angelhenderson / thesis

Common Sense Reasoning in Autonomous Artificial Intelligent Agents through Mobile Computing

Abstract

This thesis presents the design, analysis, and implementation of an autonomous artificial intelligent agent capable of planning, reasoning, rationalizing, communication, and learning through the use of common-sense reasoning. We discuss the cognitive modeling approach of our autonomous artificial intelligence and how its system design is inspired by the Artificial Cognitive Neural Framework (Crowder, 2012) as well as the biological, cognitive processes of the human brain. We cover how our autonomous intelligent agent utilizes stacked artificial neural networks, memories, and fuzzy systems to map out its cognitive process. We state our motivations for performing this research and examine the importance of common-sense reasoning in the advancements of artificial intelligence. We cover various roles and applications of our autonomous artificial intelligent agents as well as the challenges that arise when developing intelligent agents that can perceive, learn, and adapt autonomously. The outcome of this thesis includes the development of a mobile software application comprising an autonomous intelligent agent that demonstrates an understanding of context, observations, utilization of past experiences, self-improvement through reinforcement learning, and hypothesis development to reach solutions – which are presented and illustrated with examples and challenges.

Contact Information

Please contact me at [email protected] for information about this thesis