Humanoid (robot) hacking

Authors: Elizabeth Churchill
Posted: Thu, March 14, 2013 - 9:00:42

Psy’s Gangnam Style took the world by storm in 2012. YouTube was flooded with copycat dancers performing unlikely gyrations. Fabulously, it wasn’t just human beings getting in on the fun. Researchers Christophe Bartneck and Eduardo Sandoval at the University of Canterbury's Human Interface Technology Lab in New Zealand made a video of their Nao robots showing off their moves. Demoed at the Human Robot Interaction conference that just happened in Tokyo (March 3-6, 2013), the video is fun and entertaining.

As of last year, there were three main contenders for top humanoid robot: Asimo from Honda, HRP-4 from Kawada Industries Inc., and Nao from the French company Aldebaran Robotics. Without a shadow of a doubt, Nao is the cutest to my eyes. However, Roboy from Artificial Intelligence Lab at the University of Zurich, developed via open source, debuted this month at the World Congress and Exhibition of Robots, Humanoids, Cyborgs, and more. We’ll soon see how “he” fares in the ratings—from the images on the website, Roboy looks like he should be starring in a Tim Burton movie. Which is most definitely a good thing.

But let’s return to Nao, whose dancing skills also extend to segments of Michael Jackson’s Thriller routine. Launched as Project Nao in 2004, Nao stepped into the limelight as the replacement for Sony's robot dog Aibo in the Robot Soccer World Cup Standard Platform League (SPL), an international robotics competition (see snippets from Robocup 2013). Nao’s participation in the games is evidence that humanoid robots are coming along not only in their sophistication but also in terms of their general availability. It used to be that creating and programming robots was possible only for those with large budgets, typically in research labs and corporations. While still not cheap (I am still planning how to get one to play with), Nao is now at least imaginable for the hobbyist hacker. Indeed, I would have to say Nao is the ultimate, fun toy if you can afford it. In terms of Nao’s capabilities, from the Aldebaran website:

The various versions of the Nao robotics platform feature either 14, 21 or 25 degrees of freedom (DoF)… All Nao Academics versions feature an inertial measurement unit with accelerometer, gyrometer and four ultrasonic sensors that provide Nao with stability and positioning within space. The legged versions included eight force-sensing resistors and two bumpers.

The Nao robot also features an onboard Linux-powered multimedia system, including four microphones (for voice recognition and sound localization), two speakers (for text-to-speech synthesis) and two HD cameras (for computer vision, including facial and shape recognition). The robot comes with a software suite that includes a graphical programming tool ("Choregraphe"), simulation software and a software developer's kit. Nao is also compatible with the Microsoft Robotics Studio, Cyberbotics Webots, and the Gostai Urbi Studio.

There have been a number of Nao Hackathon events over the past couple of years and there is an upcoming event in London in a few weeks: David Snowdon’s Birds of a Feather session at Devoxx, UK on the 26th March in London. From this upcoming London BOF’s website some more technical details:

It ships with a middleware (called NAOqi) that provides a number of built-in behaviours (such as walking, sitting down, standing up) and provides built-in support for face detection, face & object recognition, speech synthesis, speech recognition, inverse kinematics and access to libraries such as openCV. You can develop for NAO using C/C++, python, Java, .net, Matlib & URBI.

A gloriously geeky event, no doubt, this promises to be a place where bigger questions will be provoked, if not immediately answered. 

So to the bigger question: Nao is sophisticated and fun to play with but what should we “ask” Nao to do? Aside from dancing, what has Nao been programmed to do so far? The robot has been demonstrated doing a standup routine, playing noughts and crosses, playing chess, and picking up dirty socks (always useful!). There are more examples showing the results of last year’s hackathon in London. I can't wait to see what gets built at the next hackathon, which is planned for late August 2013 in London, UK. I also note that later this year is Robotic Idol, the first robotic dance competition for K-12 education, taking place in California on October 26, 2013. While focused on (yet more) dancing, the idea is to increase students’ interest in science, technology, engineering and mathematics (STEM).

I think it would be fair to say that not all hackathon and geeky BOF events are known for critical and reflective perspectives about design and development. However, the design, development, and programming of humanoid robots are the perfect lens through which to address psychological, sociological, and ethical perspectives about being human, as well as technical issues regarding modular design, interoperable subsystems, sensors, data, biomechanics, data representation, and machine learning, to name a few topics. Once a robot is animate, all kinds of bigger questions come to the fore. Anyone who has read/seen I Robot or Blade Runner or Metropolis or a slew of other tales knows the moral dilemmas involved in dealing with an agent who looks (somewhat at least) like a human, interacts (somewhat at least) like a human, moves (somewhat at least) like a human, and performs (if not feels) emotions like a human.

Even if not humanoid, robots have always been cast as slaves, companions, and/or mirrors for us. Do our bidding, keep me company, help me reflect on myself and the nature of personhood. Robots, narcissistic delight (make them in our image), and anxiety (will they undermine or surpass us?) go hand in hand. Indeed, anxiety about robots has always been with us, since the first automata were imagined and described. In the 3rd century BC Liezi the Chinese author wrote about an account Mu of Zhou (who reigned sometime around 976-922 BC or 956-918 BC) and an a life-size humanoid automaton created by a mechanical engineer, Yan Shi, an “artificer”:

The king stared at the figure in astonishment. It walked with rapid strides, moving its head up and down, so that anyone would have taken it for a live human being. The artificer touched its chin, and it began singing, perfectly in tune. He touched its hand, and it began posturing, keeping perfect time... As the performance was drawing to an end, the robot winked its eye and made advances to the ladies in attendance, whereupon the king became incensed and would have had Yen Shih (Yan Shi) executed on the spot had not the latter, in mortal fear, instantly taken the robot to pieces to let him see what it really was. And, indeed, it turned out to be only a construction of leather, wood, adhesive and lacquer, variously coloured white, black, red and blue. Examining it closely, the king found all the internal organs complete—liver, gall, heart, lungs, spleen, kidneys, stomach and intestines; and over these again, muscles, bones and limbs with their joints, skin, teeth and hair, all of them artificial... The king tried the effect of taking away the heart, and found that the mouth could no longer speak; he took away the liver and the eyes could no longer see; he took away the kidneys and the legs lost their power of locomotion. The king was delighted.

