A Comparative Study of Multimodal Displays for Multirobot Supervisory Control

The supervisory control of ground-based mobile multirobot systems requires to perform multiple concurrent tasks under high levels of time pressure resulting in heavy workload. In this paper we present the design and evaluation of multimodal displays for a particular problem associated with the supervisory control of ground-based multirobot systems: the coordination between the platform specific robot control task, e.g. navigation and obstacle avoidance, and the mission specific payload task. The coordination requires the operator to concurrently monitor and switch attention between the robot control and the payload control tasks depending on the mission requirements. Multimodal human-robot interfaces can significantly support human information processing by communicating information across multiple channels and can therefore improve concurrent task processing. An experiment was designed and carried out with 14 participants which compares four human-robot interface configurations with a simulated two-robot ground-based multirobot system. The results show that the multimodal interfaces perform significantly better across multiple variables and have the lowest workload. Based on our gaze tracking results we can conclude that our multimodal interface has an effect on the visual scanning behaviour in the peripheral regions of the camera display.