UAVs not very successful?

I am not exactly sure how anyone comes up with these numbers, because it is often hard to tell whether someone is an insurgent, even after they are captured .....

The foreign policy community’s favorite counterinsurgency adviser, or at least their favorite Australian one, David Kilcullen, told lawmakers last week that the drone strikes targeting Al Qaeda and Taliban fighters in Pakistan are creating enemies at a far faster rate than its killing them. According to statistics he provided, the success rate of the drone bombing campaign is extremely low: just 2 percent of bombs dropped have hit targeted militants. The other 98 percent? Those killed noncombatant Pakistani civilians, he said.

http://www.dodbuzz.com/2009/05/11/drones-hardly-ever-kill-bad-guys/

AAMAS '10 CFP

Call for Papers-AAMAS 2010
The 9th International Conference on Agents and Multi Agent Systems
Toronto, Canada

Important dates:
Conference: May 10 - 14, 2010
Electronic Abstract Submission: October 8, 2009
Full Paper Submission: October 13, 2009
Author Notification: December 18, 2009

Collocated with KR, NMR, ICAPS, FOIS, (all in Toronto) and DL 2010 (in Waterloo, Ontario, Canada)
Expanded and latest information: www.cse.yorku.ca/AAMAS2010/

INTRODUCTION

AAMAS is the leading scientific conference for research in autonomous
agents and multiagent systems. The AAMAS conference series was
initiated in 2002 by merging three highly-respected meetings: International
Conference on Multi-Agent Systems (ICMAS); International Workshop on
Agent Theories, Architectures, and Languages (ATAL); and International
Conference on Autonomous Agents (AA). The aim of the joint conference
is to provide a single, high-profile, internationally-respected
archival forum for scientific research in the theory and practice of
autonomous agents and multiagent systems. AAMAS-2010 is the Ninth
conference in the AAMAS series, following enormously successful
previous conferences, and will be held at the Sheraton Centre Toronto
Hotel in downtown Toronto. See www.ifaamas.org for more
information on the AAMAS conference series.

SUBMISSION DETAILS

AAMAS-2010 seeks high-quality submissions of full papers, limited to 8
pages in length. Submissions will be rigorously peer reviewed and
evaluated on the basis of originality, soundness, significance,
presentation, understanding of the state of the art, and overall
quality of their technical contribution. Reviews will be double blind;
authors must avoid including anything that can be used to identify
them. Where submission is for full (8 page) papers only, in some
cases they may be accepted as 2 page extended abstracts. For formatting
instructions, please refer to the conference page.

In addition to submissions in the main track, AAMAS is soliciting
papers in two special tracks on robotics, and on virtual agents (see
below). The review process for the special tracks will be the same as
for the main track, but with specially-selected program committee
members.

Special Track on Robotics (Chair: Michael Beetz):
Papers on theory and applications concerning single and multiple
robots will be welcome, namely those focusing on real robots
interacting with their surrounding environments. The goal is to foster
interaction between researchers on agent and robotics systems, so as
to provide a cradle for cross-fertilization of concepts from both
fields.

Special Track on Virtual Agents (Chair: Stacy Marsella):
Virtual agents are embodied agents in interactive virtual or physical
environments that emulate human-like behavior. We encourage papers on
the design, implementation, and evaluation of virtual agents as well
as challenging applications featuring them. The goal is to provide an
opportunity for interaction and cross-fertilization between the AAMAS
community and researchers working on virtual agents and to strengthen
links between the two communities.

In addition to the conference papers, presented in parallel
technical sessions, AAMAS-2010 will include:
o Industry and Applications track
o Demonstrations
o Posters presentations for full papers and extended abstracts
o Invited talks and panel discussions

The submission processes for the demonstration and industry tracks are
separate from the main paper submission process.

ORGANIZING COMMITTEE

General Chairs: Michael Luck and Sandip Sen
Program Chairs: Wiebe van der Hoek and Gal A. Kaminka
Local Organization Chair: Yves Lesperance

A list of the other members of the AAMAS-2010 Organizing Committee may
be found at www.cse.yorku.ca/AAMAS2010/

Robot brains or robot bodies?

A few weeks ago, I had a post about roboticists taking advantage of an insect body for its mechanical capabilities and giving it a brain. Now this picture in the Guardian of a roboticist using an insect brain to control a robotic body. Maybe these researchers could get together and either make robots or some cool hybrid insects!

