Automation reliability in unmanned aerial vehicle control: A reliance-compliance model of automation dependence in high workload

Objective: Two experiments were conducted in which participants navigated a simulated unmanned aerial vehicle (UAV) through a series of mission legs while searching for targets and monitoring system parameters. The goal of the study was to highlight the qualitatively different effects of automation false alarms and misses as they relate to operator compliance and reliance, respectively. Background: Background data suggest that automation false alarms cause reduced compliance, whereas misses cause reduced reliance. Method: In two studies, 32 and 24 participants, including some licensed pilots, performed in-lab UAV simulations that presented the visual world and collected dependent measures. Results: Results indicated that with the low-reliability aids, false alarms correlated with poorer performance in the system failure task, whereas misses correlated with poorer performance in the concurrent tasks. Conclusion: Compliance and reliance do appear to be affected by false alarms and misses, respectively, and are relatively independent of each other. Application: Practical implications are that automated aids must be fairly reliable to provide global benefits and that false alarms and misses have qualitatively different effects on performance.