EPSRC Doctoral Training Partnership (DTP) Studentships

Engineering student

EPSRC logo

EPSRC Doctoral Training Partnership (DTP) Studentships

The University of Lincoln has received funding from the Engineering and Physical Sciences Research Council (EPSRC) to establish a Doctoral Training Partnership (DTP), which will provide skills and training to foster the next generation of world-class research leaders in areas of strategic importance to both EPSRC and the University of Lincoln.

Our training programme prioritises the following three thematic areas of robotics and artificial intelligence: smart energy, medical diagnosis and healthcare support systems, and bio-physics inspired robotics, in which the University has strong research groups. These research groups will provide DTP students with a rich research environment and a broad range of experienced and new researchers.

Each studentship will be associated with a specific project that will be designed to advance fundamental research in computer science or engineering within one of the thematic areas. Interdisciplinary links with other subject areas will also be expected.

We are currently providing the following 8 potential projects for the 2021-22 studentship application round and these are detailed below.

Studentship applications are now open for entry into the DTP programme, starting in October 2021.

Please make sure to check the eligibility criteria before you apply. Normally, a student must have no restrictions on how long they can stay in the UK and have been ordinarily resident in the UK for at least 3 years prior to the start of the studentship. UK students will be eligible for a full studentship, covering the costs of Home fees, a stipend to support living costs for 3.5 years, and a generous research training support grant enabling international travel and participation in the leading conferences and symposia.

Although most DTP students must be UK residents, we also have an opportunity for an international (EU and non-EU) student. The international studentship award will be subject to eligibility, and also the availability of complementary funding (to provide the differential to the international fee rate). You should get in touch with the lead supervisor before applying this award.

Closing date: Midnight, 14 March 2021.

DTP PhD Application Form 2021-22

Smart Energy

Ultra-low loading Pt electrocatalysts based on carbon supported alloys for oxygen reduction in hydrogen fuel cells and their lifetime prediction

Academic contact: Dr Guanjie He (ghe@lincoln.ac.uk), Senior Lecturer, School of Chemistry

This project aims to develop sustainable and low-cost oxygen electrocatalyst based on the carbon framework supported ultra-low loading Pt alloy clusters to realise the high performance and stable energy outputs.

With global population and demands for energy increasing, non-renewable resources such as coal, oil, and natural gas are being rapidly depleted. Global challenges, such as air pollution and climate change, require urgent technological solutions. The UK government set the 25 Year Environment Plan in 2018, emphasising the increase of resource efficiency and the reduction of pollution and waste. The hydrogen fuel cell is one of the promising candidates for next-generation clean energy suppliers, but expensive electrocatalysts (relatively high loading Pt) on cathodes limited their further application.

This project aims at developing low-cost and high-performance oxygen reduction electrocatalyst based on the ultra-low loading Pt alloys on carbon framework. The mechanism will be studied via the advanced in-situ characterisation techniques, and the lifetime of the electrocatalysts and fuel cell devices will be monitored and estimated via the developed programme. The successful candidate will work with a skilled and interdisciplinary supervisorial team (materials chemistry, computer science, chemical engineering) from University of Lincoln, UCL, and Bramble Energy Ltd. The student will be trained for chemical synthesis, materials characterisation, electrochemical evaluation, computational coding and fuel cell fabrication.

Skills the candidate can learn:

- Materials synthesis, processing, and characterisation methods
- Electrochemical performance evaluation for oxygen reduction reactions
- Fabrication of fuel cell devices
- Computational coding for lifetime prediction of fuel cells
- Collaborative skills among academic field and industry

Ideal candidate:

- Should have, at a minimum, a 2.1 degree in chemistry, materials science, chemical engineering, or a related discipline
- Can demonstrate skills and experience (or an aptitude for mastering) the synthetic chemistry, energy device fabrication, and computational coding
- Are academically curious and think deeply and creatively
- Communicate well in both written and spoken English
- Are empathetic, kind, have great social skills, and enjoy working with others from diverse backgrounds
- Take responsibility for the progress and quality of projects

Supervisory Team:

