At Mashup Camp, Geeks Plot Future of Web
CNet (01/18/07) LaMonica, Martin
Mashup developers convened in Cambridge, Mass., for the third annual
"Mashup Camp" where they discussed the current and future states of
mashups, as well as the legal barriers they face. "Mashups are the most
fun [type of Web development] just because there's so much potential," said
Alan Taylor, one of about 200 developers and technologists attending Mashup
Camp. "The biggest barriers have been artificial barriers, legal
barriers," added Taylor. Although major software vendors have created
service-oriented architecture (SOA), intricate suites of Web services,
standards, and infrastructure based on a modular format, mashup developers
have concentrated on speed and simplicity, using streamlined methods for
patching together different Web sites. "This is taking the SOA idea and
applying it to the mess that is the real world," explained Web developer
Joe Radcliff. "You can grab pieces from here and there. It makes it a lot
more informal, which increases adoption." Mashup developers are now being
courted by established technology providers, as the success of mashups
compared with the relative struggles of Web services and SOAs was stressed
at the conference. Application programming interfaces (APIs) can normally
be taken from Web sites without asking permission, but the biggest fear of
a mashup creator is a cease-and-desist order. Ironically, mashups normally
get shut down when they become very popular, since they begin to use up too
much of a Web service's resources, such as bandwidth. As a result, the
field has taken a great interest in licensing and commercialization, even
dedicating a session at the conference to these concerns. "The idea is
that we want to make this ecosystem sustainable, not just go to fun
conferences in cool places," said Eran Shir, co-founder of a company that
helps mashup developers take data from Web sites that do not offer APIs.
Click Here to View Full Article
to the top
Touch Screens for Many Fingers
Technology Review (01/18/07) Greene, Kate
The unveiling of the Apple iPhone presented the public with a touch screen
that allows more than one finger at a time to be used, but Jeff Han
believes his own multi-touch screens will make the technology more useful.
This month, his new company, Perceptive Pixel, will ship its first
wall-sized touch screen this month, which allows several people to gather
around the screen and "become collaborators," according to the New York
University consulting research scientist. "The new iPhone is too small to
be a very interesting multi-touch device," Han adds. The inexpensive
screens Han has developed can accommodate as many as 10, 20, or more
fingers, and he envisions them becoming embedded in tables and digital
walls. Multi-touch technology has been explored since the 1980s, and while
it never completely went away, Han's work has shown that the technology is
emerging in new ways. Han's displays consist of a clear piece of
six-millimeter thick acrylic illuminated with infrared light by
light-emitting diodes attached at the edges. Light from the diodes
exhibits internal reflection within the acrylic, but when a finger or other
object makes contact with the acrylic, the internally reflected light
diffuses at this point and scatters outside the acrylic. A camera behind
the surface of the acrylic captures this light, which is then translated in
real time as touches or strokes, using simple image-processing software.
"For almost two decades, we've been trapped by the tyranny of the screen,
the mouse, and the keyboard," says Northwestern University professor and
author of "The Design of the Future" Don Norman. "It's nice to think we're
breaking away from that and going toward touch-screen manipulation in the
real physical world." Some researchers are now working on utilizing
haptics to provide the user of a multi-touch screen with sensations
corresponding to their contact inputs.
Click Here to View Full Article
to the top
Research Removes Major Obstacle from Mass Production of
Tiny Circuits
EurekAlert (01/17/07)
A Princeton-led research team has found a way to possibly eliminate the
tiny air bubbles that have troubled an innovative technique used to form
intricate circuits, marking a major step in the direction of producing
smaller, less expensive microchips. Princeton University Joseph C. Eglin
professor of engineering Stephen Chou, who led the research, developed a
method for patterning computer chips using a nanometer-scale mold in the
1990s, known as nanoimprinting. This technique made it possible to produce
circuits and devices with features measuring about one nanometer long, over
10 times smaller than the mass-produced chips today, and less expensive.
