AI to the Rescue
|
AI to the Rescue |
 |
Leading-edge science in the battle against terrorism: AI-related technologies for disaster prevention and recovery.
|
|
|
|
|
|
|
Elsewhere on the Web
|
• Robots to the Rescue
• Rescue in Pockets of Rubble
• Inside Echelon
|
|
|
| |
|
News of the devastating terrorist attacks on the World Trade Center and Pentagon quickly swept across the Internet last Tuesday. Early reports indicate that the terrorists used knives to take control of their planes, resulting in experts calling for improved technology and surveillance at the nation's airports. AI-related technologies could provide us with some high-tech tools to prevent further acts of violence. Various law enforcement agencies have been using basic bomb disposal robots for years. However, the National Institute of Justice Final Report on Law Enforcement Robot Technology Assessment reviewed some of the currently available machines, indicating that "a disciplined systems engineering approach, coupled with a good understanding of the user needs, will close the gap between robots that are commercially available today and those developed specifically for the civilian law enforcement and bomb disposal communities." Several research efforts are currently underway, with a common goal to greatly enhance the capabilities of the next generation law enforcement/bomb disposal robots. Researchers at Sandia National Laboratories and Remotec, Inc. recently unveiled a police robot that is better equipped to make decisions "on its own", allowing the operator to concentrate on more high-level tasks. An older, but very informative and interactive site from NOVA Online describes various dangerous situations where hazardous duty robots could be used.
The greatest problem with Urban Search and Rescue (USAR) activities is illustrated by the fact that workers have only 48 hours to find trapped survivors in a collapsed structure, otherwise the likelihood of finding victims still alive is very small. Several robot competitions are focusing on helping search-and-rescue teams save lives in earthquake-damaged or bombed-out buildings. This year's Robot Rescue was jointly sponsored by the American Association for Artificial Intelligence and RoboCup, an annual robotics soccer competition. Each team's robot had 25 minutes to negotiate the course over different kinds of obstacles, locate dummies simulating injured persons and return to report the location of each victim. Another similar competition, The First Rescue Robot Contest, was held in Japan to emphasize the importance of rescue activities in disaster situations.
Many novel robot designs were introduced recently to overcome the difficulties encoutered in the USAR activities. Research teams around the world are making great efforts to develop robots that can successfully cope with environments that could not be fully anticipated nor programmed into the robot in advance.
Serpentine mechanisms (or "snake robots" in layman's terms) have many more degrees of freedom than conventional robots and rescue machinery, with a small cross-sectional area. These many degrees of freedom enable hyper-redundant mechanisms to thread through tightly packed volumes reaching locations otherwise inaccessible to conventional robots and people. Going a step further, a modular robot can even change shape by moving its modules around. This type of robots can offer advantages over traditional locomotion systems in rough environments and on unstructured terrain. BBC recently reported on Mark Yim's research at Xerox Palo Alto Research Center: If all goes according to plan, modular robots could enter unstable buildings after an earthquake to search for victims and then form themselves into a protective dome around the person. Crawling across a battlefield, robots could collect sensitive data about troop and landmine positions without risking human life.
Professor Robin Murphy of the University of South Florida used more down-to-earth approach and quickly deployed some of her robots to help search the wreckage at the WTC, as described in this article on National Geographic magazine. Her concept of "marsupial" robots is especially interesting, and refers to the teams of "mother" and "daughter" robots. The mother is designed to quickly carry large amounts of battery power and communications equipment into a disaster site. Once there, the mother ejects one or more daughter robots small enough to poke even deeper into the rubble, searching for evidence of survivors. They then report their sightings to the mother robot, which notifies human rescuers. The experimental robots that did prove useful in New York use tank-like threads for locomotion: similar rugged robot, Urbie, is produced as a joint effort of NASA JPL, iRobot Corporation (check out this one for a great selection of commercial robots), the Robotics Institute of Carnegie Mellon University, and the University of Southern California Robotics Research Laboratory. New Scientist reports on R2D2 lookalike robot, designed to be first on the scene after a traffic accident, with its video camera and mini TV screen allowing injured people to talk to emergency personnel while waiting for human help to arrive. Micro-air robots present another option for sophisticated mobile military reconnaissance and rescue attempts. The list could go on and on...
Robotics isn't the only AI-related way to go in preventing terrorism. An ongoing discussion in KM forum laid out some of the benefits of knowledge management and how it could apply to this tragedy. At the same time, CNet features an excellent article describing many advanced technologies that could improve security and prevent large-scale terrorist attacks. Security experts are particularly bullish on face-recognition technology and other so-called biometric security devices. Biometrics is an umbrella term which describes automated methods of establishing someone's identity from their unique physiological or behavioral characteristics. But there's more to biometrics than just face recognition: finger scans, hand geometry, retina scans, iris scans... The best known installation of face-recognition technology at Tampa's Raymond James Stadium during the last Super Bowl revealed 19 criminals hiding in the crowd. These technologies will raise privacy advocates' concerns, but they could also prove to be an efficient way to save many innocent lives. The same concerns surround Echelon, a global system of electronic eavesdropping that is thought to be used for commercial purposes as well as military spying. It monitors millions of communications, using advanced technologies like voice recognition, voiceprint identification and information extraction to target those of interest. Target messages are tagged and forwarded to NSA headquarters in Fort Meade for analysis and action. Unfortunately, it didn't prevent last week's attacks, as Bin Laden and his group recently went low-tech.