Intelligent behavior modeling in virtual environments

• Main Author: Jain Manik
• Other Authors: Ong Yew Soon
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence; DRNTU::Engineering::Computer science and engineering::Computing methodologies::Computer graphics
• Online Access: http://hdl.handle.net/10356/59264
• Institution: Nanyang Technological University

Artificial Intelligence is increasingly influencing human life, becoming a part of our daily activities. Intelligent agents have been employed in virtual environments for a long period of time. Recently, the expectations from intelligent agents and their capabilities have risen to such an extent that they are meaningful participants in the virtual experience. This project aims to create a virtual environment of infinite possibilities and employ artificially intelligent agents that can supplement user interaction with the world to create a more engaging experience. Furthermore, it aims to make artificial nonplayable character (NPCs) more autonomous beyond the conventional areas of their use, such as being the user’s followers or being characters that the user could interact with but who seldom had autonomous interaction with the game environment. Thus, intelligent agent behavior modeling in virtual environments is generally an under-appreciated aspect of building such environments, with creators often content with giving the player complete control over how the world is shaped. With virtual gaming getting increasingly popular however, there have been more forays into creating smart, helpful AI elements in virtual environments. This project aims to use a popular game engine and an NTU-developed AI engine to create a programmable AI in a vast, ever-expanding world. The author explores different scenarios and methods of implementing an AI system in an open-world sandbox game modeled on the popular imaginative game Minecraft.