Congratulations to the 2023 Winners!
Experimental Fault Analysis in DFIG Wind Turbines
by Blake Rowden [Supervisor: Ruifeng (Richard) Yan]
This project focuses on the Low Voltage Ride-Through (LVRT) capability of Doubly Fed Induction Generators (DFIG), which play a pivotal role in maintaining grid stability during voltage sags and preventing cascading failures.
The project involved the design and execution of a reproducible laboratory testing process, with a focus on safely testing a DFIG under terminal faults and validating its LVRT response. Additionally, a simulation was developed to closely replicate transient responses observed in the laboratory, enabling a robust comparative analysis between practical experiments and theoretical simulations. Detailed analyses included exploration of reactive current injection, stator DQ currents, and examination of DC bus and control dynamics under varied fault scenarios. The project not only validated critical DFIG behaviours but also adeptly integrated black-box practical observations with white-box simulations to comprehend and predict DFIG responses during disturbances.
Winner of: Best Project in Power and Energy Systems (Tied)
Sponsored By: 
Is DOE the Solution to Fixing Australia's Demand Curve?
by Thomas Jellett [Supervisor: Ruifeng (Richard) Yan]
The current electricity grids were not designed for the rapid uptake of renewable injection, and renewable supply such as solar are far exceeding demand during the hours when it is not needed. This creates more frequent occurrences of negative spot prices, and this discourages suppliers from generating energy for the market as it is no longer economically viable to supply energy when demand is low causing suppliers to switch off generators. Resulting in not enough energy being supplied to the network during peak demand.
Dynamic operating envelops (DOE) are import and export limits on devices that change dynamically based on current network conditions. The benefit of applying dynamic limits to distribution energy resources (DER) will allow us to enable higher energy exports from DER and reduce flexible loads during peak times. And as we see the number of DER penetrations increase, along with the periods of renewable generation exceeding demand, DOEs will play an essential role to managing and coordinating DER grid connections. This will be essential for reaching 100% renewable energy generation by 2050.
For this project the student has developed a software capable of receiving and processing dynamic limits for DERs.
Winner of: Best Project in Power and Energy Systems (Tied)
Sponsored By:
Blockchain in Aviation Training Industry
by Tinghan Wang [Supervisor: Naipeng Dong]
This project explores the potential of smart contracts and blokchain solutions to reveolutionize the aviation training, in addressing challenges of data integrity, transparency and compliance of regulations.
Winner of: Best Distributed Ledger Technology Project (Tied)
Sponsored By: 
The Benefits of Non-Fungible Token (NFT) Technology in Music Copyright
by Jie Dong [Supervisors: Guangdong Bai and Naipeng Dong]
The emergence of Non-Fungible Tokens (NFTs) has presented a novel opportunity to address long-standing challenges in the realm of music copyright. This research paper explores the current iterations of NFTs as a potential solution for music copyright in the digital age.
Winner of: Best Distributed Ledger Technology Project (Tied)
Sponsored By:
Precise surface following with Riemannian Motion Policies
by Chelsea Davidson [Supervisor: Chelsea Davidson]
At CERN, radiation surveys of equipment and beam lines are important for safety and analysis throughout the accelerator complex. Radiation measurements are highly dependent on the distance between the sensor and the radiation source. If this distance can be accurately established, the measurements can be used to better understand the radiation levels of the components and can be used for calibration purposes. Currently, in areas where no fixed infrastructure exists to perform these surveys, technicians must enter the facilities to manually measure the radiation dose. Manual measurements are often less precise, since they are taken at a non-constant distance from the beam line, and they also put personnel at higher risk of radiation exposure.
The goal of this project is to develop Riemannian Motion Policies (RMPs) for a robot arm, whose base is mounted on a moving monorail, to maintain its end-effector at a fixed distance from the beam line. The project is specifically aimed at developing a reactive motion planner that will allow the arm to adapt its configuration to avoid obstacles detected online by its onboard range sensors.
