Use this URL to cite or link to this record in EThOS:
Title: Topological reasoning using a generative representation and a genetic algorithm
Author: Zhang, Yu
ISNI:       0000 0004 2748 7701
Awarding Body: Cardiff University
Current Institution: Cardiff University
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis studies the use of a generative representation with a genetic algorithm (GA) to solve topological reasoning problems. Literature review indicates that generative representations outperform the non-generative ones for certain design optimisation and automation problems. However, it also indicates a lack of understanding of this relatively new class of representations. Many problems and questions about the implementation of generative representations are still to be addressed and answered. The results and findings presented in this thesis contribute to the knowledge of generative representations by: 1. explaining why genotype formatting is important for the representation and how it influences the performance of both the representation and the algorithm 2. providing different crossover and mutation methods, including both existing and newly developed ones, that are available to GA when used with the presentation and, more importantly, revealing their different properties in generating new individuals 3. providing alternative ways to map turtle graphs into the design space to form the actual designs and showing the properties of these different mapping methods and how they influence the outcome of the search. In general, this thesis examines the key issues in setting up and implementing generative representations with genetic algorithms. It improves the understanding of generative representations and contributes to the knowledge that is required to further develop them for real-world use. Based on the results and findings of this study, directions for future work are also provided.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available