Brought to you by EarthWeb
IT Library Logo

Click Here!
Click Here!

Search the site:
 
EXPERT SEARCH -----
Programming Languages
Databases
Security
Web Services
Network Services
Middleware
Components
Operating Systems
User Interfaces
Groupware & Collaboration
Content Management
Productivity Applications
Hardware
Fun & Games

EarthWeb Direct EarthWeb Direct Fatbrain Auctions Support Source Answers

EarthWeb sites
Crossnodes
Datamation
Developer.com
DICE
EarthWeb.com
EarthWeb Direct
ERP Hub
Gamelan
GoCertify.com
HTMLGoodies
Intranet Journal
IT Knowledge
IT Library
JavaGoodies
JARS
JavaScripts.com
open source IT
RoadCoders
Y2K Info

Previous Table of Contents Next


3.4. ROBOTICS

Each military service must perform duties that are unduly hazardous or are extremely difficult for humans to do. Among them are entrance into chemically or nuclear contaminated areas, survey of ocean floors, explosive ordnance disarming and disposal, and reconnaissance/surveillance activities behind enemy lines. The first two describe situations where the soldiers' or sailors' performance clearly deteriorates over time and their health is at risk. The latter two describe situations where lives are at tremendous and potentially unnecessary risk. The military is looking toward robotics as providing a safer means to perform these tasks. Many robotics applications have expert system and symbolic reasoning components.

The Naval Postgraduate School has an Autonomous Underwater Vehicle (AUV) research group, whose significant on-going research effort involves the Phoenix AUV. Phoenix is primarily a research effort that studies how AUVs can perform countermine operations and coastal environmental studies. The Phoenix performs its assigned missions using a Rational Behavior Model (RBM), consisting of three layers of software. The execution layers govern the Phoenix reactive real-time control mechanisms. The tactical layer provides the Phoenix with near-real-time sensor analysis and operation. The strategic layer manages the Phoenix long-term mission planning and mission control. The RBM helps the Phoenix perform its tasks similarly to crew members aboard ships (Brutzman, 1997).

The Army has several on-going Unmanned Ground Vehicle (UGV) projects managed by the UGV Systems Joint Project Office (USAAIC, 1996). Included among them are tactical vehicles, vehicles to breach enemy minefields, perform environmental restoration activities, neutralize explosive ordinance, and perform area security and patrolling functions. The vision of the office is to eventually provide a robotic tank platoon consisting of two soldiers and four tanks. All the soldiers would ride in one tank, leaving the other tanks unmanned. The unmanned tanks would have full capability to move and to load and fire its weapons. The plans include rule-based systems to allow the tanks to recognize and avoid obstacles, stay in formation during maneuver, and engage enemy targets.

However, the military's interests in robotics extends beyond supporting conventional warfare. The Colorado School of Mines recently announced some success in robot-assisted Urban Search and Rescue (USAR). Rescue teams can deploy small intelligent robots that are able to seek out trapped victims in a collapsed building. The expert system component encodes the knowledge of the building's pre-collapse structure and deduces likely locations of void spaces. These spaces would have the greatest chance of housing survivors of the collapse. (Blitch, 1996). As the military continues to perform foreign humanitarian aid and disaster relief missions, using intelligent robots for dangerous missions such as USAR will be a factor in savings lives and resources.

3.5. INTELLIGENCE APPLICATIONS

Intelligence gathering and interpretation are complex and difficult. Depending on areas of responsibility and expertise, an analyst might have to sift through over 700 messages daily. The analyst does not know up front for what he is looking. Many of these messages are unformatted narrative text. Often on short suspense, the analyst must search for and extract important information, establish spatial and temporal relationships when possible, and provide a concise analysis to superiors.

Research into expert systems and emerging Natural Language Understanding (NLU) techniques is beginning to provide useful synthesis tools to the analyst. The Intelligent Analyst Associate (IAA) project at the U.S. Air Force Rome Laboratory is one example. By extracting and visualizing simple entity and event data from text, IAA provides visual cues to the analyst. These cues help narrow down what the analyst must read and pinpoint related pieces of information. IAA is in its early stages of development, in part because NLU is still only an emerging technology. As the state-of-the-art of NLU rises, IAA will be better able to handle the identification of more complex events.

Another project that is not specifically military in nature but has military implications is a NASA project to identify violations to nuclear treaties (Mason, 1995). Obviously, countries intending to violate a ban on nuclear testing will go to great lengths to conceal that fact. Nuclear testing still leaves important seismic clues that sensors can isolate. The rule base performs two interpretations:

  • The "partial interpretation" helps identify that an interesting event has occurred. This helps weed out the regular seismic activity that constantly occurs throughout the world. This is done by employing a belief system that looks at input from various seismograms and indicates that it believes or disbelieves that they constitute the same event. As more information comes in, the strength of belief increases or new information can contradict what is already believed.
  • The "full interpretation" takes the interesting events and tries to identify the likely cause and narrow down the source of the event. Seismic events consist of a series of segments where the intensity of the waves increase and decrease. Expert system rules identify the segments, and subsequently identify the likely event.

This concept is applicable to other intelligence applications -- such as those that seek to interpret troop movements and other activities on satellite photograph.


Previous Table of Contents Next

footer nav
Use of this site is subject certain Terms & Conditions.
Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details.