Indicative PhD project suggestions
The indicative project information presented on this page is intended to help prospective PhD applicants who are self funded to start a dialogue with our academic staff in an area of research that interests them. During the application process options for part funding can be explored, dependant on a number of factors, but the inclusion of an indicative project in this list does not guarantee that funding will be available.
Click on a project to see more information..
Machine hearing for industrial fault diagnosis - Dr Eve Zhang and Dr Dan O'Boy Ref: YZAAE14
Industrial engineers are naturally tuned to listen to different healthy and faulty conditions during machine operation. This emerges often in real-world engineering testing practice, e.g. for rail defect inspection or automotive engine examination. With substantial development of sensor and computing technologies nowadays, acoustic signals, along with many other sensor measurements, can be utilized effectively for industrial fault diagnosis applications.
This project aims to develop an automatic acoustic fault diagnosis system that can mimic the industrial engineers’ “listening and diagnostic” capability. As demonstrated from literature, human auditory system possesses exceptional capability to extract and simplify characteristic information of audio signals. To model this functionality, the proposed machine hearing methodology will include several stages. First, the information extraction method of the human cochlea based on nonlinear processing mechanism will be analyzed, to achieve robust extraction of the acoustic features. Then, in order to simulate human auditory central nervous system to effectively synthesize all time-frequency information output from the “artificial cochlea”, a deep artificial neural network model will be developed based on real human auditory signal processing framework. Temporal memory elements will also be included in the deep neural network modelling. The developed technique will be validated using acoustic data collected from microphones, mobile phone and/or acoustic emission sensors for fault diagnosis experiments through composite structure tests and/or automobile engine tests.
Through the PhD project, based on the characteristics of human auditory system, a machine hearing technique, i.e. a bio-inspired “artificial ear” neural network model will be designed, to listen to and classify different types of industrial faults. The outcome of this research is the establishment of an intelligent acoustic diagnosis system using the new machine hearing technique, which then can be widely applied to various industrial sectors, where industrial engineers can listen to the machine to diagnose faults, e.g. for aeronautical, automotive, railway and manufacturing applications.
Proposed by: Dr Eve Zhang Primary Supervisor
Secondary Supervisor: Dr Dan O'Boy
Project ref: YZAAE14
Smart Wheelchair for the Elderly/the Disabled - Dr Jingjing Jiang and Dr Cunjia Liu Ref: JJAAE13
Smart wheelchairs are able to increase mobility and independency for the elderly and the disabled with cognitive impairment and memory issues by providing navigation support. However, the current navigation technologies implemented in smart wheelchair need to be customized to each consumer, which makes it expensive and difficult to expand to a business.
This project aims to develop an assistive navigation algorithm, which can be easily adapted to various end-users, together with a human interface for smart wheelchairs, based on sensor information from Sonar, Infrared Range Finder, 3D LiDAR, camera, etc. This is a multi-discipline project, involving programming, control design, AI technique implementation, sensor fusion and robotics.
Proposed by: Dr Jingjing Jiang Primary Supervisor
Secondary Supervisor: Dr Cunjia Liu
Project ref: JJAAE13
Autonomous navigation, guidance and control of an agricultural vehicle in challenging environments - Dr Matthew Coombes and Dr Cunjia Liu, Ref: MCAAE12
While autonomous vehicles in urban environment have received a lot of attention and rapid development in recent years, there is a large number of open challenges for autonomous driving in off-road environment. One representative yet high impact problem is the agriculture robots for smart farming.
One of the most basic requirements of an agricultural robot is to navigate and guide itself around complex farm environments. Farms are dynamic environments, often with muddy uneven terrain and unexpected situations. With recent advances in learning-based control, this project aims to co-design autonomous perception and navigation functions that will enable a ground robot to guide itself around a farm through crop rows, while avoiding objects such as livestock and ditches.
