DODGING THE TREES AND BUSES: CURRENT NOE SIMULATION

B. James Voorhies

Michael A. Cosman

Evans and Sutherland Computer Corporation

Many previous rotorcraft nap-of-earth visual simulators were designed by the "seat of the pants" method: infant technologies mated with "best guess" estimates of what a visual simulation scene should contain.  Experience with the successes and weaknesses of these earlier systems (many still in valuable active use) provided the right questions, and recent empirical quantitative studies have offered first and second generation answers.  Unfortunately, the technology's solutions have too often been, "This is as good as we can do, and so we'll prove it's good enough."  It is now possible to approach the problem from the researcher point of view.  Maturing hardware and emerging software technologies have made possible the design of a system that provides the content demanded by the research, not just to marginal but to desirable levels of performance.

The nap-of-earth visual simulation project had as a practical objective tailoring a visual database designed for rotorcraft to the requirements of a comprehensive research study.  The projected performance envelope was designed to be nap-of-earth and contour flying from five feet up and from hover to 100 knots.  Empirical issues to be addressed in the implementation included: mix of 2D and 3D cueing, and total cue densities to provide optimum visual flow; elimination or minimization of detail in the aircraft performance envelope; and scene reality versus training value.  Other perceived deficiencies of some earlier systems that are addressed by this system are multiple, properly occulting, dynamic ground and air threats, and special weapons effects (cannon, rockets, FLIR, etc.)  In addition, consideration was given in the structuring of the database to provide maximum flexibility as new research and experience dictate modifications.  What has evolved is a good example of visual flight simulation specifically designed for the Nap-of-earth regime.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

SELECTIVE SCENE MANAGEMENT IN FLIGHT

SIMULATOR VISUAL SYSTEMS

Jack W. Newhard

Staff Scientist

The Singer Co., Link Flight Simulation Division

Michael R. Nicol

Visual System Engineer

U.S. Air Force Systems Command

Aeronautical Systems Division

Wright-Patterson AFB

A variety of mission-dependent tasks are practiced in military flight simulators.  A flexible way of meeting the diverse scene-content requirements employs a composite database from which the appropriate feature models are displayed (and the less appropriate models may be excluded).  Importance codes in the feature data are the basis for feature discrimination by the real time scene management software.  This software monitors image generator loads, computes demand-variables based upon their deviations with respect to specified limits, and computes several control variables as functions of the maximum demand-variable.  These control variables regulate the flow of feature data to or within the image generator.  "Closed-loop" l9oad optimization is thereby effected.  Imagery and performance data from four example test flights (over the same path) contrast unmanaged, unselectively managed, and selectively managed image generation for two tasks.  A given database can be reoptimized by replacing importance code assignments.  The prospects appear favorable for user tuning of real time databases to tailor them for particular training tasks.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

BUILDING DIGITAL IMAGE GENERATOR OBJECTS WITH

MULTIPLE TEXTURED PLANES

Pi-Yun Cheng, Senior Staff Engineer

Schlumberger/Benson

Nicholas Szabo, Director

Research and Development

Link Flight Simulation Division, The Singer Company

New capabilities are being introduced into digital image generators that permit the use of photographically derived images to increase the detail and realism of training scenarios.  The new methods have evolved from techniques for applying pseudo-random texture to surfaces, and permit not only photo images, but also variations in the translucency of surfaces.  These methods permit the construction of natural objects with irregular silhouettes, like trees.  Nonetheless, the patterns are each essentially two-dimensional.  To construct objects that look more three-dimensional, patterns can be projected on multiple intersecting planes, thereby forming representations of trees and bushes that provide more realistic three-dimensional cueing than simple single-plane projections.  Problems of data collection and illumination are discussed, and examples are presented to illustrate the techniques.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

AN INTRODUCTION TO ARTIFICIAL INTELLIGENCE IN TRAINING SYSTEMS

Marguerite Moreno

Member of the Technical Staff

VERAC, Incorporated

Artificial Intelligence (AI) researchers are developing techniques with which to study/simulate intelligent behavior, e.g., understanding language, problem solving and learning.  Application of these techniques to the are of computer-based training technology appears promising.

For example, in the tactical training environment instructors are often overwhelmed by the complexity of configuring realistic threat scenarios, while simultaneously monitoring and reacting to trainees' actions.  Applying AI production system (rule-based) technology to build computerbased intelligent adversaries could considerable improve the tactical training situation.

Other potential AI applications in the training area include:

1)          developing detailed realistic expert models, i.e., enemy tactics;

2)          using the explanatory capabilities of an AI system that embodies "instructor" knowledge to tutor a trainee,

3)          incorporating AI concepts in computer assisted instruction, and

4)          Configuring generic training systems using a rule-based expert system similar to existing successful systems.

However, because AI is in its developmental stage, its techniques are still experimental.  As such, the feasibility/applicability of some techniques have not been clearly determined.  This paper examines these technologies and assesses the realistic potential and limitations of AI in the training environment.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

AN EXPERT SYSTEM AS A REPLACEMENT FOR A TEAM MEMBER

IN AN ASW SIMULATION

Donald L. Johnston

Member of the Technical Staff

Tactical and Training Systems Division

Logicon, Inc.