Winner of: Emesent Diversity Prize
Sponsored By: 
Tiny Giant Robot 2: Striking a Balance
by Deren Teo [Supervisor: Pauline Pounds]
The TGR combines the balancing capabilities of advanced dynamical bipeds with the affordability and simplicy of drones. The second generation TGR is superior in every way: smarter, faster, stronger, and better-dressed. Making TGR2 balance, however, required critical enabling advancements in control firmware, a breakthrough in light-weight gimbal design, and careful application of good old fashioned systematic control theory princples to make the system stand robustly.
Winner of: Best Project in Avionics & UAV
Sponsored By: 
Near-field Microwave Probes for Skin Cancer Detection
by Kexin Liu [Supervisors: Amin Abbosh & Syed Naqvi]
This project conducts a feasibility study on the utilization of near-field probes for skin cancer detection. It encompasses the exploration of various scenarios for single and multi-sensor applications in the context of tumor detection within a simulation environment. Additionally, the project delves into the interface used to acquire dielectric data through near-field probes, with a focus on enhancing its functionality.
Winner of: Best Biomedical Device or Systems Prize
Sponsored By: 
FPGA packet filter with Ethernet MAC and web server using a RISC-V softcore processor
by Matthew Gilpin [Supervisor: Matt D'Souza]
Just like any other computer connected to the broader network, edge networks must also be safeguarded from malicious bad actors. Field programmable gate arrays (FPGAs) offer the flexibility of custom hardware that can be designed to incentivise low latency, high throughput yet efficient wire-speed firewalls. While current hardware firewalls exist in todays markets, they often come at a cost and high power usage rendering them unsuitable in edge networks. To address this, we designed a FPGA firewall that fulfils these criteria.
Winner of: Best Cyber Security and Data Privacy Project
Sponsored By: 
Air Quality Monitoring Flower
by Wilko Grunefeld and Chelsea White [Supervisor: Matt D'Souza]
This project involves using a collection of sensors to estimate overall air quality, and to present this in a user-friendly manner; a flower that opens as an easy visual for overall air quality, and a PC interface that shows detailed sensor data.
Winner of: Best Non-Thesis Project
Sponsored By: 
Detecting and Classifying Neurofibromas using Deep Learning
by Georgia Power and Jinghao Lim [Supervisor: Brian Lovell]
Neurofibromatosis type 1 (NF1) is one of the most common single-gene inherited disorders globally, with an incidence of 1/2500 individuals. While several phenotypes are associated with the disorder, the most common manifestation is cutaneous neurofibroma. The majority of adults develop these distressing cutaneous tumours (cNF), which increase in severity with age. Adult patients report cosmetic disfigurement due to cNF as the greatest burden of living with NF1. There is no way to predict tumour severity which can range from < 100 to thousands. Youth and families experience reduced quality of life due to concerns about this uncertain future. We don't yet understand why this condition is so variable or have any effective medical treatments.
In the proposed research, we will assemble a consortium of internationally recognised experts in NF1 with access and capacity to recruit and phenotype patients to drive the largest genome-wide association and epigenome-wide association studies of the modifier gene networks driving the cutaneous phenotypic variance in NF1. We will then use individualised pharmacological annotation of these networks to identify precision treatment options to mitigate the most distressing and life quality damaging aspects of this devastating illness.
Winner of: Best Electrical Engineering Project
Sponsored By: 
Tiny Giant Robot 2: Striking a Balance
by Deren Teo [Supervisor: Pauline Pounds]
The TGR combines the balancing capabilities of advanced dynamical bipeds with the affordability and simplicy of drones. The second generation TGR is superior in every way: smarter, faster, stronger, and better-dressed. Making TGR2 balance, however, required critical enabling advancements in control firmware, a breakthrough in light-weight gimbal design, and careful application of good old fashioned systematic control theory princples to make the system stand robustly.