RoboCup Rescue robots

Some interesting robot designs from 2009 RoboCup Rescue League. It feels like roboticists are starting to really understand that robot designs don't need to look like traditional vehicle designs. Many more pictures, courtesy of Amir Soltanzadeh, here.

Goldman Sach's MAS

There is a theory among, presumably bitter, AI researchers that a technology is called AI until it works and then it gets called something else. Voice recognition, OCR, face recognition in digital cameras are examples of this. Goldman Sach's intelligent, distributed trading platform seems to be continuing the tradition for MAS. From what I can glean from the various newspaper articles, they have a bunch of agents that monitor trading flows and autonomously and very quickly place trades to exploit market conditions.

http://rawstory.com/08/news/2009/07/04/taibbi-nyse-ends-transparency-to-protect-goldman-sachs/

http://zerohedge.blogspot.com/2009/07/guest-post-manipulation.html

I think people overlook how MAS and MRS can process so much information, collected from a very large number of sources and make so many decisions so quickly. Humans simply can't compete. The above stories hint that this might be the case here. Assuming what Goldman is doing is legal (I have no idea), then their advantage is due to a reasonably simple MAS that can simply process more information and make more decisions faster than any human can.

I suspect that we will have to get used to this type of thing happening. It won't be just in markets, it may be in many different areas.

VTOL UAV

Here is a fun little UAV made by MLB Co. and expected to be available for about $50K within a year. I think we'll see more radical designs like this as UAV technology matures and designers understand the implications of not needing to deal with limitations of a human on board.



And for the price, I think we will see a lot of vehicles like this in use, very soon.

UCAS "Video"

Very artistic (and no doubt expensive) visualization of how UCAS might fit into next generation warfare. The video is a couple of years old, completely focused on missions and scenarios that don't feel relevant any more, but no doubt big contractors are still doing their best to sell it.

UAV Sniper

The Army is testing an autonomous helicopter capable of acting as a sniper. Not much information about the stability package for the weapon, but it must be impressive. Not sure how a helicopter is going to achieve the stealth required of a sniper? After the much publicized messes the Predators have made with their much less precise weapons, this is a good step forward.

http://www.wired.com/dangerroom/2009/04/army-tests-new/

Insects for chemical and biological detection

Controlling small UAVs has turned out to be so difficult (even understanding the way insects fly was/is really hard) that DARPA is again looking to outfitting insects instead of building vehicles.

http://www.popsci.com/military-aviation-amp-space/article/2009-06/insect-wifi-network-tracks-down-wmds

I think it was in the movie Get Smart that Max smashed an expensive prototype something like this into pieces.

Nuclear armed UAVs?

Two interesting UAV articles posted on the same day. One talks about the possibility of nuclear armed UAVs, the other notes that the USAF will train more UAV pilots than real pilots for the first time this year. This is partly in response to Gates insisting that the Air Force goes to more automation in their UAVs in response to higher crash rates than the Army had for their UAVs.

Together, these two articles show just how quickly the U.S. military is moving to UAVs. The F-22 and Stealth Bombers and Fighters might be the last major manned military aircraft. End of an era.

http://eideard.wordpress.com/2009/06/16/air-force-intends-to-rely-on-unmanned-aircraft-rather-than-pilots/

http://www.armedforcesjournal.com/2009/06/4040349

Solar powered robots?

The reigning North American Solar Challenge car made it from Texas to Canada in 50 hours. This is with a driver, adding weight and imposing constraints on the overall shape. Without those constraints, it would seem reasonable to think that a solar powered robot vehicle could move pretty effectively, reliably and quickly over long distances. I don't recall seeing any solar powered robotic vehicles (except on Mars), I wonder why.

http://arstechnica.com/science/news/2009/06/0-to-60-in-full-sunlight-revving-up-for-world-solar-challenge.ars

Spy Chips Guiding CIA Drone Strikes

Story in Wired about the possibility that the U.S. military is getting locals to plant RF chips on or near militants which are used to guide UAVs in onto their targets.

