Postgraduate Research Projects

Real-time road vision: Optimization and Implementation on FPGA.

Safety for traffic participants is a major concern for car makers and policy makers. Many technologies have been employed to improve the safety of drivers as well as pedestrians. Vision-based safety features
have been investigated for many years. Despite many progresses in this area have been made, many of the ideas are yet to be implemented in practice. This research will investigate in how to speed up the implementation of hardware (FPGA) real-time vision-based systems onto automotive vehicles. Even though this project focuses more on the implementation side of the system, novel approaches toward improving the system's performance are also undertaken.

Wireless Control of Next Generation Powerfold Actuators

The focus of this research will be on actuator mechanisms for mechatronic systems used in automotive engineering, their selection, control and application of wireless technology.
The major interest of this research is the area of actuators and their control systems. It appears the brushless, brush and ultrasonic motors are the best choice. Another research idea is to try to apply wireless technology in order to control actuators, carrying information from the transmitter to the receiver and finally transfer of energy from the central power supply to the motor.
This research will be conducted using methods of experimental work, including computer simulation and all the results will find a strong application in particular industry projects.

Building a real-time augmented map for road risk assessment

Decentralised information systems based on inter-vehicular communications have drawn increasing attention in recent years. In that context this research aims at defining a real-time crash risk assessment system by building an augmented map combining different information (full or partial) from local maps in order to obtain a complete, reliable, accurate view of the driving environment. Information is exchanged via vehicle-to-vehicle and vehicle-to-infrastructure communication systems. Existing data fusion techniques are numerous but of variable reliability and accuracy. This research will identify existing techniques adapted to our specific research problem and propose optimisations to these techniques. The cost benefits of our research in terms of its ability to identify risks in real-time will be assessed and validated. Specifically we will evaluate the risk assessment’s quality and complexity versus the current driving context. Additionally this research will also introduce a novel approach into data fusion techniques: auto-diagnostic. Auto-diagnostic implies a system able to determine its state and check for the presence of failures among its information sources, whether local or networked, or within its own software. Such a capability would allow real-time online assessment of a system’s reliability. With these prospects auto-diagnostic could become critical to future real-time automotive safety applications.

Development of gradient index optical coatings by plasma polymerization

Materials such as polycarbonate have the advantage of being both lighter and more shatter resistant than glass. These superior properties could increase automotive efficiency, through weight reduction, and provide added safety to occupants. Furthermore, production methods of polycarbonate are more versatile, where techniques such as injection moulding, will allow manufacturers to produce more sophisticated designs relatively inexpensively in comparison to glass. However due to the polymeric nature of polycarbonate it can be easily scratched which is a major disadvantage. In addition when non index matched protective coatings are applied to polycarbonate they give rise to interference fringes which would prohibit polycarbonate from ever replacing glass in automotive applications. Therefore this project aims to address this issue by creating a protective film that has a tailorable refractive index. Such an approach would eliminate cosmetically unacceptable interference fringes, whilst still maintaining the structural integrity of the substrate / coating system

Carpet acoustics

The study and prediction of sound insulation and absorption of various materials, which includes carpets, trims, heavy sound isolating layers and films, has always been of great significance and interest for acoustic consultants, involved in design of sound noise insulations packages, especially for vehicle acoustic package designs.

Statistical Energy Analysis is one of the recommended approaches to predict such a complex acoustical system. This research hereby focuses on developing an acoustically improved vehicle carpet system by mainly deriving a theory, based on wave equations, applicable for 1) Measurement of various acoustic properties of materials, in such a way that they can be used for Statistical Energy Analysis modelling 2) Evaluation of the Statistical Energy Analysis model of the acoustical system, and get the theoretical reasons behind the differences observed between the model and real system measurements.

Nanostructured pigments for Automotive Industry.

