Mechanical Engineering

Constitutive behaviour of polymeric foams for high performance sandwich structures

Supervisor

Assoc. Prof. Mark Battley, Dr Tom Allen
Faculty of Engineering
Project code: ENG070

Composite sandwich structures consist of two outer layers, or facesheets, surrounding a central low-density core, and are used in a variety of applications in marine, aviation, automotive and sporting industries. The foams increase the flexural stiffness and strength of the sandwich composite without substantial gain in mass, allowing them to be an ideal solution as lightweight strong materials. In many situations these are dynamically loaded, including the hulls of boats hitting waves or land and air vehicle crashes. The foam core materials used can have quite different failure modes, ranging from brittle to ductile depending on their formulation and how quickly they are loaded.

This project aims to develop and validate mathematical constitutive models to predict the failure process for typical structural polymeric foam core materials such as Polyethylene Terephthalate (PET), Polyvinyl Chloride (PVC) and Styrene-Acrylonitrile (SAN). Physical testing of core materials will be undertaken with Digital Image Correlation based full-field strain measurement at quasi-static and high-rate loadings. Existing foam constitutive models will be reviewed, and relevant models implemented in non-linear and dynamic Abaqus finite element models. The accuracy of the models will be determined relative to the experimental data, and additional refinements and extensions developed, and implemented as user defined UMAT material models. The new constitutive models will be used to predict the onset and propagation of damage for water and hard-object impact of sandwich panels.

Collaborative manipulation via haptic communication

Supervisor

Prof. Peter Xu
Faculty of Engineering
Project code: ENG071

Due to advancements in technology and ageing populations, robots are envisioned to play a greater role in both industry and society. We can expect greater degrees of physical interaction between people and robots. To ensure safety and efficiency in interaction cases we must have a sound understanding of how humans interact via haptic (physical) communication. Current literature has implemented robotic partners for task completion with a human. However, none have created a robot that increases performance over a human-human pair. Due to the increasing use of robots expected to interact safely and efficiently with humans, research into optimised robotic performance is necessary for effective human-robot interaction.

The objectives identified for this project are: (1) assessing the changes in performance of a human-performed task when a robot helper is utilised; (2) assessing the efficiency of utilising robots in the task completion and (3) potentially using machine learning to instruct robot behaviour.  

Speech enhancement on audio recording using UAVs

Supervisor

Dr Yusuke Hioka
Faculty of Engineering
Project code: ENG072

Unmanned aerial vehicles (UAVs) have recently gained huge popularity across a wide range of applications, including filming, search and rescue and surveillance. Such applications take advantage of capturing visual information (i.e. video and imagery) that are otherwise impossible without making use of UAVs. On the other hand, audio signals are also one that should not be overlooked. It is common to encounter environments that are often remote and harsh, which can easily render visual information unusable. This is not the case with audio. However, audio recording using UAVs have shown to be challenging due to the high noise levels radiated from the UAV rotors. This significantly affects the quality of the audio signals to aid with any application.

This problem was approached by the Acoustics Research Centre (ARC), University of Auckland, for which a UAV system, equipped with an array of microphones and a signal processing algorithm, was developed to effectively record desired audio in-flight while removing the UAV rotor noise. Recently, a method based on machine learning was used to explore possibilities of predicting UAV rotor noise with a hybrid of microphone and non-acoustical information. However, a common problem with such data-driven system is the lack of transparency between the inputs and the result it produces. To this end, studies has been made to unravel these ambiguities with the help of analytical modelling.

This project will focus on incorporating these analytical findings to optimize the current signal processing algorithm. This involves developing a suitable machine learning system, such as exploring different neural network architectures that best suit the analytical inputs. If necessary, experimental testing would also be carried out to obtain required parameters.

Suitable candidates should have some fundamental knowledge on digital signal processing with reasonable familiarity with MATLAB programming. Experience with neural networks and/or Python programming is a plus.

Noise source localisation using a microphone installed on a rotating disc

Supervisor

Dr Yusuke Hioka
Faculty of Engineering
Project code: ENG073

Sound source localisation using audio signals has been a key research area in signal processing because of its various applications. One of the applications is localising noise sources. Microphone arrays (array of more than one microphones) have been commonly utilised to achieve the purpose, however, reducing the number of microphones employed in the system has been a common interest of practical engineers. Previous students investigated feasibility of existing algorithm by building an experimental rig as a proof of concept. This project further explores the practical performance of the algorithm by focuses on collecting data in various acoustic environments.

Successful candidate should have a background in mechatronics or electronics/computer engineering and have experience in digital signal processing (e.g. completing MECHENG370 and MECHENG705, or ELECTENG733) and programming skills using Matlab and embedded C.

Dynamic simulation of a refrigeration system with an expander

Supervisor

Dr Alison Subiantoro
Faculty of Engineering
Project code: ENG074

One of the most promising methods to improve the energy efficiency of vapour compression refrigeration systems is by recuperating the expansion work with an expander. However, this modification is expected to alter the cycle dynamics. In this project, a dynamic simulation model of a refrigeration system with an expander is to be developed.

A literature review is to be carried out to investigate the established and the state-of-the-art simulation approaches. A new simulation model is to be proposed, developed and used to study the dynamics of a refrigeration cycle with an expander.

The candidate should have strong thermodynamics background and interest.  

Interactive engine assembly with augmented reality technologies

Supervisor

Prof. Xun Xu, Dr Yuqian Lu
Faculty of Engineering
Project code: ENG075

This project aims at using Augmented Reality technologies to assist factory workers with assembling complex products by providing on-demand product assembly instructions. The student will develop an AR product assembly environment using Microsoft HoloLens, CAD engine and image processing technologies.

