Intelligent Inspection of Railways Infrastructure and Risks Estimation by Artificial Intelligence Applied on Noninvasive Diagnostic Systems

The proposed work focused on a methodological approach to perform inspections by means of non-invasive diagnostic devices, based on Ground Penetrating Radar (GPR), laser scanner and standalone temperature sensor technologies. The data acquired from the inspections were processed by using a platform which estimated the risks connected to the infrastructure, including the predictive mode. The algorithms, namely Fast Fourier Transform (FFT) and Artificial Intelligence (AI), i.e. Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), were applied to monitor ballast fouling and to predict dangerous operating conditions as in the case of a train which collides into a tunnel, railway track deformation, and other potential structural failures. The work was carried out within the framework of a research industrial project, which aimed at the development of an informatic platform for the geolocation of the risk maps.