Evaluation of the Structural Health Conditions of Smart Roads Using Different Feature-Based Methods

Flexible road pavements are damaged by traffic and environmental conditions. At the same time, traditional structural health assessment and monitoring methods are often destructive and expensive. Consequently, the main objective of this study is to further develop innovative non-destructive structural health monitoring method for road pavements, which aims at identifying the presence of damages into road sections based on vibro-acoustic signature analysis. For this reason, an experimental investigation was carried out using two non-destructive monitoring systems, i.e. one traditional (based on Falling Weight Deflectometer, FWD) and one innovative (based on the aforementioned innovative method). In particular, the traditional one was used to estimate the elastic moduli of damaged and undamaged road sections, and to produce impulsive loads that, in turn, produced vibration and noise detected by the sensors of the innovative system. Signals were recorded and processed. Then, two feature-based data analysis methods (i.e., one already used in the past, herein called method 1, and one used for this application for the first time, herein called method 2) were used on the signals recorded by the innovative system mentioned above. Results show that method 2 (i.e., the chromatic method) allows obtaining better results than method 1. Indeed, thanks to both 2D and 3D graphical plots, it allows improving the ability of the innovative system to discriminate damaged road pavements from undamaged ones. This allows obtaining a solution that can be easily scaled and used in current and future smart roads.