Applying model checking to agent-based learning systems

Kirwan, Ryan F. (2014) Applying model checking to agent-based learning systems. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2014KirwanPhd.pdf] PDF
Download (5MB)
Printed Thesis Information: https://eleanor.lib.gla.ac.uk/record=b3031231

Abstract

In this thesis we present a comprehensive approach for applying model checking to Agent-Based Learning (ABL) systems. Model checking faces a unique challenge with ABL systems, as the modelling of learning is thought to be outwith its scope. The practical work performed to model these systems is presented in the incremental stages by which it was carried out. This allows for a clearer understanding of the problems faced and of the progress made on traditional ABL system analysis. Our focus is on applying model checking to a specific type of system. It involves a biologically-inspired robot that uses Input Correlation learning to help it navigate environments. We present a highly detailed PROMELA model of this system, using embedded C code to avoid losing accuracy when modelling it. We also propose an abstraction method for this type of system: Agent-centric abstraction. Our abstraction is the main contribution of this thesis. It is defined in detail, and we provide a proof of its soundness in the form of a simulation relation. In addition to this, we use it to generate an abstract model of the system. We give a comparison between our models and traditional system analysis, specifically simulation. A strong case for using model checking to aid ABL system analysis is made by our comparison and the verification results we obtain from our models. Overall, we present a framework for analysing ABL systems that differs from the more common approach of simulation. We define this framework in detail, and provide results from practical work coupled with a discussion about drawbacks and future enhancements.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Model Checking, Agent-Based Learning, Autonomous robots, Abstraction, Verification, Agent-centric abstraction, SPIN, PROMELA.
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Colleges/Schools: College of Science and Engineering > School of Computing Science
Supervisor's Name: Miller, Dr. Alice
Date of Award: 2014
Depositing User: Dr Ryan F. Kirwan
Unique ID: glathesis:2014-5050
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 28 Mar 2014 15:03
Last Modified: 28 Mar 2014 15:18
URI: https://theses.gla.ac.uk/id/eprint/5050

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year