IJRR accepting "Data papers"

I think this is a great idea, get peer-reviewed data on the web both for reducing start up costs for robotics research and for good comparison of techniques.

" From the Editorial Board of the International Journal of Robotics
Research,

The International Journal of Robotics Research announces a new genre
of paper – the data paper – for significant and peer reviewed archival
data sets. Data papers build on the success of archival multimedia
content for papers, another IJRR innovation that was introduced in
July 2001.

Our goal in creating data papers is to facilitate and encourage the
release of high quality, peer reviewed data sets to the robotics
community. We believe that the widespread availability of large data
sets (order of Gbytes) will overcome a significant entry barrier in
robotics which currently slows algorithmic development.

Benefits will come from eliminating the effort and expertise required
to collect these very large experimental data sets, and from
facilitating direct comparison to previously published work.

Further details are available in the editorial at http://ijr.sagepub.com/cgi/reprint/28/5/587"

Autopilot safer for Predators

It will be no surprise to researchers working in the field of robot teleoperation to hear that autopilots on UAVs are substantially better than human pilots at keeping their aircraft safe. Apparently the U.S. Air Force's highly trained "real" pilots, have crash landed $80 million worth of Predators, compared to none by the "Xbox trained" Army techs.

http://www.stripes.com/article.asp?section=104&article=62334

The upshot is that the Air Force is being forced to go to autonomous vehicles as quickly as possible.

Man injured by malfunctioning robot

"A Swedish company has been fined 25,000 kronor ($3,000) after a malfunctioning robot attacked and almost killed one of its workers at a factory north of Stockholm."

Apparently the man thought the robot was disconnected from power, but was not and managed to confuse him for a rock.

http://www.thelocal.se/19120.html

Nice application for Multi Robot System

MIT is working on a "Robot Gardener" that might do plant and harvest your vegetables for you. Sounds like a perfect basis for an interesting MRS.

http://www.technology.am/robot-gardener-plants-tends-and-harvests-114746.html

Highlight: "This robot pulls up to a docking station and “pees” on a tomato plant to water it. The robot moves to another tomato plant and uses the Web cam to recognize a tomato and grab it with the gripper."

Interesting TED talks

A couple of short videos on TED that may be of interest to MRS or MAS researchers:

Bonnie Bassler discovered that bacteria "talk" to each other, using a chemical language that lets them coordinate defense and mount attacks. The find has stunning implications for medicine, industry -- and our understanding of ourselves.

http://www.ted.com/index.php/talks/bonnie_bassler_on_how_bacteria_communicate.html

JoAnn Kuchera-Morin demos the AlloSphere, an entirely new way to see and interpret scientific data, in full color and surround sound inside a massive metal sphere. Dive into the brain, feel electron spin, hear the music of the elements ...

http://www.ted.com/index.php/talks/joann_kuchera_morin_tours_the_allosphere.html

Winner of IFAAMAS-08 Victor Lesser Distinguished Dissertation Award announced

Ariel Procaccia receives the 2008 Victor Lesser Distinguished
Dissertation Award sponsored by IFAAMAS, the International Foundation
for Autonomous Agents and Multiagent Systems (http://www.ifaamas.org).

The selection committee for the IFAAMAS-08 Victor Lesser Distinguished
Dissertation Award is pleased to announce that Dr. Ariel Procaccia has
won the award. Dr. Procaccia received the Ph.D. in Computer Science
from Hebrew University of Jerusalem. His thesis contains fundamental
contributions to computational social choice theory, covering a wide
variety of topics including the complexity of election manipulation,
the robustness of voting rules, the learnability of voting rules, and
incentive-compatible machine learning. Among other contributions, it
introduces a new way of analyzing the manipulability of social choice
functions, which has become influential in AAMAS and related research
communities.

The committee is also pleased to announce two runners-up: Dr. Matthew
Taylor, who received his Ph.D. in Computer Science from the University
of Texas at Austin, and Dr. Yevgeniy Vorobeychik, who received his
Ph.D. in Computer Science and Engineering from the University of
Michigan. Dr. Taylor's thesis concerns transfer learning and Dr.
Vorobeychik's thesis studies empirical mechanism design. Both
dissertations also make innovative and influential contributions to
the field of autonomous agents and multiagent systems.

Previous winners of this award were Radu Jurca (2007) and Vincent
Conitzer (2006).

The selection committee members were Vincent Conitzer, Les Gasser,
Radu Jurca, Sarit Kraus, Sandip Sen, and Makoto Yokoo (chair).

Military vision

Interesting (but somewhat unambitious) view of how autonomous systems and humans will work together in the future.

Check out the video.

Some of the technology already exists and has been demonstrated, but a lot of what is left to do for this particular vision is some messy engineering.

Scientists losing their jobs to automation?

Two articles have appeared in the popular science media in the last week talking about software that automatically made new scientific knowledge. In one case, the knowledge was completely new, while in the other case, the software autonomously learned something we already knew.

Here are the first and second articles.

It makes sense to me that this type of technology will do more over time. Science is 1% inspiration and 99% perspiration. That 99% often involves running large numbers of experiments, processing reams of data and designing slightly changed experiments based on the previous results. When the execution of the experiments can be automated, e.g., when a simulator is used, software is ideally suited to run the experiments.

Artificial intelligence is sometimes described as the science of search. Thus, the techniques developed in that field for 50 years, should be ideally suited to performing routine experiments. The design and processing of experiments and their results is essentially search over some complex, sometimes physical space.

But getting software to do that "1% inspiration" is likely a long way off. That leads to two possible outcomes: (a) Scientists starting actually working 1% of the time (about 20 minutes a week) or (b) Scientific discovery speeds up 100 fold. I know where my money is.

Robotic Fish for Pollution Monitoring

These 5ft (1.5m) long fish won't be a swarm in the intuitive sense, but there will be multiple of them. Nice application.

"EU Commissions Swarm of Robot Fish to Sniff Our Pollution in Ports

The systems the EU is buying for a pollution control project are one of the rare robotic systems shaped like animals for practical reasons, not just for looks.

Researchers in Southern England have won part of a European Union contract with a sophisticated, autonomous, marine-locomotive pollution detector that looks astonishingly like a fish.

The 2.5 million-pound ($3.6 million) contract was awarded by the EU’s 7th Framework Programme for SHOAL a project whose name looks like an acronym but isn’t, whose goal is to use robot fish to sniff out and report pollution in Europe’s harbors and rivers."

Read more.