WHAT MAKES A DISTANCE EDUCATION PROGRAM FIRST RATE?

Camille K. Fareri
Electronic Data Systems, Distance learning Solutions

 This paper addresses the issue of quality in Distance Learning ("DL") programs.  It attempts to examine the issue of quality in distance learning from different angles. Everywhere you look, new "electronic" institutions are offering distance learning courses.  The proliferation of DL courses means that colleges, businesses and the military now have choices in selecting distance learning programs and can implement first rate programs. At issue in determining quality in distance learning is "through whose eyes is the quality determined?"  On one hand, technologists built the systems and networks on which DL programs operate and view quality in mostly technological terms: access, successful transmissions, download time, etc.  On the other hand, educators, who also view access as a criterion for quality, concentrate more on the program's ability to elicit learning. Educators are more interested in the conditions of learning than bandwidth size.

One of the major components of a first rate DL program is the DL strategy employed by a specific institution. About a decade ago, the first, crude attempts at "distance learning" incorporated various methods to teach people who were widely dispersed geographically. In the early days, video broadcasts presented lectures and early attempts at computer-based learning consisted of throwing text onto the computer screen.  Electronic books were merely poor imitations of their print counterparts. For the most part, these first approaches were rather unimaginative. In the last decade, the continuum of distance learning strategies progressed from the simple -- Web pages with text delivered over the Internet, Computer Based Training (CBT) delivery and one-way teleconferences/ videoconferences – through more advanced – synchronous instruction using white boarding available online; two way, interactive synchronous teleconferences; asynchronous videoconference supported by online materials with student collaboration and interaction -- to the more mature technologies of today: online synchronous and asynchronous delivery of instructor-developed curricula and multi-media instructional objects, artificial intelligence including various avenues for student interaction/ collaboration and total virtual campus solutions integrating DL courseware with other school functions and student support applications. Regardless of the DL strategy, courseware online needs to motivate, interest and fully involve the students in the learning process.  When a boring campus lecture course is converted to monotonous text scrolling across the screen, even the most dedicated and motivated students zone out. The primary test of a course’s inherent quality is if it fully engages the students and elicits the desired learning outcomes.  This paper explores what research has indicated are the components of effective courseware, how distance education programs can meet the conditions of learning and how to determine quality.

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

LEARNING WITH REFLECTION: 
PROJECT PRAXIS

Bradley Goodman
Information Management and Instructional Systems Department
The MITRE Corporation

Luciano Iorizzo
Distance Learning Office
United States Army Intelligence Center

 Classroom learning improves significantly when students participate in structured learning activities in small groups of peers. As the U.S. military moves from schoolhouse instruction to web-based distance learning, the student loses this important contact with other students. The educational value of student collaboration has led to the use of conventional groupware tools, such as chat and email, in distance learning environments. While these tools can enrich learning, they require at least two participants who are available at the same time and cannot guarantee the quality of assistance. Students in a web-based environment require high-caliber instructional support on demand. A simulated learning companion, acting as a peer in a distance learning environment ensures the availability of a collaborator and encourages the student to learn collaboratively, while drawing upon the advantages of distance learning. The learning companion we designed for PRAXIS encourages the student to reflect on and articulate past actions, and to discuss future intentions and their consequences.                      

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

 
EXPERIMENTS IN DECISION ANALYSIS TECHNIQUES FOR SIMULATION BASED ACQUISITION

Peter Eirich
 The Johns Hopkins University / Applied Physics Laboratory

  Simulation Based Acquisition (SBA) is an emerging approach for DoD systems acquisition.  SBA can be applied to a number of acquisition areas, and could be considered as a candidate strategy or best practice for training systems acquisition.  In particular, since training systems often directly include a simulation component, the potential benefits from SBA may be even more significant for training systems acquisition than for other types of systems.  It is generally accepted that SBA must be supported by a collaborative information technology environment, built around integrated design tools, product and process databases, models, and simulations.

 The author’s paper published in the 1999 I/ITSEC conference proceedings (see Reference 1) described the preparation of an experimental environment to evaluate candidate data analysis and decision-making techniques that appeared promising for use within SBA.  The experimental focus included techniques for the post-analysis of model results, and an evaluation of the desirable characteristics for tools and techniques that could be used for shaping, defining, and quantifying the "decision space" very early in the analysis and design process.  The 1999 paper discussed some insights gained during the preparation process for the experiments, but the experimental results were not available in time for inclusion. This paper is a continuation that presents the experimental results.

