Andrea Bianchi's blog

A young Steve Jobs, at the beginning of his career, used to say that computers are like a bicycle for our minds. If we consider the efficiency of locomotion for various species on the planet, humans perform pretty poorly. However, if we allow humans to use bicycles, the result is very different and humans tops any other animal. With this example Jobs pointed out that what really sets apart humans from other species is that they are tool builders (remember the famous "The Dawn Of Man" scene from Kubrick's 2001: A Space Odyssey) and computers are the most remarkable tool that human kind ever came up with - bicycles for our minds-.

Every year the Pohang Institute of Intelligent Robotics hosts the Korean Robot Grand Challenge Competition, where autonomous robots from several teams compete in sort of scavenger-hunt game. Robots try to reach a specified floor of a building by taking the elevator, find a specific object (pack of milk, ball, pen...) in one of the rooms and bring it back to the starting point. Needless to say that no team ever succeeded yet in completing this challenge (though similarly complex challenges as the DARPA one, has been completed in the past). What is remarkable about this competition is not much the results but the challenge it proposes (a great challenge that is still unsolvable), highlighting and emphasizing few avenues of research necessary for the Human Robotic Interaction. In fact, although people give often for granted that robots can easily navigate in a space meant for humans, this is often not the case, as I have also discussed in my previous post. Indeed, autonomous navigation remains still a complex task and an open research area.

Social robots are soon going to drastically impact on our lives, changing the way we work and spend our leisure time at home. Though I discussed this theme often in my previous posts, I have until now not spoken of one very important --yet often ignored-- aspect: mobility and accessibility. It is impossible to serve humans or engage in any type of human-robot interaction if the robots are not near to humans or able to reach them, this is obvious! Even though pop-culture stereotypically shows robots that can autonomously go everywhere, the reality is very different and a simple task like walking or climbing stairs can be very challenging.

Mr. Choi, robotic research scientist at the Robot and Cognitive System Department of ETRI (one of the most famous Korean research institutions), is working with his team to enhance robot mobility. The challenge is how to make robots robustly and autonomously navigate thorough environments that are dynamic and unpredictable because they are meant for humans, as for example a street with cars or a room full of people. At ETRI researchers have worked on different strategies based on hardware and software technology in order to simplify the environmental randomness and allow robust navigation.

A few years ago when I was still working in the video-game industry with Nick Fortugno, one day Nick came to the office and shared some thoughts about the Hi-Res and Lo-Res debate that took place at the IGDA meeting in New York (May 2005).

The word res commonly stands for resolution, in terms of the graphical and audio assets used in video-games: Hi-Res games are exemplified by titles like Crysis (the game is as realistic as possible to create an immersive experience), while Lo-Res games are exemplified by the revolutionary Starfox, where the virtual word is represented more abstractly but the interaction remains realistic. People supporting Hi-Res have the idea that there is a direct correlation between resolution (or fidelity) and realism while the Lo-Res folks think that through abstraction “the user becomes less concerned with the question ‘is this real?’ and is better able to concentrate on other game aspects like actually having fun.”

If you have read my posts in this blog over the last few months, you might already be familiar with the idea that "not all robots look like robots". People like Don Norman (see my previous interview) and Frederic Kaplan (see his talk at Lift Asia 2008) have been advocating this for many years now; however, when they speak about robots, they often intentionally refer to that category of robots that populate (or will populate) our homes: these robots are called social robots, in contraposition to the already well establish domain of industrial robots. In this post, I take the liberty to speak about a third category of robots, which are not used to produce other goods or in home settings. I am not even sure I would call them robots at first, but they are undeniable physical, they move through actuators and serve humans intelligently for training and entraining purposes: I am talking about the animatronics, often used in amusement parks.

A few weeks ago I visited the Daegu Safety Theme Park, an amusement and educational park for children located in Daegu, South Korea. In 2003 Daegu was unfortunately the theater of one of the most tragic subway disasters of all times. A sequence of mistakes, a lot a of misfortune and inadequate emergency equipment turned the suicidal attempt of a dissatisfied arsonist into an unprecedented fire disaster that caused the death of about 200 people in one of the metropolitan subway stations of Daegu. After the disaster, the Korean government started a training program in schools to teach safety awareness to children and, as part of the program, the government also opened a Safety Theme Park in Daegu.

The park is absolutely brilliant and deserves a trip if you are in Daegu. In fact, more than a memorial park, the Safety Theme Park is a training camp for children and their parents, where they can learn about safety through direct experience. The park has the double objective to educate and entertain the audience, targeting in particular children, but it is undeniably instructive and fun for adults too. The audience doesn't really get in direct contact with robots or animatronics, but it experiences them indirectly as part of a plot of a staged experience.

