Supervisor(s): Gianluca Demartini

The Wikipedia editors' community has been actively pursuing the intent of achieving gender equality. To that end, it is important to explore the historical evolution of underlying gender disparities in Wikipedia. This project presents the Wikipedia Gender Dashboard, a tool designed to enable the interaction with gender distribution data in subclasses of individuals (e.g. Astronauts) over time.

Supervisor(s): Ryan Ko,Omar Jarkas,Taejun Choi,David Gaul

- The research question of this project is whether Large Language Models can be utilized to de-anonymize Bitcoin transactions to catch the cybercrimes exploiting the anonymity of Bitcoin - I found the best parameters and configurations of Llama2 to de-anonymize Bitcoin transactions

Supervisor(s): Ryan Ko,Omar Jarkas,Taejun Choi

Cryptocurrencies, particularly Bitcoin, has attracted wide range of users, including cybercriminals who exploit the pseudo-anonymity these currencies offer by using the service for illicit activities. This project aims to deanonymize cryptocurrency transactions by linking them to real-world identities, with a focus on estimating the origin IP address of the transaction initiator.

Supervisor(s): Nilufar Baghaei

The project allows users to customise their own AI shopping assistant to accompany them in their Virtual Reality Shopping. Users will be able to see and talk with their assistant in the VR space. The goal of this project is to identify potential differences in behaviour when users are allowed to customise the assistant and when they are given an assistant without choice.

Supervisor(s): Dr. Chandima Ekanayake

Investigation into the risk of fires within household PV systems. Focused on the risk posed by DC isolators in PV systems installed before 2021.

Supervisor(s): Mark Utting

How can we enable novice users to reliably generate correct code from an LLM? This 'Iterative Assitant' tool helps them to converge towards correct code just by asking them about interesting examples.

Supervisor(s): Mithulan Nadarajah

This project investigates various EV fast charging solutions for deployment in capacity constrained distribution networks. The project focuses on an NRMA proposed charging site and compares different charging solutions for various charger capacities from both a technical and economical perspective.

Supervisor(s): Pauline Pounds

The Robotics Design Lab has developed a method to allow VR users to walk through virtual environments without translating in real life. This means that they can go arbitrary virtual distances without crashing into non-virtual walls. This project involves developing a "works like", "looks like", "feels like" prototype that matches a real final product as closely as possible.

Supervisor(s): Ben Rose,Ben Matthews

This project aims to design a virtual cox for the rowing community to enhance the interactivity, immersion, and teamwork of indoor rowing training.

Supervisor(s): Ben Rose,Ben Matthews

Prototype mobile application designed to connect immigrant parents within their school community.

Supervisor(s): Tim Miller

SmartBins is a COMP1110 (Introduction to Software Innovation) project that aims to help people put waste in the right bins. It is an AI-based computer vision system using a webcam and screen that sits behind a bin and to classifier waste objects to educate people about which bins should be used for different types of waste.

Supervisor(s): Jen Jen Chung,Pipar Automation

The AIMS-ASV is a fully autonomous surface vessel capable of supporting various surveying payloads to facilitate the Australian Institute of Marine Science's goal of quality data collection on the Great Barrier Reef for ongoing preservation. This project developed mechanical sonar scanning strategies for underwater obstacle avoidance to navigate the AIMS-ASV on hazardous and unmapped reef terrain.

Supervisor(s): Jen Jen Chung,Brendan Tidd,Yifei Chen

Designing optimal reward functions for RL is a challenge, especially for complex tasks like walking. Recently, LMMs have gained attention for their remarkable capability to translate natural language into machine-level instructions. ViTLearn explores the use of large multi-modal models (LMMs) to automatically generate rewards for robotic tasks.

Supervisor(s): Tapan Saha

This thesis project examines the impact of removing available system inertia from the grid on the future stability and reliability of the network, specifically looking at frequency and the ancillary services that maintain this stability measure.

Supervisor(s): Dr Konstanty Bialkowski

Vector Network Analyser (VNA) calibration is an essential part of ensuring accurate measurements for frequency dependent devices. The project conducted is the design of an inline electronic VNA calibration device that utilises radio frequency (RF) switches.

Supervisor(s): Tim Miller

This is a COMP1100/7110 (Intro to Software Innovation) project in which the team has implemented a 'group buy' app to allow sellers to sell bulk items such as food, and buyers to 'team up' to purchase these in bulk

Supervisor(s): Guangdong Bai

This study focuses on evaluating and identifying potential security vulnerabilities within the ChatGPT plugin system. By analyzing the interaction between plugins,

Supervisor(s): Dan Kim

Advancements in AI and deep learning have improved image and video generation but introduced security risks. This study strengthens deepfake detection by applying multi-scale training and noise augmentation, leading to a 9% performance boost. The results highlight the effectiveness of the proposed approach in enhancing detection across varying resolutions and augmented images.

Supervisor(s): Ruifeng (Richard) Yan

This research aims to develop a control system based on Siemens S7-1500 programmable logic controller (PLC) for managing and controlling inverters in large-scale renewable energy (solar or battery) farms.

