The Darmstadt Dribblers have some of the most impressive humanoid robots in the RoboCup tournament. For the second year in a row, the team from the Technische Universität Darmstadt, Germany, took the title in the kid-size humanoid league (for robots 30-60 centimeters tall). How did they do it?
IEEE Spectrum's Harry Goldstein went to RoboCup 2010 in Singapore to find out. Watch the video below to see these amazing little robots playing and also an interview with Dribblers team member Dorian Scholz. Then visit their YouTube channel for more more videos, including this year's kid-size final.
Does this blushing fake face freak you out? German engineers at Bielefeld University hope it's more cute than creepy. It's a robotic head called Flobi and it's designed to express emotions while overcoming the creep factor associated with the Uncanny Valley, when robots resemble zombies due to design issues.
There have been a number of robotic heads developed over the years in an attempt to replicate human interaction and emotional expression. Flobi, described by Ingo Lutkebohle and colleagues in a paper for the ICRA 2010 conference, is notable in that it has a magnetic mouth actuator system as well as a modular construction that lets an operator perform a sex change of sorts. In only a few minutes, Flobi's plastic face and hair features can be swapped for male or female.
Flobi, about the size of a human head, was designed to be effective in communication as well as sensing. It has 18 actuators, and is equipped with microphones, gyroscopes, and high-res cameras. As seen in the photos and video below, it can express a variety of emotions such as anger and surprise by moving its eyes, eyebrows and mouth. It also has four cheek LEDs to mimic blushing.
The obvious design features are that the hardware is almost completely concealed, there are no holes in the face, and its appearance is cartoonish, not unlike Philips' human-robot interaction platform iCat.
"The comic nature is far enough from realistic not to trigger unwanted reactions, but close enough that we can take advantage of familiarity with human faces," Lutkebohle and colleagues write. "One advantage we have already explored is the use of 'babyface' attributes."
These features were very deliberate choices so that people interacting with Flobi would not notice the degree to which it falls short of human. That gap often results from silicone rubber-skin robots such as CB2, a mechatronic baby built at Osaka University, that can seem creepy.
"While I wouldn't say that plastic is inherently superior to other materials, it is instantly recognizable as a human artifact," says Lutkebohle. "Furthermore, rigid plastics match current actuation capabilities in contrast to latex, for instance, which is hard to actuate without triggering the Uncanny Valley effect."
So does Flobi overcome the Uncanny Valley? That's highly subjective, but at least one roboticist unaffiliated with the project thinks it does.
"There is value in robotics making aesthetics and design a first class citizen," says Andrea Thomaz, an assistant professor of interactive computing at Georgia Institute of Technology, who wrote about Flobi on her blog. "How the robot looks will directly impact people's perceptions of it, and in turn will impact any studies of interaction done with the platform. The Flobi project demonstrates a couple of interesting ideas with respect to robot design: magnetic features to limit awkward holes on the face, and swappable features to easily change the gender/style."
Will plastic, and not rubber, become the dominant robot design medium? We'll see whether Flobi will inspire other robots to overcome the Uncanny Valley.
Every week I grab New York Magazine and flip to the last page to see their despicably funny Approval Matrix. I like it so much in fact that I decided to shamelessly rip it off -- robot style.
Did you notice there's been a proliferation of robot infants, robot toddlers, and robot children in the past few years? It seems that roboticists enjoy becoming parents to bionic babies.
They build them for a good reason: These bots help researchers not only learn more about robotics but also investigate human cognition, language acquisition, and motor development. Some Japanese researchers even say robot babies could help introduce young people to the wonders of parenthood and boost birth rates. Yes, robot babies help make real babies!
So check out our reality matrix below, where we rated each robot according to its similarity to humans and its technical capabilities. What's the coolest? Cutest? Creepiest? And did we forget any? Let us know.
Click on the image to see a larger -- readable! -- version.
