Year: 1992

Authors: John Larson, Michael Rolecki, Michael Wiedemer

Abstract:

For a training system to be effective, it must accurately represent the operational equipment responses and performance as they will be perceived by the trainee. In modern avionics systems, response and performance are largely determined by the avionics software. The inherent flexibility to enhance the performance of modern systems has historically resulted in frequent software changes. This dynamic nature of avionics software has posed a significant challenge to training systems to keep up, or remain concurrent, with the avionics.

The B-1B is one example of such a training system. The offensive and defensive avionics response and performance are largely determined by the embedded software. Furthermore, the training system includes two devices, the cockpit procedures trainer (CPT) and the weapon system trainer (WST), which employ different approaches to avionics software simulation. The CPT employs trainer-unique software to model the avionics while the WST employs actual aircraft avionics processors and embedded software. Development and upgrade of these two training devices provides an excellent comparison between the "simulate" and "stimulate" approaches and their capabilities for aircrew training concurrency.

This paper will use the B-1B training system, now fielded for over two years, to illustrate the challenges presented by concurrent development and upgrade of the avionics and training systems. Early concurrency planning and trade studies will be reviewed and compared with actual program results. The technical and management mechanisms employed to maximize concurrency of the fielded training system will be described. Finally, applicability of B-1B concurrency concepts to future training systems will be discussed, including lessons learned and development risk factors not considered in the early trade studies.

Year: 1992

Authors: Omeed Alaverdi, Michael Warner, Jose Abanero

Abstract: In support of the Special Operations Forces Aircrew Training System MC-130E simulator development, a flight test program was conducted to collect flight dynamics data for use in both model fidelity improvements and simulator acceptance testing. Due to the lack of several essential test signals, it was necessary to embark on an extensive data reconstruction and calibration task prior to applying the data for simulation validation or upgrade. This paper outlines the flight test planning and data calibration process needed to develop and validate a high fidelity aerodynamic model. The use of optimal estimation techniques for reliable and kinematically consistent data calibration is emphasized.

Year: 1992

Authors: Grace Mak-Cheng, Kenneth Doris, Frank Zinghini

Abstract:

With the emergence of technical standards for networking defense simulations, a means to evaluate the applicability of the new DIS PDU's and the draft DIS networking requirements to joint services applications is needed. This paper describes a rapid prototyping testbed which was developed to network two simulating devices using a fiber distributed data interface (FDDI) local area network (LAN). A Distributed Interactive Simulation Interface Unit (DIU) was developed to interface each simulating device to the FDDI LAN. The intent of the DIU was to off-load the host processor and to minimize the changes to existing trainers as a result of networking. The DIU performed the dead reckoning and translation of the data from(to) the DIS PDU protocol to the host computer format. The DIU was developed using both Ada and C and the simulating devices were developed in Pascal.

The DIS PDU protocol was also used as the communication protocol between the simulating device and the DIU. Thus, this prototype implemented the DIS PDU protocol in three different languages. The "lesson learned" from this DIS implementation and suggestions to undeveloped areas, such as Simulation Management and Electronic Warfare, in the DIS standard are discussed.

Year: 1992

Authors: Frank Zinghini, David Berkman

Abstract:

Distributed Interactive Simulation (DIS) is the standard network protocol that promises to provide a common ground for the interplay of a variety of simulators in a single, integrated exercise. Many new training systems procurements require DIS support and, of course, mandate that such support be written in Ada. This in itself can be daunting, because Ada's rigid structure does not easily accommodate the complexity and diversity of information that constitutes the protocol. However, if the concept is to succeed it must also take into account the vast array of simulators that already exist, which will require implementation of DIS in all of the programming languages that preceded Ada.

This paper will examine the DIS protocols from the perspective of software implementation, and will discuss the problems and pitfalls of supporting those protocols in the variety of languages used in today's simulators. The features and quirks of these languages - strong typing, variable-length record support, dynamic allocation, and so on - will be explored for their effect on implementing the protocols. Specifically, Grumman's experience with implementing DIS in Ada, C, Pascal, and FORTRAN will be discussed. Each language has its own peculiar way of defining and manipulating data structures, and each has its good and bad points when applied to DIS.

Year: 1992

Author: Melinda Carlie

Abstract:

Much of the challenge associated with the application of computer image generation has shifted from the development of technical capacity to the implementation of mission training strategies that optimize the technology. Since current imagery capability can create realistic visual scenes, the challenge is met by balancing the desire for near exact correlation between the operational environment and the modeling of mission data bases against the training requirements of an ever broader range of training tasks.