Beyond computer chips, nanoimprinting allowed for advancements in
manufacturing nanodevices used in optics, magnetic data storage, and
biotechnology. A modification of this technique led to dispensing-based
nanoimprinting, where liquid droplets on the surface of a silicon wafer are
pressed into a pattern, which rapidly solidifies to form circuitry.
Although dispensing-based nanoimprinting was very attractive to
manufacturers because it did not have to be done in a vacuum chamber, its
widespread implementation was obstructed by the formation of gas bubbles
that can disturb an intended pattern. Chou's team studied the cause of
these bubbles and found that by increasing the pressure of imprinting or
using liquids with higher air solubility, the likelihood of the bubbles
dissolving before the liquid solidified could be significantly increased.
"This is an important step because to benefit from the technology of
nanoimprinting you need to be able to use it in mass manufacturing at low
cost," said Chou.
Click Here to View Full Article
to the top
A Good Year for the IT Professional's Salary
Computerworld Australia (01/17/07) Tay, Liz
Australia's Minister for Education, Science, and Training, Julie Bishop,
is considering spending $52 billion (AUD) over the next half-decade to
raise the public profile of science and technology professionals. Her plan
comes at a time in which the country is facing an IT skills shortage, and
observers believe many students are shunning mathematics and science
studies because they do not have a healthy view of the contributions of
scientists and technology professionals to society. According to the
latest job sector update from APESMA (Association of Professional
Engineers, Scientists, and Managers, Australia), first-year tertiary IT
enrollments are down by up to 50 percent, and nearly two-thirds of
graduates plan to look for better job opportunities abroad. Young students
continue to express concern about the local IT job market, according to
observers, and the cost of pursuing science and technology-related studies
does not help the situation. Science studies trail only medicine and law
in cost. Companies in need of IT workers have embraced increased annual
leave, retention payments, part-time and work from home contracts and
overseas exchange as benefits. APESMA also reports that $85,610 is the
average IT salary, with private and public sector pay having risen 5
percent and 3.8 percent, respectively.
Click Here to View Full Article
to the top
U.S.: No Net Governance Changes Expected
CNet (01/16/07) Broache, Anne
Senior White House officials said yesterday that they do not see U.S.
control over Internet governance as being a major issue in the future.
Their optimism was fueled by comments made by new International
Telecommunications Union Secretary-General Hamadoun Toure, who said that
under his leadership the organization will not seek "to take over the
governance of Internet." Rather, Toure says ITU will focus on
cybersecurity and bridging the digital gap. As recently as last fall's
Internet Governance Forum, representatives from developing nations
expressed their discontent with the current system of governance, saying
the U.S. held too much sway and calling for international oversight of the
system. U.S. Ambassador David Gross and Assistant Secretary of Commerce
John Kneuer said Toure's comments assure that the United Nations will let
ICANN continue to oversee technical management of the Internet. Gross, in
charge of information policy and coordinating international communications,
said the ITU's new position is "very much in harmony with our views."
Meanwhile, Kneuer confirmed that the U.S. still plans to eventually
privatize ICANN's operations. Kneuer said "coordinating the transition of
the [domain name system] to the private sector...remains important for
us."
Click Here to View Full Article
to the top
Ethics Dilemma in Killer Bots
The Australian (01/16/07) P. 29; Argy, Philip
Guard robots being deployed on the northern border of South Korea are
capable of firing on human targets without receiving any direct commands
from humans, and have brought up many important ethical questions. Each
Intelligent Surveillance and Security Guard Robot will be equipped with a
daylight camera capable of identifying targets within a 4-kilometer radius,
and an infra-red night vision camera that has a range of 2 kilometers.