 The research involved a series of experiments in which groups of experts applied different pre- and post- analysis methods to a small scale but realistic design problem – in this instance, the design of a notional missile.  Data believed typical of what may be expected from future SBA environments were presented to experts in missile design, who then used the data to reach missile design trade-off decisions.  A number of information displays were programmed, and in the course of a series of decision problems, the experts’ preferred display formats became apparent.  Suggestions for making the preferred displays even more useful were recorded during the sessions.  This paper reviews the types of information display formats utilized, indicates the ones found to be most useful by the decisionmakers, and identifies their proposals for further improvement.  In addition, to help prepare for the graphical data presentation sessions, structured group decision analysis techniques were employed in advance to assess the relative importance of several of the decision factors.  This paper summarizes additional insights for SBA decisionmaking based on employing this alternative decision analysis approach.

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

 
IMMERSIVE VIRTUAL ENVIRONMENTS TO SUPPORT SYSTEM DESIGN
AND ACQUISITION

Grace M. Bochenek, Ph.D. 
Kenneth J. Ciarelli

U.S. Army Tank Automotive Research, Development, and Engineering Center

 Engineers at the U.S. Army Tank-Automotive Research, Development, and Engineering Center (TARDEC) are developing and applying high end projection-based immersive virtual reality tools and engineering-fidelity simulations to meet Army customer demands for simulation-based evaluation of ground vehicle designs, technology, and proposed product improvements throughout a vehicle's life cycle. As part of TARDEC’s continuous improvement of its Simulation-based Development Processes, TARDEC acquired two projection-based immersive visualization facilities (i.e., CAVE   , PowerWall   ) to permit multi-functional integrated concept/product teams to assemble and solve design problems with the assistance of high-end computer visualization tools. These technologies involve real-time simulation and interactions through multiple human sensorial channels making users believe they are interacting with real vehicle systems when in actuality they are only interacting with computer generated replicas.

 Within this synthetic environment team members can simultaneously enter a virtual product design world and jointly evaluate design issues, ideas and parameters, each from their own experience, perspective, and functional responsibility. This paper will describe our virtual product design process, the visualization toolset assembled, a  summary of the customization necessary, highlights of our experiences to date in a series of user applications, changes and effects on the Army acquisition process, and future research directions. These visualization tools are being used in the Army to evaluate technologies that will significantly change the user’s role in the operation of its vehicles. It will also have application in the development and evaluation of technologies going into the Army’s Brigade Combat Team and Future Combat Systems.

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

DEVELOPMENT OF A VIRTUAL DISTRIBUTED COLLABORATIVE ENVIRONMENT

Grace M. Bochenek, Ph.D.1
 David A. Brown2
Robert Heinlein4
Abhinav Kapoor3
Yair Kurzion, Ph.D. 5
Alexandre Naaman5
Thomas D. Rikert5
Mark Sokolik2
Jenny Zhao5

1U.S. Army Tank Automotive Research, Development, and Engineering Center,2EDS VR Center, 3MultiGen-Paradigm Inc.

4EDS Federal, 5SGI OpenGL Performer 

 As a dual use science and technology effort, the U. S. Army Tank Automotive Research, Development, and Engineering Center (TARDEC), in partnership with EDS, SGI and MultiGen-Paradigm, is developing a unique suite of software tools that provides the capability for geographically distributed teams or individuals to conduct engineering level design reviews and analysis within a common synthetic environment, a virtual distributed collaborative environment (VDCE). This paper describes the impact of collaborative virtual environments on Army acquisition processes, issues related to collaboration, methodologies used to develop technical solutions, an overview of the technical architecture, and results of experimentation and applications to military system acquisition.  The VDCE technology and its application to Army processes has the potential to improve Army acquisition processes, to improve system product quality, and to reduce system development costs.              

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

DEVELOPMENT OF A VIRTUAL POVING GROUND USING HIGH-RESOLUTION TERRAIN

Alexander A. Reid, Ph. D.
Researcher Electrical Engineer

Stacy A. Budzik
Mechanical Engineer, U.S. Army Tank-automotive, Armaments Command,

