Computational Models to Support Human-Machine Interaction
Jacob W. Crandall
We often talk of the ever-elusive holy grails associated with human-computer interaction, including “human-computer collaboration,” “adjustable autonomy,” “mixed initiative,” etc. These phrases capture our goals to develop human-machine systems that leverage the complementary strengths of humans and automation to form a synergistic team in which automation becomes an adaptable and “intelligent” entity capable of providing time-critical assistance. To reach such lofty goals, automation must build and reason with complex computational models of the complete human-machine system. Such models must include (and combine) models of the humans in the system, the automation itself, and the interactions between them. In my talk, I discuss our work toward constructing and using such models in the context of human supervisory control of multiple robots.
Jacob received the B.S., M.S., and Ph.D. degrees in Computer Science from Brigham Young University, Provo, UT, in 2001, 2004, and 2006, respectively. He is currently a postdoctoral associate in the Department of Aeronautics & Astronautics at the Massachusetts Institute of Technology. His research interests include human-machine interaction, machine learning, multi-agent systems, and game theory.