Faculty of Engineering 1

Siganto Foundation Medal

SAJJAD, Muhammad Mohtasim

Thesis Title

Inter-Slice Mobility Management Protocols for 3GPP Network Slicing Framework

Supervisors

• Dr Carlos J Bernardos  (External Supervisor)
• Dr Dhammika Jayalath  (Principal Supervisor)
• Professor Yuchu Tian  (Associate Supervisor)

Citation

This thesis addresses the complex challenge of managing user mobility in a 5G network featuring network slicing – a new technology introduced to support virtually unlimited services and use cases, such as autonomous vehicles, smart factories, smart homes, etc. As the very first work on this problem, the thesis contributes towards introducing protocol architectures of the established mobility management standards in the 3GPP network slicing framework. The proposed protocols are capable of effectively supporting diverse mobility requirements of users for service types such as enhanced Mobile Broadband, Ultra-Reliable Low Latency Communications, Massive IoT, and Vehicle-to-Everything.

Outstanding Doctoral Thesis Awards

ABEYSIRIWARDENA, Kodikara Arachchillage Tharindu Malinga

Thesis Title

Cold-Formed Steel Stud Walls Subject to Distortional Buckling Failures Under Ambient and Fire Conditions

Supervisors

• Dr Anthony Deloge Ariyanayagam  (Associate Supervisor)
• Professor Mahadeva Mahendran  (Principal Supervisor)

Citation

Structurally efficient cold-formed web-stiffened steel studs are increasingly used in light steel framed walls, but they are prone to distortional buckling. This research investigates the distortional buckling behaviour of gypsum plasterboard sheathed studs under both ambient and fire conditions. Using experimental, numerical and analytical studies, the composite behaviour of steel studs and sheathing subject to distortional buckling was investigated and improved design guidelines were developed allowing for safe and economical design. Overall, this research has significantly enhanced the knowledge and understanding of the behaviour and design of cold-formed steel stud walls subject to distortional buckling under ambient and fire conditions.

Doctor of Philosophy

AMAYA CATAÑO, Jorge Alberto

Thesis Title

A 3D Printed Hydrogel Microfluidic Vascular Model for Studying the Interplay Between Atherogenic Hemodynamics and Vascular Cells

Supervisors

• Professor Zhiyong Li  (Principal Supervisor)
• Professor Yi‑Chin Toh  (Associate Supervisor)
• Adjunct Professor Prasad Kanaka Durga Yarlagadda  (Associate Supervisor)

Citation

This work presents a new approach for patient-specific vascular models, capturing anatomy, hemodynamic, and cell arrangement. MRI images were utilised to create 3D models of carotid vessel bifurcations. Computational fluid dynamics was employed to investigate hemodynamics. Replicas of the carotid vessel were produced via 3D-DLP bioprinting, resembling cellular organisation with embedded human smooth muscle cells and seeded endothelial cells. The study demonstrated cell viability, proliferation, and arrangement in the perfused channels comparable to traditional methods. The model allowed for the study of hemodynamic impact on biologically relevant anatomies, enabling experimental validation of the progression of vascular diseases such as atherosclerosis.

BAI, Jinshuai

Thesis Title

Physics-Guided Deep Learning Framework for Computational Mechanics

Supervisors

• Professor YuanTong Gu  (Principal Supervisor)
• Professor Emilie Janette Sauret  (Associate Supervisor)
• Professor You‑Gan Wang  (Associate Supervisor)

Citation

This thesis developed a novel, effective and robust numerical framework based on the physics-guided deep learning technique for a wide range of mechanics modelling. In the thesis, thorough investigations regarding the proposed framework have been conducted from both theoretical and numerical aspects. It has been demonstrated that the proposed framework has great advantages over the traditional numerical methods when facing challenges, such as nonlinearity and free-surface tracking problems. The proposed framework demonstrated possibilities of integrating state-of-the-art deep learning techniques into computational mechanics and opened a new avenue for mechanics modelling.

BATUWATTA GAMAGE, Chanaka Prabuddha

Thesis Title

A Novel Machine Learning-based Computational Framework for Predicting Microscale Morphological Changes of Plant Cells During Drying

Supervisors

• Professor YuanTong Gu  (Principal Supervisor)
• Professor Azharul Karim  (Associate Supervisor)
• Dr Charith Malinga Rathnayaka Mudiyanselage  (Associate Supervisor)

Citation

This thesis proposes a novel and effective physics-informed machine learning framework to explore microscale variations in plant-based foods during drying. By initiating fundamental numerical investigations at the cellular level—which significantly influence bulk-level changes—this research highlights the framework's robustness and flexibility over traditional methods like FEA and meshfree particle-based methods. The work demonstrates the considerable potential for employing this machine learning-based computational modelling technique for complex, nonlinear investigations showcasing its superiority in analysing and predicting the drying process of plant-based foods efficiently and accurately.

