The future of air combat is expected to evolve significantly to include new technologies and novel concepts of operation. The Manned-Unmanned Teaming concept involves low cost, attritable Unmanned Aerial Vehicles (UAVs) that could be deployed along with a manned aircraft. The UAVs act as a complementary asset and bolster offensive air operations. Given the complexity of future operating environments, the degree of autonomous control required for pilots to concurrently operate multiple UA Vs and their own aircraft is one area of concern. To determine the amount of autonomous control abstraction that has the largest impact in reducing operator workload and increasing system performance, a predictive workload model was developed using the Improved Performance Research Integration Tool (IMPRINT). This research concluded that maned-unmanned teams can increase mission performance and maintain the pilot's cognitive workload at a manageable level by utilizing higher levels of human control abstraction, where unmanned systems have greater degree of autonomy.

Effective teams coordinate their actions to achieve shared goals. In Human-Agent teams, the Artificial Intelligent Agents (AIAs) struggle to coordinate effectively due to a lack of understanding of their human teammate's intent. This places a burden on the human teammate to extensively communicate explicitly what goals they are pursuing and how they are pursuing them. To improve the AIAs ability to coordinate, we have proposed Operationalized Intent as a means to explicitly model how an operator qualitatively desires a task to be performed. In this paper, we report the results of a study to track operator intent through a tactical scenario. The focus of this paper is on the dynamics of intent and it's cohesiveness across operators. The study employed an immersive, advanced research, remotely piloted aircraft (RPA) simulator to study intent in a synthetic task environment. Using operational pilots and sensor operators in realistic scenarios we were able to elicit their intent under naturalistic conditions in the midst of challenging tactical situations to study the real-time dynamics. Analysis indicates that the method models intent which is dynamically responsive to changes in the situation and the data are suitably cohesive across operators to generalize to an operator role. When the intent data is coupled to situated data from the simulator it provides a labeled data source for future AIAs to estimate intent in real-time.

Loss of user engagement with automation in increasingly complex environments, such as driving a car, can have increasingly unexpected and consequential impacts. This paper presents a three-step process for designing cooperative automation into existing systems to maximize system performance with the constraint of maintaining user engagement. The first step identifies beneficial areas for automation through user testing, focusing on areas where human capabilities constrain performance. Next, an automation is created to perform a portion of the task leading to the constraint. User tests at this point compare different automation types to choose an automation that improves on human-only performance. Lastly, the automation trigger is adjusted to allow the human to maintain engagement with the system at a level desired to avoid the risks associated with loss of user engagement. This process is demonstrated through three human-in-the-loop experiments which illustrate the trade-space between maintaining human engagement and improving system performance.

Levels of Autonomy (LoA) provide a method for describing authority granted to operators and autonomous system elements. Unfortunately, LoA does not provide the user interface designer a clear method to distinguish interface concepts which impose varying levels of operator workload or result in human or system performance changes. The current research suggests an alternate classification framework for vehicle control, referred to as the Level of Human Control Abstraction (LHCA). LHCA describes how an operator controls a system based on the control tasks performed and the level of detail of decisions made by the operator. The proposed framework consists of five levels: Direct Control, Augmented Control, Parametric Control, Goal-Oriented Control, and Mission-Capable Control. It is suggested that as the level of detail of control is reduced through progression from Direct Control to Mission-Capable Control, the level of human attention, and workload will be reduced.

Adaptive automation has been proposed, where changes in automation are triggered based upon operator state to mitigate automation-induced problems such as complacency. This research sought to understand the effect of a weighted method, as compared to a method in which all tasks are weighed equally, when triggering changes in automation within a multitasking environment. The weighted method considered the priority and frequency of each task when computing a measure of operator performance on which to trigger changes in automation. Although overall system performance was not statistically different between the two trigger implementations, the participants with the priority based triggering scheme rated the level of automation change as more aligned with their performance and were less surprised by changes in automation level.