Abstract

Cognitive Model-enabled Simulation-based Training
of Aerospace Operations Center Operators

Azad M. Madni, Ph.D. and Carla C. Madni
Intelligent Systems Technology, Inc.

Sharon K. Garcia, Ph.D.
AFRL/HEPC

Abstract

The Aerospace Operations Center (AOC) is a weapon system for the command and control of joint forces in the dynamic battlefield as we go into the future. The training of command and control (C2) personnel, who man the AOC, continues to lag because of personnel turnover and lack of cost-effective "anytime, anywhere" training tools. The challenge in today's compressed personnel assignment cycle is to develop training that satisfies an operator's individual and collaborative decision making needs, and to provide that training in a manner that rapidly raises the learner's proficiency to a level that enables the learner to perform effectively in the operational environment. Simulation-Based Training (SBT) is the approach of choice to achieve these objectives. However, the development of SBT systems tends to be time-consuming, labor-intensive, and expensive. Once developed, these systems tend to be difficult to extend or change. In part, these characteristics are a consequence of the fact that most SBT systems in the military tend to be built from or on top of legacy components which generally tend to be relatively high fidelity and, consequently, expensive. In recent years, the SBT community has begun to look for new and innovative approaches to slash both development time and product costs while ensuring simulation scalability. This paper presents ProcessTrain™, a cognitive model-inspired SBT system that is both cost-effective from a development perspective and scalable in terms of supported users and simulated entities. ProcessTrain is built on four pillars: cognitive models; outcome-driven simulation; instructional vignettes of graduated difficulty; and the principle of selective fidelity. ProcessTrain harnesses cognitive models to: represent the expert's conceptualization of the domain and the decision making processes; guide scenario/vignette design as well as the design of feedback, remediation, and explanation facilities; assist in the specification of learning events and required interaction among roles; and facilitate understanding of learner errors. The outcome-driven simulation strategy ensures that the simulation unfolds along only those branches/decision paths that are rich in learning experiences and relevant to the attainment of learning objectives. ProcessTrain employs instructional vignettes of approximately five to ten minutes duration to keep the learner engaged. These vignettes, which are metadata tagged in terms of underlying knowledge and skills and difficulty level, are dynamically selected for presentation based on the learner's performance and scenario timeline constraints. The principle of selective fidelity, which emphasizes observational fidelity over computational accuracy, is used to simplify the simulation, and reduce costs. The paper concludes with a summary of benefits and high-payoff applications for the system.

From:

Madni, A.M., Madni, C.C., and Garcia, S.K., Proceedings of the 11th International Conference on Human-Computer Interaction, Las Vegas, NV, July 22-27, 2005.