Autonomous vehicles have become an integral tool for military use on the battlefield, and across the naval seaboard. Manning these vehicles has traditionally required the dedication of three or more people to successfully pilot, navigate, and plan the various waypoint and targeting solutions involved during the Unmanned Aerial Vehicle (UAV)’s mission. Ironically, this strategy requires more manpower to leverage a system designed to reduce the staffing footprint.

  As UAV technology has improved, more of the higher level tasking, prioritization, and target vectoring decisions have shifted away from manned personnel to the UAV itself. The potential now exists to replace teams controlling each UAV with a single person who can supervise a single UAV, or even a cluster of UAVs. This would also introduce an increased risk of human error however, as one person would now be responsible for keeping several UAVs on track. In order to transition over to this new supervisory model, steps must be taken to mitigate the risk of human error.

  The Naval Research Laboratory (NRL) has been proactively engaged in collecting human subject data through the use of its Supervisory Control Operations User Testbed (SCOUT) system. This system is designed to capture several different aspects of human response while on the job. Through observing UAV operators as they work, the system can detect changes in the individual’s workload and engagement levels, as assessed via a combination of physiological and task performance data. Physiological data is gathered from eye tracking, heart rate and respiration rate sensors which are synchronized with the simulation data. Performance metrics are determined based upon the user’s reaction times, accuracy and completion rates of tasks within the simulation, as well as plan quality for assigning objectives to UAVs. By testing a large sample size, correlations in these scores and job performance will begin to emerge, and a baseline can be established. With baseline data it becomes possible to predict the limits of one single operator’s abilities and assign tasks and UAVs to other operators if necessary. Right now this data is stand-alone, but the true value lies in real time integration within the 3D battle space environment.

  To simulate and train personnel on the usage of UAV interaction in real world engagement, the Navy has developed a prototype 3D simulation environment known as the Fleet Integrated Synthetic Test and Training Facility (FIST2FAC) Lite, herein referred to as FIST2FAC. The name was taken from the Navy’s reconfigurable laboratory aboard Ford Island in Pearl Harbor, HI, Naval Undersea Warfare Center (NUWC) Detachment Pacific’s “Fleet Integrated Testing and Training Facility” where it was first demonstrated in 2013. FIST2FAC is structured around Unity 3d as the principal image generator, and other custom software to accurately represent the different objects, terrain weather, and waypoint data for each mission. Through interaction with this environment, it is possible to run through various scenarios graphically, and more closely train towards the dynamic nature of the threat in an integrated fashion.

  With the integration of the SCOUT performance data into the 3D environment of FIST2FAC, it would be possible to better understand the role of workload and engagement amongst sailors during simulated training exercises. Additionally the integration will introduce supervisory control of UAVs into the battlespace environment. Ultimately, the two systems working in unison and in real time will be able to provide valuable data that can be used to build baseline data on operator performance, for various operators as well as UAV operators in a supervisory role. This will increase the level of confidence in one-to-many UAV control model to the level that is needed to justify the safety margin involved with completing the transition from the current model. Furthermore, it will also introduce an additional mechanism to assess combat readiness during training exercises, by way of collecting biometric data on workload and engagement.

  A team of graduate students from the George Mason University (GMU) Systems Engineering and Operations Research (SEOR) Department will work with a team of GMU Game Design undergraduate students to develop an Integration Roadmap to investigate and simulate the integration of SCOUT and FIST2FAC. This GMU project comprises of the first step toward building the integrated system. The team will produce a feasibility analysis for integration, which will also include a recommendation for implementation of the integration or present alternatives should integration seem unrealistic. This documentation will provide a path forward to implement the integration between the systems.

  The Concept of Operations (CONOPS) describes how the integrated system would conceptually work, should the feasibility analysis result in a recommendation to integrate. Storyboarding through operational scenarios assisted in the development of an integration design. This document is also intended to create a clear picture of the desired end state of the integrated system so that the audience and users can gain a clear understanding of the value of the project.