BEERAMOOLE, Prithvi Bhat

Thesis Title

A Multi-Objective Framework for Extensive Hypothesis Testing During the Estimation of Discrete Choice Models

Supervisors

• Professor MD. Mazharul Haque  (Associate Supervisor)
• Professor Alexander Paz  (Principal Supervisor)
• Adjunct Professor Bouchra Senadji  (Associate Supervisor)

Citation

This research proposed an extensive hypothesis testing framework for estimating discrete choice models. Discovery of meaningful discrete choice specifications was investigated as an optimisation problem while testing for potential explanatory variables, their functional forms, nonlinearities, heterogeneous effects, latent classes, and correlations. Multiple performance measures were used to evaluate models for both interpretability and prediction. Experiments were conducted to illustrate the ability of the proposed framework in improving efficiency and objectivity of hypothesis testing. The new methods were shown to capture superior behavioural inferences compared to conventional approaches. Searchlogit- an open-source software was implemented based on the proposed framework.

BELLIE SUBRAMANI, Vishal

Thesis Title

Valorisation of Cotton Gin Trash Through Lignin-First Approach and Co-Liquefaction Studies for Value-Added Products

Supervisors

• Dr Luqman Abiola Atanda  (Associate Supervisor)
• Dr Lalehvash Laleh Moghaddam  (Principal Supervisor)
• Associate Professor Darryn Wallace Rackemann  (Associate Supervisor)
• Dr Changrong Shi  (Associate Supervisor)

Citation

This project investigated a sustainable biorefinery process to catalytically convert cotton gin trash (CGT), an underutilised agro-industrial waste originating from the cotton ginning process into a range of valuable products such as bio-aromatics, 5-chloromethyl furfural (CMF), and hydrocarbons. Moreover, the study incorporated plastic wastes to enhance the quality and stability of bio-oils via co-liquefaction with cotton gin trash (CGT), improving the value chain competitiveness within the cotton industry and biorefineries, simultaneously elucidating potential pathways for plastic waste management.

BURGESS‑LIMERICK, Benjamin

Thesis Title

Manipulation On-The-Move: Mobile Manipulation in Dynamic, Unstructured Environments

Supervisors

• Distinguished Professor Peter Corke  (Principal Supervisor)
• Dr Jesse Logan Haviland  (Associate Supervisor)
• Dr Christopher Francis Lehnert  (Associate Supervisor)

Citation

Humans are expert mobile manipulators - they not only manipulate in unstructured, dynamic, and unknowable environments but will also generally perform manipulation on-the-move, subconsciously optimising for future tasks. For instance, a human might grasp an object from a table while walking past or flick a light switch as they enter a room. Comparatively, current approaches in robotics are slow and inelegant - a mobile manipulator would instead sequentially drive to the target, stop, and then reach out to complete the task. This thesis addresses the capability gap and demonstrates robust manipulation on-the-move in real-world environments.

CLEGG, Julien Taylor

Thesis Title

Exploring the Potential of StarPEG-Heparin Hydrogels as an In Vitro Drug Testing System for Breast Cancer

Supervisors

• Associate Professor Laura Jane Bray  (Principal Supervisor)
• Associate Professor Larisa Margaret Haupt  (Associate Supervisor)
• Professor Erik Walter Thompson  (Associate Supervisor)

Citation

This thesis characterises a 3-dimensional hydrogel as a model for breast cancer with applications for novel drug testing, understanding drug-cell interactions, and future validation as a clinical tool. Using cell lines and patient-derived cells and tissue, the hydrogel model demonstrated that different anti-chemotherapeutic drugs could be used for treatment of encapsulated cells. The drugs interacted differently with the hydrogel, some being retained for longer durations than others, allowing for continual dosage to cells. Results demonstrate that this model can be used for drug testing and development for future applications as a precision medical drug screening tool for breast cancer.

COLLEDGE, Gabriel Thomas

Thesis Title

Development of Titania Based Materials for Fluoride Remediation in Water and Wastewater

Supervisors

• Professor Graeme John Millar  (Principal Supervisor)
• Dr John Gregory Outram  (Associate Supervisor)

Citation

Excessive intake of fluoride ions from groundwater can result in skeletal diseases especially in countries such as India. Consequently, there is a need to discover a cost effective and high performing fluoride sorbent which is practically easy. The hypothesis was that modification of commercial alumina sorbent could meet the project challenges. Indeed, integration of titania into alumina sorbents produced stable, selective materials with higher fluoride capacity than alumina alone. Significantly, the alumina/titania sorbent when bound into industry acceptable beads not only exhibited superior performance, but also inhibited formation of dissolved alumina which can lead to Alzheimer’s disease.