Dr Guanjie He, University of Lincoln
Dr Shouyong Jiang, University of Lincoln
Prof Dan Brett, University College London

Artificial intelligence for energy management optimisation in smart grid

Academic contact: Dr Shouyong Jiang (sjiang@lincoln.ac.uk), Senior Lecturer, School of Computer Science

Have you considered to optimise your household energy usage to reduce your bills and save the planet? Are you looking for approaches to schedule your energy consumption in response to energy pricing schemes imposed by energy providers? Smart grid technology represents an unprecedented opportunity to move the energy industry into a new era of reliability, availability, sustainability, and efficiency that will contribute to our economic and environmental health. This project aims to develop artificial intelligence approaches to efficient energy management that achieve demand-supply balance for stable, cost-effective operation of the smart grid system, maximising different stakeholders’ interests. In particular, the project will investigate AI approaches, including machine learning and/or bio-inspired optimisation, to optimise energy management from energy supply to energy consumption. It will look for novel solutions to trade-off the profit of energy suppliers and the bills of energy users while maintaining supply-demand balance of energy and lowering carbon emission.

The overall aim of the project is to develop a AI-based decision-making tool to improve energy management of smart grid that balances different stakeholders’ interests including energy suppliers’ profit, end users’ energy consumption and bill, and governments’ net zero carbon goal, while ensuring demand-supply balance. In order to reach this aim, the project has the following key objectives:

- Investigate computational models of electricity markets of smart grid systems at different scales, including smart home, microgrid, or distributed resource systems.

- Develop machine learning algorithms for demand response, including dynamic pricing, from energy suppliers’ perspectives.

- Develop machine learning algorithms to support smart scheduling of power consumption at peak time from energy users perspectives.

- Multi-criteria optimisation of energy management from both energy suppliers’ and end users’ perspectives.
Test and evaluate the proposed approaches in real-scenario experimental platforms.

You will be working with the Machine Learning group, within the School of Engineering and School of Mathematics and Physics at the University of Lincoln. This is an exciting opportunity for developing a career in AI for smart energy and deliver world-leading research that impacts our economy and society.

Skills the candidate can learn:

Discipline specific knowledge, including machine learning, data analysis, modelling and optimisation, and smart energy technology. The candidate can develop the ability to gather and interpret information, and ability to analyse data. They can also develop problem-solving skills, project management skills, and oral and written communication skills, alongside the ability to work independently and as a team member. The candidate will also receive a broad set of training on scientific research and transferable skills.
Ideal candidate:

Interested applicants should carry, at a minimum, a 2.1 degree in AI, computer science, mathematics, engineering or any other relevant discipline and are encouraged to demonstrate any skills and/or experience relevant to the project subject area(s) of interest.

They must have the ability to engage in scientific research and to work collaboratively as part of a team.

They must have a research interest and experience in at least one of the following areas: data analysis, artificial intelligence, machine learning, mathematical modelling, game theory, energy management, operational research, and optimisation approaches (evolutionary computation, and multi-criteria decision making, etc).

They are expected to have good communication skills in written and spoken English in order to work with both computer scientists and engineers, to present research findings in workshops/conferences and to publish papers in high quality journals.

Supervisory Team:

Dr Shouyong Jiang, University of Lincoln
Dr Matthew Watkins, University of Lincoln
Prof Chris Bingham, University of Lincoln
Dr Sepehr Maleki, University of Lincoln

Medical Diagnosis and Healthcare Support Systems

AI enabled biomarker identification from exhaled breath condensates for early detection of secondary infection in chronic obstructive pulmonary disease (COPD) patients

Academic contact: Dr Wenting Duan (wduan@lincoln.ac.uk), Senior Lecturer, School of Computer Science

This project aims to develop software approaches to identify biomarkers for secondary infections in COPD using non-invasive exhaled breath samples. Using this approach, we aim to identify specific biomarkers that can be used to predict/diagnose the onset of a secondary infection prior to the patient becoming symptomatic as part of routine clinical monitoring. With this approach we hope to enable proactive clinical management of COPD patients through the early detection of secondary infections to avoid exacerbations of COPD which can lead to hospitalisation.

