Supervisory Control of Multiple Robots

Abstract

Objective: A military multitasking environment was simulated to examine the effects of an intelligent agent, RoboLeader, on the performance of robotics operators. Background: The participants’ task was to manage a team of ground robots with the assistance of RoboLeader, an intelligent agent capable of coordinating the robots and changing their routes on the basis of battlefield developments. Method: In the first experiment, RoboLeader was perfectly reliable; in the second experiment, RoboLeader’s recommendations were manipulated to be either false-alarm prone or miss prone, with a reliability level of either 60% or 90%. The visual density of the targeting environment was manipulated by the presence or absence of friendly soldiers. Results: RoboLeader, when perfectly reliable, was helpful in reducing the overall mission times. The type of RoboLeader imperfection (false-alarm vs. miss prone) affected operators’ performance of tasks involving visual scanning (target detection, route editing, and situation awareness). There was a consistent effect of visual density (clutter of the visual scene) for multiple performance measures. Participants’ attentional control and video gaming experience affected their overall multitasking performance. In both experiments, participants with greater spatial ability consistently outperformed their low-spatial-ability counterparts in tasks that required effective visual scanning. Conclusion: Intelligent agents, such as RoboLeader, can benefit the overall human-robot teaming performance. However, the effects of type of agent unreliability, tasking requirements, and individual differences have complex effects on human-agent interaction. Application: The current results will facilitate the implementation of robots in military settings and will provide useful data to designs of systems for multirobot control.