|
Research & development
November 2007
Research Agency Wants Help Solving The Seemingly Impossible
By Stew Magnuson
 ANAHEIM, Calif. — The Defense Advanced Research Projects Agency has a reputation for taking on challenges that sometimes seem to defy the laws of physics — or at least common sense.
The term “DARPA hard” refers to the problems its researchers attempt to solve. Sometimes their projects work and they make a profound impact. The Internet is one example. Sometimes the challenges are too tough. But that’s okay. The agency was created to push the research envelope, its staff says.
“Please tell us that it’s something that can’t be done. It’s science fiction,” said Brett Giroir, director of the defense sciences office. “That is a challenge we can’t resist.”
Every 18 months, during the three-day DARPATech conference, the agency trots out is program managers who outline the top items on their wish lists. Scientists, researchers and engineers from throughout the country come to see if they can contribute and maybe snag a contract with the agency.
Missed the conference? No problem. National Defense lists the toughest nuts DARPA needs help cracking. Have a solution to one of these dastardly hard problems? If so, DARPA would like to hear it.
Transparent Walls and Dirt
One of the most basic military tactics is to successfully hide from adversaries. Potential U.S. enemies have created thousands of underground structures to hide missiles, weapons of mass destruction, command and control facilities or they may use tunnels to launch attacks. Insurgents may also hide in buildings.
DARPA wants to defeat rock, concrete and plaster walls — not by blowing them up — but rather by making them “transparent,” said strategic technology office program manager Joseph Durek.
The agency is creating a suite of sensors to map the inside of buildings, tunnels, caves and underground facilities.
“I want to know how to make these physical shields transparent,” he said.
The goal is to make three-dimensional blueprints without being inside the structures. Sensors that measure gravity can be used to find empty spots beneath the earth. Heat signatures can detect cave entrances. In an urban setting, these signals are not easy to detect and the environment is cluttered, he said. And the sensors must be discreet. They can’t be easily discovered or destroyed.
The “Visi-Building” project, for example, is looking into radar to probe a building, but it must untangle the complicated scattering of energy inside the building, he said.
Once underground facilities are detected, DARPA wants to penetrate and destroy what’s inside without having to resort to nuclear weapons.
If soldiers are sent underground, they need devices to map, communicate and navigate without access to the Global Positioning System, Durek added.
Building Simulators Without Computer Programmers
 The areas of expertise the military is asking its war fighters to master are rising exponentially. Convoy protection, signals intelligence, medical response and avionics require computer programs that simulate these tasks.
“This list goes on and on. There’s just not enough time and money to build these,” said Daniel Kaufman, a program manager in the defense sciences office.
Holding 20 meetings to write a 100-page request for proposals for a computer simulation that results in 1,000 pages of specifications delivered four years later “has consistently failed our military, and will continue to do so,” he added.
The “Real World” program’s goal is to allow a soldier to build his own computer-based simulations so they can be used immediately, he said. The program should “empower the war fighter, not the software developer,” he added. The tool would be what Microsoft Word is to writing and PowerPoint to slides, he said.
A soldier, for example, could return to base after a mission where there was an ambush, input his observations onto a map of the street and reproduce what took place. It could be used for training, debriefings and mission rehearsals, Kaufman added.
To accomplish this, DARPA needs tools that will quickly reproduce 3-D images of streets and terrain, but more importantly, simulate human interactions.
Replicating the physical world is relatively easy, Kaufman said. Creating models of how a civilian from another culture or a mob might react is the “DARPA hard” problem.
If the program succeeds, “the entire business of software development will change,” Kaufman predicted.
Inner Armor
“Not all of the threats encountered by our soldiers are inflicted by the enemy,” said defense sciences office program manager Michael Callahan.
Soldiers need protection against chemical, biological and nuclear weapons, but naturally occurring threats such as high altitudes or diseases create casualties as well, he noted.
The goal of the “Inner Armor” program is to provide comprehensive protection to troops in harsh, exotic environments where deadly diseases lurk.
“The objective is to fortify the entire soldier against attack from the enemy or the environment,” Callahan said.
The plan has two objectives — “environmental hardening” and “kill proofing.”
Environmental hardening would protect soldiers from extreme heat, cold and high altitudes. Clues as to how to give soldiers immediate protection from altitude sickness could be found in the DNA of Sherpas who live in the Himalayas or animals that adapt to living above treelines, he said.
Kill proofing seeks to protect soldiers from chemical or radiological threats. Scientists might draw from the zoological world where micro-organisms live and thrive in radiological or chemical environments.