Clockwise from top left: Aldebaran Robotics; Thomas Bregardis/AFP/Getty Images; RobotCub; Kokoro; Erico Guizzo; Koichi Kamoshida/Getty Images; University of Tsukuba; Tony Gutierrez/AP Photo; Sankei/Getty Images; Georgia Institute of Technology; Yoshikazu Tsuno/AFP/Getty Images; University of Bonn; Erico Guizzo
It's late on a Friday afternoon at Gamma Two, a robotics development company in Denver, Colo., and that can only mean one thing: It's time for BeerBot.
"Wilma?" Jim Gunderson tells the robot next to him, a cabinet-shaped machine that seems straight out of a 1970s science fiction movie.
"What do you want me to do?" the robot responds.
"What do you want delivered?"
After driving itself to the kitchen and muttering along the way, Wilma the robot delivers the drink to its owner.
Gamma Two is run by husband and wife team Jim and Louise Gunderson, whose pride and joy these days lies in two autonomous mobile robots named Wilma and Basil.
The Gundersons designed the robots as personal servants that they now plan to commercialize. The machines can respond to voice commands and perform tasks such as delivering drinks and ... well, that's pretty much all they can do now.
But the Gamma Two couple has ambitious plans for the bots. Jim Gunderson, whose title is "cognitive systems architect," says many people would benefit from robots that could work as a nurse's aide or a discreet butler.
People with physical limitations, for example, could use a robot to keep track of their schedules, carry the laundry or groceries, or to follow them around with pills, glasses, phone, or a TV remote control. The robots could also check whether their owners took their pills and if not, send a text message to a caregiver, or even call for help.
The robots [see photo below], have have multiple microcontrollers and a computer running Linux. They use voice recognition to respond to their owner's commands, which according to the Gundersons enables "a very smooth interaction, much like giving commands to a very intelligent, well trained dog."
The Gundersons even programmed the robots with personalities and tempers, so they respond better if you're nice to them, saying "please" and "thank you."
To sense the environment, the robots use arrays of sonars. The sonars send data to the robot's "cybernetic brain," as the couple calls it, which processes the data to identify objects such as chairs, tables, corners, and most important, people. The robots can also recognize places like a kitchen or living room, because each has a specific sonar signature.
Equipped with two motors, the robots can move at up to 9.6 kilometers per hour [6 miles per hour], though they drive much slower when they sense objects around them. An encoder keeps track of their position.
Gamma Two plans to build the robots as customers order them. Each will cost from US $12,000 to 20,000 and take six to eight weeks to assemble. You can also rent them for parties and events.
Jim and Louise, who've been married for 27 years, form a two-headed entity with a mean IQ and deep knowledge of things as varied as tiger salamanders and extreme programming. They hold several patents and have presented numerous papers in IEEE conferences.
"A lot of our ideas come out of animated discussions between my husband and I," says Louise, the president and CEO. "We are a team up to and including the fact that we team code together."
The two met in San Francisco in the '70s while Louise was studying chemistry at UC Berkeley. They later moved to Denver where Louise received her master's in environmental science. Twenty years later, they traveled to the University of Virginia to pursue doctorates in computer science (Jim) and systems engineering (Louise). In 1997 they formed Gunderson and Gunderson, which later became Gamma Two.
The Gundersons believe there's a major obstacle preventing robots from becoming practical in daily applications. The problem, they claim, is a chasm between a robot's ability to sense the world and its ability to reason about it.
Humans and animals can perceive things, create mental models, perform manipulations and computations using those models, and then plan and execute actions. A gap between sensing and reasoning is preventing robots from doing the same. The two wrote a book detailing their ideas titled, "Robots, Reasoning, and Reification,” published by Springer in 2008.
Louise, who also studied biology, applies her insight about living systems to her study of robotics. The robots she and Jim design have hardware and software components structured as in a living creature, with a "robotic cerebellum" to handle movement and a "reification engine" to classify objects, for example.