It was memorable for example when we walked on a perfect reproduction of a subway carriage parked in a station and they set it on fire to simulate an accident (the fire was fake, but there was real smoke). After having shown us a video of the 2003 subway disaster and the remains of one of the carriages from the 2003 accident, a guide directed us to a functioning reproduction of a carriage, where we were told how to depressurize, unlock and open the emergency doors, how to protect our faces from the smoke and how to run in the underground tunnels (following a luminescent trail on the ground) to find an exit. Similarly, we also experienced a simulation of a 8 degree Richter scale earthquake: they put us on a platform on which there was a reproduction of the interior of a kitchen, with various appliances and furniture, and then they activated the platform that shook us for about 20 seconds in order to reproduce a real earthquake: they taught us what to do in such a case and our goal was to follow the directions we were given in the briefing. Finally they led us to a room reproducing a real forest during a flood and they taught us how to behave in such a situation.

Each one of the these simulations (and more that I did not describe here) made perfect use of animatronics and robots in order to simulate staged dangerous situations like a subway fire, an earthquake or a flood. Interestingly, these robots, like on a movie set, are completely invisible and have the only role to create the atmosphere and a realistic and immersive experience for the audience. Although 3D technology and virtual reality is often used to "drag" the audience in virtual and immersive worlds, I suspect that at times animatronics serve better the cause of simulating a real experience by dragging the audience not to a virtual world, but a real and physical world that looks as similar as possible to reality. Obviously the "robots which make the magic happen" (e.g., the platform of the earthquake, the fake subway wagon) are invisible or only partially visible to the audience, so that with a bit of imagination the audience can really pretend to be in the middle of a critical and dangerous situation.

When we think about robots, we inevitably associate them with machines, computers and engineering. Professor Myung-suk Kim, Ph.D., from the Industrial Design department at KAIST (and also Director of the PES laboratory and adviser for the Korea Robotics Society) agreed to be interviewed to show us that this is a misleading and quite obsolete concept.

If we consider the case of Korea, until about 10 years ago design had only a marginal role in the engineering of robots. In fact, in most cases engineers sought help from designers when they had to make their robots look appealing, by designing and creating a pretty shell around the hardware that the engineers had developed. Although this work is necessary, robot designers should have a more active role since the the beginning of the work, at the conceptualization level.

In Professor Kim's words, design has a specific role in robotics: “Thanks to advancements in robotics technology, the groundbreaking results of academic research, and the growth of the robotics market brought on by rising demand, robotics is moving beyond the realm of engineering and becoming an active field of research in the humanities and social sciences. Robot design, in particular, is gaining increasing importance as a means of presenting robots, commonly regarded as technocentric hardware, to new users as software products that combine communication media with services suited for everyday life.”

Thanks to the work and perseverance of people like professor Kim, this is in fact what is beginning to happen. We are seeing more designers involved in the design of robots starting in the early stages of their creation, contributing with usability scenarios and design prototypes. "We make products for ordinary customers, not technology, and we always think of the market: usability is our primary concern", say the researchers at PES lab. I asked them to show me how they normally work in the early stages of robot creation and they showed me some of their approaches.

A common approach of many robotic companies when interacting with designers is to ask for look-and-feel prototypes: this type of prototype helps to envision what the final product will look like and, at times, might drive the choices for the hardware. The look-and-feel prototype is not an empty shell to make a robot appealing, but rather an exploration of what type of emotions the robot can convey to the public: it dictates the personality of the robot at its roots.

Another common approach used by designers is to implement functional prototypes. In this case, the aesthetics of the robots is irrelevant because what matters is what the robot does and the interaction it can establish with users. For instance, making a robotic TV would drastically impact the ecology of current users' homes, requiring substantial alteration of the environment (in this case the goal would be to study the social role of robots in specified settings).

As better explained by Professor Kim, we are urged to think of design in a more general sense (like a "design thinking" attitude) rather than - a usual misconcept - a tool to make robots look pretty: “Because previous work on robotics aimed primarily at technological achievement, the focus tended to be almost exclusively on enhancing the skills and capabilities of robots from a developer’s point of view, rather than on improving user-friendliness. For this reason, there are few products that have succeeded in winning market popularity […]. To address this problem, active effort has begun to be expended on the study of the social attributes of robots.”

Is it possible to train people up to the point where you permanently overwrite the way they walk or swing a golf club? Professor Sunil K. Agrawal Ph.D., from the Mechanical Engineering Department of the University of Delaware recently gave a seminar talk in KAIST showing that it is indeed possible to change - to a certain extent - a deeply embedded behavior, such as the human gait, with some help from the assistive robotic technology.

Professor Agrawl's idea is conceptually very simple: take a group of patients who have suffered a stroke and have only partially recovered. Those patients usually walk more slowly than average, have an asymmetric gait, lack of ankle flection with consequent foot drop-toe dragging and are traditionally rehabilitated by human personal trainers. Repeatedly train them using a robotic exoskeleton which, by means of a haptic force feedback, constrains the trajectory of the walking feet within a virtual tunnel (see the picture below) with the shape of an input template gait. Eventually after some training, patients learn a new way of walking and this is ascertained by tracking and analyzing the data from how the patients walk after the training, without any aid from the robotic exoskeleton. The final experience for the users is very similar to that of a human trainer manually forcing the patients' legs to follow a specific trajectory, but the advantage of using machines is in the accuracy, consistency and continuity of the exercise.

