Electrical, Computer, and Software Engineering

Project 

The Formula SAE competition is introducing an autonomous vehicle category to the control of the Formula SAE cars. To meet this challenge, we are seeking to develop a fully autonomous control system that enables the car to learn to drive itself through Reinforcement Learning. Reinforcement Learning is a machine learning approach that aims to enable a robot to learn to operate based on its interaction with the environment without human input or control.

Role

Prior work has started to develop a means of enabling a Formula SAE car to autonomously navigate through reinforcement learning-based control on a simulated F1Tenth race car. This project will extend this work to operate on a real-world F1Tenth (https://f1tenth.org/) race car to navigate a "race track" in the lab using additional sensing and control modalities. We will endeavour to learn autonomous racing and overtaking maneuvers against state-of-the-art methodologies.

Requirements

This project will require strong programming skills, specifically in Python, and will require the students to come frequently into the robotics lab to work on the project. This project is not suitable for remote development due to the requirement of working with and testing on the physical vehicle. Prior experience with Pytorch or ROS (1 or 2) would be beneficial but optional, as we will teach you those tools as part of the project.

Project 

Digital Educational Engineering (DEE) is about using an engineering approach to design, build, and evaluate a software-based solution that will address some education-based problem.

Role

This project will involve developing a software application or software tool, which can utilise a range of digital technologies. The particular project selected will be determined closer to the time, after meeting the allocated student, in order to carry out a project that they are technically confident in.

Requirements

To be successful in this project, you should be a strong programmer confident in using technologies such as HTML, CSS, JavaScript, React, Python, etc.

Project 

As wireless charging for electric vehicles evolve, there is a growing interest in improving the design of various parts of the system for manufacturability, as well as interoperability. Decoupled multi-coil pads have been proven to be a viable electronic solution to magnetic and power class interoperability.

Role

The supervisor has a new innovative design which needs to be built and tested. The new design promises improved magnetic interoperability, power class interoperability, and on paper, is more easily manufacturable compared to existing decoupled multi-coil designs.

Aim and requirements

The aim of this summer research scholarship is for the research student to understand and construct such a pad for further evaluation. The student should possess good electronics knowledge, and be willing to work with high power equipment (high voltage - 800V, high currents - 200 A). This project needs someone who primarily focuses on Electronics, and Magnetics, but is able to come into the lab at least 5 days a week to do practical work. The candidate will learn how to conduct robust tests and collect relevant data for publications. If all goes well there will be opportunities for publications.

Project 

In an increasingly mobile and interconnected world, the demand for efficient, convenient, and versatile power solutions has never been higher. Wireless charging, a technology that enables the transfer of electrical energy from a power source to a device without physical connectors, is at the forefront of meeting this demand. This innovative approach leverages principles of electromagnetic fields to provide power to a wide array of devices, from smartphones and wearables to medical implants and electric vehicles.

Wireless charging offers several advantages over traditional wired methods, including improved durability of devices, enhanced user convenience, and the potential to power devices in environments where traditional cabling is impractical or unsafe. However, the technology also presents challenges such as efficiency losses over distance, the necessity for precise alignment, and potential safety concerns regarding electromagnetic exposure.

Role

This research project aims to explore the development of novel magnetic materials for wireless charging, investigate current technological advancements, and evaluate their applications and limitations. Through a combination of theoretical analysis and practical experimentation, the project seeks to contribute to the growing body of knowledge in this field and to explore innovative solutions that could enhance the efficiency and applicability of wireless power transfer technologies.

Project 

With the introduction of LED lighting, a new opportunity for control is available. This control allows a near-linear response of the luminous flux output versus electrical power. This feature enables the efficient use of multiple control functions.

Role

In this project, the development of the energy model will be undertaken. The model will provide a basis for quantifying energy consumption under various conditions (weather, occupancy, peak load, etc.). The model will utilize the latest edition of Dialux Evo software.

Requirements

Applicants should have knowledge of lighting modeling, electronics, and 3D imaging.

Project 

This project will develop a mobile robot navigation for indoor environment, i.e. university and hospital. We have multiple mobile robots (Turtlebot, Husky, Silbot, Pepper and GoCart) for this project. We will use a LiDAR sensor for map building and localization.

Role

This project will integrate the currently working prototype modules, i.e. speech understanding and generation, face detection, etc. You will work with other researchers who have done this project.

This project aims to:
– Develop a ROS based navigation SW
– Develop a map building SW
– Develop a localization SW
– Integrate with different AI modules, i.e. face detection, if needed
– Evaluate its performance

Project scope will be decided after the meeting with supervisors.

Ideal student

You will bring to the role a passion for research and engineering, excellent computing skills (including a high level of programming ability), and a strong sense of responsibility.

Project 

This project will develop an interactive chatbot system for social robot that talks with visitors at the university reception. We have a working version using DialogFlow, and will develop a better chatbot. You may use reinforcement learning and/or Deep Neural Network (DNN) if needed. So when the text is given, chatbot generates its reactive speech by considering history of conversation. This project is related to ongoing 5 years research project, SHRI, and will work with PhD and professional staff. All necessary devices will be provided.

This project aims to:

– Do a literature survey on interactive conversation of human
– Find the way to consider relation and history of conversation
– Develop a chatbot
– Apply it to one of robot platform if possible

Project scope will be decided after the meeting with supervisors.