Gamma Two is located in the Arts District of Denver and on the first Friday of every month the Gundersons open up their lab as if it were a gallery. For over a year they've been loading trays of canapes onto Wilma and Basil, which mingle among guests serving crab puffs, cheese, and crackers.
If you are in Denver this August, be sure to stop by the Gamma Two labs. To celebrate the life of Andy Warhol, Jim and Louise will paint their robots' panels with Warhol-like designs. Be the first of your friends to say you've tried robotically served hors d'oeuvres.
Outside of New York City’s Eyebeam studio, an artist's hub dedicated to the convergence of art and technology, two women pause to see a pen doodling across a canvas behind a window. When they touch little circles on the glass, the pen changes direction.
“What’s this?” they ask. Then they read the description. This is a SADbot.
SADbot, or Seasonally Affected Drawing Robot, is a solar-powered, interactive drawing machine created by Eyebeam artists Dustyn Roberts and Ben Leduc-Mills. The contraption was on display this month at Eyebeam's Window Gallery in Chelsea.
"People are only happy when it's sunny," says Roberts. "Just like our robot."
When the sky is dark, SADbot stops doodling and "goes to sleep." But when the sun is out, SADbot lets people interact with it and doodles across a large canvas.
SADbot uses an Arduino microcontroller, four photocell sensors, a battery, and two stepper motors to control two cables attached to a pen. The electronics gets power from solar panels on the building's roof. But light not only powers the installation -- it also affects SADbot's behavior.
The interactive part occurs when a person stands in front of SADbot and covers up one of its photocell sensors, which the SADbot registers and then changes its drawing direction. By covering the sensors in a determined sequence, a person could do his or her own drawings.
But after checking the gallery's window where SADbot was to be installed, Roberts and Leduc-Mills noticed a problem. The window doesn't get much sunlight -- which would make SADbot, well, sad.
No problem. The artists built a rooftop mirror array to direct sunlight to a fixed mirror hanging off the ledge, which reflects light down to the gallery window.
If none of the photocells are covered, SADbot draws according to the programmed algorithm -- in the current case, small movements in random directions.
"At the moment its aesthetic is very small, random movements, or doodles," says Leduc-Mills. Since the project has been up, they've been filling up one canvas with doodles per day, which tells them that SADbot has received a lot of interaction.
Leduc-Mills wanted to create an interactive project that people could influence from the sidewalk so he took his ideas to Roberts, a mechanical engineer, and SADbot was born.
To build SADbot, the duo raised over US $1,000 in funding on Kickstarter.com, an innovative project-funding site, which paid for all of the bot's components. Depending on the size of the donation, backers of the SADbot project received SADbot drawings, mini SADbot DIY kits, and fully built miniSADbots.
SADbot uses open source platforms like Arduino, Processing, and EasyDriver motor boards, so it's easy for you to build your own SADbot!
What if your desk lamp could not only shine light but also project online content onto your workspace? LuminAR is an augmented reality project from MIT's Media Lab that combines robotics and gestural interfaces in an everyday household item.
Developed by Natan Linder and Pattie Maes from the Fluid Interfaces Group, the device consists of two parts: a bulb and a lamp. The LuminAR Bulb can be screwed into a standard incandescent light fixture and contains a pico projector, camera, and a compact computer with wireless access to the Net. The lamp fixture, meanwhile, is a a rotating base with a multi-jointed robot arm that can move to different positions by following user gestures.
The bulb's camera tracks hand positions while the projector streams online content to different areas of the desktop. The two turn a desk into an interactive surface. The robot can also be taught to remember preferred areas to project content or digital tools such as an email application or a virtual keyboard, as seen in the video below.
The project is similar to the Sixth Sense by Pranav Mistry, also of the Fluid group, and other gestural interfaces that combine hand tracking with content projection. The difference is the form factor. The LuminAR Bulb could have wider appeal because it can be used with any ordinary desk lamp, though it would then lack robotic functions.
