern smart servo by tweaking the holding current, but the compliance obtained in that way doesn’t store and recover energy, as our own tendons do.

COMAN adds compliance through the use of series elastic actuators, which connect the servo to the joint via springs.  These allow the joints to absorb reaction forces, for example when the foot strikes the ground during walking, in a way that not only saves energy, but simplifies the control algorithms.  Compliant joints also make the robot safer to be around, since it’s less likely to crush things that get in its way.

While other robots (such as the new Baxter industrial robot) make use of series elastic actuators or other forms of compliance, COMAN is the first humanoid to have compliance in all limbs.

Check out the new video showing off COMAN’s ability to stand on a moving platform, deal with an incline, and handling pushes without falling down.

[via IEEE Spectrum]

DARPA says:

DARPA’s Autonomous Robotic Manipulation (ARM) program is developing software, hardware and sensors to enable robots to semi-autonomously grasp and manipulate objects in unstructured environments (meaning, “outside of a laboratory”) with human operators providing only task-level instructions. For example, rather than dictating step-by-step every movement a robot makes, a human can give DARPA’s ARM robot a high-level command like “Open the door” or “Screw in the bolt.” Performers on the ARM program have already demonstrated success using one arm and hand to manipulate objects. Now DARPA is having teams test two arms and hands on tasks that require bimanual manipulation, like the robot changing a tire shown in this video. If DARPA is successful with grasping and manipulation, while also making robots more adaptable to changing environments and driving down the cost of production, robotic manipulation systems can be applied to a wide range of potentially dangerous Department of Defense applications, including defusing improvised explosive devices and searching bags.

A recent New York Times article talks more about the program.  For more detail, visit the DARPA ARM program home page, and for some related work, see our posts on the DARPA Robotics Challenge.

I just found out (thanks to this Makezine post) about a new children’s club called the Maker Scouts.  From the Maker Scouts International website:

The Maker Scouts is a weekly scouting program for young children (4-10) who are ready and interested in Tinkering and Making with different materials and tools. The Maker Scouts program provides young Makers learning opportunities in STEAM [Science, Technology, Engineering, Art, and Math] education through four 12 week sessions, each of which focus on critical thinking, design, and building with hand-on activities.

Activities are designed to help the youngest makers become safe, proficient users of making tools and materials on their way to discovering their passion and language of self expression. The program also provides young people learning opportunities that foster the core values of the Innovator Mindset– embracing failure, frustration tolerance, communication, collaboration, creativity and empathy.

Like other scouting groups, Maker Scouts receive badges (which they can collect on their work aprons) for making skills, as well as demonstration of a very sensible and inclusive set of Core Values.  Unlike some other scouting groups, Maker Scouts welcomes all interested children between the ages of 4-10.  Their recent meeting of Maker Scouts Los Angeles attracted equal numbers of girls and boys, who built and played musical instruments.  Other activities span a wide range of skills:

Maker Scouts MAKE robots, sew upcycled clothing , play with circuits, invent games, use e-textiles, build with wood, metal, plastic, make cardboard creations, play with food, tinker with real tools, shoot off rockets, invent real world solutions to real world problems and so much more.

For older kids, a sister organization called Hacker Scouts is available.

My hat is off to everyone involved in these organizations.  STEM jobs are growing so fast that employers often can’t find people to fill them.  And even kids who don’t grow up to work in a STEM career will benefit greatly from the critical thinking and problem-solving skills that STEM activities build.  Working with kids can be challenging, but we owe it to them to give them every opportunity we can to stretch their minds and build stuff.

Maker Scouts is just getting started, so give some thought to how you can help them in your area!