DMELLO, Alvin Thomas Arthur

Thesis Title

Enabling TCP/IP Communications Over High Frequency Communication Links

Supervisors

• Adjunct Associate Professor Ernest Foo  (Associate and External Supervisor)
• Dr Dhammika Jayalath  (Principal Supervisor)
• Adjunct Associate Professor Jason Frederick Reid  (External Supervisor)

Citation

High frequency (HF) communications have been the only wireless communication medium for beyond line of sight communication in a satellite denied environment. Defence and Emergency Disaster Recovery organisations have been the primary users of HF for decades. Due to its unreliable nature, HF has been primarily used for voice and low speed basic data. This research enables the most widely used IP based protocol, TCP/IP, over HF networks enhancing collaboration and flexibility in operations utilising HF systems. It examines the limitations of IP-based communications in HF channels and proposes an improved HF-TCP model.

FAN, Xiwei

Thesis Title

Spatial Phenotyping of Osteoarthritis through Macro, Micro, and Molecular Multiscale Approach

Supervisors

• Professor Ross William Crawford  (Associate Supervisor)
• Dr Indira Prasadam  (Principal Supervisor)

Citation

Osteoarthritis (OA) is a disabling disease affecting the whole joint. Traditional bulk analysis, averaging measurements across entire samples, limited our understanding of OA. This project employed multiple advanced microimaging techniques to overcome the limitation. The project illustrated that OA tissue exhibits distinct reactions at spatial levels from structural, elemental, mechanical, multi-omics and enzymatic perspectives. It also developed a novel protocol for preserving metabolites, proteins, and mineralisation while maintaining histological morphology. The project provides a solid foundation for personalised treatment development in OA by leveraging state-of-the-art imaging techniques and subcellular analysis methods.

GALVEZ SERNA, Julian Andres

Thesis Title

Autonomous Mission Planning and Biosignatures Image Detection for Planetary Exploration with Drones

Supervisors

• Associate Professor David Timothy Oki Flannery  (Associate Supervisor)
• Professor Luis Felipe Gonzalez  (Principal Supervisor)
• Dr Fernando Vanegas Alvarez  (Associate Supervisor)

Citation

This thesis reviews and expands current drone navigation and mission planning approaches for planetary exploration missions with drones, focusing on Mars and the autonomy needed. Recognising the importance of astrobiology in Mars exploration, this thesis proposes a high-level mission planning architecture for planetary exploration based on AI techniques such as reinforced learning, deep learning and computer vision focusing on missions to detect life signatures or biosignatures on planetary surfaces. The tools and ideas developed in this thesis aim to enable drones to help us answer the question, are we alone in the universe?

GOAN, Ethan James

Thesis Title

Practical Uncertainty in Neural Networks

Supervisors

• Professor Clinton Brian Fookes  (Principal Supervisor)
• Distinguished Professor Kerrie Lee Mengersen  (Associate Supervisor)
• Associate Professor Dimitri Perrin  (Associate Supervisor)

Citation

The adoption of machine learning technologies has grown considerably as the predictive performance of deep learning models continues to improve. Application of these systems for real-world scenarios requires not only raw predictive power, but also informative uncertainty information. Quantifying uncertainty comes at the expense of increased computation and time, and as a result most models do not aim to communicate any such information. This thesis addresses this by proposing practical means to quantify uncertainty in offline scenarios, real-time scenarios, and within existing neural networks not designed within a probabilistic framework.

HE, Beibei

Thesis Title

Availability and Mobility of Microplastics in River Sediments

Supervisors

• Professor Godwin Adefikayo Ayoko  (Associate Supervisor)
• Professor Sobana Ashantha Goonetilleke  (Principal Supervisor)
• Dr An Liu  (External Supervisor)
• Dr Llewellyn Rintoul  (Associate Supervisor)

Citation

This research study systematically investigated the presence of different microplastics in urban river sediments, assessed the factors which influence microplastics distribution, and evaluated the environmental risks of microplastics on sediment quality. Furthermore, the research project developed a three-dimensional mathematical modelling framework on the dispersal and transport processes of different sedimental microplastics. The outcomes of this research study will contribute to the informed management of waste plastics and formulation of effective mitigation measures

HENDAWITHARANA, Sahani Uththara

Thesis Title

Advanced Numerical and Experimental Studies of Bushfire Safe Rooms

Supervisors

• Dr Anthony Deloge Ariyanayagam  (Principal Supervisor)
• Professor Mahadeva Mahendran  (Mentoring Supervisor)

Citation

This research investigated the heat transfer in safe rooms exposed to bushfires and their bushfire resistance using full-scale laboratory and out-door experimental studies of external wall systems and a safe room and associated advanced numerical studies. It has advanced the knowledge and understanding of the heat transfer and the performance of the critical external building envelope and the safe rooms with different configurations exposed to bushfire flame conditions. Suitable solutions and methodologies have been proposed, addressing various complexities and challenges, which can be used in any bushfire-prone country like Australia. This will improve life safety and community resilience during bushfires.