Winner of: Les Dobson Memorial Prize: Best Systems & Software Engineering Project
Sponsored By: 
Designing a Soil Respiration Chamber
by Thomas Salpietro [Supervisor: Matt D'Souza]
The aim of this project is to is to develop a soil respiration chamber that uses a carbon dioxide (CO2) sensor to measure the CO2 emissions from soil. This project is supported by the UQ Aquatic Research Group and is imperative to improving their current soil respiration sensing system. The need for “smart” monitoring of carbon dioxide is constantly growing in numerous applications, such as agriculture, plant gas exchange, atmospheric gas monitoring and, most importantly, soil carbon dioxide flux
Winner of: Best Hardware Solution
Sponsored By: 
Sprout 🌿
by Caleb Ang, Lachlan Benson, Eugene Chong, Li-Sung Ou, Hoang Ly Phung, Nisha Vashist [Supervisor: Mashhuda Glencross]
This project addresses the brief on self-mental awareness through biofeedback. The ultimate objective is to design a wearable watch paired with a mobile application for patients with bipolar disorder. Our Arduino prototype has been equipped with biofeedback sensors to detect:
- Movement (via gyro/accelerometer)
- Sweat levels
- Steps (pedometer)
- Heart rate (note: the sensor is currently malfunctioning)
These sensors collectively identify shifts in the user's mood and potential negative mental hetitleh patterns based on specific parameters. Data is presented on a wearable watch display, which is now simulated by a React web app. The user is then prompted to verify if they are experiencing a ""negative"" mental loop. If confirmed, they are navigated through a recovery process, pre-determined by their medical practitioner. This process offers them activities designed to assist them in navigating and managing their negative mental state.
Winner of: Best User Experience Design
Sponsored By: 
Reinforcement Learning for Active Search and Grasp
by Thomas Pitcher [Supervisor: Jen Jen Chung]
When humans need to find and retrieve hidden objects they are able to fluently reason between changing their viewpoint and moving objects within the space. For robots, this is a mechanical search problem. While we have tools for autonomously synthesizing grasps on cluttered scenes, as well as methods for actively improving scene reconstruction when the target object is only partially visible, the goal of this project is to extend these capabilities to situations where the target object is fully occluded at the start of grasping and interactive search is needed. In particular, this project will look at how the task can be formulated as a reinforcement learning problem where the main challenge is in properly comparing the value of moving the camera to a new viewpoint vs. physically removing objects to reveal more of the scene.
Winner of: Best Software Project
Sponsored By: 
Applications of Software Defined Radios on Unmanned Aerial Vehicles for Intelligence, Surveillance, and Reconnaissance
by Mark Nothdurft [Supervisor: Konstanty Bialkowski]
The use of UAVs have an ever-growing presence during modern conflicts. This project explores the use of SDRs for ISR mounted to UAVs for countering the adversaries use of UAVs. The thesis looked to develop a functioning passive radar System, capable of performing direction-finding passive radar and bistatic passive radar. Additionally the ability of an unknown active radar signal to be decoded was explored.
Winner of: Best Project in Microwave, Photonics and Communications
Sponsored By: 
Sprout 🌿
by Caleb Ang, Lachlan Benson, Eugene Chong, Li-Sung Ou, Hoang Ly Phung, Nisha Vashist [Supervisor: Mashhuda Glencross]
This project addresses the brief on self-mental awareness through biofeedback. The ultimate objective is to design a wearable watch paired with a mobile application for patients with bipolar disorder. Our Arduino prototype has been equipped with biofeedback sensors to detect:
- Movement (via gyro/accelerometer)
- Sweat levels
- Steps (pedometer)
- Heart rate (note: the sensor is currently malfunctioning)
These sensors collectively identify shifts in the user's mood and potential negative mental hetitleh patterns based on specific parameters. Data is presented on a wearable watch display, which is now simulated by a React web app. The user is then prompted to verify if they are experiencing a ""negative"" mental loop. If confirmed, they are navigated through a recovery process, pre-determined by their medical practitioner. This process offers them activities designed to assist them in navigating and managing their negative mental state.
