AN35 Conception de documents
Activité de groupe et rédaction finale
Vous êtes au service communication de la division robotique d’une importante société
informatique. Rédigez et imprimez une présentation pour une réunion d’information chez
Airbus, intégrant les progrès décrits dans les articles ci-dessous. Vous reprendrez à votre
compte les innovations et données des documents pour proposer votre collaboration avec cette
société. Format environ 7 à 8 écrans en présentation type Powerpoint.
Robotic assistants may adapt to humans in the factory
New algorithm allows robots and humans to work side by side.
Jennifer Chu, MIT News Office
In today’s manufacturing plants, the division of labor between humans and robots is quite clear: Large,
automated robots are typically cordoned off in metal cages, manipulating heavy machinery and
performing repetitive tasks, while humans work in less hazardous areas on jobs requiring finer detail.
But according to Julie Shah, the Boeing Career Development Assistant Professor of Aeronautics and
Astronautics at MIT, the factory floor of the future may host humans and robots working side by side,
each helping the other in common tasks. Shah envisions robotic assistants performing tasks that would
otherwise hinder a human’s efficiency, particularly in airplane manufacturing.
“If the robot can provide tools and materials so the person doesn’t have to walk over to pick up parts
and walk back to the plane, you can significantly reduce the idle time of the person,” says Shah, who
leads the Interactive Robotics Group in MIT’s Computer Science and Artificial Intelligence Laboratory
(CSAIL). “It’s really hard to make robots do careful refinishing tasks that people do really well. But
providing robotic assistants to do the non-value-added work can actually increase the productivity of
the overall factory.”
A robot working in isolation has to simply follow a set of preprogrammed instructions to perform a
repetitive task. But working with humans is a different matter: For example, each mechanic working at
the same station at an aircraft assembly plant may prefer to work differently — and Shah says a robotic
assistant would have to effortlessly adapt to an individual’s particular style to be of any practical use.
Now Shah and her colleagues at MIT have devised an algorithm that enables a robot to quickly learn an
individual’s preference for a certain task, and adapt accordingly to help complete the task. The group is
using the algorithm in simulations to train robots and humans to work together, and will present its
findings at the Robotics: Science and Systems Conference in Sydney in July.
“It’s an interesting machine-learning human-factors problem,” Shah says. “Using this algorithm, we
can significantly improve the robot’s understanding of what the person’s next likely actions are.”
Taking wing
As a test case, Shah’s team looked at spar assembly, a process of building the main structural element
of an aircraft’s wing. In the typical manufacturing process, two pieces of the wing are aligned. Once in
place, a mechanic applies sealant to predrilled holes, hammers bolts into the holes to secure the two
pieces, then wipes away excess sealant. The entire process can be highly individualized: For example,
one mechanic may choose to apply sealant to every hole before hammering in bolts, while another may
like to completely finish one hole before moving on to the next. The only constraint is the sealant,
which dries within three minutes.
The researchers say robots such as FRIDA, designed by Swiss robotics company ABB, may be
programmed to help in the spar-assembly process. FRIDA is a flexible robot with two arms capable of
a wide range of motion that Shah says can be manipulated to either fasten bolts or paint sealant into
holes, depending on a human’s preferences.
To enable such a robot to anticipate a human’s actions, the group first developed a computational
model in the form of a decision tree. Each branch along the tree represents a choice that a mechanic
may make — for example, continue to hammer a bolt after applying sealant, or apply sealant to the
next hole?
“If the robot places the bolt, how sure is it that the person will then hammer the bolt, or just wait for the
robot to place the next bolt?” Shah says. “There are many branches.”
Using the model, the group performed human experiments, training a laboratory robot to observe an
individual’s chain of preferences. Once the robot learned a person’s preferred order of tasks, it then
quickly adapted, either applying sealant or fastening a bolt according to a person’s particular style of
work.
Working side by side
Shah says in a real-life manufacturing setting, she envisions robots and humans undergoing an initial
training session off the factory floor. Once the robot learns a person’s work habits, its factory
counterpart can be programmed to recognize that same person, and initialize the appropriate task plan.
Shah adds that many workers in existing plants wear radio-frequency identification (RFID) tags — a
potential way for robots to identify individuals.
Steve Derby, associate professor and co-director of the Flexible Manufacturing Center at Rensselaer
Polytechnic Institute, says the group’s adaptive algorithm moves the field of robotics one step closer to
true collaboration between humans and robots.
“The evolution of the robot itself has been way too slow on all fronts, whether on mechanical design,
controls or programming interface,” Derby says. “I think this paper is important — it fits in with the
whole spectrum of things that need to happen in getting people and robots to work next to each other.”
Shah says robotic assistants may also be programmed to help in medical settings. For instance, a robot
may be trained to monitor lengthy procedures in an operating room and anticipate a surgeon’s needs,
handing over scalpels and gauze, depending on a doctor’s preference. While such a scenario may be
years away, robots and humans may eventually work side by side, with the right algorithms.
“We have hardware, sensing, and can do manipulation and vision, but unless the robot really develops
an almost seamless understanding of how it can help the person, the person’s just going to get
frustrated and say, ‘Never mind, I’ll just go pick up the piece myself,’” Shah says.