HERATH, Buddhi Yushendra

Thesis Title

Design and Fabrication of Multi-Structural Scaffolds for the Treatment of Critical-Sized Bone Defects

Supervisors

• Mark Knackstedt  (External Supervisor)
• Associate Professor Judith Paige Little  (Associate Supervisor)
• Dr Beat Schmutz  (Associate Supervisor)
• Dr Marie‑Luise Wille  (Principal Supervisor)
• Adjunct Professor Prasad Kanaka Durga Yarlagadda  (Associate Supervisor)

Citation

This thesis developed a standardised workflow for designing patient-specific bone scaffold implants to treat large bone defects. The design workflow combined generative design principles, numerical modelling, and 3D printing and was validated through application to both, retrospective and ongoing clinical cases. The resulting workflow enabled the user to design complex scaffolds in a time efficient manner and they demonstrated excellent patient fit.

HEWA THONDILEGE, Akila

Thesis Title

Multimodal Image Correspondence

Supervisors

• Professor Clinton Brian Fookes  (Associate Supervisor)
• Dr Kien Nguyen Thanh  (Associate Supervisor)
• Professor Subramanian Sridharan  (Principal Supervisor)

Citation

Multimodal images are used across many application areas including medical and surveillance. Due to the different characteristics of different imaging modalities, developing image processing algorithms for multimodal images is challenging. This thesis proposes effective solutions for the challenging problem of multimodal semantic correspondence where the connections between similar components across images from different modalities are established. The proposed methods which are based on deep learning techniques have been applied for several applications including epilepsy type classification and 3D reconstruction of human hand from visible and X-ray image. These proposed algorithms can be adapted to many other imaging modalities.

HOWLADER, Md Mohasin

Thesis Title

Traffic Conflict Techniques Based on Extreme Value Theories: Investigation of Right-Turn Crash Risks at Signalised Intersections

Supervisors

• Professor Ashish Bhaskar  (Associate Supervisor)
• Professor MD. Mazharul Haque  (Principal Supervisor)
• Dr Shamsunnahar Yasmin  (Associate Supervisor)

Citation

This study developed a comprehensive safety assessment framework for opposing-through crashes at signalised intersections by applying Traffic Conflict Techniques. The framework combines Artificial Intelligence-based analytics and Extreme Value Theory for microscopic investigation of opposing-through traffic conflicts. This investigation addresses the identification of suitable traffic conflict measures, time-varying crash risk heterogeneity, severity-based crash risk estimation, transferability, correlation between traffic conflict types, evaluation methodology for engineering treatments, and the application of crash risk estimates in real time. The outcomes of the study open avenues for integrating safety into the operational framework for signalised intersections.

JIMENEZ FORERO, Javier Alexander

Thesis Title

Valorisation of Waste Biomass and Tyres Through Thermochemical Processing: Upgrading Strategies and Carbon Functionalisation for Sustainable Resource Recovery

Supervisors

• Dr Luqman Abiola Atanda  (Associate Supervisor)
• Dr Lalehvash Laleh Moghaddam  (Associate Supervisor)
• Professor Kostyantyn Ostrikov  (Associate Supervisor)
• Associate Professor Darryn Wallace Rackemann  (Principal Supervisor)

Citation

In response to the global challenge of escalating industrialisation and waste, this project explores thermochemical solutions for the sustainable management of abundant waste sources, such as scrap tyres and agricultural residues. Key findings include efficient demineralisation strategies for tyre derived char, production of high-value carbon composites, and optimisation of solvent systems for agricultural waste processing, The study underscores the critical influence of feedstock selection and processing methods on the properties of resulting carbon materials, exploring their distinct catalytic applications and environmental implications. The project contributes to advancing sustainable waste management practices, emphasising material recovery and environmental impact reduction.

