Doctor of Philosophy (PhD), Wright State University, 2021, Computer Science and Engineering PhD

A significant issue in cognitive systems research is to make an agent formulate and manage its own goals. Some cognitive scientists have implemented several goal operations to support this issue, but no one has implemented more than a couple of goal operations within a single agent. One of the reasons for this limitation is the lack of knowledge about how various goals operations interact with one another. This thesis addresses this knowledge gap by implementing multiple-goal operations, including goal formulation, goal change, goal selection, and designing an algorithm to manage any positive or negative interaction between them. These are integrated with a cognitive architecture called MIDCA and applied in five different test domains. We will compare and contrast the architecture's performance with intelligent interaction management with a randomized linearization of goal operations.

Committee: Michael T. Cox Ph.D. (Committee Co-Chair); Mateen M. Rizki Ph.D. (Committee Co-Chair); Matthew M. Molineaux Ph.D. (Committee Member); Michael L. Raymer Ph.D. (Committee Member); Michelle A. Cheatham Ph.D. (Committee Member) Subjects: Artificial Intelligence; Computer Science

Master of Science (MS), Wright State University, 2017, Computer Science

A cognitive system attempts to achieve its goals by utilizing the appropriate resources present to yield the best possible outcome within a short duration. To achieve the goals in such an efficient manner, it is important for the agent to manage its goals well. Goal management not only makes the agent efficient but also flexible, more durable to the sudden changes in the environment, and self-reliant. Goal Management consists of various goal operations including goal formulation, selection, change, delegation, achievement, and monitoring. Each operation is unique and has its own significance in aiding the performance of the agent. The thesis work focuses on the implementation of two particular goal operations. These are goal selection and goal change with concentration of the former. Goal selection allows the agents to choose among its goals by using any criteria which are appropriate for the domain. Goal change allows the agent to change its current goal to another goal because of reasons like the inadequate amount of resources or detection of a discrepancy. The implementation of these operations is done within a cognitive architecture called the Metacognitive Integrated Dual-Cycle Architecture in the two problem domains of construction and restaurant. In the construction domain, the goals are to construct the towers using the resources within a provided time limit, and in the restaurant domain, the goals are to satisfy the maximum number of people by serving items ordered with a limited amount of money. After the implementation of goal selection and goal change, the work is evaluated using various methods, one of which is the comparison of the performance of MIDCA with and without those goal change operations and the other is by comparing two different goal selection methods. Several graphical depictions and mathematical formulae are presented that support the course of performance comparison.

Committee: Michelle Cheatham Ph.D. (Committee Chair); Michael Cox Ph.D. (Committee Co-Chair); Mateen Rizki Ph.D. (Committee Member) Subjects: Artificial Intelligence; Computer Science