  TARDEC, in conjunction with their Dual Use Application Program partners, is collaborating to create a realistic, engineering-level of detail, virtual environment in support of, both the Army’s Simulation Based Acquisition and Simulation Through the Life Cycle programs, along with commercial product development. The Vehicle and Heavy Equipment Virtual Proving Ground (VHEVPG) will be used by the Army and Industry to apply "proof of concept" demonstrations through use of high fidelity, motion based, human and hardware-in-the-loop simulations. This is being accomplished through the utilization of high-resolution engineering-level vehicle models, terrain and visualization, along with three of the worlds most advanced ground vehicle motion simulators. These include both TARDEC’s Ride Motion Simulator (RMS) and Crew Station/Turret Motion Base Simulator (CS/TMBS) and the National Advanced Driving Simulator (NADS) located at the University of Iowa. Environment will exploit the unique capabilities provided by each individual simulator- the high frequency capability of the RMS (up to 50 Hz), the large active payload (25 tons) of the CS/TMBS, and the sustained accelerations and large motion envelope of the NADS. The objective of these programs is to develop a high-fidelity VHEVPG comprised of dynamic models, experimental terrain techniques, enhanced graphics and associated data collection and analysis techniques across distributed, concurrently running, simulations. This environment facilitates the evaluation of vehicle and human performance, human-machine interoperability, vehicle and crew compartment design, along with the design of training simulators. The results of this program will enable the  acquisition of vehicles and their subsystems, resulting in an efficient user oriented design process.

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

21 ST CENTURY TERRAIN – ENTERING THE URBAN WORLD

Jeffrey T. Turner
Christian P. Moscoso

Program Executive Office – Intelligence, Electronic Warfare, and Systems

  This paper describes the Rapid Terrain Visualization (RTV) programs advancements in the rapid collection of high-resolution digital topographic elevation and feature data in support of crisis or contingency operations for both military and civilian users. The ability to rapidly collect high-resolution urban terrain data affords our leaders and planners the capability to implement the next generation of visualization tools and tactical decision aids.

 Information in this paper highlights the technology developed to collect this data as well as prototype applications evolving to exploit high-resolution urban terrain.             

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

DYNAMIC TERRAIN IN A DISTRIBUTED SIMULATION ENVIRONMENT WITH LOW COST PC

Rita Simons
Simulation, Training and Instrumentation Command (STRICOM)

Jesse Liu
AcuSoft

Graham Upton and Tim Woodard
Diamond Visionics

  As technology utilized in simulation has grown, so have the requirements for a realistic solution to the dynamic terrain problem in the synthetic environment.  In order to support the DoD Simulation Based Acquisition (SBA) initiative, the need for a high fidelity Synthetic Natural Environment simulation is fundamental and critical. Specifically, a realistic dynamic terrain solution is required by the Advanced Concepts and Requirements (ACR) community, and maneuver forces using simulation to support their collective training objectives. Research has previously been conducted in the area of dynamic terrain implementation, and the dynamic environment.  Dynamic terrain is not new to the simulation community, however previous efforts have required high-end computational platforms, were unable to perform in real-time, and were often low fidelity in appearance. With the fast paced improvements in the performance of Personal Computers (PCs) and image generators, the realism that is required for a dynamic terrain implementation is now achievable on a PC. The US Army STRICOM sponsored a Phase I Small Business Innovative Research (SBIR) topic addressing these requirements, which has progressed to a Phase II effort.  In the Phase I effort, Diamond Visionics Company (DVC) and AcuSoft teamed to provide a PC based technology demonstration of dynamic terrain incorporating simple soil dynamics.  Phase II objectives include the development of a platform independent software solution that has an open architecture and application program interfaces, providing the fundamental functionality required by digital synthetic environments to implement dynamic terrain in a DIS/HLA network. The developed solution will use SEDRIS (Synthetic Environment Data Representation and Interchange Specification) as the underlining data standard. This paper will address the use of dynamic terrain in a Distributed Simulation Environment utilizing low cost PC platforms.  It will examine the challenges of implementing dynamic Synthetic Natural Environment in a distributed simulation environment, specific issues related to DIS networking, and the challenges and advantages associated with HLA migration.  It will also address interoperability with simulations and systems that encompass a wide range of fidelity, resolutions and application domains.                      