JOHNSTON, Cameron James

Thesis Title

Synthesis of High Purity Alumina from Various Aluminium Rich Sources

Supervisors

• Professor Sara Jane Couperthwaite  (Associate Supervisor)
• Associate Professor Wayde Neil Martens  (Mentoring Supervisor)
• Dr Rachel Anais Pepper  (Principal Supervisor)

Citation

With the increased demand for critical mineral, such as high purity alumina, alternative feedstocks that would be typically overlooked are becoming viable alternatives.  This thesis addresses the use of aluminium rich sources for the synthesis of high purity alumina. The work shows that the fundamental understanding of aluminium chemistry can allow for different low grade feedstocks and mine waste sources and low grade ores can be successfully used for the synthesis of high purity alumina.

KARAPITIYE PATHIRANAGE, Indunil Erandi Ariyaratne

Thesis Title

Bushfire Resistant Wall Systems Using Lightweight Cement Blocks

Supervisors

• Dr Anthony Deloge Ariyanayagam  (Principal Supervisor)
• Professor Mahadeva Mahendran  (Mentoring Supervisor)

Citation

In situations where the unpredictability of bushfire severity and path makes evacuation unfeasible and fatal, the safest option is to seek shelter in a safe room or building. To withstand extreme fire conditions, it is important for the external walls of these safe rooms to have excellent fire resistance. This research study successfully developed lightweight aggregate masonry blocks with enhanced fire resistance, specifically designed for constructing walls in bushfire safe rooms and buildings located in bushfire-prone areas. Experimental and numerical studies were conducted at the material and component levels in developing lightweight aggregate masonry blocks that exhibit improved fire resistance.

KHATTAK, Nouman

Thesis Title

Characterisation and Seismic Vulnerability Assessment of Vintage Unreinforced Masonry Buildings in Queensland, Australia

Supervisors

• Dr Hossein Derakhshan  (Principal Supervisor)
• Adjunct Professor Nimal Jayantha Perera  (External Supervisor)
• Professor David Pathmaseelan Thambiratnam  (Mentoring Supervisor)

Citation

This thesis includes structural characterisation and seismic vulnerability assessment of pre-1940 unreinforced masonry buildings in Queensland, Australia. The generated data can be utilised to mitigate potential seismic risk arising from these buildings for enhanced community resilience. The thesis reports statistics from building inspections, highlighting key features influencing their seismic vulnerability. An internationally recognised methodology is adopted to assess their seismic vulnerability, with the outcomes being validated against empirical data. Utilising advanced computer simulations, seismic fragility of a typology of buildings is determined. This study lays the groundwork for understanding the typologies and the seismic vulnerability and fragility of such buildings.

KONATE, Salamata

Thesis Title

Interpretability of Machine Learning Models for Medical Image Analysis

Supervisors

• Adjunct Professor Andrew Peter Bradley  (Principal Supervisor)
• Professor Clinton Brian Fookes  (Associate Supervisor)
• Adjunct Professor Olivier Salvado  (External Supervisor)
• Dr Elliot Smith  (External Supervisor)

Citation

In the realm of medical machine learning (ML), the application of visual explanation methods to interpret model behaviour poses a significant challenge. As AI progresses, researchers strive to tackle real-world issues. This project explores the intersection of AI and medical imaging analysis, aiming to improve disease diagnosis accuracy through ML models. However, the "black box" nature of these models raises concerns about reliability and bias. By investigating saliency methods, particularly in scenarios involving biases in images, this research aims to shed light on model behaviour and limitations in medical ML.

LAUBACH, Markus

Thesis Title

Preclinical Assessment of a Novel Aspirator Device for Intramedullary Bone Graft Harvesting

Supervisors

• Dr Nathalie Bock  (Associate Supervisor)
• Distinguished Professor Dietmar Hutmacher  (Principal Supervisor)
• Dr Jacqui Anne Mcgovern  (Associate Supervisor)
• Dr Siamak Saifzadeh  (Associate Supervisor)

Citation

To repair large bone defects, the patient's own bone material is collected using the Reamer-Irrigator-Aspirator (RIA) system. The RIA system offers a way to extract bone from the inner part of long bones, such as the thigh. However, the RIA system has limitations as it is not very intuitive for surgeons to use and the quality of the bone graft is controversial in the literature. In in vitro and in vivo studies, we have investigated a novel device that aims to simplify and improve the intramedullary bone harvesting process in order to achieve improved outcomes for patients with bone defects.

LIU, Minghao

Thesis Title

Two-Dimensional Ferroelectric Materials on the Applications of Next-Generation Electronics

Supervisors

• Professor YuanTong Gu  (Mentoring Supervisor)
• Associate Professor Liangzhi Kou  (Principal Supervisor)
• Professor Ziqi Sun  (Associate Supervisor)

Citation

This project aimed to discover and design novel ferroelectric two-dimensional materials via theoretical approaches. The project was based on the state-of-the-art first-principles calculations and micromagnetic simulations. The thesis investigated two heterostructures and one monolayer with novel electronic, optical and magnetoelectric properties, rendering them promising materials platforms for next generation ferroelectricity-based information nanodevices.