Chronic obstructive pulmonary disease (COPD) is characterised by long-term breathing problems which can be exacerbated by secondary infections. Current diagnosis of secondary infections in COPD patients requires invasive sampling which makes routine screening labour intensive, unpleasant and impractical for patients. Two schools at the University of Lincoln, the School of Computer Science and the School of Life Sciences are collaborating to develop software approaches to identify biomarkers for secondary infections in COPD using non-invasive exhaled breath samples.

The main aim of the PhD project is to develop algorithms for analysis of mass spectrometry (MS) data that can detect infection in COPD from non-invasive breath samples. You will analyse MS data exploiting advanced pattern recognition, mass spectrometry imaging, and machine/deep learning techniques for infection classification and prediction.

The successful student will be associated with the Laboratory of Vision Engineering in School of Computer Science, but will also be part of a cross-discipline collaboration amongst several research groups including the Laboratory of Vision Engineering, the Machine Learning group, the Diabetes, Metabolism and Inflammation research group, and Animal Behaviour and Welfare group. By joining the Doctoral Training Partnership programme at the University of Lincoln, you will be also conducting research together with other PhD and post-doctoral students in a supportive and intellectually stimulating environment.

Skills the candidate can learn:

Whilst pursing this project, the student can gain extensive knowledge of the state-of-art of image processing, machine learning, and deep learning techniques. They will have the chance to develop advanced Matlab, Python or R programming skills, and obtain experience of cross-disciplinary working with academics, clinicians, and other researchers. The student will also receive broad training on research and transferable skills including scientific writing, presentation skills, and project management.

Ideal candidate should have:

- A first class, upper second class (2:1) or Master's qualification in Computer Science, Bioinformatics, or equivalent subject area

- Good knowledge of image processing or machine/deep learning

- Good programming skills of Matlab, Python, or R programming

- Capability to work independently and as part of a team

-  A good mathematical background

- Excellent written and oral communication skills in English

- A real passion and commitment for research

Supervisory Team:

Dr Wenting Duan, University of Lincoln
Dr Shouyon Jinag, University of Lincoln
Dr Neil Holden, University of Lincoln

Localising breast cancer risk in mammograms

Academic contact: Dr Faraz Janan (fjanan@lincoln.ac.uk), Senior Lecturer, School of Computer Science

The project aims to develop a breast cancer diagnosis support system by localising the future risk of developing cancer, in particular associated with mammographic focal densities. By analysing time series data with new mammographic image analysis framework, the project aims to propose a CAD system that can flag women likely to develop breast cancer in near future. The system should also be able to suggest among the bilateral breast, as well as the location/quadrant where the risk of developing cancer is higher.  

Applications are invited for a PhD studentship to work on a project using a range of advanced AI and image analysis techniques to study mammographic images in the context of breast cancer risk. The project will require a full-time research commitment and will be based in the Laboratory of Vision Engineering at Lincoln. It will combine mammographic density quantification with an AI based classification and pattern recognition framework (desirably Deep Learning methods). It would evaluate the methods developed on mammograms acquired from the Optimam that have negative priors, CC and MLO views available for both breasts and depict biopsy-proven cancers, as well as normal cases. This would help us to assess the effectiveness of a CAD system and its suitability in a clinical set up.

The 3-year project will be carried out in close collaboration with scientists and breast radiologists at Oxford and Lincoln. The successful candidate will be required to apply, develop and program algorithms in the area of computer vision and machine learning, while applied to mammographic images -including but not limited to x-ray and Digital Breast Tomosynthesis (DBT). Interested candidates should send their CVs (including references) to fjanan@lincoln.ac.uk.

Skills the candidate can learn:

Mammographic image analysis, clinical aspects of breast cancer detection and diagnostics, deep learning methods applied to medical imaging

Ideal candidate:

A strong academic track record with a 2:1 or higher degree in computer science, mathematics, biomedical engineering, electrical(computer) engineering’s or its equivalent if outside the UK.  The desirable candidate should have an excellent performance in a relevant postgraduate degree. The candidate is expected to demonstrate expertise of coding in Matlab and python, with good knowledge of image processing techniques. A prior working experience of deep learning methods is desirable. The candidate should be willing to work in close collaboration with clinical radiologists.