Ocean cleaning robot

Supervisor

Dr Minas Liarokapis
Faculty of Engineering
Project code: ENG076

This project will focus on the development of an ocean cleaning robot boat that will be equipped with a reconfigurable gripping net that will allow the collection of plastic bottles and other trash from oceans, in an autonomous manner.  

Robotic airship for human robot interaction

Supervisor

Dr Minas Liarokapis
Faculty of Engineering
Project code: ENG077

Miniature indoor robotic airship platforms have great potential in indoor exploration and human robot interaction applications, offering high mobility, safety and extended flight times. This project will focus on the feasibility, design, development and evaluation of such a platform. Selected commercially available envelope materials will be considered and tested in terms of their helium retention capability. The obtained envelope properties will be used in a feasibility study, demonstrating that indoor airships are environmentally and financially viable, given an appropriate material choice. The platform’s mechanical design will be studied in terms of gondola placement and rotor angle positioning, resulting in unconventional gondola arrangements. The system will be tested in a simple, path following experiment for proof-of-concept purposes, proving its efficiency in attaining the desired heading and altitude configuration.

LES modelling of turbulent natural convection

Supervisor

Dr Stuart Norris
Faculty of Engineering
Project code: ENG078

Turbulence is problematic to model, and the addition of heat transfer and buoyancy compound the issue. In this project you will model buoyant turbulent flows using Large Eddy Simulation with a parallel research code on a parallel computer. The results will be compared against experimental data.

Implementing inerters using shunted piezoelectric materials

Supervisor

Dr Vladislav Sorokin
Faculty of Engineering
Project code: ENG079

An inerter is a two-terminal mechanical device which applies force between its terminals proportional to the relative acceleration between them, similar to a spring and damper which apply forces proportional to the relative displacement for the spring and velocity for the damper between their terminals. Inerters have become a hot topic in recent years especially in vehicle, train and building suspension system. The main difficulty, however, is in practical implementation and manufacturing of such devices.

The present project is concerned with trying to implement an inerter using shunted piezoelectric materials. The idea is to use electro-mechanical coupling inherently present in these materials. The project implies theoretical study and experimental testing of the device. 

Mechanical performance of materials to be using in in-road electric vehicle charging systems

Supervisor

Dr Tom Allen
Faculty of Engineering
Project code: ENG080

This project will look to establish the mechanical performance of materials being investigated for application in in-road charging pads. Inclusion of charging systems within roadways is set to extend range and increase uptake of EVs. During this project, you will work to experimentally establish the mechanical and thermo-mechanical performance of materials for use within charging pads.  

Development of an STEM toy assessment framework

Supervisor

Dr Tom Allen
Faculty of Engineering
Project code: ENG081

While targeted efforts towards high schools may increase the enrolment numbers of underrepresented groups within engineering (specifically female, Māori and Pacific students), the career path of many prospective students have been heavily influenced prior to high school. Primary school (particularly ages 5-11) is where the foundation for an interest in engineering and the subjects necessary for pursuing a degree are laid. This project will continue the development of a framework for assessing the “STEM value” in toys for this age range as a tool to assist whānau and friends in selecting toys.

Develop an automated guided vehicle that can think like humans

Supervisor

Dr Yuqian Lu
Faculty of Engineering
Project code: ENG082

Factories of the future require autonomous AGVs that can transport goods autonomously without fixed routes. This project aims at developing smart AGVs using driverless vehicle technologies. Students will learn to use state-of-the-art driverless vehicle algorithms and hardware.

It is preferable that the candidate has some practical programming experience with Arduino and/or ROS. It is strongly recommended that the students can understand Mandarin as some technical docs are in Chinese.

Create your factory chatbot

Supervisor

Dr Yuqian Lu
Faculty of Engineering
Project code: ENG083

Artificial intelligence and big data analytics have created lots of opportunities for the future world of work. One scenario is that you can have an interactive robot working alongside with you, taking all the hassles away from you.

This project aims at developing a chatbot that you can assign challenging tasks via verbal communications. For example, you could ask the robot to write a project report for you every Friday.

As a student, you will be able to learn the latest Natural Language Processing and chatbot development techniques. It is preferable that the student has strong programming skills.

Detecting failures in mechanical power transmissions systems

Supervisor

Dr Jaspreet Dhupia
Faculty of Engineering
Project code: ENG084

Acquire vibration measurement from a motor-gearbox-generator set-up available in the Vibrations lab and process it to detect failures such as broken gears and faulty bearings in a mechanical power transmission.

Machine learning for analysis of brain recordings during anaesthesia

Supervisor

Dr Luke Hallum, Xavier Vrijdag
Faculty of Engineering
Project code: ENG085

The electroencephalogram (EEG) is a way of painlessly, noninvasively measuring human brain activity with electrodes adhered to the scalp. EEG has a range of applications in biomedical engineering, including the development of medical devices. The use of convolutional neural networks (CNNs) is a recent development in the field of machine learning (ML), with promising applications in both engineering and medicine. One potential application is this: CNNs may better enable medical devices to detect changes in a patient’s EEG recordings during surgical anaesthesia. Using ML and CNNs, we aim to improve device sensitivity to changes in level of anesthesia, specifically nitrous oxide anesthesia. This interdisciplinary project is a collaboration between Engineering and the School of Medicine.

We are looking for a student with an interest in biomedical engineering, who has programming and/or signal analysis capability, and who is eager to learn more advanced signal analysis methods.