QUANTITATIVE PERFORMANCE-DRIVEN PC-BASED IMAGE GENERATOR EVALUATION FOR VISUAL INTEGRATED DISPLAY SYSTEMS

Budimir Zvolanek
Training and Support Systems – The Boeing Company

William Paone
PureLogix Division - Westar Corporation

Ed Elking
Training and Support Systems - The Boeing Company

Tim Dwyer
Training Systems Product Group – USAF ASC/YW

 Affordability and performance of training systems devices have become key factors in improving availability of training to a broader military community.   An example of this is the Target Projection System (TPS) embedded within Boeing’s Visual Integrated Display System (VIDS) – a product delivered to a number of military training programs such as the T-38 and the USAF Distributed Mission Training (DMT) F-15C and DMT F-16 programs.  Driven by a dedicated image generator (IG), the TPS simultaneously projects multiple high-resolution images of aircraft onto VIDS screens.  Until recently, only a desk-side workstation or a full-featured, high-performance IG has had the polygon and pixel-fill performance necessary to generate the TPS aircraft imagery.  However, rapid advances in PC-based three-dimensional (3-D) graphics technology have finally offered such performance at a much lower cost.  This paper describes Boeing’s effort in transitioning PC-based IG technology into the VIDS product by quantitative measurements of PCIG performance using TPS-specific benchmarks.  IG performance requirements and their embodiment in benchmark databases and test software are described.  Available PC-based IG descriptions are provided followed by a comparison of the benchmark test results, as well as a discussion of issues with real-time image generation hardware and software integration.  Finally, a recommendation of the TPS PC-based IG is presented based on the observed performance, as well as IG features and other ‘non-performance’ factors.  Suggested PCIG applications conclude the paper. 

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

DEVELOPING INTELLIGENT INFRARED TARGETS FOR TESTING AND TRAINING (IRT3)

Primary Author:
Dan Mullally

Secondary Author
 Thomas L. Clarke
Institute for Simulation and Training, University of Central Florida

Third Author
Glenn Boreman
Center for Research, Education in Optics, Lasers, U of Central Florida

 This paper will describe the developmental steps of a Live Fire Testing and Training Initiative project to develop intelligent, interactive infrared (IR) targets for use in both training and testing. The University of Central Florida (UCF) team will develop an IR projection capability suitable for providing live-fire targets for testing and training with IR systems in the 8-12 micron band. The system will use a Computer Generated Forces (CGF) system to control the IR projector imagery to provide intelligent, shoot-back capable, IR targets projected onto a fountain of water. Current IR Target systems are unsatisfactory. Conventional approaches use IR targets physically heated with heating strips that are constantly being "blown away" when used in live fire.  In addition heating strips have slow response time and cannot provide fast changing and moving imagery.  Scanning laser projectors are not suitable since their interaction with the scanning mirrors in Forward Looking Infra Red (FLIR) sensors  produces the appearance of a cloud of butterflies. The unique developmental approach detailed in this paper is based on the Texas Instruments (TI) video projector Digital Light Processor (DLP) technology. The project is designed to produce the full range of military targets on unique reusable and renewable water-based projection screens.                       

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

ESTABLISHING A UNIT CONTROL METHODOLOGY TO SUPPORT EMBEDDED SIMULATION

Vanna McHale
Applied Software and Systems Engineering Technology Group
Science Applications International Corporation

 The Inter-Vehicle Embedded Simulation Technology (INVEST) program is dedicated to providing onboard simulations in support of training exercises for tactical vehicles. The Synchronized Player Model (SPM) portion of the INVEST program was conceived to reduce the wireless communications bandwidth between the embedded simulations used in a coordinated training exercise. Current research of the SPM project focuses on the development of a Unit Control Language (UCL) used to provide the virtual models of a live unit.This research identified a set of unit control primitives that operate as high level behaviors to facilitate synchronization between live vehicle(s) and their simulation model. This paper describes the primitives identified for successful control and the Difference Analysis Engine (DAE) developed for primitive selection.  Experiments to validate the UCL as a potential means of vehicle synchronization were executed within a Java testbed environment and generated results that were evaluated against current dead reckoning techniques. The success of this unit control language, merged with previous research in independent vehicle control provides optimal solutions for reducing bandwidth in coordinated training.  Future research includes analysis of Subject Matter Experts decision making criteria for DAE refinement.                   

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

SIMULATION DRIVEN VIRTUAL OBJECTS IN REAL SCENES

Erol Gelenbe and Khaled Hussain
School of Electrical Engineering and Computer Science

Billy Foss
Institute for Simulation and Training

Niels Lobo
School of Electrical Engineering and Computer Science, University of Central Florida

 Hubert Bahr
U.S. Army STRICOM, Orlando, FL

 STRICOM, together with the School of Electrical Engineering and Computer Science (SEECS) and the Institute for Simulation & Training (IST) at the University of Central Florida are developing a system to allow virtual objects to be placed in live images in real time. The proposed approach is simulation driven in that it will use a geometric database of the site of the live scene to drive a simulator which will be used to predict the location of the synthetic object in the real scene at each instant of time.  The research we are conducting involves object identification in the real world scene using registered overlays, registration of the real world view with the synthetic view of the virtual terrain data, placement of the virtual object with the synthetic terrain and then the natural view using simulation, and finally realistic integration of the synthetic object into the live scene. New techniques are being developed to determine the occlusion of virtual objects based on their relation to terrain features in the live scene.  This paper describes the whole process used in the project, discusses the basic algorithms and presents novel  techniques used for recognition and placement of the objects.              

