An introduction to decision making under uncertainty from a computational perspective, covering both theory and applications ranging from speech recognition to airborne collision avoidance. Many important problems involve decision making under uncertainty—that is, choosing actions based on often imperfect observations, with unknown outcomes. Designers of automated decision support systems must take into account the various sources of uncertainty while balancing the multiple objectives of the system. This book provides an introduction to the challenges of decision making under uncertainty from a computational perspective. It presents both the theory behind decision making models and algorithms and a collection of example applications that range from speech recognition to aircraft collision avoidance. Focusing on two methods for designing decision agents, planning and reinforcement learning, the book covers probabilistic models, introducing Bayesian networks as a graphical model that captures probabilistic relationships between variables; utility theory as a framework for understanding optimal decision making under uncertainty; Markov decision processes as a method for modeling sequential problems; model uncertainty; state uncertainty; and cooperative decision making involving multiple interacting agents. A series of applications shows how the theoretical concepts can be applied to systems for attribute-based person search, speech applications, collision avoidance, and unmanned aircraft persistent surveillance. Decision Making Under Uncertainty unifies research from different communities using consistent notation, and is accessible to students and researchers across engineering disciplines who have some prior exposure to probability theory and calculus. It can be used as a text for advanced undergraduate and graduate students in fields including computer science, aerospace and electrical engineering, and management science. It will also be a valuable professional reference for researchers in a variety of disciplines.

No abstract

An Interactive lntertace was used in achool-baaed research to explore the learning of kinematics through body movement, at ages 11-12,13-14 and 16 plus. The students' responses and s u ~~e s s e 5 are significant for learning at school and at Interactive Centres.

A user-friendly system that yields operational benefit results from data-driven prototype evaluations and benefits analyses that iteratively feed back into the prototype design and development. In this study, initial requirements development and field evaluations were conducted using a shadow operations technique at the center tower backup Dallas-Fort Worth International Airport air traffic control tower (ATCT). Results are discussed in reference to the design process of the Tower Flight Data Manager (TFDM) prototype. Nonintrusive measures for quantitatively validating human-systems design issues were identified for this study, including visual gaze analysis and verbal command sequence analysis. Behavioral validation of design issues simplifies the process to prioritize beneficial design changes. The iterative process used resulted in an interim TFDM prototype that was rated by active air traffic controllers as both beneficial and usable in an operational ATCT environment.

This paper presents two methods of analyzing air traffic controller activity: cognitive workload measurement through the novel comparison of controller-pilot verbal communications, and visual attention quantification through manual eye gaze analysis. These analyses were performed as part of an evaluation of the Tower Flight Data Manager (TFDM) prototype system. Cognitive workload analyses revealed that, when comparing participant controllers utilizing TFDM to a control group utilizing existing air traffic control (ATC) equipment, participants issued commands sooner than the control, and thus were perceived to have a lower workload. While visual attention data were not available for the control group, analyses of participant gaze data revealed 81.9% of time was spent in a head-down position, and 17.2% of time was spent head-up. Results are related back to system inefficiencies to find potential areas of improvement in design.