Robot-Human Crosstraining Makes Better Partners
MIT Interactive Robotics Group
A robot arm interprets the hand movements of Stefanos Nikolaidis, assisting him in a collaborative place-and-drill task.
Paramount Pictures Corporation
June 29, 2011 --
In the movie "Transformers: Dark Side of the Moon," Autobots, transforming robots from the planet Cybertron, work alongside humans to battle the Decepticons, a rival group of robots bent on destroying the universe. The latest Transformers installment may just be a work of fiction, but would real-life robots on Earth be up to the task of saving the planet from the threat of an overwhelming destructive force? Of course not, but that doesn't mean they won't get there one day. Until then, we present the next best thing in this slideshow of incredible real-life robots.
Alex Kossett and Nikolaos Papanikolopoulos, U
We begin the slideshow with an actual real-life transforming robot. Although it doesn't tranform into a sentient, humanoid machine, this robot can transform from what appears to be an elaborate motorized rolling pin to a helicopter when the terrain starts to get rough.
Take a look at this flying robot in action here. BLOG: Rolling Robot Transforms into a Helicopter
What's more terrifying than a hungry, fully grown cheetah chasing you down? How about if it were made of metal? Meet CHEETAH, a robot designed by Boston Dynamics for one purpose: to hunt you down like an animal. Once built, this robot will be fast, agile and strong enough to chase down, catch and subdue even the fastest human runners. This robot is one of a line of prototypes known as "terror bots." An appropriate name, especially if you see this guy biting at your heels as you run full speed. BLOG: TERROR BOTS BEING DESIGNED TO HUNT YOU DOWN
If this robot reminds you of Scorponok from the "Transformers" film series, you wouldn't be far off. Designed by researchers at the University of Bielefeld, this advanced walking robot is based on a rather simple creature: an insect. HECTOR, short for Hexapod Cognitive autonomously Operating Robot, has six legs and elastic joints that allow its motions to mimic muscle movement. This construction allows the robot to navigate over uneven terrain. At a little more than three feet long and weighing in at 26 pounds, this robot probably won't be involved in any world domination schemes anytime soon. BLOG: HECTOR THE WALKING ROBOT INSPIRED BY INSECTS
Snake Robot to the Rescue
Cheetahs and insects aren't the only animals inspiring robotics' engineers. Mechanical snakes are also being designed to mimic their mechanical counterparts. Unlike CHEETAH, which is made to hunt you down, this snake robot, created by researchers at Georgia Tech Univeristy, is actually designed to come to the rescue. Their unique body shape allows them to burrow through uneven soil. With this unique feature, emergency responders could deploy these robots after a particularly devastating natural disaster, such as an earthquake, when victims are buried and out of reach. BLOG: SNAKE-LIKE ROBOT SWIMS TO THE RESCUE
Robots may not yet be able to conquer the Earth, but what about the wide world of sports? They're already playing soccer and tossing baseballs. Now it looks like they're competing in marathons. (Well, robot marathons anyway.) Last February, five bipedal robots ran a non-stop 26.2-mile race on a 100-meter indoor track in Osaka, Japan. But don't expect these machines to compete with humans anytime soon. Robovie-PC, the winner of the race, finished in just under 55 hours. BIG PIC: TOY-SIZED HUMANOID WINS ROBOT MARATHAN
Ingmar Posner, Oxford Mobile Robotics Group
No this robot can't run or jump or slither or swim. So what can it do? This machine, known as Marge, has a very different ability entirely: It can read -- and it can learn. Marge may just look like a Tonka truck underneath a coffee pot, but this machine is actually smart enough to read The New York Times and BBC Online. It is even a skilled editor and can identify and correct misspellings. And because Marge's brains are built in its software, not its hardware, this same programming could make its way into other devices, such as cell phones. BLOG: ROBOT CAN READ, LEARN LIKE A HUMAN
Can anyone really tell the difference between right and wrong? Well, this robot can. This robot's ethical code is based on a software program modeled on an approach to ethics developed in 1930 by Scottish philosopher David Ross. As a result, this robot is designed to take the moral high ground -- and will tell on you if you're doing something wrong. Sure this robot doesn't have the firepower of an Ironhide or a Starscream, but a judgmental expression and a jittery nod of disapproval can be just as damaging. BLOG: ROBOT MAKES ETHICAL DECISIONS
Meet Robonaut 2, a humanoid robot that could one day be your co-worker -- or even take your job. Yes, this robot has everything any employer looks for in a diligent worker: It's capable and tireless, and it doesn't ever need lunch or bathroom breaks. In fact, you'll never guess where this robot is currently employed: the International Space Station. Robonaut 2 is currently working along astronauts, helping with basic maintenance tasks, such as cleaning. DNEWS VIDEO: MEET 'ROBONAUT 2,' YOUR FUTURE CO-WORKER
Robonaut 2 isn't the only robot on the space station with a job. Dextre, the Canadian robot that lives outside the International Space Station, has been tasked with refueling satellites while in low-Earth orbit. The robot will also be capable of performing minor repairs. In other words, this robot is essentially a space mechanic. Although a gas-pumping robot may not seem like much, Dextre could pave the way for an entirely new industry for satellite servicing. NEWS: SPACE STATION ROBOT LANDS A JOB
Humans and robots work better together if they can swap roles and learn from each other, a finding that may pave the way for the use of more robots on assembly lines, in space, on farms or in any kind of dangerous, dirty or boring operation.
“Cross-training,” or switching jobs, improves efficiency, as well as robots' confidence and the trust of human masters, researchers at the Massachusetts Institute of Technology found.
“We want robots to work alongside people,” said Julie Shah, assistant professor of aeronautics and astronautics at MIT and head of the Interactive Robotics Group in the Computer Science and Artificial Intelligence Laboratory.
“There’s a trend of removing the cages around robots and integrating them with manual work. That’s coming around with greater computation and sensing. We don’t want the robots to do the same task over and over again. They can’t if they are going to work in this dynamic environment. What we're interested is how we can make them smart enough to assist people.”
Shah and graduate student Stefanos Nikolaidis set up a simple experiment that involved 36 people coming into a lab and working with a robot arm to drill three screws into a board. The actual drilling was simulated. Just like an assembly line, they placed the screws, the board moved down the line and the robot pressed in the fasteners.
Shah said the volunteers all had different ways of working with the robot arm. “Some people didn’t like the robot; they were nervous,” she said. “They preferred to put in all three screws, stand back, and watch the robot drill. In other cases, people wanted to do it as fast as possible, so the robot drilled as fast as the screws were placed.”
By switching job roles, the robot was able to learn more quickly and the human began to trust the bot more. This human-robot cross-training worked better than giving positive or negative rewards to the robot.
“Cross training results in better teamwork,” Shah said.
The other big issue is forging trust between the two, according to Nikolaidis. This is important if the job involves tasks such as lifting heavy objects or performing dangerous work.
“If a person has learned to do something for 20 years, it’s hard to change and to place trust in a machine,” Nikolaidis said. “It’s one of the biggest barriers we are facing.”
The MIT researchers say for robots to become more like helpers who can take and perform commands than simple lifters or drillers, they will have to be able to learn the way humans do. And that means new kinds of expectations for both members of the human-robot team.
“From a teamwork perspective, the way we train robots today looks nothing like how medical or military teams train,” Shah said. “If we can make a robot train together with a person, the human-robot team will work better.”
A paper detailing their work will be presented at the International Conference on Human-Robot Interaction in Tokyo in March.