Skip to content
In high-risk, time-pressure domains, the ability to get only salient information is of paramount importance. Missing or superfluous information in these domains can detract from a decision maker’s ability to make correct judgments. Consequently, decision support systems are being developed to facilitate expert decision-making by modifying the information presented to the operator.
Interactive machine learning involves humans interacting with agents during their learning process. As this field grows, it is pertinent that non-expert users are able to have a satisfying experience in teaching the agents in order to retain usage of the agent and minimize frustration for the user. Previous work has investigated which factors contribute to the user’s experience when teaching agents in a fully observable domain.
The aim of this thesis is to gain a deeper understanding of adaptation of work strategies in human-robot teams, specifically how human-robot interaction protocols and function allocation can foster and/or limit such adaptation. To analyze these effects, this work will model adaptation to work demands in a team of humans and robots by mapping existing models of individual human adaptation to multi-agent systems.
Longer, more complex spaceflight missions require well designed human-robot teams (HRT). Therefore, we must understand: How do we design teams and allocate tasks? What design options impact different components of human-robot teaming (HRT)? Team designers should understand: Key elements of human-robot teaming; Interaction and how to shape it; Metrics to measure successful human-robot teams. This research focuses on developing an analysis tool to understand HRT.
This thesis proposal examines context dependent total energy alerting to protect against low energy unstable approaches in commercial aviation operations. Currently, many individual states are monitored independently to identify unstable approaches, rather than an integrated single assessment of total energy. An alert would also have to be context dependent, integrating the individual states along with phase of flight awareness, aircraft profile modeling, and expected human responses to ind
Shipboard-landing maneuvers in rotorcraft piloting involve a number of unique challenges. Such maneuvers can be cognitively demanding even for experienced rotorcraft pilots. To minimize risk, these maneuvers are conducted within well-defined boundaries related to weather and visibility. In order to expand this envelope, technological aids are being proposed to augment decision making capabilities and reduce pilot workload without compromising safety.
This proposal defines a research plan to investigate the effects of visibility range and runway marking on pilot performance with enhanced flight vision systems. The motivation behind this experiment is to observe the implications of sensor technologies in general aviation. This research hopes to provide some insight into the possible concerns of enhanced flight vision systems with regards to pilot performance.
Decision makers are continuously required to make choices in environments with incomplete information. This dissertation sought to understand and, ultimately, support the wide range of decision making strategies used in environments with incomplete information. The results showed that the standard measure of incomplete information as total information, is insufficient for understanding and supporting decision makers faced with incomplete information.
Cognitive Engineering Center (CEC) Georgia Institute of Technology 270 Ferst Drive Atlanta GA 30332-0150