Thumbnail Image
Publication

Unified logic maze generation: combining state graph representations and local search for diverse puzzle design

Henke, Johnathon
Citations
Altmetric:
Advisor
Mehta, Dinesh P.
Editor
Date
Date Issued
2023
Date Submitted
Keywords
local search
 logic maze
 maze characteristics
 objective function
 simulated annealing
 state graph
Collections
2023 - Mines Theses & Dissertations
Show all metadata
Files
Thumbnail Image
Henke_mines_0052N_12597.pdf
Adobe PDF, 1.42 MB
Research Projects
Organizational Units
Journal Issue
URI
https://hdl.handle.net/11124/178465
Embargo Expires
Abstract
Several logic mazes and their state graph representations are defined with the goal of generating additional instances of each maze. As Local Search proved to be an effective mad maze generation method, we focused on defining an objective function that considers maze characteristics. By representing logic mazes as state graphs, it becomes possible to employ the same objective function for scoring, generating, and contrasting various maze types. This not only simplifies implementation, but also enables the creation of common characteristic state graphs for completely unique and separate high-level puzzles. We outline desired maze features, provide implementation details, and present example objective functions for generating mazes with the targeted attributes.
Associated Publications
Rights
Copyright of the original work is retained by the author.
Embedded videos