Evaluating Human Interaction with Automation in a Complex UCAV Control Station Simulation Using Multiple Performance Metrics

Abstract

The dynamics and complexities of human–machine systems and the overwhelming amount of data that must be handled by human operators is making automation a critical factor in planning, decision-making, and in execution in many complex systems. Complex systems are characterized by uncertainty, ambiguity, ill-defined goals, dynamically changing conditions, distractions, and time pressures. A semi-autonomous system requiring significant human-centered design support are remotely operated vehicles (ROVs) such as unmanned aerial vehicles (UAVs), unmanned combat aerial vehicles (UCAVs), space maneuverable vehicles (SMVs), and unmanned emergency vehicles (UEVs). The objectives of this research are to (1) develop a simulation system that would allow investigation of human operator performance issues when supervising multiple UCAV vehicles, and (2) investigate human performance through the collection of multiple dependent measures. The simulation tool was designed to be adaptable to allow continued research on a variety of factors related to the control of autonomous vehicles. A research study using this simulation tool investigated the effects of automation and the number of UCAVs being controlled on operator performance during an identify and destroy mission. Results indicate that increasing the number of UCAVs significantly increased workload under LOW Automation and the increase in workload was reduced when HIGH Automation was introduced. This study showed that operators are able to control multiple UCAVs more effectively with appropriate automation.