Structural health monitoring (SHM) plays a critical role in ensuring the safety and performance of aerospace structures throughout their lifecycle. As aircraft and spacecraft systems grow in complexity, the integration of machine learning (ML) into SHM frameworks is revolutionizing how damage is detected, localized, and predicted. This review presents a comprehensive examination of recent advances in ML-based SHM methods tailored to aerospace applications. It covers supervised, unsupervised, deep, and hybrid learning techniques, highlighting their capabilities in processing high-dimensional sensor data, managing uncertainty, and enabling real-time diagnostics. Particular focus is given to the challenges of data scarcity, operational variability, and interpretability in safety-critical environments. The review also explores emerging directions such as digital twins, transfer learning, and federated learning. By mapping current strengths and limitations, this paper provides a roadmap for future research and outlines the key enablers needed to bring ML-based SHM from laboratory development to widespread aerospace deployment.
Scarselli et al. (2025) studied this question.