SGER: Requirements Assessment for a Multiple Robot-Multiple Human Interface

Vanderbilt University, Nashville TN

Investigators

Abstract

There exist many examples of Human-Robotic Interaction (HRI) for a single robot or a small number of robots (2-4). But HRI development for large robot teams requires novel techniques for presenting large amounts of data so as to support a human's ability to maintain situation awareness while not overloading his/her cognitive capabilities. In this project the PI will assess needs and requirements, and develop preliminary interface designs, for tools that will permit a small number of humans to supervise large robotic teams, a capability not available with current HRI methodologies. To this end, the PI will build on prior work in HCI, CSCW, and complex human-machine systems. She will interact continuously with and solicit feedback from the Metro Nashville Police Chemical, Biological, Radiological, Nuclear, and Explosive (CBRNE) search and rescue team, because a basic premise of the work is that in order to develop useful interfaces one must first understand how organizations will actually use large teams of robots, which in turn requires an understanding of how the human team currently completes their tasks (without robots), their expectations of a robotic team, and requirements for robotic team capabilities. Broader Impacts: This work will advance the fields of robotics, artificial intelligence, and human factors. Specifically, it will advance the community's understanding of how large numbers of robots can be effectively supervised, managed, and guided by a small number of human operators, thus affording more usable systems while reducing the associated manpower and exposure to potentially dangerous contaminants. Due to the project's focus on deploying robots for CBRNE attacks, the results will have broad affects on developing deployable robotic systems based upon user feedback to meet ever-increasing Homeland Security needs.