Human Autonomy in Future Drone Traffic: Joint Human-AI Control in Temporal Cognitive Work

Abstract

The roles of human operators are changing due to increased intelligence and autonomy of computer systems. Humans will interact with systems at a more overarching level or only in specific situations. This involves learning new practices and changing habitual ways of thinking and acting, including reconsidering human autonomy in relation to autonomous systems. This paper describes a design case of a future autonomous management system for drone traffic in cities in a key scenario we call The Computer in Brussels. Our approach to designing for human collaboration with autonomous systems builds on scenario-based design and cognitive work analysis facilitated by computer simulations. We use a temporal method, called the Joint Control Framework to describe human and automated work in an abstraction hierarchy labeled Levels of Autonomy in Cognitive Control. We use the Score notation to analyze patterns of temporal developments that span levels of the abstraction hierarchy and discuss implications for human-automation communication in traffic management. We discuss how autonomy at a lower level can prevent autonomy on higher levels, and vice versa. We also discuss the temporal nature of autonomy in minute-to-minute operative work. Our conclusion is that human autonomy in relation to autonomous systems is based on fundamental trade-offs between technological opportunities to automate and values of what human actors find meaningful.

Keywords: UTM; autonomy; human-centered AI; interaction design; joint control framework; scenario-based design; unmanned traffic management; visualization design.