Aviation Human Factors Division - Institute of Aviation
INTRODUCTION
Unmanned aerial vehicles (UAVs) are quickly becoming a part of the national airspace system (NAS) as they transition from primarily military and hobbyist applications to mainstream flight applications such as security monitoring, satellite transport, and cargo hauling. Before the full potential of UAV flight in the NAS can be realized, however, FAA standards and regulations for UAV operations must be established. Given the experience of the U.S. military that mishap rates for UAVs are several times higher than for manned aircraft (Williams, 2004)—over thirty times higher, in some cases (Department of Defense, 2001)—the importance of carefully designed standards and regulations is clear.
Issues related to human factors are likely to be of particular concern in establishing guidelines for UAV flight. As noted by Gawron (1998), UAV flight presents human factors challenges different from and in some ways greater than those of manned flight. These arise primarily from the fact that operator and aircraft are not co-located. As discussed in more detail below, the separation of operator and vehicle imposes a number of barriers to optimum human performance, including loss of sensory cues valuable for flight control, delays in control and communications loops, and difficulty in scanning the visual environment surrounding the vehicle. Unmanned flight also allows the possibility that a single operator might control multiple vehicles simultaneously, a task likely to impose unique and heavy workload demands.
The goal of the current work was to examine the existing research literature on the human factors of unmanned flight, and to delineate issues for future research to address. The topics discussed below are divided into the categories Automation; Perceptual and Cognitive Aspects of Pilot Interface; Air Traffic Management Procedures; and Crew Qualifications. As will be clear, however, the issues covered within the various categories are highly interrelated. Answers to questions about crew complement, for example, will be contingent on the nature and reliability of automation provided to support UAV operators. Likewise, decisions about interface design will depend on the extent to which flight control is automated, with manual flight mode demanding traditional stick-and-rudder controls and automated flight mode allowing for point-and-click menu-based control or other forms of non-traditional interface.
It is also important to note that unmanned aircraft will likely serve a range of purposes in civilian airspace, and that the demands placed on human operators will vary with characteristics of the flight mission. Proposed uses for UAVs include agricultural, geological, and meteorological data collection; border surveillance; long distance transport; search and rescue; disaster monitoring; traffic monitoring; and telecommunications relay. Furthermore, military UAVs will increasingly be required to transition through civilian airspace en route to their missions. In some of these cases, the vehicle is likely to operate solely within line-of-sight communications range and only over relatively short periods of time (i.e., on the time scale of several hours or less). In other cases, the vehicle will operate at distances demanding over-the-horizon communications, and will potentially remain airborne for many days on end. These mission characteristics will modulate concerns about communications delays between ground control station and vehicle, and about the need for transfer of vehicle control between crews. For some applications, additionally, operators will likely be required to make frequent control inputs, adjusting flight parameters or selecting new waypoints “online” in response to changing task demands or conditions. For other applications, flight path will be predetermined and less susceptible to modification, reducing the immediacy and frequency with which operators are required to intervene in flight control and allowing for a heavier reliance on automated vehicle guidance.