About the project
The aeronautical industry generates a massive volume of heterogeneous data during flight test activities, which often remains partially exploited. This project aims to bridge the gap between raw flight data and actionable technical knowledge.
By leveraging advanced Machine Learning (ML) and Artificial Intelligence (AI) techniques, such as clustering, classification, and automatic pattern recognition, this project will contribute to transform an extensive archive of flight parameters into a structured, queryable, and predictive system that will also link the raw time-histories of flight test data with the metadata stored into company databases and with the technical document management system. The final output will be a software tool to support high-fidelity analysis of power requirements, trim maps, vibrations, and fuel consumption.
The project is structured around four primary pillars:
- Automated data preprocessing: developing a robust framework for normalization, filtering, and synchronization of multi-source onboard signals (e.g., engine data, airframe parameters, vibrations).
- Automatic flight condition cataloguing: utilizing unsupervised and supervised ML to identify and group specific manoeuvres (e.g., hover, autorotation, low-speed controllability) without manual intervention, improving and expanding the capabilities of similar applications already developed by LH.
- Engineering application modules: creating targeted analytical models for critical flight regimes, such as:
- power and fuel: predictive models for power required and consumption based on variable atmospheric and weight conditions
- flight mechanics: building empirical trim maps and control position requirements
- dynamics: correlating vibration levels with aerodynamic loads and power settings
- Traceability and integration: ensuring a “white-box” approach where every aggregated result can be traced back to individual test points, maintaining full transparency and verifiability. Developments will leverage upon existing Leonardo Helicopter Division (LHD) internal tools to ease integration
Methodological principles:
- traceability-first approach: every output must be reproducible and linked to raw data
- no black-box models: preference for interpretable ML techniques where feasible
- scalability: algorithms must handle large multi-aircraft datasets
- industrial integration: alignment with Leonardo workflows and IT constraints
The School of Engineering is committed to promoting equality, diversity inclusivity as demonstrated by our Athena SWAN award. We welcome all applicants regardless of their gender, ethnicity, disability, sexual orientation or age, and will give full consideration to applicants seeking flexible working patterns and those who have taken a career break. The University has a generous maternity policy, onsite childcare facilities, and offers a range of benefits to help ensure employees’ well-being and work-life balance. The University of Southampton is committed to sustainability and has been awarded the Platinum EcoAward.