Winner of: Project with the Most Commercial Potential
Sponsored By: 
Investigating the Impact of Visual Embellishments on Elderly Users' Experience in Virtual Reality Exergames
by Suki Zhang [Supervisor: Nell Baghaei]
This project represents a significant advancement in the domain of virtual reality (VR) technology, focused on enhancing the experience of elderly users in virtual reality exergames, particularly in a body gesture-based monster hunting exercise game. By exploring the interplay of visual embellishments, this project contributes to a more profound understanding of how these elements can be strategically optimised to augment the experiences of elderly individuals in virtual reality exergames.
Preliminary findings from data collected from 27 participants revealed that the virtual reality exergame, which were conducted in an ethical manner and received approval from the ethics committee. With the incorporation of visual embellishments, resulted in higher physical exertion as well as increased heart rate with minimal discomfort or virtual sickness symptoms. This not only enhances participants' physical engagement but also fosters a heightened sense of enjoyment during their virtual reality exergaming activities. Our research underscores a commitment to pioneering user experience design and dedication to enhancing the lives of elderly users through innovative XR technologies. We look forward to uncovering deeper insights in the field's future.
Winner of: Most Innovative use of XR
Sponsored By: 
Near-field Microwave Probes for Skin Cancer Detection
by Kexin Liu [Supervisors: Amin Abbosh & Syed Naqvi]
This project conducts a feasibility study on the utilization of near-field probes for skin cancer detection. It encompasses the exploration of various scenarios for single and multi-sensor applications in the context of tumor detection within a simulation environment. Additionally, the project delves into the interface used to acquire dielectric data through near-field probes, with a focus on enhancing its functionality.
Winner of: People's Choice Award
Sponsored By: & 
2023 Project Nominations
A combined skeleton aware multi-modal & entropy metric-based key pose extraction method for sign language detection
This project aims to address the current limitations surrounding sign language detection in Auslan (Australian Sign Language), the preferred language of the Australian Deaf community. In particular, this project aims to detect a difference between sign language gestures and general hand gestures by analysing physical hand movement patterns, thus furthering the development and complexity of sign language recognition in Auslan.
Nominations
Air Quality Monitoring Flower
This project involves using a collection of sensors to estimate overall air quality, and to present this in a user-friendly manner; a flower that opens as an easy visual for overall air quality, and a PC interface that shows detailed sensor data.
Nominations
Applications of Software Defined Radios on Unmanned Aerial Vehicles for Intelligence, Surveillance, and Reconnaissance
The use of UAVs have an ever-growing presence during modern conflicts. This project explores the use of SDRs for ISR mounted to UAVs for countering the adversaries use of UAVs. The thesis looked to develop a functioning passive radar System, capable of performing direction-finding passive radar and bistatic passive radar. Additionally the ability of an unknown active radar signal to be decoded was explored.
Nominations
Automating Threat Detection with Open Source AI-SIEM
Traditional SIEM solutions depend on rules and thresholds to correlate events and recognise probable security breaches. Nevertheless, these outdated systems have difficulty detecting complex attacks. Their wide reliance on manual setup and security operators' expertise restricts the scalability and velocity of threat recognition.
To tackle these issues, this project suggests utilising an open-source AI-SIEM that merges advanced machine learning methods with traditional SIEM capabilities. The...
Nominations
Blockchain in Aviation Training Industry
This project explores the potential of smart contracts and blokchain solutions to reveolutionize the aviation training, in addressing challenges of data integrity, transparency and compliance of regulations.
Nominations
Cyber Security Activities for High School Students
Introducing cyber security to high school students is becoming more important as the world continues to shift its operations online, making it a hotspot for cybercrime. Existing solutions that aim to teach high school students about cybersecurity are game-based and often poorly balance educational value and engagement. This project researched “How to design security activities that complement the Australian Curriculum, whilst being engaging to students” Cyber Security activities have been des...
Nominations
CycleSafe
By leveraging machine learning (ML), artificial intelligence (AI) and a wearable alert system, Safe Cycle aims to alert professional athletes and non-professional cyclists alike to imminent threats to their safety.