Requirements

You will bring to the role a passion for research and engineering, excellent computing skills (including a high level of programming ability), and a strong sense of responsibility.

Project 

Artificial Intelligence (AI) piracy is the norm. The source of data (could be text, speech, images) used to train AI models are not traceable once the model has learned from the data. Currently, there are limited approaches to ensure ethical use, ownership of data and distinction of human-made vs artificially generated data. A means to ensure traceability of data is what this project proposes.

AI and more specifically Generative AI, has sparked a revolution in human creativity and knowledge. Ask any data scientist and they will tell you that the engine powering this new age is one word - data! However, an undesired impact of this AI revolution is that data ownership and privacy have become a growing concern over the last decade.

Speech and image data are central in many AI applications these days. However, speech and images are unique resources (taonga) that belong to individuals and are bound by the data sovereignty principles of the community from which the data is collected.

Role

This project aims to explore options for traceability of data (specifically speech and image data) that is used in AI applications.

The project will comprise:

1. A literature review on options to ensure data traceability and potential issues with the options.
2. Development of a  a proof of concept of one approach to ensure data traceability.

Requirements

An understanding of the role of data in AI, how models are trained, and programming (Python) knowledge are sought after for this project.

Project 

Rapid advancements in Machine Learning (ML) have resulted in widespread deployment of ML solutions in several areas. Recent advances in semiconductor device technology and hardware architectures, data processing, and computing are shifting ML solutions towards the edge, close to data sources.

Deploying state-of-the-art ML algorithms requires high-performance yet low-power hardware architectures. Heterogeneous computing platforms are introduced to optimise performance and energy efficiency.

Role

This project aims to investigate performance and energy-efficiency advantages provided by AMD AI engine technology and develop customised ML hardware accelerators on heterogeneous Versal platforms to improve both energy consumption and performance.

Requirements

This project requires a strong background in digital system design and computer system architecture. Competency with Electronic Design Automation (EDA) tools is required. Applicants should be willing to learn to work with new design tools from AMD.

Project 

Software testing is a costly yet crucial task in software engineering, ensuring confidence in the expected behaviour of implemented software. A test oracle is a mechanism to distinguish between correct and incorrect executions and it is fundamental to expose bugs during testing. An example of test oracle are assertions that predicate on the values returned by the method or class under test.

Aim

The emergence of Large Language Models (LLMs) like ChatGPT has prompted researchers to explore leveraging these models to enhance the effectiveness of automated testing. The goal of this research project is to design and explore novel methods for generating test oracles using LLMs.

Requirement

Proficiency in Java is essential for this project.

Project 

Machine Learning Testing is a recent research topic aiming at adapting software testing concepts to detect errors in ML-based software. One of the most challenging concepts to adapt is test oracle. In particular, it is challenging to automate the generation of a test oracle that can assess the quality of ML predictions beyond just passing all test cases.

The oracle problem in machine learning refers to the challenge of defining a ground truth for evaluating the predictions made by a machine learning model. In traditional software testing, an oracle can be an expected output that is used to validate the correctness of a program's behavior. However, in machine learning, the prediction accuracy based solely on a set of labeled data may not be a reliable indicator of model performance on unseen data. This is because machine learning models often make predictions based on patterns learned from the training data, which may not generalize to new examples.

Aim

This research project aims to design and develop an automated method for defining "trustworthy" test oracles. Highly motivated students will explore how to use such oracles to automatically improve a ML model.

Ideal student

This project is suitable for students who have proficiency in the Python programming language.

Project 

This project explores the ideal supercapacitor size for a solar-powered toy car operating under partial shading conditions. By testing various supercapacitor sizes, the project aims to identify the most efficient configuration for consistent performance despite varying sunlight levels.

The findings will provide insights into the effectiveness of supercapacitors in small-scale solar applications, particularly when sunlight is obstructed.

Role

This hands-on investigation will include measuring charging times, voltage, and the car's travel distance and time under different lighting conditions to determine the optimal supercapacitor size for real-world scenarios.

Project 

When an algorithm is processing some data, e.g. numbers are sorted, an image is filtered or a neural network is trained, the foremost objective has been to do this as fast as possible. For decades new processors have become faster, by executing more operations per second, and new algorithms have been more efficient with shorter runtimes.

Today, however, computers, from small devices to large data centres and supercomputers, consume a significant and growing amount of electrical energy. In order to achieve more sustainable computing, we need to reduce the energy consumption of computation. While this is being addressed on a technological level, with smaller processors, memories and networks, the algorithmic side receives little attention.

Role

In this project you will investigate and compare computing tasks, for example, processing a batch of images or training a neural network, in terms of the energy that is needed to complete the computation. For many problems there exist alternative approaches, only think of the many ways numbers can be sorted.

We want to find out which of these alternatives for a given problem needs less energy to obtain the result. This will be done by studying the literature on the topic, identifying computational problems and algorithms with alternative approaches and then measuring and comparing their energy consumption of a given lab computer of the PARC (Parallel and Reconfigurable Computing) lab.

Energy consumption can vary not only by how long it takes to compute, but by how many processors are involved, how much data is transferred and how often processors are idle. We aim to characterise patterns and approaches that can be used to identify the best techniques to be used from areas like data science and machine learning which have strong saving potential and a growing appetite for computing power.