AI for diagnosis and prognosis
Data-driven and hybrid methods for state estimation, fault diagnosis and degradation prognosis in engineered systems.
Research
Models that respect the system
My research connects machine learning with physical structure to understand degradation, uncertainty and risk in complex engineered and environmental systems.
Physically informed representations
Machine-learning representations shaped by observable physical variables, system behaviour and the real granularity of health indicators.
Complex systems engineering
System modelling, interfaces, requirements, uncertainty, risk, safety, and verification and validation across multidisciplinary settings.
Environmental intelligence
Remote sensing, signal and image processing, IoT sensing and predictive indicators for vegetation, wildfire and climate risks.
Working approach
System understanding + observable data + rigorous evaluation = models that can support engineering decisions.