Description:-Automotive industry is facing a challenge to reduce its impact on climate change by minimising fuel consumption. Fuel is consumed not only to run automobiles but also to run air conditioners to cool down the heat builds up. This heat build up is due to absorption of near infra-red radiation (NIR) of solar spectrum by coloured pigments used in exterior coating as well as in the interior polymeric
components. This project is aimed at the development of NIR reflective pigments and their formulation to enhance the NIR reflectance and emissivity leading to reduction in the heat builds up in coatings,
polymers and other material used in the automotive industry.

Developing Lightweight Vehicle Structures

The unique properties of aluminium and metallic foams promise a variety of applications in vehicle design one of which includes the absorption of impact energy, and indeed such structures are now being used in practice. However, trends in increasing usage of aluminium and metallic foams as light vehicle safety components require a thorough understanding of the properties, especially mechanical properties. In this project, systematic investigations will be made to explore structural components made of aluminium with metallic foams. In particular, enhancement in crashworthiness performance under static and dynamic lateral loadings through the use of such structures will be looked at.

Speech Enhancement for Robust Speech Recognition

Previous work in speech enhancement has looked at the problem from the angle of intelligibility and quality which are human measures of enhancement performance. This research looks at a popular single-channel speech enhancement technique from the perspective of generating better speech recognition performance from a statistical-based classifier (machine) in an automotive environment.
This has led to two distinct pathways for performance improvement - online optimisation of enhancement parameters and the use of the phase spectrum through speech (or noise) phase estimation.

Recycling of End of Life Vehicle waste 

At present there is an urgent need to recycle vehicles that have reached their end of life in Australia as there are estimates that over 500,000 of the 12.5 million vehicles on Australian roads reach their end of life each year.  This research will study the "Auto Shredder Residue (ASR)" which is the by product of shredded ELVs and discover various environmental and economically viable processes to recover useful materials from ASR waste.

Ontology-Based Integration of Design Environments

The research undertaken by UniSA investigates in practical applications of ontologies in design environments. Based on the context of multidisciplinary optimisation (MDO) in the automotive industry, we show that semantic integration of design processes and artefacts can be achieved by employing a formal ontology framework. By utilising established engineering standards such as ISO 10303, domain specific abstractions are specified, and, based on these abstractions, concrete process improvements such as simulation reuse and increased virtualisation of the development process are possible. Examples for
pivotal elements of this research are the creation of an ubiquitous language used throughout the design process - e.g. EXPRESS is utilised in our research for this purpose - annotating design artefacts with
aspects relevant for later reuse, and, code generation, using a model from a particular design domain as input and generating a subsequent model as output for a target environment.

Factors effecting attention whilst driving

This study is investigating how different characteristics of a person may influence their ability to maintain attention and avoid distraction whilst driving. Factors such as age, driving experience, perceptual style and personality are being investigated during real and simulated driving conditions. Participants will also be cognitively loaded with distraction tasks such as using a mobile phone or being involved in a simulated conversation. Driving performance measures and physiological variables will be used to quantify one’s ability to maintain attention and deal with such secondary tasks whilst driving. Results will be useful for studying driver behavior in the future.

Modelling of Fatigue Cracking in MIG Welded Automotive Steel Structures

The use of high strength steels in automotive components is steadily increasing as automotive designers use modern steel grades to improve structural performance, reduce vehicle weight and enhance crash performance. The effective use of steel in automotive structures is essential in order to maintain its competitive position against alternative materials such as aluminium and composites.

The focus of this research project is to model fatigue cracking in MIG welded Automotive steel structures with emphasis on using simple experimental test results in a numerical model capable of accurately predicting fatigue cracking for true crack growth in the slow crack growth region (as per Automotive Industry standards) for more complex structures that are uneconomical and time consuming to experimentally test.

