PhD Studentship: Enhancing Adaptive Control System Performance Using Advanced Machine Learning Techniques

Overview

• Qualification type: PhD
• Subject area: Control and Machine Learning
• Location/Campus: College Lane, Hatfield
• Closing application date: 10 June 2024
• Start date: July 2024 or as soon as possible thereafter
• Duration: three years, full time

Funding information (fully funded for UK, EU and international students)

Annual tax-free bursary of approximately £18,622 pa, plus tuition fees (£5590 for UK or £14,905 for International and EU applicants).

Project Details

In recent years, there has been a significant breakthrough in the application of Machine Learning (ML) across various domains. Emerging techniques such as DDPG (Deep Deterministic Policy Gradient), PPO (Proximal Policy Optimisation), and TD3 (Twin Delayed DDPG) have shown promising performance in dealing with control systems. However, the performance, stability, and robustness of these algorithms are still undergoing investigation, and their generalisability compared to conventional adaptive and robust controllers remains not fully comprehended.

This project aims to comprehensively analyse the performance and robustness of state-of-the-art ML techniques on control system problems. It will extract both the limitations and advantages of these algorithms. Moreover, the project will develop novel ML solutions that not only push performance boundaries but also exhibit superior generalisability compared to traditional adaptive control methods. Rigorous theoretical and statistical analysis will be carried out to prove the effectiveness of these proposed techniques. Hence, a strong foundation in mathematical and control theory is essential for conducting this research.

The applicant should have a relevant degree, ideally with a background (or strong interest in developing knowledge) in the following areas:

• Control theory (classical, adaptive, and optimal controllers).
• Machine learning theory and techniques.
• Nonlinear stability and statistical analysis.
• Programming skills in MATLAB or Python(NumPy, PyTorch, TensorFlow), or other relevant tools.

Entry requirements

Applications are invited from individuals with a first or upper second-class degree (or equivalent) in a relevant discipline such as, engineering, maths, computer science, etc. A master’s degree is essential. We are seeking applicants with very good analytical and programming skills. In some areas priority will be given to applicants with demonstrable practical engineering skills and experimental experience.

Eligibility

The studentship is open to UK/EU and international applicants.

How to Apply

Informal enquiries can be made to Dr Pouria Sarhadi, the project supervisor, or Prof. Pandelis Kourtessis, Associate Dean of Research and Enterprise.

Please download and complete an application form

Please also send with your application form:

• A research proposal not exceeding 2 pages, addressing the stated problem
• Two academic references
• Copies of qualification certificates and transcripts
• Certification of English language competence (minimum IELTS 6.5 or equivalent) for candidates for whom English is not their first language.

Email your completed application via the above ‘Apply’ button.

Closing date for applications: 10 June 2024

Interview dates: week commencing 17 June 2024

Expected studentship start date: 1 July 2024

Advert information