While humans can use a joystick and touchscreen to control the robots, they
are programmed to respond autonomously when an intruder does not provide a
correct password. The robot's responses include sounding an alarm, using
non-lethal force, or firing a machine gun or rifle; these would be the
world's first robots with such capabilities. While the manufacturer says
the robots are superior to human guards because they are immune to weather
conditions and fatigue, many point out that a human soldier could utilize
discretion and understand the consequences of his actions. Australian
Computer Society's Mike Bowern expresses concerns over the potential for
"software and hardware defects" to "influence the robot's conduct." He
also points out that little is known of the ethical considerations taken by
the robots' designers, or any code they must follow, since Korea doesn't
have an independent professional association such as the ACM or the ACS,
and the Korean Ministry of Information and Communication seems to place
greater importance on technical aspects than it does on professional or
ethical concerns. Many worry that these robots could eventually be sold to
private customers. Computer ethicist James Moor points out the robots
could not be held legally or morally responsible for their actions, leaving
such responsibility up to technology professionals.
Click Here to View Full Article
to the top
Interview With Bill Cheswick
Security Focus (01/15/07) Biancuzzi, Federico
In an interview with Federico Biancuzzi, Internet Mapping Project creator
and Lumeta chief scientist Bill Cheswick says useful information about
attacks could be culled through a combination of data about firewall probes
and other information about an assault on an organization, and he notes
that he prefers placing such logs in a big, cheap drop-safe. Cheswick
describes network intrusion detection systems (NIDS) as an ongoing network
monitoring effort, and notes that false negatives and false positives are a
recurring problem for the technology; potential subversion of the NIDS is
another significant minus. In terms of finding a solution to distributed
denial of service (DDoS) attacks, Cheswick says, "I see no theoretical
possibility of doing anything more than mitigating attacks, and ultimately
throwing large amounts of computing and network capacity at the problem."
Cheswick harbors doubts about intrusion prevention systems or reactive
firewalls, which on the surface seem logical but are actually difficult to
execute, and carry the danger of turning on their users through the
machinations of a clever attacker. Network security research that has
drawn Cheswick's attention or excitement includes a SANE paper at Usenix
that rethinks intranet design by shifting from an end-to-end scheme to
centralized control, which Cheswick thinks could be potentially useful for
military and corporate networks; a paper detailing a proactive Microsoft
project to find browser exploits on malevolent sites; and investigations
into the use of virtual machines such as Xen and VMware. Cheswick
anticipates the continued exploitation of susceptible machines for
underhanded money-making schemes such as spam email, phishing, and DDoS
extortion attacks, because the incentives remain strong. It is his hope
that there will be fewer vulnerable systems with the implementation of
Vista.
Click Here to View Full Article
to the top
The Bookshelf Talks With Steven Casey
American Scientist (01/07) Vol. 4, No. 1, Ross, Greg
Ergonomics and human-factor expert Steven Casey has recently released his
second book focusing on accidents caused by user error that might have been
prevented by more enlightened design efforts. The incidents detailed in
the new book, "The Atomic Chef," "make examples of instances in which
engineers or designers didn't incorporate human factors or ergonomics into
the design of the system or the product," Casey says. As technology
advances, Casey observes that a single person has more ability to disrupt a
system. He uses the term "design-induced error" rather than user error,
since he chooses to cite stories "where there is a deficiency in the system
or the actual interface that the operator uses could have been done
better." The story for which the book is named involves Japanese nuclear
scientists who disobeyed rules and accidentally created the same type of
deadly chemical combination that the rules were put in place to protect
against. Although the users were directly at fault, Casey says, "The whole
social aspect of that setting was as important as, say, a one-on-one
operator interface and these 'macroergonomic' issues, I think, will become
increasingly important in the future." His work regularly concerns
vehicles, where he has observed that migrating technology from one type of
vehicle to another is the major cause of design-induced error. While Casey
does not expect human error to ever be completely be removed from
technology, the effort is being made to minimize it by involving experts
such as himself, and users themselves, in the design process.
