Description:
OBJECTIVE: Develop a self contained deployable system to automatically quantify combined human and systems performance in real-time and for after-action-review by fusing output of normative models of behavior, human state, system state, and contextual situation state. DESCRIPTION: Complex weapons systems require years of training for crews to master all aspects of the system, the situations in which they are deployed, and the human to human interactions that are necessary for mission success. While such systems are developed on the basis of stringent functional requirements to meet the needs of the ultimate users in the fleet, it is possible that expert crews have become accustomed to and compensate for adverse system traits that would be unacceptable or even dangerous when the system is used by crews with less experience in demanding theater operations. Expert crews are usually very skilled operators of the systems they operate with an extremely well developed behavioral knowledge base to recognize appropriate sequences, tactics, and contexts of the situation. Such a high level of expertise is necessary to apply the system to its full potential in tactical situations with rapidly changing parameters and limited time to evaluate possible decisions. For example, experienced pilots can operate most aspects of the aircraft with a high degree of skill and focus on the mission evolution rather than being occupied by inner-loop control of the aircraft and associated systems. However, experienced pilots may inadvertently use their expertise to compensate for and overcome adverse systems traits without much effort or attention whereas less experienced pilots in the same situation may encounter the system as unacceptable or even dangerous. Aside from the fact that undesirable system traits can adversely affect flight safety, the costs for design changes to address adverse system traits increase exponentially throughout the system development cycle and for fielded systems the costs can take on catastrophic proportions. With the proper measurement tools, it would be possible to detect adverse system traits early on. What is needed is research, development, and technology transition to arrive at a self-contained deployable system that automatically supports assessment of the combined human-systems performance on the basis of intelligently fused sources of real-time information derived from normative behavior models, the human operator, the system, and the situational context. The proposed systems concept should be based upon a novel model that addresses the differences in procedural and cognitive behaviors that would be expected at varying levels of expertise. The system should be able to provide performance assessment feedback in real-time and during after-action-review and it should be based on accurate real-time system state data such as position, velocity, accelerations, control manipulations, as well as human state data such as eye tracking, heart rate variability, and neural measures of performance from existing commercially available sensors and situation context such as planned trajectories and performance envelopes. The developed system should provide detailed means to record, annotate, process, transmit, and display pertinent information derived from the source data to indicate overall systems performance, operator state including attention and loading, and performance deviations from normative expectations. The system should provide the means to generate a detailed and easy to understand record of a test sortie so that potentially adverse system traits can be identified, documented, and predictions could be made on how these traits, if left unmitigated, could affect system performance in an operational tactical environment with normal fleet user characteristics. As such a technology would be most useful if it could be deployed in operational and possibly distributed systems it is important that considerable attention be given to design of the form factor, processor miniaturization and integration, transmission protocols, and technical readiness of the proposed sensor solution. PHASE I: Define and develop a novel model that captures differences in cognitive and procedural behaviors, concept of operations, architecture, necessary algorithms, research, development, and transition plan to develop a system that can automatically quantify combined human and systems performance in real-time and for after-action-review by fusing output of normative models of behavior, human state, system state, and contextual situation state. PHASE II: Produce a deployable prototype of the Phase I concept, collect development and validation data in a high fidelity simulation test environment that closely approximates an operational system of interest to a transition customer, and demonstrate its use and benefits to transition stakeholders. Refine transition plan on the basis of study data and demonstration feedback. PHASE III: Finalize design developed in Phase II, conduct testing in an operational environment and transition to interested parties. PRIVATE SECTOR COMMERCIAL POTENTIAL/DUAL-USE APPLICATIONS: It is conceivable that non-military commercial application opportunities will exist for testing of operational systems in an industrial setting such as automobiles, aircraft, information systems, and consumer products. REFERENCES: 1. Dwyer, D. J. & Salas, E. (2000). Principles of Performance Measurement for Ensuring Aircrew Training Effectiveness. In O'Neil Jr., H.F. & Andrews, D.H. (Eds.),"Aircrew Training and Assessment"(223-244). Mahawan, NJ: Lawrence Erlbaum Associates. 2. Gevins, A. & Smith, E. (2005). Neurophysiologic measures for Neuroergonomics. In Schmorrow, D.D. (Ed.)"Foundations of Augmented Cognition"(841-850). Mahwah, NJ: Lawerence Erlbaum. 3. Stevens, S.M., Forsythe, J.C., Abbott, R.G., & Giesesler, C.J. (2009). Proceedings from the 5th International Conference on Foundation of Augmented Cognition. Neuroergonomics and Operational Neuroscience: Held as part of HCI International:"Experimental Assessment of Accuracy of Automated Knowledge Capture."San Diego, CA.
