Frequency Identification of Bridges Using Smartphones on Vehicles with Variable Features

This paper presents a crowdsensing framework employing smartphones in monitoring populations of bridges in future smart cities. In this regard, because there are a variety of vehicles with different features traveling over the bridge at different speeds, it is critical to investigate the robustness of indirect monitoring methods against vehicle features. Therefore, an experimental study was performed, including two lab-scale bridges with different boundary conditions and a robot car that was capable of maintaining a variety of speeds and suspension systems. The proposed framework focuses on the identification of frequencies of the bridge using acceleration signals recorded on the car. It was demonstrated that by using a large set of passing cars with different features, the fundamental frequency of the bridge was captured. Successful identification of the deviation of fundamental frequencies between two bridges with relatively close frequency values proved that the framework is capable of detecting damage induced frequency changes of the bridge. Since this framework relies on the use of smartphones, it provides the opportunity to efficiently monitor a plethora of bridges in a metropolitan area with minimum cost.