A Disruptive Semantically-Enhanced, Self-Optimising Intelligence Framework for Trustworthy Cyber-Physical Systems in Next-Generation IoT Environments

DISRUPT-X will introduce a novel semantically enhanced, neuro-symbolic architecture that embeds intelligence directly at the edge of CPS. By combining semantic technologies, federated and continual learning, predictive analytics, and trustworthy AI, the framework enables systems to continuously assess and repair data quality, adapt predictive models under changing conditions, and make risk-aware, explainable decisions in real time. I will be providing expertise in context-aware data quality inference at the edge and the robust self-healing pipelines

I am interested in collaborating with MSCA fellows on:

• edge intelligence deployment

• real-time CPS data pipelines

• reliability and trust estimation of predictions

• federated and continual learning under drift

• integration of AI predictions into control loops

• trustworthy CPS control design

• Long-term system optimisation and self-optimisation engines

• trustworthy AI for CPS and relevant sectors such as health and Industry 5.0

Fellows with backgrounds in machine learning, trustworthy AI, Semantic modelling, edge computing, CPS optimisation, and Applied domain expertise will find strong alignment with this opportunity.