Supervisor(s): Mashhuda Glencross,Jason Weigel,Ben Rose

SwingTheory is a tennis coaching app for beginners that analyzes users' swing posture using OpenPose and machine learning, providing corrective feedback and tracking improvement through live sections. It helps users refine their technique by comparing their form to an ideal posture.

Supervisor(s): Lu Zhang

Establish a cost-effective SDR-based testbed for interference signal localization and tracking.

Supervisor(s): Mashhuda Glencross,Jason Weigel,Angus Moore

Using cutting-edge artificial intelligence, we forecast potential flood extents and provide timely notifications if you're at risk. Stay informed, stay prepared, and protect what matters most with our accurate predictions and real-time alerts.

Supervisor(s): Associate Professor Rahul Sharma,Dr Gayan Lankeshwara

The rapid growth in distributed energy resources (DERs) such as rooftop solar and batteries poses challenges for low-voltage network management. Integrating dynamic operating envelopes (DOE) with demand response (DR) enhances DER asset utilisation for customers, while enabling network providers to maintain secure operations within statutory limits under current and future conditions.

Supervisor(s): Jarrod Knibbe

This project explored high resolution haptic systems for VR pens where the goal was to make writing in VR feel just like writing with a pen and paper. An omnidirectional ball haptic system was designed for this project and has been tested in user studies.

Supervisor(s): Alina Bialkowski

Deep neural networks demand substantial computational resources and are often seen as black boxes in decision-making. Model compression and explainability can address these issues. This project investigated the interplay between pruning, accuracy and explainability, and developed an explainability-based pruning approach that revealed key differences between human and machine vision processing.

Supervisor(s): Dr Marie Bodén

The study investigates the impact of Augmented Reality (AR) in enhancing online shopping experience within the luxury retail sector by integrating Media Richness Theory (MRT). It explores how AR impact engagement and decision-making by offering immersive try-on experiences. Through qualitative interviews with users who interacted with virtual try-on feature on a luxury retail platform, the researc

Supervisor(s): Ramesh Bonu

The rapid rise of power electronic converter-based generation units led to the evolution of virtual synchronous generators (VSGs) for power system stability. However, VSGs also suffer from instabilities due to improper control design. This thesis alleviates the instability through white box modelling technique and proposes a strategy for control design of VSGs.

Supervisor(s): Amin Abbosh,Lei Guo

This project proposed an electromagnetic probe design, incorporated with several machine learning techniques, for classifying the water purities. The result show a prominent accuracy and potential to be used in practical applications.

Supervisor(s): Matt D'Souza

This project creates a graph-based neural network framework written in the Rust for Machine Learning and AI applications.

Supervisor(s): Matt D'Souza

The design of smart RFID plant tag reader aims to create a mobile reader which can read the information of different plants, allows people to access information about plants at any time, facilitating easier management.

Supervisor(s): Matt D'Souza

This project aims to create a system that can track and display ADS-B transponders of nearby aircraft within a designated portion of the sky. By harnessing the capabilities of Software Defined Radio (SDR) and embedded systems, the goal is to provide real-time transponder data, including aircraft IDs, latitude, and longitude based on a selected portion of the sky.

Supervisor(s): Mashhuda Glencross,Jason Weigel,Lorraine Han

The TEBS (Technology-Enhanced Board Game System) combines tactile game pieces on a digital board to enhance traditional board games, creating a novel interactive gaming experience.

Supervisor(s): Nell Baghaei

In this project, we introduce virtual mirrors to an individualised Virtual Reality application for self-compassion therapy. The results shows that virtual mirrors bring slightly improvement to the therapeutic effects and UX, receiving neutral or positive feedback from nearly all participants for its role in the comforting process.

Supervisor(s): Pauline Pounds

This project develops a convertible prop-rotor that can transform between a wide flat rotor like a helicopter's and a narrow, highly pitched propeller like a conventional aeroplane's. Previous analytical work indicated that this concept was feasible, but this work puts it to the test with a practical, compact conversion mechanism. The protor is tested across a range of airflow conditions.

Supervisor(s): Justin Cooper-White

Cell and tissue engineering has had many advancements, but there are still no methods to control cells in a spatial and temporal controlled manner. This project investigates the role of electric fields in axon behaviour, independent of any other factors, and applies this to biological control.

Supervisor(s): Matthew D'Souza

Improving Radio-Frequency Communication in High Noise Environments for Applications in UQ RobotX Unmanned Surface Vessel Project

Supervisor(s): Matthew D'Souza

With a growing global population, the availability of food is becoming a critical issue. This highlights the urgent need to monitor key environmental factors that impact plant growth, such as CO2 levels, soil temperature, soil moisture, ambient light, and UV exposure. Real-time data from various waterproof monitors, transmitted wirelessly to a centralized hub, enables effective management of plan

Supervisor(s): Naipeng Dong

This project studies vulnerabilities of smart contracts and explores existing and potential technology in detecting and preventing these vulnerabilities.