Supervisory Team:

Dr Faraz Janan, University of Lincoln
Dr Louise Wilkinson, Oxford University Hospitals NHS Trust
Prof Xujiong Ye, University of Lincoln
Dr Tryphon Lambrou, University of Lincoln

AI-based multimodal sensor information fusion for fall detection and prediction

Academic contact: Dr Miao Yu (myu@lincoln.ac.uk) Senior Lecturer, School of Computer Science

Novel AI techniques for fusing information from multiple sensor modalities, to timely detect/predict falls will be developed in this project, with the aim of minimising the adverse effects of falling on vulnerable people.  Related techniques which will be developed in this project will contribute to falls prevention and the reduction of associated healthcare costs of falls, such as fractured neck of femur and prolonged hospitalisation. Moreover, it is anticipated that the techniques developed in this project will contribute to the development of novel healthcare support systems for detecting/predicting/preventing falls, as well as optimise independent living for older adults and other related stakeholders by reducing their risks of falling in their independent living.

Falls and fractures are a common and a serious health issue faced by older people in England and globally. Around a third of people aged 65 and over, and around half of people aged 80 and over, fall at least once a year. Falling is a cause of distress, pain, injury, loss of confidence, loss of independence, and mortality. Detecting/predicting the occurrence of falls and reducing falls and associated fractures are important for maintaining the health, wellbeing, and independence of older people and other related stakeholders in the health and social care sectors.

Considering the importance of detecting/predicting falls and for minimising the adverse effects of falling on vulnerable people, this project aims to detect and predict the risk/likelihood of falls based on multimodal sensor modalities (vision sensor and wearable sensor). The candidate is expected to develop novel signal processing, machine learning and deep learning algorithms, to optimally extract and fuse information from vision and wearable sensors’ recordings of a person’s movements, for detecting possible falls as well as predicting the fall risks associated with some normal daily activities (such as walking or sitting). Novel automatic fall detection/prediction techniques based on multimodal sensors are expected to be developed at the end of the project.

The successful candidate will be given the opportunity to work across disciplines and engage with both experts across different Schools in the University of Lincoln and experienced physiotherapists in the NHS and private sector. Moreover, the successful candidate may have internship opportunities in prestigious AI healthcare start-ups specialised in wearable and vision sensors for healthcare.

Skills and experience you can gain are expected to include:

- Opportunities to develop expertise in machine learning and deep learning and their applications in healthcare
- Quantitative research skills and applied statistics
- Critical thinking and problem-solving abilities
- Applied data science/machine learning/deep learning
- Strong verbal and written communication skills
- Opportunities to present research outputs at national and international venues.

Internship opportunities exist at Vitrue Health and Activinsights

Ideal candidate:

Interested applicants should hold, at a minimum, a 2.1 degree in a relevant or related discipline and are encouraged to demonstrate any skills and/or experience relevant to the project subject area(s) of interest. They must evidence an ability to engage in scientific research and to work collaboratively as part of a team. Excellent communication skills in written and spoken English are also required.

Supervisory Team:

Dr Miao Yu, University of Lincoln
Prof Simon Parsons, University of Lincoln
Ross Clifford, University of Lincoln

Bio-physics Inspired Robotics

A Miniaturised Stiffness-controllable Soft Medical Manipulator

Academic contact: Dr Khaled Elgeneidy (kelgeneidy@lincoln.ac.uk), Senior Lecturer, School of Engineering

This PhD studentship is focused on developing a miniaturised soft medical manipulator that utilises a novel controlled-stiffening mechanism to enhance tool stability and force output. This research will involve studying particle jamming through bio-physical modelling and translating that information into an engineering model. Designing this bio-inspired medical robot that can actively change its stiffness, in response to sensor data, and operate in a controlled way is the main focus of this project.

