Development of a Closed-Loop PLM Application for Vibration-Based Structural Health Monitoring of UAVs

Unmanned Aerial Vehicles (UAVs) require rigorous structural inspections to ensure safety and integrity throughout their lifecycle. Traditional visual inspection methods are labor-intensive, subjective, and inadequate for real-time fault detection. This study presents an integrated software application that enables vibration-based structural health monitoring within a closed-loop Product Lifecycle Management (PLM) framework. The system collects time-domain vibration data from UAV components during the pre-flight phase and applies deep learning architectures, including Gated Recurrent Units (GRUs), Long Short-Term Memory networks (LSTMs), and Convolutional Neural Networks (CNNs) for accurate fault classification. Communication with the UAV is handled through the DroneKit-Python API, while RESTful APIs interface with the Aras Innovator PLM platform to automate data exchange and support predictive maintenance. Upon detecting anomalies, the application triggers safety protocols, such as UAV disarming and automatic maintenance request generation. Experimental validation shows that the proposed system achieves high fault detection accuracy, confirming the feasibility of the closed-loop PLM approach. The system enhances reliability and traceability, and supports data-driven decision-making by enabling continuous feedback across the UAV lifecycle.