Click Here to View Full Article
to the top
A Nano Solution to Increasing Bandwidth
Technology Review (01/17/07) Bullis, Kevin
MIT Researchers have developed a way to remedy the fundamental problem
facing the use of photonics in communications, which could facilitate
cheaper, more intricate, and better performing optical networks. Though
photonic devices have been successfully miniaturized in the past few years,
inefficiency still plagues these devices. Light delivered through
cylindrical fiber optics gets broken up into varying orientations of light
waves, and outputs in devices at the microscale change if the light is
either horizontally or vertically polarized, so devices were made to
process only specific polarizations. For example, only horizontally
polarized light would be used, and the vertically polarized light
discarded, but this technique results in weak signals. The MIT researchers
approached this problem not by building different devices to process the
different light polarizations, but by creating a device that can convert
vertically polarized light into horizontally polarized light by gradually
rotating it. With all the light having the same polarization, it can all
be processed by identical devices, which means clear, robust signals. This
current advance only applies to photonic applications involving light with
multiple polarizations, mostly communications applications that utilize
fiber-optics. A surplus of bandwidth in past years reduced the attention
given to improving these applications, while communications demands have
been once again been increasing the need for advancement. One of the
researchers on the project, MIT electrical engineering and physics
professor Erich Ippen, explains that his team's work will help satisfy the
needs of next-generation telecommunications: "When you integrate things
like this, the complexity and the performance of the kinds of filtering we
can do are a little more advanced than the methods that are used today," he
notes.
Click Here to View Full Article
to the top
Computer Privacy in Distress
Wired News (01/17/07) Granick, Jennifer
Recent court cases have brought the question of computer privacy into the
spotlight, as it pertains to the Fourth Amendment's protection against
unreasonable search and seizure. Recent cases have proposed that border
agents can search PCs of individuals crossing the border, without
reasonable suspicion or a warrant. Though "routine" searchers are allowed
to take place without reasonable suspicion, no court has directly addressed
the question of whether searching a PC at the border is a routine or
non-routine search. Due to the amount of private information on PCs, the
length of time searches take, and the probability of finding contraband,
courts may rule that reasonable suspicion is needed for such searches.
U.S. v. Zeigler, heard in the 9th U.S. Circuit Court of Appeals, has stated
that employees of private companies have no reasonable expectation of
privacy, meaning no Fourth Amendment rights, concerning their workplace
computers. Unless defense attorneys' requests for a rehearing are granted,
the government could walk into an office without cause or a warrant and
search the entire contents of the computer of any employee. The 9th
Circuit is also trying to figure out a way to make sure authorities get the
information they need without accessing or disturbing private, unrelated
material that may be on the same disk drive. For example, in prosecuting
United States v. Comprehensive Drug Testing, the government obtained
warrants and seized databases containing drug test results for the 10
baseball players suspected of taking steroids, as well as the test results
for hundreds of other athletes, and despite a lower court ruling that said
the government must return the unrelated information, the 9th Circuit
upheld a government appeal. This case shows that warrants must not only
state what authorities can seize, but what they may not access on these
seized machines. Courts, and possibly Congress, have a complicated road
ahead in crafting a computer privacy compromise that is supported by both
privacy advocates and investigating authorities.
Click Here to View Full Article
to the top
You Talkin' to Me?
New Scientist (01/13/07) Vol. 193, No. 2586, P. 42; Buchanan, Mark
Trapping photons so that they can interact and become a quantum material
composed of light is the goal of researchers that include University of
Melbourne scientist Andrew Greentree, and such a breakthrough could perhaps
even clear a path to a practical quantum computer. Three separate groups,
including one led by Greentree, have developed models demonstrating that
the generation and testing of photon materials within several years is a
feasibility. University of Cambridge researcher Dimitris Angelakis, who
leads one of the other groups, says the approach they took involves coaxing
"light to talk to light through matter" in a manner that enables the
concurrent interaction of many photons. Their concept is the placement of
atoms inside cavities in a photonic crystal. Each cavity is tuned to
absorb precisely one photon and repel others, creating a photon blockade,
and a phase transition is caused by disturbances induced by a beam of laser
light striking each cavity. Photons can then flow unimpeded throughout the
array. A new understanding of numerous weird quantum effects could be
achieved with a "quantum simulator" comprised of interacting photons.