One of the interesting things to me was that apparently the militants had worked out a while ago that this was going on and were doing their best to stop it. That seems to indicate a pretty high level of technical sophistication.

http://www.wired.com/dangerroom/2009/06/spy-chips-guiding-cia-drone-strikes-locals-say/

In a review of "Wired for War"

"Another scientist talked about how the military came to him and said, "Oh, we'd like you to design the hunter-killer drone from the Terminator movies." Which, you know, is kind of incredibly scary, but it makes perfect sense from another perspective in that if it's effective for SkyNet, their thinking is: "Well, it could be really neat in our real-world battlefields.""

http://www.hplusmagazine.com/articles/robotics/wired-war-or-how-we-learned-stop-worrying-and-let-dystopian-sf-movies-inspire-our-

DARPA Strategic Plan

Four "Future Icons" in DARPA's strategic plan will require particular involvement of MAS and MRS researchers:

Air Vehicles: unmanned air vehicles that quickly arrive at their mission station and can loiter there for very long periods

National Cyber Range: The National Cyber Range will provide an environment for realistic, qualitative and quantitative assessment of potentially revolutionary cyber research and development technologies

Persistent Surveillance Sensors: determine, track, and neutralize elusive threats, such as improvised explosive device factories

Networks: self-forming, robust, self-defending networks at the strategic and tactical level are the key to network-centric warfare; these networks will use spectrum far more efficiently and resist disruption if the GPS time signal is unavailable


DARPA currently emphasizes research in nine strategic thrusts:
Robust, Secure, Self-Forming Networks
Detection, Precision ID, Tracking, and Destruction of Elusive Targets
Urban Area Operations
Advanced Manned and Unmanned Systems
Detection, Characterization, and Assessment of Underground Structures
Space
Increasing the Tooth to Tail Ratio
Bio-Revolution
Core Technologies

http://www.darpa.mil/Docs/StratPlan09.pdf

Richard Feyman at Thinking Machines

Great article about Richard Feyman's time at Thinking Machines. This guy was a pure scientist.

"That is positively the dopiest idea I ever heard." For Richard a crazy idea was an opportunity to either prove it wrong or prove it right. Either way, he was interested. By the end of lunch he had agreed to spend the summer working at the company."

".... we informed Richard that his assignment would be to advise on the application of parallel processing to scientific problems. 'That sounds like a bunch of baloney,' he said. 'Give me something real to do.'"

"When I got back to Boston I went to the library and discovered a book by Kimura on the subject, and much to my disappointment, all of our "discoveries" were covered in the first few pages. When I called back and told Richard what I had found, he was elated. 'Hey, we got it right!' he said. 'Not bad for amateurs.'"

http://www.longnow.org/views/essays/articles/ArtFeynman.php

Someone has been watching Transformers

I can't believe this is anything more than a publicity stunt, but this very orange and very large robot is supposed to clear forests (for fire prevention) and perhaps even fight fires. Amusing, if nothing else.

http://www.treehugger.com/files/2009/05/giant-robot-fights-forest-fires.php

Autonomous Robotics Journal - latest issue

Papers on robots with whiskers and underwater robots, sounds like we are halfway to an autonomous walrus ....

On the use of likelihood fields to perform sonar scan matching localization
Authors: Antoni Burguera, Yolanda González and Gabriel Oliver
DOI: 10.1007/s10514-009-9108-0
pp. 203-222

Contact type dependency of texture classification in a whiskered mobile robot
Authors: Charles W. Fox, Ben Mitchinson, Martin J. Pearson, Anthony
G. Pipe, and Tony J. Prescott
DOI: 10.1007/s10514-009-9109-z
pp. 223-239

Maneuvering and control of a biomimetic autonomous underwater vehicle
Author: Jenhwa Guo
DOI: 10.1007/s10514-009-9117-z
pp. 241-249

http://www.springerlink.com/content/0929-5593

Imagine if scientists cooperated

http://tierneylab.blogs.nytimes.com/2009/04/30/what-if-scientists-didnt-compete/

"What if scientists, instead of rushing to publish or perish, chose to cooperate? Sean Cutler decided to do “a little experiment,” as he calls it, and you can see the results in the forthcoming issue of Science."

I think MRS and MAS researchers cooperate more than researchers in most fields do, because there is so much to be done to make even a single application work. There is also so much "low hanging fruit", that competing hard to be the first to some some particular problem rarely seems worth the effort - just find an interesting unsolved problem that there are not so many people working on.

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.