Still, it's an innovative way to free computing from the mouse-and-keyboard box and embed it in the environment. I wonder whether the projector is powerful enough to work well on a brightly lit desktop, and whether the robotic arm might misinterpret an involuntary gesture like sneezing and do something undesirable. Or it might hand you a tissue.
The robot has different configurations, including one with wheels. The space version has a head, torso, and arms, but no wheels or legs, because it will be mounted on a spacecraft or satellite.
The goal is to use Justin to repair or refuel satellites that need to be serviced. Its creators say that ideally the robot would work autonomously. To replace a module or refuel, for example, you'd just press a button and the robot would do the rest.
But that's a long-term goal. For now, the researchers are relying on another approach: robotic telepresence. A human operator controls the robot from Earth, using a head-mounted display and a kind of arm exoskeleton. That way the operator can see what the robot sees and also feel the forces the robot is experiencing.
Justin's head has two cameras, used for stereoscopic vision, which means the operator can get a sense of depth when manipulating the arms. And the arms and fingers have force and torque sensors, to provide feedback to the operator, who are able to know if, say, a screw is tight.
Watch the video to see a reporter operating the robot, which, he quips, probably "costs much more than what I can earn my entire life."
Humans aren't the only ones playing soccer right now. In just two days, robots from world-renowned universities will compete in Singapore for RoboCup 2010. This is the other World Cup, where players range from 15-centimeter tall Wall-E-like bots to adult-sized advanced humanoids.
The RoboCup, now in its 14th edition, is the world’s largest robotics and artificial intelligence competition with more than 400 teams from dozens of countries. The idea is to use the soccer bots to advance research in machine vision, multi-agent collaboration, real-time reasoning, sensor-fusion, and other areas of robotics and AI.
But its participants also aim to develop autonomous soccer playing robots that will one day be able to play against humans. The RoboCup's mission statement:
By 2050, a team of fully autonomous humanoid robot soccer players shall win the game, complying with the official rule of the FIFA, against the winner of the most recent World Cup.
It may seem far-fetched that robots will ever be able to compete with the likes of Messi or Kaká but 40 years is a long time in terms of technology. And what's wrong about dreaming big? Just think of the days when people would say a computer would never beat humans in chess -- until IBM's Deep Blue did just that in 1997. For now researchers explore fundamental questions in robot development: How well can robots move and think on their feet? And how well can they score goals? But maybe soon they'll be building PeléBot.
So check out some of this year's top players below, and let the games begin!
1) Led by Professor Manuela Veloso, the Carnegie Mellon University team (known to some as the Brazil of robot soccer) has developed a physics based motion planning AI for his dribbling bots. Using a dual-camera overhead vision system, the roboticists have programmed their robots to take into account the physics of ball movements, so the control algorithms can better predict where the ball is going to be and then position the robots accordingly.
2) RoboErectus was developed at the Robotics Center of Singapore Polytechnic. This 12-degrees-of-freedom robot has three processors for vision, control, and AI, as well as three sensors: an USB camera to capture images, a tilt sensor to detect falls, and a compass to track direction. The vision processor performs recognition and tracking of objects like the ball, goal, field lines, goal post, teammates, and opponents. And he's got some pretty sweet moves.
3) Dutch Robotics, an initiative from TU Delft, TU Eindhoven and University of Twente, presents TUlip, a humanoid soccer robot to compete in the adult size league. TUlip, 1.2 meter tall and with 14 degrees of freedom, looks a bit unsteady but this video shows it has skills. And the music is very inspirational.
4) Led by Professor Dennis Hong, Virginia Tech's Team DARwIn is always a strong contender. DARwIn, if you're wondering, stands for Dynamic Anthropomorphic Robot with Intelligence. Their robot, DARwIn IV, stands 56 cm tall and has 20 degrees of freedom. It's powered by a CompuLab fit-PC2 and multiple Robotis Dynamixel servos.
5) The UPennalizers compete in the standard platform league, in which all teams use the same robot platform, Aldebaran's Nao humanoid, so the challenge becomes programming the best players and team strategies. Teammate Aylin Caliskan says that robots beating humans by 2050 "should be 100 percent possible."