Greentree's team has thought up an alternative method for building photon
materials by exploiting defects in the crystal lattice of thin sheets of
diamond rather than two-level atoms in cavities.
Click Here to View Full Article
- Web Link to Publication Homepage
to the top
Factories of the Future
CIO (01/01/07) Vol. 20, No. 6, P. 43; Hapgood, Fred
Adaptive manufacturing is manufacturing that makes spur-of-the-moment
adjustments based on sudden economic changes, and it requires an
infrastructure where information processes and information management
issues are prevalent, using emergent technologies such as sensor networks
and 3D printing. The interest in adaptive manufacturing is being fueled by
several things, including a shift in economists' perception of markets and
economies from predictable to chaotic systems, and the philosophy that
managers employ natural selection as a business tool. Greater
efficiency--an essential element of adaptive manufacturing--is facilitated
by technologies such as machine vision, which involves the recognition of
objects in an image and their association with specific properties.
Potential benefits of this approach include reduced maintenance and quality
control costs. Prototyping can be accelerated through the use of desktop
manufacturing or 3D printing, a process in which a product is taken from
blueprint to three dimensions in a matter of hours via the deposition of
stacked layers by a printer. High-resolution sensor and actuator networks
enable the detection of environmental changes and the introduction of those
changes: Increasing the number of actuators translates into more control
points, while boosting the number of sensors means more focused and
intelligent control; networking the devices allows control from any
physical point. These technologies add flexibility to manufacturing and
open the way for the incorporation of new managers, partners, and other
participants into the assembly process. Adaptive manufacturing allows
little time for companies to build the skill sets they may need, and
addressing this problem is the goal of a new form of collaboration in which
short-term alliances with companies that possess the needed skills are
forged.
Click Here to View Full Article
to the top
Is Your Car Smarter Than You Are?
Popular Mechanics (12/06) Vol. 183, No. 12, P. 88; Stewart, Ben
Computerized automotive control systems are beginning to supplant drivers
as the decision-makers in such critical areas as accident avoidance.
Electronic stability control (ESC) applies pressure to individual brakes to
prevent oversteering and understeering. Logically, the next step is to
incorporate within vehicles software and sensors that anticipate hazardous
conditions and take over if the drivers' avoidance tactics are determined
to be insufficient. Motorists who enjoy participating in the driving
experience do not like this prospect, but the National Highway Traffic
Safety Administration recently proposed a new federal safety standard that
calls for the inclusion of ESC in all light vehicles by the 2012 model
year. "We're developing safety systems that cut in only when critical
situations occur--when we are absolutely certain that the driver needs some
support," explains Mercedes-Benz's Dr. Joerg Breuer. ESC may become
invisible to the motorist once a certain level of sophistication is
reached: Lexus is working on a Vehicle Dynamics Integrated Management
system that facilitates smooth, concurrent operation of stability control,
steering, and braking systems. Spring 2007 will see the rollout of Lexus'
intelligent cruise-control system, which uses cameras and radar to spot
obstacles such as people and vehicles, and issues an alert if the driver
does not see them.
Click Here to View Full Article
to the top
Making Every E-Vote Count
IEEE Spectrum (01/07) Vol. 44, No. 1, P. 13; Cherry, Steven
A team of graduate computing engineering students from U.S. and Canadian
universities presented a voting system two months ago that reportedly
jettisons all the problems of commercial e-voting systems. The team is led
by cryptography researcher David Chaum, and the system, Punchscan, is easy
to explain and can be deployed with commercially available equipment.
