Robot Decision Making to Play a Tower Building Game
Loading...
Authors
Heaton, Jacqueline
Date
Type
thesis
Language
eng
Keyword
Artificial Intelligence , Machine Learning , Decision Making , Collision Avoidance , Robots , Reinforcement Learning
Alternative Title
Abstract
Robots are becoming more integrated into society as they become more advanced, and the programming behind them needs to continue to progress in order for the robots to be utilized to their fullest potential. Artificial Intelligence (AI) is one of the most versatile and quickly growing areas of robotic control, and has been used for a variety of different robots and tasks. One potential use of robotics and AI is in that of childhood development. Cooperative play has been shown to be a crucial part of childhood development, and for children with developmental disabilities, playing with other children may be difficult and frustrating, leading them to miss out on this important milestone. Cooperative play with robots has been shown to have positive educational and therapeutic effects on children with developmental disabilities, and so robots can be used as substitute players for children who have troubles playing with other children. To achieve this, an AI must be developed that can make appropriate decisions or moves for a given game, to such an extent that the children would choose to play with the robot instead of alone. This will require both high-level decision making, for deciding on the move to use, and low-level motion planning, to actually carry out the move in the real world. Both of these parts can be carried out by AI agents, which can work together to produce an AI system capable of reasoning and intelligent decision making with respect to a game. In this thesis three AIs and two algorithms will be developed to play Menara, a cooperative tower building game. The three AIs include the low level motion planning agent, the high level pillar placement agent, and the high level tile placement agent. The algorithms include the method for selecting the pillars to have available to the agent during gameplay, and how many pillars the agent plans to place in a single turn. The low level agent, trained first with collision checking only at the final location of a step, then with continuous collision checking, performed reasonably well, achieving a final success rate of 85\%. The high level tile placement agent was able to successfully balance a tile 62\% of the time, while the high level pillar placement agent was able to succeed 79\% of the time on the test dataset.
Description
Citation
Publisher
License
Queen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.