Assisting interruption recovery in supervisory control of multiple UAVS

Performance degradation due to interruptions is a critical issue, particularly when people are supervising highly autonomous systems in time and safety critical environments. Previous research in the development of automated support to help supervisory control operators resume task activities after an interruption has had limited success. This paper describes two new interruption recovery approaches that attempt to mitigate the disadvantages of previous approaches. In particular, this paper describes the design and initial investigation of a prototype assistive interface developed to support interruption recovery for supervisory control of multiple unmanned aerial vehicles (UAVs). These two 'replay'-type interruption recovery approaches enable increased user-control of the event discovery process and provide event 'bookmarks' to highlight emergent system events. The findings from this initial study provide several recommendations for the future design of interruption assistance tools for human supervisory control tasks.

Duke Scholars

Author Mary "Missy" Cummings Electrical and Computer Engineering