This paper describes the derivation of a process to meet two specific needs identified in the development of mission data bases for the currently fielded M1 Tank Driver Trainer. The first need is precise identification of task training requirements in clear and measurable terms to facilitate test and evaluation of mission data bases on specific criteria. This need is one of translation; the user's terminology and techniques for stating requirements are significantly different from those of the technology expert or engineer. The second need is to provide a path that facilitates design of mission data base solutions engineered to meet the stresses in processing load, vehicle dynamics, environmental complexity, and modeling that differ significantly for ground based training from those of flight simulators.

Year: 1992

Authors: Dennis Hribar, David May, James Probsdorfer

Abstract: Virtual Environment Training offers a cost effective alternative to high-priced simulators and training devices. It uses the advantages of multimedia technology to provide quick access to motivating video sequences, audio warnings, realistic images and detailed text and drawings to interactively train critical tasks. This paper will highlight a recent project completed at Newport News Shipbuilding to use Virtual Environment Training to improve performance of submarine crewmen. It will detail critical considerations in the use of Digital Video Interactive (DVI®) in Virtual Environment Training. This paper will also identify a Virtual Environment Training development process and highlight a software application that was developed to aid in training design. Finally, significant performance improvements from a recent test/evaluation of this virtual environment part-task trainer will be discussed.

Year: 1992

Authors: Duane Boman, Mark Schlager, Jennifer Gille, Tom Piantanlda

Abstract: This study was part of a larger effort to evaluate the feasibility of using virtual environment (VE) technology in aircraft maintenance training within the next 10 years. A survey of VE technologies evaluated present technology capabilities, projected near-term capabilities, and identified areas in which additional research efforts are needed to provide effective VE application systems. A parallel survey of training organizations identified maintenance tasks that could potentially benefit from VE training. By matching the projected VE technology capabilities to the performance requirements of the candidate tasks, timeframes were estimated for the delivery of a VE maintenance trainer for each task.

Year: 1992

Authors: D. Gould, C. McKinney

Abstract: This paper identifies an architecture and an educational program structure that would promote enhanced training effectiveness for military personnel by extending the traditional school classrooms to the military personnel's workplace and by providing a more flexible curriculum schedule. The Navy's surface ship and submarine forces responsible for sonar and weapons systems will be used to illustrate examples of how this architecture can be implemented to extend classes from the Navy's classrooms to ships, bases, offices, and remote locations. We illustrate how distance learning systems can be used as the backbone for this architecture and how the student work-stations in this architecture also can be used to accomplish Computer-Based Training (CBT), Part Task Training (PTT), and operational training, among others. We will identify technologies, such as video and audio digital compression, that provide the capabilities for the two-way interactive participation between an instructor and students at separate locations. Finally, we will address how this architecture can help eliminate problems with student availability and costs for access to the traditional classrooms, as well as facilitate a student's natural progression through an established curriculum.

Year: 1992

Author: Mark McKeon

Abstract: The multi-mission data processing capability of today's fourth generation fighters has attained an unprecedented level. To elevate the tactical level of aircrew thought, the Aircrew Coordination Support System (ACSS) of tomorrow's Strike/Fighters will process and display even more Fighter/Attack information. The vast capabilities of the ACSS will generate new group task analysis and tasking requirements. To be successful in the dynamic environment of multi-mission combat sorties, more sophisticated integrated aircrew coordination must be obtained. Problems, however, have already arisen in the orchestration of this process and will continue to present themselves. Many of the aircrew coordination problems can only be solved through the employment of on-board computer supported embedded training systems. This presentation will address the demands that drive the development and employment of airborne embedded training systems. Also addressed will be the required training support that these systems must provide the aircrew. Finally, the importance of embedded training systems in addressing multi-mission aircrew coordination problems will be presented.

Year: 1992

Author: Alfred Bishop

Abstract:

This paper presents a case study of a first-time experience in Ada software development. It identifies how design goals were established and how the technical challenges of implementation were overcome. It also provides some informative statistics on the resulting Ada code.

The Combined Arms Staff Trainer (CAST) is a battlefield training device that provides the U.S. Marine Corps with a highly realistic, integrated environment for wargaming. The trainer consists of a large, 3-dimensional terrain board, a laser system for marking opposing and friendly fire location, audio communications, and networked personal computers. CAST requires 12 types of computers, each having a complicated, menu-driven user interface. A main design goal was an integrated menu system capable of processing approximately 400 required menus and minimizing code changes resulting from minor changes in menu content or format. The intercomputer communications design was based on networked file sharing—the mailbox paradigm. A software team of six people used object-oriented design to implement 77,000 lines of Ada (332 library units). A compilable program design language formed a basis for partially automating the preparation and maintenance of detailed design documents. The problems encountered and the lessons learned as a result of the project are discussed.