While many organizations believe that the human-to-robot (m:N) ratio is a key factor driving workload during operation of uncrewed aircraft systems, does this assumption hold for the operation of the highly autonomous systems often deployed across industry today? Unfortunately, there is a lack of robust conceptual workload models that capture human performance while interacting with these highly autonomous systems. The current study addresses this gap by modeling a supervisory task where a pilot-in-command supervised a large fleet of autonomous delivery drones in a large urban metro area. Across various scenarios, while the number of supervised aircraft had a small impact on workload, it appears that other task-based cognitive and visual demands underlied observed changes in workload. These modeling efforts provide an initial baseline dataset, derived within a real-world context, to help guide subsequent investigations into what factors influence perceived workload during the operation of automated uncrewed aircraft.
Giolando et al. (2025) studied this question.