In addition, the robot will need to traverse muddy water-logged areas without getting stuck and have a means by which to judge if the area is actually traversable. Advanced control algorithms need to be developed which can use sensors on the robot and its drive system move the vehicle through this challenging terrain.
This project will have strong practical elements and will be based at the Autonomous Systems Laboratory, where we have state-of-the-art robot platforms and high-end sensors such as 32-bit Velodyne Lidar, multimodal cameras, millimetre-wave radar, etc. This exciting project will develop skills in robotics, navigation, ground vehicle control, programming and practical skills related to operating autonomous robots.
Proposed by: Dr Matthew Coombes Primary Supervisor
Secondary Supervisor: Dr Cunjia Liu
Project ref: MCAAE12
Lightweight damage tolerant composite enclosure structure for electromagnetic interference shielding - Dr Gang Zhou, Ref: GZAAE001
Conventional solid enclosure structures for electronic equipment that shield radar, radio signals and radiations are made of metallic materials. They are not only heavy but also limited in their structural performance. Although composite sandwich structures could offer a lightweight better-performing alternative, conventional composite materials are limited in their electrical conductivities. This industry-funded project is primarily intended to develop a down-scaled prototype box structure, which combines unique composite materials with improved electrical conductivity with structural design concepts, to demonstrate its EMI shielding capability. It must also demonstrate that such composite enclosure structure is damage-tolerant, excellent in corrosion resistance and fatigue life in addition to being cost-effective.
Proposed by: Dr Gang Zhou
Start up strategies for battery electric vehicles an aircraft in cold climates -Dr Ashley Fly, Ref: AFAAE003
The battery electric vehicle (BEV) has the potential to decarbonise the transport sector through the use of renewable electricity and zero tailpipe emissions, however limited vehicle range remains a significant issue. This is especially true in cold climates where batteries become less efficient and available capacity is reduced. Furthermore, conventional lithium-ion batteries cannot charge at low temperatures due to the risk of lithium plating and short circuiting. This project will research methods of battery operation strategies to maximise the range available in a BEV when starting at sub-zero temperatures. This may include combinations of self-heating through discharge, external heaters and additional energy storage devices. The methodology could be equally applicable to electric road vehicles or high-altitude unmanned aircraft.
Proposed by: Dr Ashley Fly
Lifetime simulation of lithium-ion batteries for electric vehicles -Dr Ashley Fly, Prof. Lisa Jackson, Dr Sarah Dunnett, Ref: AFAAE004
A limitation of lithium-ion batteries used in electric vehicles, smartphones and other devices is that performance degrades with time and in use. Whilst the different factors that contribute to the degradation of a battery have been identified, their interaction and contribution to the overall performance reduction in use is still not well understood.
This project will seek to address these unknowns through the development of a battery degradation model suitable for lifetime predictions of battery applications; the model will then be validated through comprehensive experimental testing and characterisation of lithium-ion cells. Once completed this model will be able to further the understanding of how to operate batteries for extended lifetimes, inform future cell design to minimise particular degradations mechanisms and facilitate the development of accelerated testing strategies.
Proposed by: Dr Ashley Fly, Prof. Lisa Jackson, Dr Sarah Dunnett
Reliability analysis of platoons -Dr Sarah Dunnett, Prof. Lisa Jackson, Ref: SDAAE005
Vehicle platoons, or convoys, where the movements of a group of vehicles is coordinated, has been investigated for several years due to their advantages such as a reduction in fuel consumption. Other potential improvements include road throughput and road safety, however these are not still fully understood as there has been limited work in these areas.
The aim of this project is to investigate the safety and reliability of a variety of different platoon configurations and strategies. With the potential evolution of fully autonomous vehicles the concept of using these in convoys will also be considered. It is anticipated that a variety of reliability techniques will be used and adapted to model the various scenarios and to evaluate their effectiveness to meet platooning objectives. This will involve adopting a ‘system of systems’ safety analysis approach and simulating various platoon configurations.
proposed by: Dr Sarah Dunnett, Prof. Lisa Jackson
Autonomous searching for chemical release sources using multiple robots - Dr Cunjia Liu, Ref: CLAAE007
Finding the location and strength of unknown hazardous releases and predicting their dispersions are of paramount importance in emergency response and environmental monitoring. Examples include responding to events such as volcanic eruptions, nuclear power accidents or chemical, biological or radiological (CBR) accidents or attacks, and even exploring methane emissions on the planet Mars.