Nominations
Data Science for Sleep Data Analysis
During NREM sleep, k-complexes and sleep spindles often occur and have been hypothesized to be involved in memory consolidation; however, their exact purpose remains to be elucidated. In Parkinson’s disease, studies have shown associations between a reduction in K-complex and spindle frequency (density) in dopamine-treated patients. Further research indicates excessive daytime sleepiness and cognitive decline where dementia is present in PD also reduces spindle density. The underlying patholo...
Nominations
Data-Driven Transformer Condition Assessment
Traditional methods for power transformer insulation assessment, such as oil tests and dissolved gas analysis, face challenges in detection accuracy and practicality. This thesis enhances transformer insulation diagnosis by integrating diverse datasets and applying advanced machine learning techniques. A comparative analysis using various algorithms was performed, followed by the design and implementation of a convolutional neural network and an ensemble learning model. These advanced methods...
Nominations
Designing a Soil Respiration Chamber
The aim of this project is to is to develop a soil respiration chamber that uses a carbon dioxide (CO2) sensor to measure the CO2 emissions from soil. This project is supported by the UQ Aquatic Research Group and is imperative to improving their current soil respiration sensing system. The need for “smart” monitoring of carbon dioxide is constantly growing in numerous applications, such as agriculture, plant gas exchange, atmospheric gas monitoring and, most importantly, soil carbon dioxide ...
Nominations
Detecting and Classifying Neurofibromas using Deep Learning
Neurofibromatosis type 1 (NF1) is one of the most common single-gene inherited disorders globally, with an incidence of 1/2500 individuals. While several phenotypes are associated with the disorder, the most common manifestation is cutaneous neurofibroma. The majority of adults develop these distressing cutaneous tumours (cNF), which increase in severity with age. Adult patients report cosmetic disfigurement due to cNF as the greatest burden of living with NF1. There is no way to predict tumo...
Nominations
Disaster Simulation using AR Sandboxes
We created an extension to an existing AR Sandbox visualisations to simulate floods and bushfires and their effects. UC Davis and RWTH Aachen University have developed sandboxes which simulate water free surface flow and geological phenomena respectively. It was determined that the RWTH sandbox is the most suitable to develop on further (under GNU GPL) due to its extensibility in python and prebuilt functionality for ArUco markers. Unreal Engine has been selected as a base fallback, in case t...
Nominations
Experimental Fault Analysis in DFIG Wind Turbines
This project focuses on the Low Voltage Ride-Through (LVRT) capability of Doubly Fed Induction Generators (DFIG), which play a pivotal role in maintaining grid stability during voltage sags and preventing cascading failures.
The project involved the design and execution of a reproducible laboratory testing process, with a focus on safely testing a DFIG under terminal faults and validating its LVRT response. Additionally, a simulation was developed to closely replicate transient responses obs...
Nominations
FAMEFAKED
FAMEFAKED is an educational game that presents the user with a video that is either deep faked or real. The user must guess whether they think the video is real or not.
Nominations
FPGA packet filter with Ethernet MAC and web server using a RISC-V softcore processor
Just like any other computer connected to the broader network, edge networks must also be safeguarded from malicious bad actors. Field programmable gate arrays (FPGAs) offer the flexibility of custom hardware that can be designed to incentivise low latency, high throughput yet efficient wire-speed firewalls. While current hardware firewalls exist in todays markets, they often come at a cost and high power usage rendering them unsuitable in edge networks. To address this, we designed a FPGA fi...
Nominations
Gardens of the Galaxy
You escaped Earth, only to crash-land in an unknown planet. You wake up to a furious alien towering over you; you ruined his farm, and we better restore it! The "Gardens of the Galaxy" is a casual space-themed farming simulator game, with weather, fishing and combat against aliens. The main gameplay involves planting, growing and harvesting plants, taking care of them and protecting them from hostile animals and the harsh climate conditions. Whilst doing so, you must ensure your hunger and he...
Nominations
Gesture Recognition (using hands and arms) with mmW Radar tracking to control an accessible music instrument.