Dedicated short range communication models, algorithms, and protocols for cooperative collision warning systems

One of the promising safety technologies for road vehicle is cooperative collision warning system that incorporates the emerging wireless technology such as wireless ad hoc networks and satellite navigation systems. A vehicle equipped with cooperative collision warning system will provide early warning to the driver to prevent accident. The system is still under active research and has not been mature enough to be deployed widely.
This research strives to improve cooperative collision warning systems, specifically addressing channel congestion and threat assessment problems by providing more accurate traffic models, prediction algorithms and efficient application-level communication protocols.

Evolution and optimisation within the Australian context

This research optimises organisational operating systems unique to the Australian context based on “the Toyota way”.
Toyotas founding dogma of jojo (gradual and steady) is evolving to embrace kakushin (revolutionary change or radical innovation).The 2008 Bracks report precurses an Australian paradigm shift.
As global supply chains grow massive, highly integrated and mutually interdependent, Australian suppliers may become financially stressed, with diminished R&D investment, IP stronghold and market attractiveness.
Research outcomes will overcome Australian value chain challenges, by pragmatically leveraging both emerging and previously ideologically antagonistic methodologies, to address: Scale, Fragmentation, “Lumpy” launches, Discontinuities, Regional investment competition and Exchange movement (and capital investment).

A dynamic model for automotive eco-design in Australia

The research will produce a dynamic model that can be used as tool for Eco-Design in the Australian context. The model will enable automotive companies to assess the effect of changes in critical variables on design and manufacturing. The critical variables include energy supply, resource availability, market demand (and public perception) for environmentally friendly vehicles and vehicle performance. The project will have a strong 'materials technology' focus and, along with the design outcomes, will enable sustainability through increased recycling and reuse, improved energy efficiency and reduced emissions. Although focussing on the Australian context, the originality of the dynamic model will have international significance and be adaptable to design problems in other areas of the world.

Development of novel crash structures

The recent focus of automobile manufacturers towards the manufacture of light-weight components for achieving good fuel economy has opened up great challenges towards designing light-weight yet crashworthy energy absorption structures. Composites have great potential as a light-weight energy absorbing material when compared to metals. The focus of this research is to understand in depth the energy absorption characteristics of composites for different impact conditions, geometry, material parameters and crushing speeds. This will aid in designing an efficient crash structure by implementing the results obtained from the physical tests into commercial finite element analysis package.

Developing Numerical Model for Flow-Induced Vibration Excitation Mechanisms towards Virtual Laboratory Simulation

Flow induced structure vibration features in numberless industrial systems, especially in Automotive engineering application. Unfortunately, due to the complexity of its physical phenomenon, numerical model which is capable to resolve the non-linear interaction between fluid and solid parts are still limited within literature. Knowledge of excitation mechanism of flow-induced structural vibration still remains elusive. This research aims for developing numerical tool for studying Flow-induced Vibration mechanism. The results and knowledge obtained from this current work would help to gain more fundamental insight of flow-induced structural vibration problems which could certainly be applied in future automotive designs and hopefully lead to successes of Australian automotive industry.

Reliable and Secure Wireless Ad-Hoc Network for Vehicle-to-Vehicle Communication

This research aims to improve reliability of vehicle-to-vehicle (V2V) communication based on Dedicated Short Range Communication (DSRC) technology. V2V communication has unique characteristics like adverse communication environment, dynamic changes in network topology, high node mobility, delay constraints, absence of centralised administrative entity and frequent network disconnections. This research will focus on the development of a novel adaptive routing protocol to achieve reliable and delay tolerant networking among vehicles, addressing broadcast storm problems.

Tool wear prediction model

Advanced high strength steels (AHSS) are increasingly used in sheet metal stamping in the automotive industry. In comparison with conventional steels, AHSS stampings produce higher contact pressures at
the interface between draw die and sheet metal blank, resulting in more severe wear conditions, particularly at the draw die radius. Developing the ability to accurately predict and reduce the potential tool wear during tool design stage is vital for shortening lead times and decreasing production cost. This project investigates the influence of draw die geometry on the wear distribution over the draw die radius and presents a methodology for optimising the draw die geometry to reduce wear using numerical and experimental methods.