Throughout this project, the PhD candidate will work closely with clinical partners to guide the development of the soft manipulator. The internal supervisory team from the University of Lincoln combines expertise in soft robotics (Dr Khaled Elgeneidy), medical robotics (Prof Mini C. Saaj), and computational physics (Dr Fabien Paillusson). The candidate will carry out research in the Bio-robotics and Medical Technologies Laboratory at the University of Lincoln. There will be opportunities to work collaboratively with Lincoln Medical School. Additionally, the project will be supported by external clinical co-supervisor, Dr Mohamed Thaha, from St Bartholomew’s Hospital, Queen Mary University of London as well as Dr Fulvio Forni from the University of Cambridge, who will act as an external academic co-supervisor.

Supervisory Team:

Dr Khaled Elgeneidy, University of Lincoln
Prof Mini C. Saaj, University of Lincoln
Dr Fabien Paillusson, University of Lincoln

AI enabled haptic-guided teleoperation: a study on Human-Robot Interaction

Academic contact: Dr Amir Ghalamzan Esfahani (aghalamzanesfahani@lincoln.ac.uk), Associate Professor, Lincoln Institute for Agri-Food Technology

This project aims at developing and testing novel teleoperating robot manipulators and assess their effectiveness. You will be specialising in Human Robot Interaction and will study how various robotic features (such as intelligent haptic guidance) impacts performance, operators’ mental workload, subjective experience, and physiological and brain data.

We are offering a fantastic multi-disciplinary PhD opportunity, fully funded, working with the Lincoln Centre for Autonomous Systems (LCAS), the interactive technologies lab (intlab), and the School of Psychology.
While we have extensively studied haptic-guided shared control for telemanipulation for remote handling in extreme environments, this PhD student will have the opportunity to study this in a wider context and application domain, such as agri-food robotics, and medical robotics. The student will have the opportunity to work with state-of-the-art robotic systems/equipment (e.g. Franka-Emika manipulators and Hpatin Virtuose™ 6D) and benefit from the setups at our group developed during several EU-H2020 and UK-EPSRC funded collaborative projects.

This interdisciplinary project is related to the design, development and implementation of AI, machine learning, and/or robot control solutions for haptic-guided shared control to facilitate teleoperating a robotic manipulator for human users. This project also involves evaluation techniques to assess operators’ mental workload and effort based on physiological and brain data by using portable Brain-Computer Interfaces and data science techniques to reveal the effectiveness of each developed technology. Full training will be provided.

By joining the Doctoral Training Partnership programme at the University of Lincoln you will participate in the fully funded 3.5 years programme supported by a multidisciplinary supervisory team across 3 different schools in the College of Science (School of Computer Science and the intLab, School of Psychology, and Lincoln Institute of Agri-food Technology).

Training opportunities:

- The PhD student will work with the state-of-the-art robotic equipment.
-  The can gain deep knowledge in Robotic Manipulation, Teleoperation, Control, ML and data analysis and apply them in the real experimentations
- They can gain significant knowledge for Human Robot Interaction studies and Data Analysis
- Develop analytical, problem-solving and research skills
-  Develop coding skills (C++, Python and ROS)
- Work in collaboration with an active team of researchers within L-CAS and across collaborating universities, including the University of Cambridge, Birmingham, Bristol and QMUL.

They will be working in a collaborative and stimulating environment, strengthened by the cohort-driven activities, where knowledge sharing and joint problem solving is the norm. The multidisciplinary nature of the programme will make you think about problems from a whole new perspective and explore innovative ideas.

We are seeking talented candidates with:

- MSc (first or upper second-class) in Robotics, AI, Human Factors, Human Computer Interaction, Computer Science, Mechanical Engineering, Electronics, Cybernetics, or related scientific discipline. Applicants who graduated in the UK with a strong first degree, or complementary industry experience are also considered.

- Good knowledge of robotics, control, ML, AI, Human Robot Interaction.

- Programming skills (Proficient C++ and/or Python)

- ROS experience is considered a strong plus

- Oral and written fluency in English

- Good communication, team working and presentation skills

- Experience with real robotic systems is a strong plus.

