Yiannis Papadopoulos

[£20,780per annum] This fully funded PhD studentship at the University of Hull focuses on developing SafeML-based confidence generation and explainability methods for UAV-based anomaly detection of blade surfaces in offshore wind turbines. The project addresses the growing use of unmanned aerial vehicles (UAVs) for equipment anomaly and fault detection in offshore wind energy, where image quality and decision confidence are critical for reducing maintenance costs and downtime. The research aims to propose a methodology that generates confidence in decisions made from drone-captured images, using the SafeML tool—a novel open-source safety monitoring tool. The approach involves measuring statistical differences between new images and trusted datasets (validated by experts during model training) to assess confidence in anomaly detection outcomes. This methodology will enhance deep learning explainability and interpretability, providing insights for wind farm owners, system designers, and third-party UAV operators regarding the causes of incorrect diagnoses, algorithmic responsibility, and image quality issues. Supervised by Dr Zhibao Mian, Dr Koorosh Aslansefat, and Professor Yiannis Papadopoulos, the project offers interdisciplinary training opportunities, including introductory MSc modules in AI and Data Science and a dedicated Safe AI module delivered by the supervisor group. These will equip the candidate with advanced digital and data science research skills, preparing them for careers in data science, safe AI, or further research addressing future technological challenges. Eligibility requires a First-class Honours degree, or a 2:1 Honours degree and a Masters, or a Distinction at Masters level with any undergraduate degree (or international equivalents) in engineering, computer science, or mathematics and statistics. Non-native English speakers or those requiring a Student Visa must provide evidence of English proficiency (IELTS 7.0 overall, minimum 6.0 in each skill). The studentship provides funding of £20,780 per annum. Applications should be submitted via the project link. For further information, contact Dr Zhibao Mian at [email protected]. The application deadline is 5 January 2025.