Without all the swearing, pushing, and penalties, RoboCup might not be as exciting as the World Cup in South Africa, but it's hard not to be inspired by this bunch of ball-kicking bots.
With only a small team of developers and a week's worth of development, the PR2 can now play pool! The "Poolshark" team started last Monday and began making shots on Friday. The PR2 won't be hustling you in pool halls anytime soon, but it pocketed five shots on Friday before the team decided it was time to celebrate.
The Poolshark team dealt with numerous technical challenges throughout the week: engineering a special grip and bridge so the PR2 could hold the cue, a ball detector, table localization, visualizations and input tools, shot selector, and more.
A big thanks goes to Alon Altman for his open-source FastFiz billiards library. FastFiz is a physics and rules engine for billiards that the Poolshark team used to select which shots the PR2 should take. The Poolshark team has released its own code in the billiards stack.
June is "Hackathon" month, so there are two more one-week hackathons to come: pushing a cart and fetching a drink from a refrigerator. It's one down, two to go!
These look like promising applications. Or as Evan Ackerman at BotJunkie aptly puts it:
A robot that can play pool and fetch me beer? Hellooooo new best friend.
When Canadian filmmaker Rob Spence was a kid, he would peer through the bionic eye of his Six Million Dollar Man action figure. After a shooting accident left him partially blind, he decided to create his own electronic eye. Now he calls himself Eyeborg.
Spence's bionic eye contains a battery-powered, wireless video camera. Not only can he record everything he sees just by looking around, but soon people will be able to log on to his video feed and view the world through his right eye.
Spence and his collaborators -- Kosta Grammatis, John Polanski, Martin Ling, Phil Bowen, and camera firm OmniVision -- managed to get a prototype working last year. Time magazine named it one of the best inventions of 2009. Now the group is developing a version that offers a clearer picture.
I recently met with Spence in Toronto. In unreleased footage (see screenshot below) that Spence gave me, he demos the prosthetic as a colleague filmed. The feed from the eye-cam is picked up by a wire antenna that Spence held to his cheek, and relayed to a flatscreen TV in the background.
The prototype in the video provides low-res images, but an authentic experience of literally seeing through someone else's perspective. The image is somewhat jerky and overhung by huge eyelashes; a blink throws everything out of whack for a half-second.
Rob Spence demonstrates his bionic camera eye. A wireless camera in a specially designed prosthetic relays a video feed to an antenna that he presses to his cheek. The feed is shown on the screen behind him as a colleague films. Image: Rob Spence
The bionic eye is simply designed, and components are constantly changing. It basically contains a 1.5mm-square, low-res video camera, a small round printed circuit board, video transmitter, and a 3-volt rechargeable Varta microbattery. The components are contained in resealable clear acrylic used in false eyes, but it has two holes for wires to recharge the battery.
"I can recharge my eye via USB off my laptop," says Spence.
The Eyeborg prototype in the video, the third, can only work for an hour an a half on a fully charged battery. Its transmitter is quite weak, so Spence has to hold a receiving antenna to his cheek to get a clear signal. He muses that he should build a Seven of Nine-style eyepiece to house it. He's experimenting with a new prototype that has a stronger transmitter, other frequencies and a booster on the receiver.
"Unlike you humans, I can continue to upgrade," Spence quips. "Yes, I'm a cyborg. But I think that any technology -- even clothing -- makes people cyborgs."
Spence loves to ham it up as Eyeborg, installing a red, laser-like LED light in one version of the prototype and pulling on a 1970s track suit to become Steve Austin (see the video below). But he's serious about using his camera eye to get Internet users to view the world through his eye, and is developing an Eyeborg app that may feature augmented reality functions.
"In today's world, you have Facebook and camera eyes," he says. "Tomorrow, we'll have collective consciousness and the Borg. It's a collective robot consciousness. I believe that's a genuine modern concern."
Eyeborg seems content to lead us into our robot future.