Skip to content
Future air traffic concepts of operation may couple autoflight functions to ADS-B targets in high traffic density airspace operations. However, selection (and de-selection) of other aircraft raises a host of issues. The selected traffic may then require special designations. The pilot may require assistance with identifying aircraft that subsequently needs to be de-selected.
Talk on Option and Constraint Generation using Work Domain Analysis: Implementation for Reinforcement Learning Algorithm part of the Interactive Machine Learning Project sponsored by ONR.
Short presentation given at the 2014 Human Factors and Erogonomics Society Annual Meeting.
Engineering education research (EER) is a relatively young field of inquiry, established with the intent to improve the academic experiences of young and emerging engineers. While many researchers’ perceptions of how to improve engineering education stem from traditional classroom experiences, a select group of researchers belong to EER-oriented departments, labs, and research centers.
Decision makers are often required to make decisions with incomplete information. In order to design decision support systems utilizing restrictiveness and guidance to assist decision makers in these situations, it is essential to understand how certain decision making strategies are affected by incomplete information.
Ecological rationality is the study of when certain decision making strategies exploit specific environments so efficiently that further information and computation would not necessarily increase accuracy. This perspective challenges many of the normative accounts of rationality by arguing that heuristics, i.e. decision making strategies that ignore some available information, in some environments can make decisions faster, more efficiently, and/or more accurately than analytic decision making strategies. This presentation introduces a new environmental structure, a measure of distribution of incomplete information called complete attribute pairs, and shows how this structure indicates when a well-studied heuristic decision making strateigy, Take-the-Best (TTB), is ecologically rational under conditions of incomplete information.
We investigate the use of Work Domain Analysis (WDA), a technique from the field of cognitive engineering, to inform the creation of options and constraints for Reinforcement Learning (RL) algorithms. The micro-world of Pac-Man is used as a tractable and representative work domain. WDA was conducted on individuals familiar with Pac-Man and an Abstraction Hierarchy (AH), a means-ends representation of their understanding of the game, was created for each individual.
Aparna presented her paper on this topic on October 9th, 2014 at the 33rd Digital Avionics Systems Conference in Colorado Springs, CO. The paper focuses on estimating faults in complex large-scale integrated aircraft systems, especially where they interact with, and control, the aircraft dynamics. A general assumption considered in the reliability of such systems is that any component level fault will be monitored, detected and corrected by some fault management capability. However, a reliance on fault management assumes not only that it can detect and manage all faults, but also that it can do so in sufficient time to recover from any deviation in the aircraft dynamics and flight path. Thus, this paper examines how faults can ripple through a system by simulating the key dynamics within the system before and after any fault management may resolve them, using a case study of adaptive control’s response to an elevator control reversal.
Alexandra Coso presented her PhD Thesis on Friday, March 28th, 2014. The intent of the doctoral work was to (1) examine how stakeholder considerations are currently integrated into aerospace vehicle design practice and curricula, (2) design empirically-informed and theoretically-grounded educational interventions for an aerospace design capstone course, and (3) isolate the characteristics of the interventions and learning environment which support students’ integration of stakeholder considerations.
Marc Canellas presented his work on Accuracy and Effort of Decision Making with Incomplete Information at the 2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support (CogSIMA) in San Antonio, TX.
Cognitive Engineering Center (CEC) Georgia Institute of Technology 270 Ferst Drive Atlanta GA 30332-0150