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

TRACKING TECHNOLOGIES FOR VIRTUAL REALITY TRAINING APPLICATIONS: A CASE STUDY

Malachi J. Wurpts
Tammy L. Swanson
Rolan Tapia
Southwest Research Institute

 In response to continuous reductions in available funds, time, personnel, and facilities dedicated to training, the United States Marine Corps (USMC) is avidly exploring supplementing its current training with virtual reality (VR) training applications. The goal of these applications is to provide a computer-mediated experience in which trainees can perceive and interact with a synthetic or simulated battlefield and simulated objects in a realistic manner. VR training devices have the potential to meet the flexibility, portability, and reconfigurability training requirements that are now necessary to prepare for ever-increasing USMC operational demands. Recent advancements of VR technologies are increasing the potential for more realism in VR training devices. Despite the tremendous advancements in VR technologies, one major question remains. Are these advancements significant enough to support the levels of realism required for training purposes? To answer this question, the USMC through the Office of Naval Research (ONR) is conducting several VR initiatives. One of these initiatives is the Small Unit Tactical Training Advanced Technology Demonstration (SUTT ATD) Program. The top-level goal of the SUTT ATD is to demonstrate how VR technologies can be used to support current and future USMC training requirements. These requirements range from the training of individuals operating in close combat situations to the training of a crew of individuals operating in a wide array of combat vehicles. One of the major SUTT ATD tasks is the evaluation and assessment of various VR technologies. These include computers (hardware and software), tracking systems (for human and/or weapon motion), locomotion systems (for traversal through a virtual environment), and graphical display technologies.

 This paper focuses specifically on the SUTT ATD tracking technology studies. With the emergence of many new tracking technologies (e.g., mechanical, acoustic, inertial, magnetic, and optical) over the last few years, it is extremely difficult to determine the appropriate tracking solution(s) for various training applications. Because tracking requirements may differ significantly for different applications, selecting the appropriate tracking system for a specific application becomes even more challenging. While technical specifications such as accuracy and resolution may be useful indicators of tracking system performance, for comparative purposes they can be misleading. In most cases, these specifications are tied to environmental conditions that are ideal for the specific technology. In addition to conducting accuracy measurements in more representative environments, these studies address other qualitative measures such as cost, ease of use, footprint, reliability, expandability, and technical support. They are also intended to present an approach to resolving the technical issue of finding and selecting appropriate tracking technology solutions.                     

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

AN EVALUATION OF THE TRAINING EFFECTIVENESS OF VIRTUAL ENVIRONMENTS

Barbara Barnett
Katrin Helbing
Glenn Hancock
Raymond Heininger
Bruce Perrin
The Boeing Company

 During the Boeing Joint Strike Fighter Concept Development Phase, we investigated the feasibility of using three-dimensional (3D) solid models, implemented within a Virtual Environment (VE), as a low-cost partial replacement for conventional hardware mockup trainers for aircraft maintainers. Currently, there are few studies directly comparing performance using VE-based training to more conventional methods. This paper summarizes the results of  several empirical studies conducted to evaluate the effectiveness of aircraft maintenance training within VEs. In these studies, trainees were taught a simple remove-and-replace maintenance procedure. The task, while not complex, required a number of ordered steps involving visual and physical obstructions. Training effectiveness was assessed with a written test of task procedures and with an objective assessment of task performance on a hardware mockup. Measures of performance included task completion time and procedural errors (e.g., incorrect action, wrong tool). The initial study compared hardware mockup training to two alternative display formats: solid model-based VEs and 3D line drawings implemented as computer-based displays (CBD).  Within each of these display formats, we compared passive "hands-off" training with user-interactive training. Results of the study indicated that as realism in the virtual training environment increased, performance approached that achieved with the more costly, time-intensive hardware mockup training. Another study assessed immersive Virtual Reality (VR) for task training. Participants in this condition trained for the maintenance task in an immersive VR, wearing a head-mounted display and interacting with task components using a 3D mouse. Results indicated that training time for the immersive condition was longer than the other CBD training methods, with a diminished task performance. Finally, using participants from the initial training study, we addressed the effectiveness of using Ves for maintenance rehearsal three months after completing initial training. Rehearsal involved a review of the task using one of two CBD methods. The first rehearsal condition was a review with annotated technical drawings; the second was the solid model-based interactive VE. A third group, the control, had no rehearsal. Performance for all participants was evaluated as before – with a written test of task procedures and performance on the hardware mockup. Results of this study showed a trend for better performance after interactive VE rehearsal over that of the other two conditions.  Collectively, these studies indicate that solid model-based VEs provide a potentially significant alternative to hardware mockup based training, resulting in savings in training time and cost. Further research is needed to identify the types of training scenarios for which VEs are most effective.             

