Year: 1988

Author: Gary Sackett

Abstract:

Embedded training, defined as the use of actual system equipment with specialized training software, is being specified for more and more systems. The Advanced Automation System (AAS), designed for the Federal Aviation Administration (FAA) by Hughes Aircraft Company, leads the way in the application of embedded training and embedded simulation capabilities to solve a complex air traffic training problem. This paper addresses the approach taken by Hughes to develop training system requirements, convert these requirements into design requirements using embedded training/simulation capabilities, and implement the design. Two specific modes of embedded simulation capabilities will be described, system supported and standalone training. These modes take advantage of the system design which uses a large, central computer for certain global or strategic functions and smaller common computers distributed throughout the system for localized or tactical processing.

System supported training uses the entire air traffic control system to provide the training capabilities. Consoles, resident on the system local area network (LAN) and designated for training, use a special simulation Computer Software Configuration. Item (CSCI) in the central computer which stimulates system software to provide an overall simulation of the entire air traffic control environment. This full-scale simulation is designed to train air traffic controllers to full proficiency.

Standalone mode uses single air traffic control consoles, detached from the system LAN, controlled by an on-board computer. These consoles execute a simulation CSCI which replaces part of the console-resident software. This CSCI interfaces with other unmodified operational software (a CSCI in a second processor in the console) to fulfill its training mission: teaching operational and maintenance procedures.

This paper shows how, through effective early design of training capabilities, the overall system design was influenced to make the embedded training features possible without adding substantial hardware and software cost to the program. The final result is an integrated approach to training which maximizes use of existing resources to provide critical training without impacting on going air traffic control operations.

Year: 1988

Author: John O'Reilly

Abstract: Tactical missions are very demanding scenarios for both pilots and instructors. The pilot expects a flight simulator to provide a realistic simulation while the instructor requires an efficient teaching environment. Artificial intelligence (AI) provides both of these expectations: AI, in the form of an expert system, produces realism through fuzzy logic and it delivers a scenario which is easily generated and modified. The realism goal is achieved by the expert system which uses a set of protocols or rules which define the operation of threats within the tactical mission. These rules are generally English-like statements which the instructor can manipulate to form the necessary mission. This paper describes an IR&D project which combines the use of the Prolog programming language with an intelligent threat reaction system. Prolog is ideal for this project because the language furnishes a rapid development environment, and the expert system, written in Prolog, provides for easy modifications to the scenario. The paper also discusses expert systems in general and fuzzy logic processes specifically as they pertain to the main goal and design of the project.

Year: 1988

Authors: Richard Economy, John Ellis, Robert Ferguson

Abstract:

Photographic texture has been a feature of Computer Image Generators for several years. When applied to models such as aircraft, trees, and buildings, photographic texture has produced extremely realistic results. Current methods of applying texture to the terrain surface of a simulator data base use either synthetic or self-repeating photographic texture patterns, neither of which produce the same degree of realism as that achieved with photo-textured models.

This paper describes techniques which have been developed to apply real imagery, either from aerial photographs or satellite data, to the terrain surface of a simulator data base. A data base, with extensive photo-texturing, has been built to demonstrate the results.

Year: 1988

Author: Charles Murton

Abstract: The Basic Trade Training for Royal Air Force, ground electronics technicians is the subject of a total course redesign using the Integrated Job Performance Training method. This has shown a requirement for the students to carry out practical fault-finding exercises in a way representative of operational methods. The equipments which the new course is designed to cater for are the modern, modular digital electronic equipments, which are extremely expensive and inherently unsuited to deliberate fault insertion for training purposes. Therefore a requirement has arisen to provide a method of training which teaches many of the same skills as would be learned using real equipment, but without the actual presence of such equipment. This has given rise to the concept of an ‘emulator’, a device which is capable of representing the characteristics of a number of different real equipments for training purposes. The paper discusses this concept, together with the different ways in which it might be implemented.

Year: 1988

Authors: Mark Coe, Robert Kiggins

Abstract: The acquisition of training evaluation data for inclusion in the Army's Integrated Training Management System (ITMS) generates great challenges. The nature and scope of the Army's training domain place new stresses on field acquisition devices. The volume of training data generated from squad level up and its importance within ITMS dictate that an automated acquisition solution must be found. Barcode scanning and voice recognition technologies are examined as non-computational or "acquisition-only" field devices and are found to provide advantages for many types of training activity, from garrison to National Training Center.