But throughout history, there has been no greater threat to soldiers than infectious diseases, Callahan said. Vaccines currently protect soldiers against only seven of 44 highly contagious pathogens. “I envision that we will preposition universal immune cells that are capable of making anti-bodies that neutralize not tens, but hundreds of threat agents,” he said.
This will require the ability to predict pathogen evolution, he added. Once a natural or manmade “bug” is identified, the military must be able to use these universal pathogens to produce 3 million doses within 12 weeks. Each dose should cost “pennies.” In addition, the vaccine cannot be tested on humans. And little white mice are not acceptable, he added. To that end, DARPA is attempting to reproduce the human immune system in a laboratory setting.
Computational Social Science
 There are two things Americans generally don’t do very well: mastering foreign languages and understanding outside cultures. DARPA wants to let computers do the thinking for officers conducting counterinsurgency or peacekeeping operations.
Even if the U.S. education system did suddenly begin producing legions of highly trained, culturally sensitive linguists, they could never keep pace with the tremendous variety and volume of foreign languages as well as the sources needed to be monitored such as television and radio broadcasts, road signs, graffiti and enemy written and spoken communications.
DARPA has been at the forefront of automated translation capabilities, but understanding language is only one element of what Charlie Holland, director of the information processing office, called “decision dominance.”
Commanders need cognitive computer programs that reason, seek alternatives and identify the best solution for a given situation, he said.
DARPA is developing computer systems that can take massive amounts of data and organize them into “useful organized knowledge with a minimum of human assistance, and adapt [to] new situations.”
To make effective decisions, commanders must “gain a rich understanding of the complex and dynamic political, cultural, and economic landscape in the area of operations,” said Sean O’Brien, a program manager in the office.
But a combatant commander’s decision to take action in a community may have unintended consequences. DARPA wants to develop simulators that predict how a community will react to a particular action or strategy.
“What options do I have at my disposal to meet mission objectives and what are the likely outcomes?” O’Brien asked.
Agent-based computer models can simulate social situations by using “objects,” social groups for example, which are set against each other and behave and interact using a simple set of rules. Such simulations have been used to predict traffic flows and macro- and micro-economic effects.
There is an explosion of new data sources on communities, who they are and what they believe, down to tribal, neighborhood and social group levels, he said. In addition, there is a proliferation of cultural and behavioral theories. However, many of these theories are in conflict, he noted.
DARPA wants to create “agents” that both mirror actual communities, and leverage the hundreds of social, cultural and behavioral theories that govern social interactions.
It will rely on accurate data on communities including their characteristics, beliefs and standards of behavior, census data, political leanings and attitudes towards their own and the U.S. government, he said.
The challenge is bringing together the most compelling social science theories, formalizing them, integrating them and writing them into a computer simulation.
The simulation would need to be populated with real-time data as an operation approaches.
If DARPA can accomplish this goal, “we may revolutionize the social sciences along the way,” O’Brien asserted.
Chemical Mapping of Urban Environment
Peter Haaland, a program manager with the strategic technology office, foresees a day when information on the myriad numbers of chemicals found in a city is as handy and accessible as a Google map.
The problem is that current chemical sensors are based on preconceived threats. They focus on only a few dozen agents — most of them falling under the chemical weapons category. The presence of such common, but potentially deadly chemicals, as chlorine and hydrogen chloride, are not collected.
There are about 80,000 commercially available chemicals, and many of them are toxic. DARPA wants to create a map showing where and if they appear in a given area.
These comprehensive chemical maps would show forces what building might hide chemical laboratories or contain weapon caches.
“DARPA is looking for toolkits to revolutionize awareness of the chemical environment,” he said.
Kerosene fumes would be common, as would nitric acid vapor, but detecting them together with other chemicals in remote locations might indicate someone was attempting to extract plutonium from spent nuclear fuel, he said.
One idea is to replace sensors with nanotechnology-based samplers that extract chemicals — other than nitrogen, oxygen and water vapor — from the air. The samplers, with a GPS device attached, could be placed on helicopters, military vehicles or secretly placed on a delivery truck making the rounds through a city. The sampler would be retrieved and the data extracted.
But extracting and analyzing that data quickly will be the challenge, he said.
“Creating a system that will perform this analysis quickly and accurately will not be easy, but think of what we’d learn if we could map a composition of metropolitan region in two days.”
Gas leaks, illegal toxic waste, chemical weapons and drug laboratories could be displayed on a map, he said. Gaining “accurate insight into our chemical environment would be worth the trouble,” he added.
Please email your comments to SMagnuson@ndia.org
Back To Top
|
Search Tips
