Research Project

Organizing Teamwork: Understanding Operator and Multiple-Robot Team Performance

Motivation/Research Problem
Current space exploration and subterranean missions rely upon autonomous / semi-autonomous mission types wherein humans and machines work together towards a common goal. The multiple-robot teams used in urban search and rescue missions are an example of these mission types. The use of this technology allows for human operators to supervise multiple robots in a hazardous environment, at the same time. A human-to-robot ratio of two to three operators per robot has been shown to be a reliable baseline for human-robot teamwork (Fincannon et. al., 2013). In 2019, Hong and colleagues found that operators can effectively maneuver up to eight robots before experiencing decreases in performance. Manipulating variables such as robot reliability and autonomy have been shown to influence the efficiency of a human-robot team and alter the human-to-robot ratio (Hong et. al., 2019). In order to organize these teams and increase performance, architectures such as shared pooled or sectored teams have been tested as methods for distributing workload amongst operators (Lewis et. al., 2006). A pooled architecture wherein teammates communicate with each other and supervise multiple robot teams collaboratively was found to be more efficient than a sectored architecture wherein humans supervise multiple robot teams separately. In human-human teaming literature and organizational behavior management literature, researchers have examined team lead directiveness as a variable to increase human-human team efficiency (Somech, 2006). Their findings indicated that teams led by highly directive leads were generally more productive than teams with leads who exercised low directiveness. Currently what is lacking in the literature is a human-robot team architecture that focuses on minimizing workload and increasing team efficiency.
Research Team
Collaborators:

  • Amir Rahmani: primary JPL contact, directly involved in the development of work
  • Suzanne Sheld: ARCS Co-PI, collaborator
  • Kevin Zemlicka: ARCS Co-PI, collaborator
  • Gina Masequesmay: ARCS Co-PI, collaborator

Faculty: 

  • Nhut Ho: ARCS Director, CSUN Professor, advise team
  • Li Liu: ARCS Associate Director, CSUN Assistant Professor, advise team
  • Amiel Harman: ARCS/CSUN faculty
  • Ellie Kazemi: ARCS Co-PI, CSUN professor, directly supervise ARCS Fellows

Student Team:

  • Adisa Ptah
  • Eric McCoy
  • Helina Mekonnen
Alignment, Engagement and Contributions to the priorities of NASA’s Mission Directorates
Having a human-robot teaming architecture that focuses on team lead directiveness would minimize workload and increase team efficiency. Additionally, designing a chain of command for multiple-robot teams to communicate error or overload to the operator may support mission coordination (Fincannon et. al., 2013). This architecture would allow organizations, such as NASA, to effectively organize human-multiple-robot teams, increase productive communication amongst operators and agents, and may help complete missions more efficiently. By reviewing the best practices in human operation of multi-agent systems, robot-robot teaming, this research will propose an architecture for minimizing Workload in human operators and increase efficiency.
Research Questions and Research Objectives
This study will look at the variables that influence the human-to-robot ratio and propose strategies for increasing the number of robots that human operators can supervise at one time. We will conduct an experimental study with two different conditions in which we manipulate multiple-agent systems and the team architectures. The primary purpose of this study is to establish an architecture for organizing human-robot teamwork and identify the variables that influence the human-to-robot ratio.
Research Methods

we will conduct an experimental study using a computer interphase to simulate human-robot team missions. We will manipulate the structure and communication of human-robot operators to determine if we can establish similar outcomes (team performance and NASA Task Load) with fewer human operators.

Research Deliverables (Publications, Presentations and other Products)
Anticipated deliverables

  • A list of features of interest identified by key stakeholders
  • A focused list, generated by subject matter experts at NASA/JPL, for extending the NASA Task Load
  • Research results of survey, interviews, and focus group
  • Research results of the experimental study
  • Extension of NASA Task Load
  • Conference presentations
  • Publication

Anticipated Publications

  • We are still considering appropriate outlets but will likely submit to the journals that we most reference. We are also looking into appropriate outlets for convergence of psychology and autonomy.

Conference Presentations

  • Same as above. Additionally, we will likely present at the California Association for Behavior Analysis and the Association for Behavior Analysis International.
Research Timeline
January 2020  —

  • Colloquium #1, Research Plan, Present Research Plan, Seminar at JPL, Literature Review (Human-robot teamwork)

February 2020  —

  • Literature Review (Human-robot teamwork)

March 2020  —

  • Literature Review (Human-robot teamwork)

April 2020  —

  • Literature Review (Human-robot teamwork)

May 2020  —

  • Build human-robot teamwork architecture

June 2020 —

  • Build human-robot teamwork architecture

July 2020   —

  • Build human-robot teamwork architecture, Develop Experimental Design, Literature Review (Urban Search and Rescue Simulation Software)

August 2020  —

  • Develop Experimental Design, Seminar #2, Literature Review (Urban Search and Rescue Simulation Software), Start IRB submission

September 2020  —

  • Complete experimental design.

October 2020  —

  • Submit NSTGRO21 Grant, Train team in development of simulation

November 2020  —

  • Train team in development of simulation

December 2020  —

  • Train team in development of simulation

January 2021  —

  • Train team in development of simulation, Revise Human-robot teamwork architecture, Develop simulation, Submit for IRB approval and implement feedback

February 2021  —

  • Revise Human-robot teamwork architecture, Develop simulation, Submit for IRB approval and implement feedback

March 2021  —

  • Conduct experiment, Run data analyses, Start drafting manuscript
Are there other activities (e.g., proposals or additional projects) that you have developed or anticipate based on your NASA ARCS project?
Because of COVID-19, we have been very busy revising the projects we have. Both fellows applied for NSTGRO21 grant. Our collaborator, Dr. Rahmani, is invested in our collaborations.

Impact of Project Partnership with NASA:

The name recognition and prestige associated with NASA and JPL is attractive for recruiting strong students that would have historically not considered pursuing research careers or growth in STEM.

I continue to be invited by journals, conferences, and podcasters in the field of behavior analysis to discuss my efforts on this project because the convergence of behavior analysis and autonomy is new and exciting for expanding the reach of behavioral science.