Traditional music instruments such as piano or violin require players to have a high degree of dexterity to use. This leads to instruments being difficult to use or simply inaccessible to those with dexterity related physical disabilities. Even more non-traditional instruments such as modular synths use inputs such as buttons or knobs, which still require some degree of fine motor skills. Some alternate motion controlled instruments already exist, such as the Theremin (using RF) and Realsense...
Nominations
I Would Walk 500 Virtual Miles (More)
Walking around in VR sucks. Really - you run out of space so quickly, especially if you live in a shoebox apartment like us. You can just feel the walls closing in... Long story short: this project explores a new technique for allowing a user in a VR environment to feel like they have unconstrained walking range, without actually having unconstrained walking range. How, you ask? An excellent question... the answer to which will have to come visit to discover! Is this the next step in vir...
Nominations
I Would Walk 500 Virtual Miles (More)
Walking around in VR sucks. Really - you run out of space so quickly, especially if you live in a shoebox apartment like us. You can just feel the walls closing in... Long story short: this project explores a new technique for allowing a user in a VR environment to feel like they have unconstrained walking range, without actually having unconstrained walking range. How, you ask? An excellent question... the answer to which will have to come visit to discover! Is this the next step in vir...
Nominations
Insect Classifier for FarmBot
Insect pests have long been a problem for agricultural production, with chemicals and natural enemies historically used for control. However, climate change has exacerbated the challenge of pest management, causing 40% of the world's crops to be lost to pests annually. Effective and sustainable methods for insect pest management are critical, and the proposed project aims to address a gap in knowledge by designing an interoperable acoustic classifier subsystem for FarmBot-based crop managemen...
Nominations
Intruder detection and classification based on mmWave radar
This project explores the use of mmWave radar as a sensor to perform intruder detection and identification. For this project, the intruder is a human with different kinematic parameters and profiles. The radar sensor acts as an invisible fence operating at millimetre wave frequency band, in which the output of the project will propose the best operating frequency to get maximum information of the intruders and optimum coverage range. The project work comprises data collection through experime...
Nominations
Investigating the Impact of Visual Embellishments on Elderly Users' Experience in Virtual Reality Exergames
This project represents a significant advancement in the domain of virtual reality (VR) technology, focused on enhancing the experience of elderly users in virtual reality exergames, particularly in a body gesture-based monster hunting exercise game. By exploring the interplay of visual embellishments, this project contributes to a more profound understanding of how these elements can be strategically optimised to augment the experiences of elderly individuals in virtual reality exergames. Pr...
Nominations
Is DOE the Solution to Fixing Australia’s Demand Curve?
The current electricity grids were not designed for the rapid uptake of renewable injection, and renewable supply such as solar are far exceeding demand during the hours when it is not needed. This creates more frequent occurrences of negative spot prices, and this discourages suppliers from generating energy for the market as it is no longer economically viable to supply energy when demand is low causing suppliers to switch off generators. Resulting in not enough energy being supplied to the...
Nominations
Ken
Household chore distribution system. An app that uses various NFCs around households to allow individuals to coordinate household chores.
Nominations
Magnetic Resonance Image (MRI) Segmentation in the Latent Space
The proposed model pipeline aims to perform image segmentation and reconstruction of MRI in the latent space. The model takes an MR image as input and generates a reconstructed segmentation mask of the image. The pipeline comprises a pretrained VQ-VAE encoder, a latent U-Net trained using mean squared error loss, and a custom VQ-VAE decoder trained using dice squared loss. The trained model is able to successfully segment a brain MRI with a validation mse loss of 0.332 and a validation dice s...
Nominations
MICROBOTICS: Microangelo's Magnificent Miniature Masterpieces
Renowned tiny renaissance man, Microangelo, is famous for his feud with the equally famous Leonardo Di Minutive. Each rival seeks to create smaller and smaller artworks, expertly drawn and tinier than the last. In the latest artistic exchange, Leonardo set the bar even smaller by painting his magnum opus. the Mini Lisa, a mere 2 cm high - not to be outdone, Microangelo must respond with something even smaller! Your task: construct Microangelo's robot, capable of drawing an accurate image o...
