Agentic AI for Complex Environments

Summary

Research Context

Positioned within Ulster University’s School of Computing, this research theme focuses on agentic AI systems that can reason, adapt and act within complex real-world environments.

Projects span trust-aware digital twins for autonomous IoT and smart-home assistance, multi-agent intelligence for optimising complex organisational processes, and explainable model recommendation based on dataset complexity.

The research brings together agentic AI, digital twins, multimodal sensing, language models, process mining, reinforcement learning and meta-learning. Across these applications, the theme seeks to develop intelligent systems that make context-aware and explainable decisions, adapt through feedback, and operate safely and efficiently in dynamic environments.

The projects directly support priorities in Software/Cyber by developing advanced capability in agentic AI, digital twins, explainable AI and intelligent automation, contributing to Programme for Government priorities around economic productivity and the digital transformation of services.

Trust-Aware Agentic Digital Twins for Autonomous IoT Device and Service Management

Supervisor Names: Dr Joseph Rafferty

This PhD will explore how agentic AI, digital twins and multimodal sensing can support people living independently in smart-home environments.

The project will combine mmWave radar, power monitoring, PIR and privacy-managed camera data to recognise what an inhabitant is trying to achieve and identify when assistance is needed.

Digital-twin agents will maintain contextual models of the person, environment, devices and activities, while LLMs and smaller language models will support semantic interpretation and planning. Assistance will be delivered through personalised speech, visual prompts and adaptive video guidance, with formal verification used to ensure actions remain safe, explainable and appropriate.

Multi-Agent Intelligence for the Optimisation of Complex Business Processes

Supervisor Names: Dr Zeeshan Tariq

Agentic Process Intelligence is an emerging area that moves beyond traditional process mining by enabling AI systems not only to analyse and predict operational issues but also to recommend actions that improve performance.

This PhD will develop intelligent agents that learn from organisational process data, evaluate alternative interventions and propose next‑best actions to optimise key indicators such as duration, resource use and service outcomes.

Combining process mining, reinforcement learning and adaptive decision‑making, the research builds on established industry‑informed work with BT Ireland Innovation Centre and contributes to advanced AI capability for complex digital service environments.

Data2Model: Explainable Model Recommendation through Dataset Complexity Profiling

Supervisor Names: Dr Muhammad Shafi

Data2Model will develop an agentic and explainable AI framework for recommending suitable machine learning models based on dataset characteristics. It will analyse properties such as dimensionality, class imbalance, sparsity, correlation and non-linearity to create dataset complexity profiles.

Using benchmark datasets, the project will apply meta-learning to identify relationships between dataset characteristics and model performance.

An agentic AI layer will autonomously explore and refine model recommendations using evaluation feedback, while explainable AI will provide clear reasons for each recommendation. The framework aims to make model selection faster, more transparent and computationally efficient.

Applying to Multiple Projects: Applications for more than one PhD studentship are welcome, however if you apply for more than one PhD project within Computing, your first application on the system will be deemed your first-choice preference and further applications will be ordered based on the sequential time of submission.

If you are successfully shortlisted, you will be interviewed only on your first-choice application and ranked accordingly. Those ranked highest will be offered a PhD studentship.

In the situation where you are ranked highly and your first-choice project is already allocated to someone who was ranked higher than you, you may be offered your 2nd or 3rd choice project depending on the availability of this project.

The School of Computing at Ulster University holds Athena Swan Bronze Award since 2016 and is committed to promote and advance gender equality in Higher Education.  We particularly welcome female applicants, as they are under represented within the school.

Essential criteria

Applicants should hold, or expect to obtain, a First or Upper Second Class Honours Degree in a subject relevant to the proposed area of study.

We may also consider applications from those who hold equivalent qualifications, for example, a Lower Second Class Honours Degree plus a Master’s Degree with Distinction.

In exceptional circumstances, the University may consider a portfolio of evidence from applicants who have appropriate professional experience which is equivalent to the learning outcomes of an Honours degree in lieu of academic qualifications.

• Sound understanding of subject area as evidenced by a comprehensive research proposal detailing background, research questions, aims and methodology
• A comprehensive and articulate personal statement demonstrating interest in the subject area

Desirable Criteria

If the University receives a large number of applicants for the project, the following desirable criteria may be applied to shortlist applicants for interview.

• International applicants should hold, or expect to hold, a First Class Honours Degree in a relevant subject area
• International applicants should hold, or expect to hold a relevant Masters Degree at a level equivalent to a distinction at Ulster
• Work experience relevant to the proposed project
• Track record of peer reviewed publications

Equal Opportunities

The University is an equal opportunities employer and welcomes applicants from all sections of the community, particularly from those with disabilities.

Appointment will be made on merit.

Funding and eligibility

This project is funded by:

• Department for the Economy (DfE)
• Vice Chancellor's Research Scholarship (VCRS)

PhD scholarships are available for this project to all applicants worldwide, regardless of residency or domicile, and will cover:

• Tuition fees.
• A maintenance allowance of approximately £22,000 (exact rate is to be confirmed) per annum for three years, subject to satisfactory academic performance.
• A Research Training Support Grant (RTSG) of approximately £900 per annum.

Applicants who already hold a doctoral degree or who have been registered on a programme of research leading to the award of a doctoral degree on a full-time basis for more than one year, or part-time equivalent, funded from any source, including self-financing students, are NOT eligible to apply for an award.

Students in full-time employment are not eligible for an award.

Part-time PhD scholarships may be available at a 0.5 pro rata rate of the full-time scholarship, which would require a six-year registration period.

Due consideration should be given to financing your studies.

The University will award a successful candidate one of the following scholarships for the project, depending on the funding available at the time an admissions offer is made:

1. Department for the Economy (DfE) PhD Scholarships
2. Vice Chancellor’s Research Scholarships (VCRS)

Please note that the Terms and Conditions for scholarships are subject to change for 2027/28.