This research was supported in part by Boeing Research and Technology and conducted in
collaboration with ABB.
Freed From Its Cage, the Gentler Robot
Rethink Robotics
The Baxter robot from Rethink Robotics.
By ANNE EISENBERG / Published: March 30, 2013
FACTORY robots are usually caged off from humans on the assembly line lest the machines’ powerful
steel arms deliver an accidental, bone-crunching right hook.
William Litant/M.I.T.
Prof. Julie Shah of M.I.T. and two graduate students, Ron Wilcox and Matthew Gombolay, ran a cross-training experiment.
But now, gentler industrial robots, designed to work and play well with others, are coming out from behind their protective
fences to work shoulder-to-shoulder with people. It’s an advance made possible by sophisticated algorithms and
improvements in sensing technologies like computer vision.
The key to these new robots is the ability to respond more flexibly, anticipating and adjusting to what humans want. That is
in contrast to earlier generations of robots that often required extensive programming to change the smallest details of their
routine, said Henrik Christensen, director of the robotics program at the Georgia Institute of Technology.
“Researchers in labs worldwide are building robots that can predict what you’ll do next and be ready to give you the best
possible assistance,” he said.
One of those researchers is Julie A. Shah, an assistant professor in the department of aeronautics and astronautics at the
Massachusetts Institute of Technology. Dr. Shah once taught robots to do tasks the old way: by hitting a button that
essentially told them “good,” “bad” or “neutral” as they did each part of a job. Now she has added a technique called crosstraining, in which robots and humans exchange roles, learning a thing or two from each other in the process.
In a recent study, Dr. Shah and a student had human-robot teams perform a chore borrowed from the assembly line: the
humans placed screws and the robots did the drilling. Then the teammates exchanged jobs and the robots observed the
humans drill.
“The robot gathers information on how the person does the drilling,” adding that information to its algorithms, Dr. Shah
said. “The robot isn’t learning one optimal way to drill. Instead it is learning a teammate’s preferences, and how to
cooperate.”
When the cross-trained teams resumed their original roles, both robots and people did their jobs more efficiently, the study
found. The time that the humans were idle while waiting for the robot to finish a task dropped 41 percent and the time that
humans and robots worked simultaneously increased 71 percent, when compared with teams working with robots trained
the old way.
“This is a fascinating application of cross-training,” said Andrea Thomaz, an assistant professor of interactive computing at
Georgia Tech. “By learning the human’s role, the robot can better anticipate actions and be a better partner, even if in the
end it will only do one role.”
The humans on the teams also improved their teamwork skills, said Illah R. Nourbakhsh, professor of robotics at Carnegie
Mellon University and author of the book “Robot Futures,” published this month by M.I.T. Press. “In the future, this idea of
cross-training will turn out to be really important as robots start to work shoulder-to-shoulder with us,” he said. “We are not
very good at adopting the point of view of a robot. This study showed that we can learn, though, with the right signals.”
Dr. Christensen of Georgia Tech said: “Robots of the future won’t just be in manufacturing. Almost any area could have a
robot that would help make our life easier,” whether “lifting patients in hospital beds or helping at home.
“But they have to be safe, and they have to have the kind of anticipation that Julie Shah is working on, because they have to
be able to automatically figure out what we need help with,” he said.
Gentle, helpful robots aren’t just being created in labs; they are also arriving in the marketplace. Since January, Rethink
Robotics of Boston has been sending customers its two-armed robot called Baxter, which can work uncaged, moving among
people. “We are shipping robots every day and have a backlog of orders of about three months,” said Rodney Brooks,
Rethink’s founder, chairman and chief technology officer.
Baxter, which costs $22,000, can lift objects from a conveyor belt. “You don’t have to tell it the exact velocity,” Dr. Brooks
said. “It sees objects and grabs them, matching its speed to the speed of the object.”
Baxter is used in manufacturing plants and shops of varying sizes. One example is the Rodon Group, a plastic injection
molding company in Hatfield, Pa., where Baxter packs boxes on the factory floor.
Baxter’s cameras inspect what is to be lifted, recognizing an object from many angles. In the coming year, Baxter will be
able to grab objects not only from above, but also from the side, putting them into a milling machine, for example, and
pressing the “go” button. It will also be able to connect with other machines, to synchronize tasks.
“Baxter is a great starting point for this new generation of robots,” said Dr. Christensen of Georgia Tech, who has no
connection to Rethink Robotics’ work, “making the technology accessible to companies that before would have had to pay
hundreds of thousands of dollars.”
“He’s opening up a new market,” Dr. Christensen said of Baxter’s work.
Baxter is not the only unfenced robot on the assembly line. A Danish company, Universal Robots, for example, sells a onearmed robot for $33,000 that can also be used without a cage.
IMPRESSIVE as the new robots are, they will soon have even more advanced skills, said Stefan Schaal, a professor of
computer science, neuroscience and biomedical engineering at the University of Southern California and a director of the
Max Planck Institute for Intelligent Systems in Germany. In the future, robots will be able to go onto the Internet and
exchange information, leading to vast gains in what they can accomplish.
“It will take time before we get there,” he Schaal said, “but it will happen.”
E-mail: novelties@nytimes.com.