Improved Human-Robot Teaming through Facilitated Initiative

Abstract

This paper evaluates collaborative tasking tools that facilitate dynamic sharing of responsibilities between robot and operator throughout a search and detection task. The goal is to arbitrate human and robot initiative such that the user can provide input at different levels without interfering with the robot's ability to navigate, avoid obstacles, plan global paths, and achieve task goals. A real-world search and detection experiment is used to compare standard shared mode (SSM), where robot behavior unfolds from the robot's perception of its local environment and the attainment of task goals are the result of continuous operator supervision, to a collaborative tasking mode (CTM), where operators input mission level tasks and the system dynamically constrains user and robot initiative based on the task element. Participants who utilize CTM do not experience a significant performance penalty, yet benefit from reduced workload and fewer instances of confusion. In addition, CTM participants report a higher overall feeling of control as compared to those using SSM