An Ilities Tool for Minisatellite Software Validation

Loading...
Thumbnail Image
Files
Brown_Mason_Final Thesis.pdf
Embargoed until 2025-12-16
File version
Primary Supervisor

Dey, Sharmistha

Other Supervisors

Tuxworth, Gervase

Bernus, Peter

De Souza Junior, Paulo A

Editor(s)
Date
2024-12-16
Size
File type(s)
Location
License
Abstract

Space debris caused by satellite failure is a growing issue. Validating functional and non-functional requirements improved the likelihood of satellite mission success. Tools such as MATLAB, GitLab, and Atlassian Jira have been developed to validate functional requirements. While some tools can aid the validation of quality attributes or 'ilities' (including scalability, adaptability, affordability, and reliability), such as Microsoft Excel, Docker, and Jenkins - these tools mainly assist in tracking validation rather than performing validation themselves. The aim of this thesis is to close this gap to reduce space debris caused by satellite failure. To date, no open-source software tool to validate ilities for the space sector has been released. Contractors have developed their own standards to define quality attributes, but these standards are inconsistent, or even in conflict, across the industry. Furthermore, tracking tools have only been designed for very large and expensive satellites or very small and cheap satellites; there are no publicly available quality tracking tools for the current most popular launch class, the minisatellite. Likewise, no software tool to calculate quality attributes has been developed. Research question one of this thesis asks: what ilities are desirable to improve or to assure mission success? Grounded theory was used to identify the most desired ilities for minisatellite missions, drawing on a survey, a two-year minisatellite case study, a literature review and taxonomy. The findings indicate that reliability, survivability, and scalability are the most desired attributes for any minisatellite mission. Research question two addresses the issue of how to define specific ilities. This is a critical issue given the range of terminologies in current use across the industry. Accordingly, a Quality Attribute Taxonomy tool was developed with the aim of reducing the subjectivity and ambiguity when selecting and defining quality attributes to inform design. The tool comprises a database that aggregates definitions from active standards used across the space segment; it does not impose a 'universal standard definition' but, rather, informs the user of the existence of a variety of definitions so they can make an informed decision regarding which ilities are desirable for their purpose. The third and final research question asks: how can desired ilities be calculated? Again, the research identified that the current environment lacks quantification methods when it comes to ilities. Therefore, a Minisatellite Quality Attribute Calculator was developed using decision theory. The calculator uses Bayesian estimation to determine the expected utility of reliability, survivability, and scalability when it comes to mission success. The tool allows a program manager to select a course of action that will normatively yield the most value for the current mission based on prior mission data. The Bayesian network was constructed and tested using a National Aeronautics and Space Administration (NASA) Ames Research Centre (ARC) publicly available dataset and a synthetic dataset, each containing satellite mission successes and failure results. In summary, this research has produced two quality attribute tools, the Minisatellite Quality Attribute Taxonomy Tool and the Minisatellite Quality Attribute Calculator Tool, designed for use by any satellite program manager. The tools require a program manager to select a desired quality attribute and enter mission-specific data. The Minisatellite Quality Attribute Taxonomy Tool aggregates various definitions of the selected quality attribute from existing standards, while the Minisatellite Quality Attribute Calculator Tool uses mission-specific data and Bayesian estimation to calculate a quantitative measure of how well the selected quality attribute, such as reliability, survivability, or scalability, is achieved for the mission. These tools will be especially valuable for university or small to medium industry missions that likely do not have the expertise or time availability to consider multiple standards or develop proprietary tools. Importantly, these tools can also assist satellite program managers in establishing priorities for non-functional requirements in the design and implementation of satellite systems, and can serve as a platform to accomplish educational, scientific, and military objectives.

Journal Title
Conference Title
Book Title
Edition
Volume
Issue
Thesis Type

Thesis (PhD Doctorate)

Degree Program

Doctor of Philosophy

School

School of Info & Comm Tech

Publisher link
DOI
Patent number
Funder(s)
Grant identifier(s)
Rights Statement
Rights Statement

The author owns the copyright in this thesis, unless stated otherwise.

Item Access Status
Note
Access the data
Subject
Persistent link to this record
Citation