Year: 1988

Authors: M. Dunnam, W. Hosler

Abstract: Link has recently applied state-of-the-art instructor operator station (IOS) technology to an existing full mission training device. An intelligent workstation, designed to focus on the instructor's specific needs for each training task, has been added to an F-16 Full Mission Simulator, replacing an IOS representing an earlier technology. This system was conceived with the participation of the user, taking advantage of past experience with F-16 training requirements and methodologies. The new IOS technology addresses the needs of the instructor to accomplish specific training tasks by providing displays and controls optimized for specific tasks. Windows, color, icons, and merged text and graphics contribute significantly to the system's ability to support the instructor. Maintainability has been enhanced by decentralizing the IOS software in commercial off-the-shelf workstations. Instructors using the system indicate that it is meeting its training design goals, and that it is significantly more effective than the earlier design.

Year: 1988

Author: John Smith

Abstract: Baseline Correlation Matrix (BCM) is a fairly new program management tool in the acquisition business that provides traceability and comparison of the user's requirements, developer's specifications, and operational tester's evaluation criteria. Its primary purpose is to relate and align those requirements, specifications, and evaluation criteria to ensure orderly system development and test. A major problem encountered in the test and evaluation arena has been the agreement on the parameters selected for test evaluation. As an example, during the Initial Operational Test and Evaluation (IOTE) of the EF-111A Operational Flight Trainer (OFT), problems surfaced that obviously showed discontinuity on what the developer specified. Because of this disconnect, the Training Systems System Program Office (SPO) (ASD/YW), Aeronautical Systems Division, Wright-Patterson AFB, researched the BCM concept and developed a process for application to our training system programs. The paper will lay out the methodology used to apply the BCM to training system programs. The areas of the BCM will be summarized with a short history and how it was developed. The paper will discuss the approach and criteria used for selecting specific training programs for the BCM process and provide a guide for preparation and approval. Future test management objectives of the Training Systems SPO using BCM techniques will be outlined.

Year: 1988

Author: Lynn Thompson

Abstract:

Real Time software configuration control continues to be a problem area despite many guidelines, procedures, and regulations. Software traceability is particularly frustrating throughout the development phase. This paper will describe a "common sense" approach to software configuration management and a process of software migration during product development. The concept is independent of languages, compilers, and machine implementation.

First, a general classification scheme is developed, and controlling mechanisms are discussed. Second, the author defines a software development and migration process. This discussion will include software engineering development, test and integration, and operational and baseline areas. Next, this "common sense" approach will be related to MIL-STD 2167A. And last, an implementation of the "common sense" approach with lessons learned is presented.

While details of implementation must be developed for a specific configuration, the concept provides for traceability and controlled development with few constraints on the individual software engineer, a logical migration of software to a deliverable product, and a high degree of confidence in having "what you think you have."

Year: 1988

Author: Benjamin Covington

Abstract: The increasing role of the U.S. Army Reserve Components (RC), in strategic conventional deterrence, warfighting and contingency response capability requires optimization of RC training management, delivery and support systems. Optimization of training must focus on individual and unit readiness. It must begin with a fundamental understanding of the unique RC training environment and result in an interactive, tailored training system aligned to the realities and constraints of that environment. It must provide a modernized training system to support a modernizing RC force. The objective system to provide training to the RC must be technology based and distributed, through telecommunications and other means to RC units and soldiers at their home stations to the maximum extent practicable. The Reserve forces of all services have similar and compatible needs. In some cases, joint or multi-service applications of distributed training are both appropriate and economically advantageous. Industry will play a vital role in providing contract services for delivery of training and training support products, development of technology based, training devices, simulators and simulations, and in providing telecommunications and support for distributed training.

Year: 1988

Authors: Richard Reynolds, Leslie Fiedeldey, Brenda Hoskin, William Jorgensen

Abstract: A number of embedded training systems are currently in use or under development by the U.S. Navy. Consequently, there is a need to consolidate the experiences and findings of these embedded training development efforts, to evaluate the effectiveness of various high-level features of in-place embedded training systems and to assess the applicability of those features to future embedded training systems and sub-systems. This paper documents the lessons learned from the development and use of 15 operational training systems in the Navy. The choice of systems to be studied was based upon accessibility of the systems, their capabilities, general applicability, and technological and instructional complexity. The instructional features of the systems selected were identified and described and a taxonomy was created. Effectiveness evaluation criteria were developed and on site collection of data was accomplished by interviewing users of in place embedded training systems and administering a standardized assessment instrument. The analysis and evaluation of these training systems found few systems which could be considered true embedded training systems. Many systems made use of test target generators or data input devices which provide only rudimentary tools for training. Recommendations for the design of future embedded training systems are presented. These guidelines address the following areas of ET design: configuration, training characteristics, support, and policy.