Among the key problems with commercial systems that Punchscan addresses are
ballots that cannot be recounted in disputed elections; vulnerability to
malware and hackers; and the possibility of election rigging through the
exploitation of secret computer code contained in commercial e-voting
systems. The Punchscan ballot is designed so that it can be torn in half,
with candidates' names and assigned letters on one half and a set of holes
on the other that correspond to the letters, which show through when the
ballot is folded. A unique number is assigned to the ballot and is printed
on both halves, and the voter indicates the candidate of their choice with
a special pen that marks both the hole on the top sheet and the number on
the bottom sheet; either half of the ballot can be used to record the votes
via a portable scanner, while the other half is shredded. Since letters
are randomly assigned to candidates, no one can determine the voter's
selections by studying just one half of the ballot, but Punchscan can
because the random assignment is recorded in a database keyed to the ballot
number. No database connects the ballot number with the name of the voter,
so the voter's personal choices are kept private. Since at no point in the
voting process do the computers contain more than half the data needed to
know how someone voted, there is no need to physically safeguard the
machines.
Click Here to View Full Article
to the top
The Discover Interview: Marvin Minsky
Discover (01/07) Vol. 28, No. 1, P. 14; Kruglinksi, Susan
MIT professor and artificial intelligence pioneer Marvin Minsky, who
co-founded MIT's Artificial Intelligence Laboratory, wishes to build a
thinking machine capable of self-reflection, which he outlines in his book,
"The Emotion Machine." He believes the only way to analyze a tool as
complex as the brain or mind is to simulate it, and what does and does not
work in the simulation should be considered. Minsky cites several
initiatives to give AI human-like common sense, but what he thinks is
really needed "are the right kind of answers to questions that a 3-year-old
child would be filled with." The MIT professor [who received ACM's A.M.
Turing Award in 1969] disagrees that a failure to deliver on early promises
caused the AI field to go bust, when what actually happened is the
depletion of high-level thinkers; "Nowadays everyone in this field is
pushing some kind of logical deduction system, genetic algorithm system,
statistical inference system, or a neural network--none of which are making
much progress because they're fairly simple," Minsky remarks. He adds that
research support to flesh out radical new system concepts is lacking
because of funders' overriding emphasis on practical applications. Minsky
sees a need for intelligent robots because of the population explosion; in
addition, the aged will need automated caretakers, while artificial
physicists and scientists could help solve problems that are beyond the
capacity of human researchers.
Click Here to View Full Article
- Web Link to Publication Homepage
to the top
Robot Gives Snipers Owl Eyes
Popular Science (01/07) Mone, Gregory
A robotic head that can identify snipers using both audio and visual cues
could soon be deployed in Iraq. The device, know as RedOwl, is controlled
by a laptop and can read a nametag from a distance of 100 yards and
identify the make and model of a rifle from a mile away, just from the
sound of the blast. When RedOwl hears a gunshot, it analyzes the sound to
tell if it is enemy fire, and if it is the head swivels in the direction
the shot came from. Then a thermal imager picks out the person firing and
an infrared spotlight with a range of one mile allows troops using
night-vision to see the sniper without the sniper knowing he has been
spotted. RedOwl could be placed on top of a tank robot built by iRobot,
which can be steered using a modified video game controller, to lead troops
into buildings. The robot's ears consist of four microphones that receive
incoming sound waves, and a processor that immediately analyzes the data to
locate the exact source and identity of the sound. The eyes consist of a
central camera that shows RedOwl's operator where the robot is going, and a
powerful zoom that allows the operator to observe potential enemies from a
distance. RedOwl uses a laser rangefinder that bounces a beam off a target
to calculate the distance to the object, and since it is aware of its own
GPS location, the robot can inform troops as to exact locations of targets
up to 3,000 feet away.
