Effects of Imperfect Automation on Operator's Supervisory Control of Multiple Robots

Abstract

Abstract : A military multitasking environment was simulated to examine the performance of robotics operators whose task was to manage a team of four ground robots with the assistance of RoboLeader, an intelligent agent that could coordinate the robots by changing their plans based on developments in the mission environment. The reliability of RoboLeader s recommendations was manipulated to be either false-alarm prone (FAP) or miss prone (MP), with a reliability level of 60% or 90%. The visual density of the targeting environment was also manipulated. Results showed that the type of RoboLeader unreliability (FAP vs. MP) affected operator s performance of tasks involving visual scanning (target detection, route editing, and situation awareness). There was a consistent effect of visual density for multiple performance measures. Participants with higher spatial ability performed better on the two tasks that required most visual scanning (i.e., target detection and route editing). Participants self-assessed attentional control was found to impact their secondary tasks (communication and gauge monitoring) more than their primary tasks (target detection and route editing).