LOUIS, Alan Jan‑Fay

Thesis Title

Control of Energy Storage Systems for Support of Voltage and Flow in Long Radial Distribution Networks

Supervisors

• Adjunct Professor Gerard Francis Ledwich  (Associate Supervisor)
• Associate Professor Yateendra Mishra  (Associate Supervisor)
• Associate Professor Geoffrey Ralph Walker  (Principal Supervisor)

Citation

Long radial distribution networks are vital to communities worldwide in providing cost-effective electricity supply across large areas; however, they are expensive to construct, have poor communications, and are prone to power quality issues. This research investigated the application of modern energy storage systems to support voltage and power flow in these networks through dynamic power control and energy storage. A holistic approach was developed, which targeted three key areas: local voltage control in resistive networks, multi-objective optimisation of voltage and flow to maximise network benefits, and the development of distribution state estimation to enhance local control schemes.

MADDUMA KANKANAMALAGE DON, Darshana Priyasad

Thesis Title

Improving Information Fusion in Deep Learning

Supervisors

• Dr Simon Paul Denman  (Associate Supervisor)
• Professor Clinton Brian Fookes  (Associate Supervisor)
• Professor Subramanian Sridharan  (Principal Supervisor)
• Dr Tharindu Romesh Fernando Warnakulasuriya  (Associate Supervisor)

Citation

As humans, we effortlessly use information obtained from multiple sensory perceptions when making decisions or responding. However, even with the explosive growth in the area of deep machine learning, Artificial Intelligence systems struggle to effectively fuse multisensory and multi-modal information to improve the overall performance of intelligent systems. This research has contributed to the creation of new knowledge with respect to information fusion by demonstrating the applicability of the novel fusion techniques that we developed to a wide range of tasks in the fields of human-computer interaction and automated diagnosis of diseases.

MANDEL, Nicolas Johann

Thesis Title

UAV-based Navigation and Grid Mapping Using Semantic Observations

Supervisors

• Professor Luis Felipe Gonzalez  (Principal Supervisor)
• Professor Michael John Milford  (Associate Supervisor)

Citation

This research was a step towards enabling Unmanned Aerial Vehicles to use semantic information for navigation. A navigation framework was established, a method to address constraints that UAVs face during early development stages was introduced and additional steps towards more effective grid maps for semantic navigation were taken by introducing a novel mapping approach and a map representation that respects UAV flying height. This thesis supports UAV autonomy by taking steps towards higher-level navigation using meaningful representations of environments.

MUHL CASTOLDI, Natalia

Thesis Title

Modelling Spinal Growth and Remodelling: Integrating Experimental Data in Multiscale Modelling and Simulation to Enhance Understanding of Healthy and Pathological Conditions

Supervisors

• Dr Maria Antico  (External Supervisor)
• Professor Laura Susan Gregory  (Associate Supervisor)
• Dr Dermot O'Rourke  (Associate Supervisor)
• Professor Peter Pivonka  (Principal Supervisor)
• Vittorio Sansalone  (External Supervisor)

Citation

Humans have an unique upright posture allowing bipedal locomotion. However, this posture has led to spinal diseases such as adolescent idiopathic scoliosis and vertebral compression fractures due to osteoporosis. The relevance of growth-related pathologies highlights the need for a good understanding of the respective changes in the spinal shape (growth) and bone tissue structure (remodelling). To this, this thesis focuses on the age-related evolution of vertebral shape and density through a vertebral statistical shape and appearance model, a discrete bone remodelling model, a growth model responsive to mechanobiological factors, and a new framework to model healthy vertebral growth.

NANJAPPA, Vinitha

Thesis Title

Fate and Distribution of Per and Poly Fluoroalkyl Substances (Pfas) in the Urban Environment

Supervisors

• Professor Godwin Adefikayo Ayoko  (Associate Supervisor)
• Associate Professor Weerawickramage Prasanna Egodawatta  (Principal Supervisor)
• Professor Sobana Ashantha Goonetilleke  (Associate Supervisor)

Citation

Per and poly fluoroalkyl substances (PFAS) are manufactured organic chemicals with a strong carbon and fluorine bond and thus extensively used in domestic and industrial applications, including firefighting foams. Eventually, PFAS reach the surface water contaminating water and sediment matrix. The distribution of PFAS in different creek and river systems demonstrates that residential and agricultural land uses and point sources, such as military bases and traffic hubs, are chief contributors. The outcome of this study is beneficial in monitoring PFAS in the environment and the partitioning behaviour assists in developing methods for the removal of PFAS.