Click Here to View Full Article
to the top
Advisory Council Seeks Tighter Cyber Security Net
GovExec.com (01/16/07) Marino, Jonathan
The National Infrastructure Advisory Council will send a report to the
White House that declares the need for greater cooperation between private
and public interests in order to establish a cybersecurity network that can
defend against an increasing terrorist threat. The report states that
sufficiently critical cybersecurity is needed by 2015, and that the
Department of Homeland Security (DHS) must work with infrastructure owners
and operators to build sector-specific maps that could help organize
efforts should a disaster or attack take place. Council member Margaret
Grayson said during her presentation that regulatory oversight might be
required to make sure the mandated tasks are being carried out
satisfactorily by both the public and private bodies involved. The report
follows the Homeland Security Advisory Council recommendation that DHS
Secretary Michael Chertoff extend the department's research to look into
how a terrorist attack could utilize the Internet to obstruct homeland
security.
Click Here to View Full Article
to the top
Novel Interfaces Pursued for Increased Content
EE Times (01/16/07) Mokhoff, Nicolas
Three groups presenting interface devices at the International Conference
on Consumer Electronics displayed innovative ways to access large amounts
of content on both stationary and mobile computers. A joint project
between the Hitachi Human Interaction Laboratory in Tokyo and the
Interaction Design of the Royal College of Art in London produced a
book-like device that allows a user to turn pages to find desired content.
Using light-dependent resistors, the device can detect three states: Open
or closed, the page the book is opened to, and if the book is facing up or
down. The intention is for the device to act like a book, in that a user
can easily find a favorite, "well-creased" page and can simply place the
book face down to save a page for later. Media on a hard-disk can be
linked to the pages, and although the device was displayed as a TV remote
control, it could be used for radio, digital photos, or Web sites.
Researchers from Mitsubishi Electric Microcomputer Application Software and
Ryukoku University showed an interface that allows a user to change the
display by moving the device itself, without the need for any additional
input tools. The interface works like a magnifying glass, letting users
choose what is highlighted on a crowded mobile screen. The third device,
developed by Kyungpook National University's Department of Information and
Communications and School of Electrical Engineering and Computer Science,
is capable of recognizing a user's hand gestures, such as throw, push, or
pull, as well as the user's posture, by analyzing the output signals from a
two-axes accelerometer. The presenters said their approach was suitable
for low-end mobile devices since the accelerometer can be easily embedded
into them.
Click Here to View Full Article
to the top
Humanoid Robots Gain Ability--Slowly But Steadily
Nikkei Weekly (01/08/07) Vol. 45, No. 2267, P. 16; Matsuda, Shogo
The University of Tokyo is collaborating with Toyota Motor, Matsushita
Electric Industrial, Mitsubishi Heavy Industries, and four other major
companies on research to integrate robotic technologies and information
technologies so that practical humanoid robots can be developed and put to
use. The project is expected to consume approximately $8.4 million
annually, and the plan calls for robot caregivers within 10 years, to be
preceded by incremental breakthroughs such as robots that can help put
things away, bed-making robots, and robots that can carry people and
provide other kinds of assistance. Toyota's effort is focused on improving
robotic leg movement and dexterity; developments in this area include a
one-legged hopping robot and a prototype with a motor in the torso linked
to limbs via wires, allowing for lighter, faster-moving arms and legs. A
research group led by University of Tokyo professor Yasuo Kuniyoshi has
devised a humanoid machine that can adjust its movements according to the
condition of the ground it rests on as well as get to its feet from a
supine position using a "skin" of tactile sensors. Enhancing humanoid
robots' precision in action is the goal of a group at the National
Institute of Advanced Industrial Science and Technology, which has
developed a jointed robotic hand that can mimic the finger and thumb
movement of a real hand. Meanwhile, professor Makoto Shimojo of the
University of Electro-Communications is concentrating on the creation of
robots that can adjust their actions by determining the state of the person
or object they are caring for or conveying. His group has developed a
sensor-equipped robot hand that can securely grasp and hold objects through
adjustment of finger strength.
Click Here to View Full Article
- Web Link to Publication Homepage
to the top