Research Projects

Collaborative Judgement

A shared mental model is defined as a mental construct that includes a shared understanding of the task that is to be performed, the involved teamwork, the environment, team goals, and the constraints that they must operate within. Humans work best when shared mental models are present. This research will focus on human-computer interaction (HCI) to see if humans can work with AI, proving that shared mental models humans and AI are beneficial.

Impact of Shared Mental Models on Human-AI Interaction and Mission Effectiveness

Human teams are most effective when the members of the team utilize a shared mental model (SMM), meaning a shared perception of goals and actions through effective communication and an understanding of their fellow team members' goals and likely methods. Currently humans and AI teams share no such model.  At best, humans working closely with AI begin to anticipate what the AI can do and when it can be trusted, as is the case in medical decision making.

Models, Metrics, and a Mathematics of Interactional Trust for Humans and Automation

At the very core of technological acceptance is human-machine trust and its fragility. In this project we have proposed and are testing models of human-machine trust that includes its antecedents such as faith in technology, familiarity, and situation awareness. Our research also incorporates expectations that the automation will be cooperative and perform ethically, legally, and abide by norms, drawing from many fields from human-robot interaction to game theory, social psychology, and management and information science.

Shared Human-AI Mental Models in Flight Planning

Shared Mental Models play a crucial role in a number of different aspects of teamwork. Shared mental models are more than simply an understanding of the state of the exterior world, they also incorporate information about who within a team has both the ability to perform certain tasks as well as the responsibility to see that they are performed correctly.

Assisting Time-Critical Decision-Making in the Space Situational Awareness Domain

The space environment is cluttered with significant amounts of debris, which poses a threat to weather satellites, communication satellites, and other space assets given their high velocities. The Joint Space Operations Center (JSpOC) keeps a track of thousands of space objects with priority given to active satellites. Keeping the space object catalog updated is a primary objective of JSpOC, and it takes up a significant portion of the resources available.

Advanced Guidance and Cueing Algorithms for Rotorcraft Shipboard Landing Envelope Expansion

While landing maneuvers in vertical flight can be cognitively challenging, such maneuvers are more challenging in shipboard operations. Shipboard landing maneuvers for rotorcraft involve spatial confinement, deck movement, and air turbulence issues, which can be exacerbated by degraded visual conditions. The project is an interdisciplinary effort and it aims at developing advanced guidance and cueing to improve the efficacy of shipboard landing operations in the navy.

Technologies for Mixed-Initiative Plan Management for Human Space Flight

NASA’s future missions will push the bounds of human-space exploration and challenge the mission designers and engineers to create automated systems that will enable the joint human-automation teams to operate more autonomously as they move further from terrestrially based mission control and the time lag of communication becomes a challenge.

Objective Function Allocation Method for Human-Automation/Robotic Interaction using Work Models that Compute

Future manned space missions will require astronauts to work with a variety of robotic systems. To develop effective human-robot teams, NASA needs objective methods for function allocation between humans and robots. This study develops an objective methodology for function allocation between humans and robots for future manned space missions. Some problems that need to be addressed in function allocation include: (a) monitoring of agents, (b) agents waiting on other agents (idle time), (c) high task load of agents, (d) excessive amount of communication required.

NSF-CPS: Adaptive Intelligence for Cyber-Physical Automotive Active Safety System Design and Evaluation

The main objective of this research is to use techniques and models from human factors, computational neuroscience, and adaptive and real-time optimal control theory in order to investigate the effects of the introduction of learning and adaptation to the next generation of ASCS. In particular, we will:
(a) Learn the driver’s habits, driving skills, patterns and weaknesses.
(b) Model his/her current cognitive state along multiple dimensions such as attentiveness, aggressiveness, etc.

NASA - Flight Deck Display Counter-Measures to Disorientation

The most common causes of aircraft incidents and accidents today are pilot spatial disorientation and/or loss of energy situation awareness. To re-assess the underlying mechanisms of SD and/or LESA, we are building a computational model of human pilot. Having a computational model allows us to create fast-time simulations instead of relying on the real-time human-in-the-loop simulations. The purpose is to utilize this novel model for large-scale evaluations spanning wide range of potential conditions and variations in flight crew behavior.

Pages

Map of Cognitive Engineering Center

Cognitive Engineering Center (CEC)
Georgia Institute of Technology
270 Ferst Drive
Atlanta GA 30332-0150