PEIRIS, Mithum Chamara Shan

Thesis Title

Behaviour and Design of LSF Walls Subject to Combined Compression and Bending Actions Under Ambient and Fire Conditions

Supervisors

• Dr Anthony Deloge Ariyanayagam  (Associate Supervisor)
• Professor Mahadeva Mahendran  (Principal Supervisor)

Citation

Light steel framed walls used in the cold-formed steel construction are subjected to combined loading in several applications. This thesis presents detailed studies on the behaviour and design of steel compression members and plasterboard sheathed wall panels subjected to combined compression and bending actions under ambient and fire conditions. Both small-scale and full-scale tests were conducted, while simple and advanced numerical models were developed, validated, and used in parametric studies. This thesis has proposed improved design methods for unsheathed and sheathed light steel compression members under ambient and fire conditions for possible adoption by industry and cold-formed steel design standards.

RAKHMETOVA, Olga

Thesis Title

Thermodynamic Methods in the Process Modelling of Biocrude

Supervisors

• Professor Richard John Brown  (Associate Supervisor)
• Associate Professor Thomas James Rainey  (Principal Supervisor)

Citation

This thesis is a step forward in addressing the challenges associated with process modelling of biocrude and assessing the accuracy of thermodynamic models in the context of biofuel production. The comparative techno-economic analysis developed in this study has demonstrated the financial feasibility of fractional distillation as a biocrude upgrading step. This research highlighted the influence of thermodynamic methods on process modelling outcomes, which in turn have profound implications for the technical and economic aspects of biofuel production.

SHAHIN, Alireza

Thesis Title

Structural Performance of Timber-Steel Composite Beams

Supervisors

• Dr Craig Cowled  (Principal Supervisor)
• Dr Sabrina Fawzia  (Mentoring Supervisor)

Citation

Constraints on high-quality timber resources have created an imperative for the timber industry to improve the utilisation of their huge existing low-grade timber resources. One of the most promising and innovative methods to address this need is combining low-grade timber and steel plate using mechanical fasteners to manufacture a cost-efficient and sustainable composite section that benefits from both materials' structural advantages. To address this industry's need, the focus of this research was to develop a novel layered timber-steel composite (TSC) beam by efficient use of materials and applying a simple fabrication method to reduce the time and cost of manufacture.

SHARMA, Shubham

Thesis Title

Enhancing Dynamic Origin-Destination (OD) Matrix Estimation for Large Networks Through Feature Engineering: A Comprehensive Framework

Supervisors

• Professor Ashish Bhaskar  (Principal Supervisor)
• Professor Richi Nayak  (Associate Supervisor)

Citation

This thesis proposes novel algorithms for the efficient online estimation of dynamic travel demand in large-scale urban transport networks. Dynamic travel demand refers to the number of people travelling between various zones in a city at different times of the day, captured using Origin-Destination matrices. This data is vital for understanding travel behaviour, optimising traffic flow, and reducing congestion. However, their estimation for large networks is computationally challenging. This thesis developed a novel framework that leverages historical data and machine learning to reduce the computational requirements of the estimation process and improve accuracy, enabling design of better traffic management strategies.

SIVAKANTHAN, Subajiny

Thesis Title

Development and Characterisation of Rice Bran Oil and Sesame Oil-Based Oleogels for Healthy Food Applications

Supervisors

• Dr Sabrina Fawzia  (Associate Supervisor)
• Professor Azharul Karim  (Principal Supervisor)
• Terrence Madhujith  (External Supervisor)
• Sagadevan Mundree  (External Supervisor)

Citation

Trans fats, found in partially hydrogenated oils, are linked to increased risks of heart diseases, contributing to over 278,000 deaths annually worldwide. Their use in processed foods such as margarines and shortening poses a significant public health challenge, prompting global efforts to eliminate trans fats from the food supply. This thesis explores developing trans fat-free, bioactive-enriched oleogels as alternatives for high-trans fat margarines and shortenings. This study aims to develop and optimise healthy oleogels by tuning their structural and nutritional aspects. This investigation contributes to formulating food products that cater to the growing demand for healthier and functional dietary options.

