May ‘17 – Jul ‘17
At the Human Experience and Agent Teamwork Laboratory, Harvey Mudd College, Claremont, California
I worked with one other person to determine what aspects of scheduling are most significant for human-robot teamwork. While robots are able to execute events with precision and consider detailed constraints, humans are adaptable and often use innate heuristics to improvise when unexpected events occur. We decided to design an experiment to see whether it is better for the human or robot to take the lead in a collaborative task.
Once we ran experiments, we evaluated the effects of a human or robot lead on the team’s efficiency and fluency. I ran statistical tests and found a significant difference between the human- and robot-led groups, verifying that when a robot takes the lead in dictating a schedule, there are gains in team efficiency without loss of humans’ perceived comfort.
I also took the opportunity to mentor an Upward Bound high school student who joined our project.
Lastly, I co-authored and presented a paper* at the AAAI Fall Symposium ’17.
*Who Takes the Lead? Automated Scheduling for Human-Robot Teams. Castro, B., Roberts, M., and Mena, K. and Boerkoel, J. In Proc. of Artificial Intelligence in Human Robot Interaction AAAI Fall Symposium Series (AI-HRI 2017). Presented Nov. 10, 2017 by Castro, B. in Arlington, VA.