Development of an Autonomous Manager for Dynamic Human-Machine Task Allocation in Operational Surveillance

Surveillance environments require simultaneous performance of multiple complex tasks. These increasing demands require the effective leveraging of automated teammates. However, the degree of assistance autonomous systems can provide is still limited for cognitively complex tasks. Balancing the workload requires intelligent and dynamic redistribution of tasks between humans and autonomy. To address this challenge of maintaining performance in a Human-Machine Team (HMT), we developed an Autonomous Manager (AM) to dynamically redistribute tasks between a human and automation based on task performance and physiological indicators of workload. We tested the AM's decision logic across multiple scenarios using simulation, allowing us to examine the benefits and limitations of the AM.