Show Me What's Wrong: Impact of Explicit Alerts on Novice Supervisors of a Multi-Robot Monitoring System

With the rise of autonomous multi-robot systems, the role of the robot operator shifts from controlling and observing a single robot to that of a supervisor overseeing multiple robots. Previous studies suggest that timely warnings of problematic events improve a user’s ability to monitor multiple robots, however, research has not examined the influence alerts have on user monitoring behavior and their perceptions of the system. We present a 2x2 study design where we manipulate the operator’s cognitive load through the number of simultaneous robots under their control and the level of support through the absence or presence of warnings. Our findings suggest that users are more likely to fixate on alerts when shown explicitly, and task difficulty influenced the user’s willingness to allow the robots to act autonomously. Our research offers insights for advancing the design of autonomous multi-robot interfaces, emphasizing strategies to enhance the simultaneous monitoring of numerous robots.