Nominations
MICROBOTICS: Mini Matchbox Molecular Marble Mobile Manipulators
The eccentric A/Professor Pounds has lost her marbles! Now, this is not especially unusual for an academic, but A/Prof. Pounds' marbles are very small… very, very small. Finding and returning these tiny spheres to their proper receptacles will require more deft manipulation than a simple pair of tweezers will accommodate. Send in the micro robots! Students must build tiny robots to sort marbles into pockets in a complex arena, but be small enough to fit in a matchbox.
Nominations
MoodMates
Commuicating subtle information between remote offices. The project has developed a mood-based availability service between remote colleagues. This is done through a soft-toy like interface that shows various light colours to show availability (Red=Busy; Green=Available) and showing various moods through emojis.
Nominations
Near-field Microwave Probes for Skin Cancer Detection
This project conducts a feasibility study on the utilization of near-field probes for skin cancer detection. It encompasses the exploration of various scenarios for single and multi-sensor applications in the context of tumor detection within a simulation environment. Additionally, the project delves into the interface used to acquire dielectric data through near-field probes, with a focus on enhancing its functionality.
Nominations
Orange Crocodile
Peer-to-peer resource sharing tool. This is a mobile phone app that allows households (in a shared student accomodation or locality) to share their resources (such as vacuum cleaners) amongst one another. The app lets individuals reserve tools from others and share their own. The gamification and visualization aspects of the app are really apealing.
Nominations
Precise surface following with Riemannian Motion Policies
At CERN, radiation surveys of equipment and beam lines are important for safety and analysis throughout the accelerator complex. Radiation measurements are highly dependent on the distance between the sensor and the radiation source. If this distance can be accurately established, the measurements can be used to better understand the radiation levels of the components and can be used for calibration purposes. Currently, in areas where no fixed infrastructure exists to perform these surveys, t...
Nominations
Project Sterling
Project Sterling is a novel digital currency concept that enables electronic payments and money exchange through modalities unavailable to traditional online payments, bank transfers or cryptocurrencies. This project includes a demonstration of the complete end-to-end payment transfer system using secure cryptographics.
Nominations
Reinforcement learning for active search and grasp
When humans need to find and retrieve hidden objects they are able to fluently reason between changing their viewpoint and moving objects within the space. For robots, this is a mechanical search problem. While we have tools for autonomously synthesizing grasps on cluttered scenes, as well as methods for actively improving scene reconstruction when the target object is only partially visible, the goal of this project is to extend these capabilities to situations where the target object is ful...
Nominations
RGB-D Camera based Auslan Dataset Curation
Sign language translation (SLT) aims to convert a continuous sign language video clip into a spoken language. Considering different geographic regions generally have their own native sign languages, it is valuable to establish corresponding SLT datasets to support related communication and research. Auslan, as a sign language specific to Australia, still lacks a dedicated large-scale dataset for SLT. To fill this gap, we curate an Australian Sign Language translation dataset. With a collectio...
Nominations
Room Occupancy with Radar Tracking
The project aims to develop a system that can accurately detect and track the number of occupants in a room using mmW radar technology. The system will be interfaced with different devices such as ESP32, STM32 Discovery, Jetson Nano/NX, or Linux/windows-based PCs. A system that uses machine learning algorithms will be developed to provide real-time information of indoor room occupancy. The system will also relay the information back to an online dashboard accessible via a web browser or mobil...
Nominations
Secure Chrome Extension for Password Improvement
This project creates a secure chrome extension that helps users to improve their passwords, while keep in easy to remember.
Nominations
Secure Communications for Networked Embedded Systems
This project investigates the tradeoffs in performance and power consumption for a range of network security protocols available for light-weight IOT devices.
Nominations
Sensing the world using EM signals: enhancing passive radar with higher BW signals
This project will look at using modern sensors like mmWave radar, and/or passive radar to enhance road infrastructure by providing a complete situational awareness.
Nominations
Space Buns
An asymmetric serious game that teaches high school students (aged 13-18) health and safety objectives in a fun and interesting manner.