This project aims to develop an integration solution using multiple mobile robots to search the area and collect the data and using Bayesian framework with machine learning techniques to make sense of the data. This is a multidiscipline research project, involving robotics, control engineering, sensor fusion and machine learning. It is also expected the PhD student will closely work with other industrial partners in this project.
Proposed by: Dr Cunjia Liu
Solid State Gas Sensors for Automotive Exhaust Monitoring - Dr Jung-Sik Kim, Ref: JSKAAE008
The prospective postgraduate research project calls for the evaluation, design and development of sensing devices based on solid state gas sensing technologies. The aim/scope of the project encompasses the extension and/or replacement of the air-fuel (so called ‘Lambda sensor’) sensor based on solid oxide electrochemical technologies, with the potential to investigate other exhaust gas components to be monitored and analysed in real time. The proposed sensing methodologies should be tested and evaluated against existing solutions and indicate the viability of the solid state sensors to form the basis of a real-time, in-situ on-board diagnostics system for vehicle exhaust gas emissions analysis.
Nonlinear dynamics and control of UAVs operating in urban environments - Dr James Knowles, Ref: JKAAE010
Uninhabited Aerial Vehicles (UAVs) have the potential to improve society in a variety of ways. Recent work on the ‘flying high challenge’ (http://flyinghighchallenge.org/ ) identified several urban-based scenarios, such as the delivery of emergency medical supplies, as potential future use cases for autonomous urban UAVs. One technical challenge associated with operating UAVs in urban environments is that the vehicles need to be highly agile, in order to respond more quickly to wind disturbances (e.g., gusts). Current autonomous flight systems, which assume the linearity of an aircraft, cannot perform agile and robust manoeuvres. Furthermore, the computer models used to help design flight control systems often struggle to capture relevant nonlinear dynamic behaviour, particularly for less conventional hybrid aircraft concepts (e.g., tiltrotor).
This project will consider the challenge of modelling the nonlinear flight dynamics of a UAV, and controlling it to perform agile manoeuvres.
Proposed by: Dr James Knowles
Safe- and Eco-Driving Control for Connected and Autonomous Vehicles - Dr Dezong Zhao, Ref: DZAAE011
This PhD project aims to develop the key algorithms for speed advisory systems in connected and autonomous systems (CAVs). Thanks to the development of communication technologies fused with onboard sensors, vehicles have been equipped with connectivity and autonomous technologies over the past years. CAVs have easier access to the required traffic information, therefore they can be controlled more precisely compared to human-driven vehicles. However, the energy saving benefit may not be achieved for personal CAVs because the sub-optimal path planning profiles. The conflicting objectives of speed/distance tracking and energy efficiency should be merged in a single problem to be optimised in real time.
This project will propose an integrated safe- and eco-driving control framework based on dynamic programming and model predictive control. The vehicle safety information and energy consumption will be both considered to make the best decision on path/speed planning.
If you don't see a topic in the list of indicative projects that matches your requirements then more information on a the wider range of our research can be found by viewing our individual research areas:
- Rolls Royce University Technology Centre (UTC)
- Caterpillar Innovation and research centre (IRC)
- National Centre for Combustion and Aerothermal Technology (NCCAT)
- Vehicle Aerodynamics
- Intelligent Mobility and Autonomous Systems
- Hybrid Vehicles and Advanced Propulsion
- Risk and Reliability
If you would like an informal conversation about any of the research areas or projects, you can either contact the named academic directly, or : Dr Sarah Dunnett email: S.J.Dunnett@lboro.ac.uk