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

EMPLOYING AUGMENTATION IN VIRTUAL ENVIRONMENTS FOR MAINTENANCE TRAINING

Axel Hintze
Marco Schumann
Stefan Stuering
Department Of Planning and Visualization Techniques – Fraunhofer Iff

 The application of conventional training methods displays many disadvantages particularly for highly complex equipment. Significant improvements and cost reductions can be achieved by means of three-dimensional computer visualization and animation of technical scenarios. This paper identifies with a list of common problems which will be improved in the Virtual Training environment.  The main idea is to develop a general modeling methodology that can be utilized in a wide variety of scenarios, while minimizing the need for programming simulation source code.  The different layers of information used to define a training scenario are then described in detail.  Both the author’s view and the trainee’s view of the developed prototype are presented. Finally, the paper concludes with a description of the goals of a most recent research project which will adapt these scenario structures for utilization in an Augmented Reality environment                 

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

FINITE-STATE GRAMMATICAL MODEL AND PARSER FOR AIR TRAFFIC CONTROLLER'S COMMANDS

Jorge L. Ortiz, Ph.D, PE
Electrical & Computer Engineering Department, College of Engineering
University Of  Puerto Rico

 This paper presents a grammatical model for the air traffic controller's (ATC) commands using finite-state transition networks (FSTN). The grammatical representation is used by a syntactic parser and recognizer for the analysis of the grammatical structure of the commands. A grammatical description using FSTN is proposed for the ATC's commands assigning word categories and syntactic structure that can be followed by a syntactic parser for recognition and parsing. This paper, also, presents an innovative model using "skip loops" for the implementation of a syntactic parser using finite-state transition networks to delete and remove incorrect or out of syntax words. These words could be the effect of mixed streams of words or errors in the conversion from spoken language to characters. The skip loop is an arc that allows the finite-state transition automata (FSTA) to delete a word that does not match the grammatical structure of the sentence, and continue the recognition process without affecting the syntactical definition of the sentence. This particular approach is especially useful in areas such as the command language of the air traffic controller (ATC). The model uses FSTN with skip loops to model and recognize ATC’s command language. The use of skip loops allows deleting words that may be present in the statement that are unrecognized or that do not fit into the grammatical structure of the ATC's language. This technique facilitates the recognition of the statement minimizing the possibility of declaring the statement as ill-formed. Two syntactic parser prototypes are implemented using Prolog and CLIPS. These techniques are useful in applications like military tactical environments that are exposed to rapidly changing commands, streams of information, and different sources of background noise. Many critical decisions have to be made extracting the correct information from multiple input streams making difficult and uncertain the selection of the correct input information. The method presented introduces a certain degree of intelligence using current AI techniques to obtain an intelligent syntactic parsing of the input information. The parser syntax can be defined to dynamically adjust its model to follow a particular stream of information that sounds or looks appropriate for the particular context. The purpose of the parser will be to model the process that resembles the human ability to follow a single dialog in an environment where there is many conversations and background noise.              

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

A SPEECH-CONTROLLED INTERACTIVE VIRTUAL ENVIRONMENT FOR SHIP FAMILIARIZATION

Stephanie S. Everett
David L. Tate
Tucker Maney
Kenneth Wauchope
Naval Research Laboratory

 This paper describes an interactive virtual environment (VE) designed to help Navy personnel become familiar with the layout of a ship. The system combines a 3D VE model of the ship with a spoken natural language interface that enables the user to issue verbal commands and queries to the model. By allowing the user to ask "Where am I?" or "Where is the Communications Center?", or tell the system to "Show me how to get to the Control Room from here" the virtual environment becomes more than just a passive representation of a 3D space – it becomes an active training aid that may help speed the learning process.

 The paper also discusses the need for better integration of graphic representation and object identification information to support future interactive VE systems.                      

