Investigating Human-Algorithm Interactions in Supervisory Control Settings

Missy Cummings
Massachusetts Institute of Technology

Supervisory control is the process by which a human operator intermittently interacts with an automated system to effect change in a controlled process or task environment, e.g., military command and control, air traffic control, and process control. Central to all of these environments is the critical role that algorithms play in these automated processes. The systems must include varying degrees of automation and autonomy for safe and effective operation, yet human judgment and knowledge-based reasoning are also equally critical. This presentation will detail a series of experiments that examined if and how humans and algorithms could complement one another in supervisory control settings

Presentation (PDF File)

Back to Machine Reasoning Workshops III & IV: Mission-Focused Actions/Reactions Based on & System Integration of Information Derived from Complex Real-World Data (by invitation only)