CAREER: Understanding and Enabling Human-Robot Collaboration with Groups of People

点击翻译按钮获取中文摘要

How well a team performs depends on its internal dynamics. As robots become more and more integrated into teamwork in areas that span surgery, space exploration missions, search and rescue, or manufacturing, it is important to understand how robots impact team dynamics and a team's ability to perform. The goal of this project is to investigate the impact that a robot has on a team's dynamics and then to use this understanding to develop novel ways for robots to directly shape those dynamics to improve a team's likelihood to succeed. Findings from this work will advance our understanding of robot assisted teamwork. Further the project will develop educational activities that allow children and students to develop the skills necessary to succeed in a work environment that includes robots. While most work is performed by groups and teams, research on human-robot collaboration has focused predominantly on dyadic interactions, i.e. a single human collaborating with a single robot. The effectiveness of teams depends on the quality of its interpersonal interactions. Transforming human-robot collaboration from a dyadic towards a group focus raises novel questions about such teaming. In particular: 1) how a robot's behavior influences both human-robot interactions and human-human interpersonal interactions in a group of people, 2) how robots can deliberately shape human-human interpersonal interactions, and 3) what the consequences of such influence are for a team's overall effectiveness. This project investigates how a robot's resource allocation behavior impacts people's interactions with each other and their ability to perform as a team. Allocation behavior refers to behavior associated with a robot's allocation of resources and tasks between several people (e.g. handing over a tool or building block to a team member, allocating social attention and gaze, or providing assistance). This project will: 1) use lab and field studies to develop basic understanding about allocation behavior exhibited by people and robots and how such behavior affects a group's interaction, 2) use an iterative design approach combined with an online and laboratory based task to develop and evaluate allocation behaviors for robots to shape a group's interaction in desirable ways, and 3) examine how a robot's resource allocation behavior during collaborative tasks affects all dimensions of team effectiveness.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.