KAIST Professor Dong-Soo Kwon, Ph.D, is probably one of the most well-know robotic experts in Korea: he is the vice-president of the Korea Robotic Society (KROS) and the Director of the Human-Robot Interaction Center at KAIST, where he and his team research about Human-Robot Interaction, Medical Robots and Haptics. Because of his reputation both in academia and in industry, I have decided to ask him what are, in his opinion, the prospective of future robot business: will home robots ever take off as commercial goods?

Let's compare the robot industry to other industries which share similar technologies and purposes, such as the car industry (purpose: mobility), the computer industry (purpose: information technology) and the mobile-phone industry (purpose: communication).

The car industry dates back to the Benz Patent-Motorwagen (1886), but it is only from 1908 with the Ford Model T and the advent of mass production that we can talk of a real car industry. Furthermore, it took until 1972 to reach saturation point (1 car per driver in the U.S.A.) and the point in which the market became completely mature: from the start to the saturation point it took 64 years. The computer and mobile-phone markets followed a similar evolution - introduction, mass production, saturation point. So for instance, the time elapsed between the first mass-produced computer by IBM in 1981 (IBM 5150) and today (we have now finally reached the point in which we have 1 computer per person in the U.S.A.) is about 30 years.
Similarly, the mobile phone market started with the Motorola DynaTAC in 1983 and can now be considered mature, after about 27 years.

So, what about the robotics industry? We could say that robots started being mass produced in 1975 with the PUMA Unimation, which would make robots present on the market for more than 30 years. According to the paradigm followed by similar industries, robots should hence already be a mature market. This is in fact true, but it is only limited to the case of industrial robots: any modern factory uses some kind of production industrial robot, confirming that the market is already mature. However, when we talk about personal robots, we can probably say that only in 2002 with the Roomba, robots made their appearance in the consumers' home as a mass produced good. Taking 2002 as a symbolic date, we can infer that the personal robot market is still very young (about 8 years old) and that if the paradigm of other industries still holds for robots, we should have a mature home-robot market only in 2035 (a common symbolic date in many sci-fi books and movies, including I, Robot).

The next questions to ask are what we will use robots for and how much will they cost. Professor Kwon has no doubts: we will use personal robots in a variety of situations for a multitude of purposes. Personal robotic toys will entertain and babysit children, teachers will use telecontrolled robots to teach in remote places (telepresence), while restaurants and touristic destinations will make use of service robots to better guide and serve customers: in many cases though these robots will not be completely autonomous but simply partially guided remotely by other humans (the service providers), allowing them to serve multiple customers simultaneously. In terms of cost, we can again compare the robotic industry to similar ones and try to drag some conclusions from that data. If the price of an ordinary car is in the $10,000-$100,000 range, the price of a computer in the $500-$3,000 range and of a mobile phone in the $50-$1,000 range, we can probably expect personal robots (which share some technology with each one of these products) between $3,000 and $10,000, a price that should be cheap enough to make robots appealing for consumers and assuring profitability to the industry.

Further readings:

Roti by KAIST and Rastech Co.: English Tutoring Telepresence Robot

As a child, my first encounter with robots was through Japanese animations, in which futuristic armored exoskeletons fought against all sorts of giant evil monsters. Since then, I have always been fascinated with robots and I have wondered why they are so present in East-Asian pop culture and what their unique and distinguishable characteristics in Japan and Korea are? I have asked these question to an expert in robotics and culture, Dr. Zoz Brooks, who is also an engineer and artist from MIT Media Lab's Robotic Life group (Dr. Cynthia Breazeal, the leader of this lab, is a confirmed speaker for Robolift 2011).

Recently I had the chance to visit Theo Jansen’s Animals Modular exhibition in Seoul, where I could finally admire in person his beautiful creatures. Theo Jansens is a very well known Dutch artist, famous since the early 90s mostly because of his amazing kinetic sculptures, large animal-like plastic skeletons which move autonomously on beaches with the power of wind (check out Theo Jansen’s BMW commercial for seeing his sculptures in action). Apparently, one day some people in Netherlands saw these huge creatures wandering around the beach, taped them on video and shared it: that was the beginning on Theo Jansen’s publicity.

I have seen these creatures in action, walking back and forth the hall of the museum in Seoul, and I have to admit that it is impressive to see them in motion. From an artistic point of view, it is, in fact, the motion which makes these creatures really alive and credible; from a technical point of view, understanding the skills involved to engineer the motion, storing the energy from the wind in bottles, sensing the water to avoid drowning in the sea are simply amazing brain teasers. It is also fascinating to see how the whole system has been conceived by the artist as part of a continuous and natural evolution: the creatures are in fact classified according to different evolutionary traits (hence, the choice to name them in Latin). There is much to say about these topics, but I leave readers the words of Theo Jansen himself (watch his TED talk).