Richard W. Obermayer

Executive Scientist

Vreuls Research Corporation

In an investigation of Artificial Intelligence (AI) techniques applied to automating instruction in an emergent, team environment, a computer model has been created to replace one of the players in an Anti-Submarine Warfare (ASW) simulation.  The simulation environment is that of a Combat Information Center but concentrates on the interactions of three team members: the operations coordinator (ASWOC), the aircraft controller (ASAC), and the fire control officer (ASWFCO).  The model is an expert system for the ASWFCO that runs in real time along with the simulation.  The ease of adding and modifying rules for the model has shown the power of prototyping that is possible using an AI expert system approach.  Replacing a human by a model enables training efficiency by reducing the need for manpower in training simulations.  Developing a player model is a necessary first step toward automating performance measurement and instruction in an emergent environment.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

TRIO–AN EXPERT SYSTEM FOR AIR INTERCEPT TRAINING

W. Feurzeig and W. L. Ash

BBN Labs

G. Ricard

NTEC

TRIO is an expert system for training F-14 interceptor radar operators in the basic tactics of high-speed air intercepts.  It introduces artificial intelligence methods into real-time training.  The TRIO task environment supports simulations of airborne raiders, interceptor and target aircraft, and weapons models.  It provides dynamic displays of heading, bearing, and displacement vectors, radar screens, flight instruments, intercept parameters, missile envelopes and interceptor/target aircraft ground racks.  It incorporates real-time speech recognition and synthesis subsystems including highly advanced capabilities for recognition of naturally articulated extended utterances.  TRIO supports three instructional modes: pre-flight demonstrations, in-flight monitoring and guidance, and post-flight debriefing.  The instruction employs an articulate expert to demonstrate and explain intercept tactics, a set of daemons for real-time performance monitoring, and a knowledge-based performance analysis program to detect and diagnose student errors.  TRIO is currently implemented on a personal LISP machine, the BBN Jericho computer.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

A LIFE-CYCLE COST STRUCTURE FOR DEFENSE TRAINING PROGRAMS

Mark I. Knapp and Jesse Orlansky

Institute for Defense Analyses

The subject of this paper is a cost-element structure (CES) that identifies, defines and structures a list of cost elements that is intended to describe fully the life-cycle cost of any formal program, course, or device for individual training of military personnel.  It was developed to satisfy a widely recognized need for consistent and credible evaluation of cost in cost-effectiveness analyses of alternative methods of training.  The cost-element structure is based upon authoritative and widely used cost guides promulgated by the Services and the Office of the Secretary of Defense, and many potential users contributed to its development.  Accordingly, the general use of a comprehensive CES such as this offers the following advantages.  It should (1) ensure that all elements of life-cycle costs are accounted for, (2) reveal gaps in essential data, (3) permit making credible and equitable comparisons among training alternatives, (4) identify "cost drivers" for trade-off analysis or cost reduction, (5) enable resource specialists to focus on elements of interest, while observing the impact of those resources in a total-program context, (6) disclose significant variables for the development of cost-estimating relationships and, (7) improve communication and understanding among officials at various levels in the Services and the Office of the Secretary of Defense whose decisions affect the conduct of military training.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

CONTRACTOR OPERATION AND MAINTENANCE OF SIMULATORS

Johnnie A. Butler and Maurice Winsor

Simulator Operation and Maintenance Analysts

Naval Training Equipment Center

In its continued effort to streamline and make its operations more efficient and cost effective, the Navy investigated various methods of obtaining operation and maintenance support for simulation equipment.  After extensive research, the concept of contractor support was formalized into an active program titled Contractor Operation and Maintenance of Simulators (COMS).  This new initiative was implemented in October 1982.  This paper takes a constructive look at the short history of COMS to assess effectiveness and explore the following areas:  (1) the training device support program under the Navy's organic support posture, (2) immediate changes which resulted from the initial implementation of the COMS program and related problems encountered, and (3) areas for future study.  The paper predicts future changes in procurement strategies and concepts that will reshape the training device support concept for years to come.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.

MEASURING CONTRACTOR PERFORMANCE UNDER THE NAVY PROGRAM FOR CONTRACTOR OPERATION AND MAINTENANCE OF SIMULATORS

Gordon Steven Dow

Simulator Operation and Maintenance Analyst

Naval Training Equipment Center

The Navy Contractor Operation and Maintenance of Simulators (COMS) program, which replaces the disestablished TRADEVMAN rating, began in October 1982.  The Contractor Performance Factor (CPF) has been formulated as a quantitative measure of performance to help government personnel monitor COMS contracts.  The CPF is measured against a pre-established baseline requirement.  Three types of CPF being employed by the Naval Training Equipment Center have been applied to major training systems, multi-station training systems, and miscellaneous portable training equipment.  The CPF is reported on each training system by the contractor, via the Contracting Officer's Technical Representative (COTR), to the procuring activity.  The CPF feedback is used by management to make budgeting, contracting, and planning decisions.

This paper is available on the I/ITSEC Compendium CD-ROM.

Order it from I/ITSEC’s Website.