SWARNKAR, Priyanka

Thesis Title

Transition Metal-Based Electrocatalysts for Electrochemical Water Splitting Influenced by Cerium

Supervisors

• Adjunct Professor John Marcus Bell  (Associate Supervisor)
• Professor Anthony Peter O'Mullane  (Associate Supervisor)
• Associate Professor Thomas James Rainey  (Principal Supervisor)
• T Sundararajan  (External Supervisor)

Citation

This project was a step forward in developing a material catalyst that has high potential for the overall water splitting process in an alkaline medium. The method is based on the electro-deposition, flame heat treatment of the selected catalyst onto the stainless steel substrate. The thesis investigated an electrodeposited, flame-heat-treated electrocatalyst for over-all water splitting using different electrochemical measurement techniques.

THILAKASIRI DISSANAYAKAGE, Pubudu Theekshana Dissanayake

Thesis Title

Deep Machine Learning for Biosignal Analysis

Supervisors

• Dr Simon Paul Denman  (Associate Supervisor)
• Professor Clinton Brian Fookes  (Principal Supervisor)
• Professor Subramanian Sridharan  (Associate Supervisor)
• Dr Tharindu Romesh Fernando Warnakulasuriya  (Associate Supervisor)

Citation

The application of deep learning to biosignal data is a rapidly evolving research domain in the Medical AI field, with many applications such as cardiac anomaly detection, seizure prediction and pneumonia diagnosis. Research conducted in this thesis presents a series of solutions for biosignal applications, focusing on models' generalisability, generation, interpretation, and design aspects. To evaluate the universality of the solutions, this thesis presents benchmark studies on various biosignal modalities such as electrocardiograms, phonocardiograms, lung sounds, and electroencephalograms. Furthermore, it focuses on the interpretability of the designed models and provides solid medically sound explanations for predictions made.

WALKER, Violet Maeve

Thesis Title

Enabling Multi-Agent UAV Search and Target-Finding in Partially Observable Environments

Supervisors

• Professor Luis Felipe Gonzalez  (Principal Supervisor)
• Dr Fernando Vanegas Alvarez  (Associate Supervisor)

Citation

Autonomous searching and target-finding within uncertain environments is a common requirement of robotic agents within search and rescue problems. Coordinating multiple unmanned aerial vehicles (UAVs) to search and target-find within partially observable environments is an actively explored problem within UAV robotics research. This thesis proposes and evaluates a software framework that enables multiple UAV agents to search and target-find within the kinds of environments commonly found in the search and rescue domain.

WEEKES, Angus Bret

Thesis Title

Off-the-shelf Vasculature: Biofabrication of Compliant Decellularized Vascular Grafts

Supervisors

• Dr Jason Jenkins  (External Supervisor)
• Professor Travis Klein  (Principal Supervisor)
• Professor Zhiyong Li  (Associate Supervisor)
• Dr Christoph Adalbert Meinert  (External Supervisor)
• Dr Nigel Pinto  (External Supervisor)

Citation

Regenerative medicine is a promising approach to produce vascular grafts for treatment of cardiovascular disease. However, current small diameter tissue-engineered vascular grafts (TEVGs) possess unsuitable mechanical properties, reflected by low patency rates in vivo. In this thesis, TEVGs were biofabricated using biomimetic melt-electrowritten (MEW) tubular scaffolds as substrates for cell culture. Through dynamic culture, TEVGs developed collagen-rich extracellular matrix and physiological mechanical performance. TEVGs were successfully decellularised, toward the development of non-immunogenic off-the-shelf products.

WILSON, Samuel James

Thesis Title

Post-hoc Feature-based Out-of-Distribution Detection for Real-World Conditions

Supervisors

• Dr Feras Dayoub  (External Supervisor)
• Dr Tobias Fischer  (Associate Supervisor)
• Dr Dimity Ann Miller  (Associate Supervisor)
• Professor Niko Suenderhauf  (Principal Supervisor)

Citation

This thesis addresses the challenge of detecting when vision models make erroneous, yet confident, predictions on new unknown inputs. The contributions proposed in this thesis detect these errors by monitoring the internal behaviour of the vision models during deployment, signalling when the model has behaved abnormally for a prediction. These contributions are researched in the context of mobile robot deployments in the real-world, emphasising the importance of computational efficiency, accuracy and risk reductions for the robot and anyone nearby.

YOUNG, Amy Rose

Thesis Title

Developing Professional Identity for Engineering Undergraduates

Supervisors

• Professor Leslie Anthony Dawes  (Principal Supervisor)
• Adjunct Professor Bouchra Senadji  (Associate Supervisor)

Citation

Engineering professional identity is a central construct which encompasses engineers' perceptions of themselves as professionals and is shaped by their education, work experience, personal values, and ethical principles. This research explored the complex development of engineering students' identity development using journey maps and highlighted the significance of practical experiences, peer connections and external recognition. The study identified key developmental stages during education and career progression and proposed practical strategies to support professional identity development.