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

SPEECH RECOGNITION IN NOISY MILITARY TRAINING ENVIRONMENTS

Stephen G. Boemler
R. Bradley Cope
Naval Air Warfare Center Training Systems Division, Orlando

Henry L. Pfister
SOS Inc., Shalimar, FL

 A common problem using speech recognition in simulated training environments is that computer speech recognition often fails at even moderate noise levels. Although training realism is increased when simulated noise environments such as a carrier landing deck are incorporated, the intense noise from military activities makes speech recognition highly problematic. This research developed adaptive noise canceling digital filters to enhance computer speech recognition in high noise environments. This particular design is specific to the Landing Signal Officer Trainer (LSOT FY 99 Block Upgrade was initiated under Contract N61339-97-D-0003, Delivery Order 010, 7 April 1999 by NAWCTSD project manager Ron Cole), but the high noise speech recognition technology adaptation is applicable to a large number of training, simulation, and operational systems. The project focused on removing the additive noise in the environment with adaptive digital filters inside a speech recognition algorithm. The adaptive filter relies on a recursive algorithm that is self-adjusting, which allows the filter to perform in situations where complete knowledge of the signal is not available. It is a process where the parameters of the adaptive filter are updated from one iteration to the next, and the parameters become data dependent. The design of the adaptive filter employs two signal estimates, one for the noise and the other for the combined speech and noise signal. The adaptive filter identifies the noise signal and looks for similar spectra in the speech signal. It then removes any matching noise signal from the speech signal.The speech recognition algorithm source code was modified to incorporate the adaptive filter after the acoustic signal processing which consisted of a 256 speech sample (real part of the Fourier Transform). The recognition is triggered by a push to talk microphone and the initial signal period is used as the noise estimate (or in the case of the LSO trainer, the synthesized aircraft noise is used since the noise is known and does not require an estimation). A Least Mean Squared adaptive filter is performed and the noise reduced complex Fourier coefficients are processed by the speech recognition system. This avoids the distortion encountered when transforming the filtered data back to the time domain. The LSO trainer environment simulates an aircraft carrier landing deck with an ambient noise level of 80 dB. The speakers use handsets which feed into a PC based speech recognition system running the Entropic HTK speech recognition system. This system was modified to incorporate the adaptive filter in the frequency domain for noise cancellation.  Examples of the LSO noise environment, speech input, and recognition performance are presented for this research.  This research was funded by a combination of Small Business Innovation Research (SBIR) contracts from the Naval Air Warfare Center Training System Division in Orlando and the Air Force Research Laboratory.  The NAWCTSD SBIR contract N61339-98-C-0017 was initiated by program manager Robert Seltzer and the Phase II effort was completed on 31 July 2000.              

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

AGGREGATION OF ENTITIES FOR ENTITY-AGGREGATE SIMULATION INTEROPERABILITY

Mr. Lawrence A Rieger
Mr. Emmett Beeker

HQ, TRADOC  TPO OneSAF, GRC International, Inc., An AT & T Company

 The continual evolution of military simulations has provided much of the technology for the exchange of entity data between the environments.  In particular, the extensive development in the resolution and granularity of aggregate simulations, combined with the entity data and ownership transfer capabilities of the High Level Architecture (HLA), has broken much of the virtual to constructive barriers for meaningful and productive data exchanges. So much so that in simulations, the real division has changed to be between the entity and aggregate simulations environments.  The real challenge is no longer moving entity data between the live, virtual and constructive environments, but rather the movement of entity data between the entity and aggregate environments.  This paper discusses eliminating the artificiality of aggregate state casualty resolution and assessment tables and the aggregation and de-aggregation of entities when passing ownership between entity and aggregate simulations.                  

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

ON THE FIDELITY OF SAFs: CAN PERFORMANCE DATA HELP?

Amy Henninger
Avelino Gonzalez
William Gerber
Michael Georgiopoulos
Ron DeMara
School of Electrical Engineering and Computer Science
University of Central Florida

 A recent report developed by the National Research Council (NRC) for the Defense Modeling and Simulation Office (DMSO) encourages the use of real world, war-gaming, and laboratory data in support of the development and validation of human behavioral models for military simulations. Also encouraged in this report is the use of interdisciplinary teams embracing the disciplines of the psychological, computer, and military sciences to create such models. This paper describes the use of an artificial intelligence modeling framework, observational learning, to support these objectives. This framework combines the research methods of experimental psychology with the machine learning methods of computer science to develop behavioral models from data generated by military experts participating in live and/or simulated exercises. To date, research has demonstrated that behavioral models developed through this framework can be integrated into popular Semi-Automated Force (SAF) systems to enhance their performance. However, there has been no known investigation as to what the benefits of this approach are with respect to behavioral model fidelity. This paper introduces the interdisciplinary nature of observational learning by briefly surveying its history with respect to computer science and psychology and by illustrating how it can be used in conjunction with military experts. Next, this paper examines experimental evidence to determine whether a significant difference exists between SAF performance and human performance for a low-level, skill task. Finally, this paper demonstrates how behavioral models developed through human performance data generated by military SMEs can be used in conventional SAF systems to make SAF performance more "human-like".                

