Decision Support Visualizations for Schedule Management of Multiple Unmanned Aerial Vehicles

Abstract

Unmanned aerial vehicles (UAVs) are quickly becoming indispensable in military operations, particularly in time-critical missions. Although UAV systems are currently controlled by a team of people, in the future increased automation could allow one person to supervise multiple UAVs. These time-critical, complex single-operator systems will require advance prediction and mitigation of mission schedule problems. One challenge in designing an interface for the human/multi-UAV system is informing the operator of long-term consequences of potential mission schedule changes he or she may make. This paper presents two different decision support visualizations, StarVis and BarVis, designed to show the operator current mission schedule problems as well as the consequences of requesting schedule changes. An experiment tested these two visualizations against a no visualization control in a multiple UAV simulation. Results showed that StarVis produced the best performance and lowest subjective workload across different operational tempos, while BarVis supported lower but consistent performance and perceived workload under different operation tempos. This research effort highlights how different information provided in decision support can have different effects on performance and workload in a multiple UAV human supervisory control task.