Enabling Human-Aware Autonomy Through Cognitive Modeling and Feedback Control

Across many sectors, ranging from manufacturing to healthcare to transportation, there is growing interest in the potential impact of automation that is truly collaborative with humans. Realizing this impact, though, rests on first addressing the fundamental challenge of designing automation to be aware of, and responsive to, the human with whom it is interacting. While a significant body of work exists in intent inference based on human motion, a human's physical actions alone are not necessarily a predictor of their decision-making. Indeed, cognitive factors, such as trust and workload, play a substantial role in their decision making as it relates to interactions with autonomous systems. While these factors have long been studied by cognitive psychologists and human factors experts, consideration of them in the context of closed-loop interactions between humans and autonomous systems remains a nascent area of research. In this chapter, parameterized and data-driven modeling frameworks are presented for quantifying input–output relationships as they relate to dynamic cognitive states, particularly human trust. Innovations in human subject experiment design needed to not only generate essential training data but also validate models and subsequent control policies are also discussed. The third topic of discussion is the design of algorithms to govern closed-loop interactions based on calibration of human cognitive states. Finally, the chapter concludes with a discussion of open challenges and opportunities for further investigation on these topics.