Supervisory Team:

Dr Amir Ghalamzan Esfahani, University of Lincoln
Dr Horia Maior, University of Lincoln
Dr Julia Foecker, University of Lincoln

Collective behaviour of autonomous organisms: from bio-particles to robotics

Academic contact: Dr Fabien Paillusson (fpaillusson@lincoln.ac.uk), Senior Lecturer, School of Mathematics and Physics

This project aims to develop physical models of collective behaviour derived from biological systems comprising self-propelling agents, such as schools of fish and flocks of birds, and to then transform them into swarming behaviours of mobile robots. The ultimate aim is to improve the design of complex behaviours for robot swarms by integrating elements of collective intelligence inspired by biological agents.
Active Matter is an emerging interdisciplinary field in physics and applied mathematics that refers to systems comprising interacting agents that can drive their own motion (such as birds, fish, insects, “smart” artificial micro-particles, or bio-mimicking robots). Active Matter systems are opposed to Inert Matter systems, whose behaviours are entirely determined by the mechanical interactions between the agents. Consequently, in the past two decades Active Matter models have demonstrated complex collective behaviours such as the formation of active clusters, obstacle induced phase separation, and organised flocking motions, which are usually not achievable in assemblies of inert agents.

These newly found “living structures” can in turn be implemented in real life with collections of bacteria, artificial micro-particles, or bio-mimicking robots for industrial, medical or agricultural applications making use of their self-assembling properties and resilience to external influences. In Active Matter systems, simple sets of rules can lead to many rich phases of collective behaviours. There is ample opportunity to develop new classes of rules that can give rise to never-before-seen phases, and ultimately provide insights into how to reverse-engineer rules for targeted goals.

Active Matter systems are of particular interest with respect to swarm robotic systems where the behaviour of individual robots is affected by the fluid dynamics of their environment; for example, aerial drones whose position may drift when buffeted by gusts of wind, or surface-water/underwater vessels that may be dragged along by water currents. These swarm robotic systems must coordinate their movement despite the influence of the fluid environment they inhabit. By translating the mechanics of Active Matter systems to swarm robotic systems, we hope to improve the performance of embodied agents used in real-world applications. This novel approach would be particularly beneficial in GPS-denied environments such as deep-sea exploration, where individual robots must remain aggregated without the aid of an external frame of reference.

This interdisciplinary project at the interface of physics, computational modelling, and robotics will develop new theoretical and computational models for such systems and validate them on physical robotic swarms. It will broadly consist of three main tasks:

- Formulating a set of microscopic rules deduced from a set of bio-inspired Active Matter systems, and formulating a statistical physics description of a large groups of such Active Matter systems

- Deriving continuous field models starting from the microscopic rules defined in task (1) to describe the collective behaviour of Active Matter

- Validating the developed models through their implementation on aerial/surface-water robot swarms – initially in the widely-used ARGoS multi-robot simulator, and then on physical hardware

This project combines complementary expertise from two schools: the School of Mathematics and Physics and the School of Computer Science, building on existing research. The successful student will be associated with the Centre for Computational Physics and the Lincoln Centre for Autonomous Systems (L-CAS), of which the supervisory team are members.

Skills the candidate can learn:

- Theoretical and computational modelling
- Design of control algorithms for robot swarms
- Interdisciplinary collaboration skills
- National and international collaboration skills
- Technical oral presentation and written communication skills

Ideal candidates:

Interested applicants should hold, at a minimum, a 2.1 degree in either Physics, Mathematics, Computer Science, Engineering, or a related discipline. Applicants with a relevant Master's degree are particularly welcome. The candidate is expected to have good communication and teamwork skills and must be motivated to learn new things. Interested applicants are encouraged to demonstrate any skills and/or experience relevant to the project subject area(s) of interest.

Supervisory Team:

Dr Fabien Paillusson, University of Lincoln
Dr Alan Millard, University of Lincoln
Prof Andrei Zvelindovsky, University of Lincoln

As an institutional Athena SWAN Bronze Award holder, we are committed to advancing gender equality in STEM, therefore female applicants are strongly encouraged to apply.