FUNCTIONAL REQUIREMENTS FOR EFFECTIVE DECISION MAKING IN HUMAN- HUMANROBOT TEAMS: LESSONS LEARNED FROM OPERATIONAL SETTINGS

摘要

Many environments require humans and robots operating together to accomplish complex and dangerous tasks, but technology-centered designs often support robot navigation but not the mission goals of the organization using the robot. Urban Search and Rescue (USAR) is a particularly valuable domain to identify general functional requirements for effective HRI, and our purpose in this paper is to demonstrate how a CSE approach can lead to valuable design guidelines that more effectively support decision making within Human-Robot teams. Our analysis of HRI in USAR lead us to identify several important guidelines for supporting effective coordination for Human-Robot teams: including (i) enable individual problem holders to have direct control over point-of-view to facilitate active information seeking, and (ii) create common reference frames on shared imagery so different problem holders can remotely coordinate information and actions. Designs based on these guidelines will support a broad class of coordinated activities between team members.