While this particular dream robot’s anatomy may be a little confusing, what is more interesting is the king’s jealousy at this flirtatious creature. The robot stirred not just curiosity but anger and fear—enough to spur control and destruction of the automaton’s “body,” faculties and agency. Think about it: Nao may be cute now, but there could be threat lurking in the shadows. Forget doom narratives of robots taking over the world. Let’s address some closer to home fears about loss: Will Nao steal your boyfriend or girlfriend? Or, is it creepier to imagine oneself falling in love with a robot? In Do Androids Dream Of Electric Sheep? by Philip K. Dick, we are told….“Some female androids seemed to him pretty; he had found himself physically attracted by several, and it was an odd sensation, knowing intellectually that they were machines but emotionally reacting anyway.”

As I said, I think Nao is cute, but I am rather far away from that worry as of right now. But as robot displays of emotion get more sophisticated, this will change. It is well worth checking out the TED video from Hanson Robotics (old but still compelling) of a robot with Einstein’s face demonstrating its capabilities for facial expression. Only the most controlled, cold, or socially challenged of us would not get drawn in to at least some degree by these emotional expressions. It is also well worth checking out Angela Lim’s work, based on her Masters thesis work, on models of emotion for robots and for the design of trustworthy, user-friendly robots. Beyond the immediate, in-the-moment interactional issues, there are larger-scale issues that are discussed in an excellent book from MIT Press that came out last year, Robot Ethics: The Ethical and Social Implications of Robotics.

It is always the case that what gets built reflects not just the technical possibilities and limitations of the robot development platform itself, but also what the designer/builders believe to be the right kinds of things to have a robot do. What we have robots do, our imaginings turned into code and robot action, says more about us than it says about robotics. Household chores and menial tasks are always high up on people’s list when I ask them this question. Baby-sitting the kids seems to be a little more problematic, but finding human baby-sitters to have that elusive night off seems to be a pain point for parents. And while I am well aware of the landscape out there, I will not even venture into the territory of fembots and sexbots. Ask yourself: What would you want your humanoid robot to do? Thought experiments about what our design ideas and ideals for humanoid robots say about us, the designer/developers, are fascinating. It is most interesting perhaps to ask people, what ripple effects would bringing a robot into your home or workplace have? I may not need to take the trash out anymore, but now I need to spend time tending to my robot or debugging issues that arise or cleaning up after it if the trash-task inexplicably fails. We will shape the robots and thus the settings in which they act, and they in turn will shape us and our actions. Further, if my car is anything to go by, things will get less reliable as time goes on. And if the apps on my phone or the updates to my computer software are an indication, my robot may well spend as much time being updated and upgraded as I spend sleeping and eating. Less concrete but nevertheless important, from the aforementioned MIT Press text, Allan and Wallach suggest that we need to carefully consider what “functional morality” machines such as robots have. They say we need to spend "more time thinking about the contexts in which (ro)bots operate and about human responsibility for designing those contexts."

I would not be me if I didn’t comment on how weird it is that robots are assigned genders so easily. Asimo is routinely referred to as “he” by people on first encounter, people with no prior knowledge of him/it. Even the non-humanoid robots in the film WALL-E are assigned genders. From the Wikipedia page we see “He (WALL-E) falls in love with another robot named EVE, who also has a programmed task, and follows her into outer space on an adventure that changes the destiny of both his kind and humanity.” What on earth is it about these non-humanoid robots that suggests gender assignments? Watching the hackathon videos, Nao, I note, is sometimes a “she” to her programmers. That said, happily to date, in my (albeit limited) viewing, I have not discerned what I would consider to be stereotypically gendered behaviors programmed in by developers.

In sum, I am joining the herd of folks who believe that robots are fun and educational, and educational at all levels, inspiring the development of technical skill sets to prompting consideration of social sciences and issues raised in the reflective, philosophically grounded humanities. I’d love to see schools adopting robot hacking as a way to teach young people engineering and computer sciences, the fundamentals of social interaction dynamics, basic issues in philosophy/ethics, and the practice of reflective design processes—all in one fell swoop. But for now, I am working on getting a Nao for myself, and then, given my current passion for Latin Jazz, I’ll be working out how to get Nao to calypso and double tuck. 

Some useful references:

Patrick Lin, Keith Abney, and George A. Bekey (eds.), Robot Ethics: The Ethical and Social Implications of Robotics, MIT Press, 2012, ISBN 9780262016667.

See also the Journal of Human Robot Interaction.

Elizabeth F. Churchill is Director of Human Computer Interaction at eBay Research Labs in San Jose, California. She is also vice president of ACM SIGCHI.

Posted in: on Thu, March 14, 2013 - 9:00:42

Elizabeth Churchill

Elizabeth Churchill is a director of user experience at Google. She has been a scholar and research manager focused on human-computer interaction for over 20 years. A Distinguished Scientist of the ACM, her current work focuses on HCI aspects of the social web and the emerging Internet of Things.
View All Elizabeth Churchill's Posts

Post Comment

No Comments Found