Nominations
Sprout 🌿
This project addresses the brief on self-mental awareness through biofeedback. The ultimate objective is to design a wearable watch paired with a mobile application for patients with bipolar disorder. Our Arduino prototype has been equipped with biofeedback sensors to detect:
- Movement (via gyro/accelerometer)
- Sweat levels
- Steps (pedometer)
- Heart rate (note: the sensor is currently malfunctioning)
These sensors collectively identify shifts in the user's mood and potential negative...
Nominations
The Benefits of Non-Fungible Token (NFT) Technology in Music Copyright
The emergence of Non-Fungible Tokens (NFTs) has presented a novel opportunity to address long-standing challenges in the realm of music copyright. This research paper explores the current iterations of NFTs as a potential solution for music copyright in the digital age.
Nominations
The Challenges of Undergraduate Students Learning to Program
Learning to program is one of the first courses Engineering and Computing students face when starting their university degree. Many students find it challenging as they are new to study at university and have three more new subjects on the schedule. This project seeks to understand “What are the struggles and expectations of students learning to program for the first time at University?” With the aim of understanding what challenges and learning experiences students face, and to gain insight ...
Nominations
The Impact of Humanoid Avatars on Body Ownership within Individualised Virtual Reality: VR Applications in Mental Health Therapies
Individualised virtual reality (iVR) offers a user more control by affording users the ability to customise their experience leading to better outcomes. iVR has been shown to yield benefits in the application of self-compassion therapy, which has undergone several studies, revealing more about iVR therapy at each iteration. This latest project is a replication of a previous iVR study to determine if inexpensive avatars could be used as a cheaper alternative to financially, and often, computat...
Nominations
Tiny Giant Robot 2: Striking a Balance
The TGR combines the balancing capabilities of advanced dynamical bipeds with the affordability and simplicy of drones. The second generation TGR is superior in every way: smarter, faster, stronger, and better-dressed. Making TGR2 balance, however, required critical enabling advancements in control firmware, a breakthrough in light-weight gimbal design, and careful application of good old fashioned systematic control theory princples to make the system stand robustly.
Nominations
Transformer replacement priority tool
A software tool was developed to address the challenge of aging electrical transformers in power grids. This tool translates theoretical models and transformer failure data into actionable insights for replacement priority determination. Originally an offline version, it was later advanced with Python’s Django for real-time data updates. The software offers diverse analysis, including risk cost, enabling utilities to rationalize their replacement strategies. Through an intuitive interface, th...
Nominations
Using Embedded Machine Learning Edge Computing for Calorific Estimation Of Food
The purpose of this project is to develop a user-friendly device that can estimate the calories in food using machine learning techniques. By using an edge device equipped with a camera and depth sensor, the device will allow Australians to monitor their food consumption and make smarter dietary choices, which will reduce the likelihood of overeating and lower the rate of obesity in Australia. The device is intended to eventually be integrated into a mobile phone application for convenient an...
Nominations
VESSELBOOST: A Method Combining Data Augmentation and Test-time Adaptation to Improve Small Vessels in Human Brain Magnetic Resonance Angiograms
High-resolution magnetic resonance angiography (MRA) data can provide unique insights into vascular diseases such as small vessel disease, and help understand vascular contributions to neurodegenrative diseases. Segmenting brain vessels from MRA is crucial for quantifying and analysing the characteristics of human brain vasculature, but accurate segmentation is challenging due to the high number of small vessels contained in these images. VesselBoost addresses this fundamental issue by using ...
Nominations
Weather Station Rankings for Dynamically Rated Transmission Feeders
This thesis project created a weather station ranking system for Powerlink Queensland. Powerlink Queensland has 23 dynamically rated transmission feeders, where ratings are calculated based on real time weather data measurements sourced from BOM weather stations. Currently, Powerlink's method for allocating weather stations as data sources to feeders is inconsistent, unreviewed and subjective. To solve this problem, I designed a weather station ranking system that assists Powerlink line desig...
Nominations