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

USE OF ACTIVE NETWORK TECHNOLOGIES FOR DISTRIBUTED SIMULATION

Dr. Stephen Zabele
Litton-TASC, Inc. 

Thomas Stanzione
Litton-TASC, Inc.

  While distributed simulation infrastructures have evolved dramatically over the past several years to provide ever increasing levels of flexibility, abstraction, and interoperability, the scalability and performance of the simulation infrastructure continues to be a critical limiting factor. In particular, it is now becoming apparent that the limitations of the supporting networking technologies are a significant impediment to achieving needed levels of scalability and performance. Advancing the state-of-the-art for large-scale distributed simulations therefore requires significant advances both in the underlying network technologies and in the ability of simulations to exploit these new capabilities. Under the Specialized Active Networking technologies for Distributed Simulation (SANDS) project sponsored by the Information Technology Office (ITO) of the Defense Advanced Research Projects Agency (DARPA), TASC and the University of Massachusetts, Amherst (UMass) are developing Active Networks-based capabilities to improve significantly the performance of network-based distributed simulations . Our primary objective is reducing the substantial amounts of irrelevant network traffic delivered to simulation hosts in order to both improve bandwidth efficiency and to reduce the considerable overhead associated with reading and discarding unneeded data. Our approach involves installing dynamic packet filters within the network that act on behalf of each host to eliminate unneeded packets as early as possible. Our goal is a seamless integration with the High Level Architecture (HLA) Declaration Management (DM) and Data Distribution Management (DDM) services. Use of Active Networks to provide interest management services offers several important benefits to large scale simulations: (i) Because each entity can install its own filters, information filtering is accomplished in a "receiver-driven" manner, allowing each entity to customize its filters according to its own need. This decentralized approach allows active filtering to scale well as the number of entities grows large. (ii) Because active filtering is performed at a routing point, filtering can also be dependent on the state (e.g., congestion-level) at that router. In particular, this allows both entities and network routers to determine which data should be shed in times of congestion overload, and provides an effective means for mediating among the conflicting demands of different entities.

This paper is available on the 2000 I/ITSEC CD ROM. Order it from I/ITSEC'S Website

HIGHER-LEVEL INTEGRATED TEAM TRAINING ENVIRONMENT FOR  SPACE (HILITE)

Timothy Choate
John Friskie
Sytronics, Inc.

 Eric Loomis
Ball Aerospace & Technology Corp.

 Dr. Barbara Sorensen
Air Force Research Laboratory

 As the Air Force continues the Expeditionary Aerospace Force (EAF) implementation, two factors are paramount to its employment: 1) training the geographically-separated, yet organizationally-related, EAF units for their area of responsibility prior to their on-call window, and 2) incorporating space-based systems into all EAF training and operations to gain the force enhancement effects fundamental to successful EAF employment. These two factors will yield the light, lean, and lethal force possible through the EAF concept.  However, to realize the full combat potential of the aerospace team, the EAF plan requires units to train as they would fight, despite the limits imposed by financial constraints and geographic separation.  Providing this training capability is Distributed Mission Training (DMT), the concept the Air Force is pursuing as the means to train aerospace teams using realistic synthetic battlespaces.  DMT is an overarching approach applied to several domains – DMT-Air, DMT-Space, DMT-Special Operations, and DMT-Command and Control – with each domain having its own unique issues preventing a universal solution. This paper focuses on DMT-Space (DMT-S) and examines using DMT to conduct team training for space crews as part of the overarching EAF paradigm.  We examine the team training requirements and system capabilities needed for such an approach, and present the results of our efforts to design and implement a prototype DMT-S training environment using the High Level Architecture and other distributed simulation technologies. Our prototype system provides the simulation assets that are needed to deploy a simulation environment for space operators involved in space missile warning activities. Further, the HILITE environment enables realistic, real time interaction between space-based system operators and dynamic digital threat environments.  This allows operators to train effectively at any time and from any location. In addition, we examine the potential connectivity and interplay between the DMT-S and DMT-Air segments to determine the requirements and possible scenarios for a fully heterogeneous and multi-system battlespace capability.  This information will be of significant interest to the I/ITSEC community as it focuses on strategic training initiatives and provides a unique and timely perspective on how DMT technologies can be applied to support space operator training requirement