Vehicle-Based Sensing for Bridge Scour Damage Detection

Drive-by monitoring using instrumented vehicles enables indirect detection of bridge damage, offering a practical alternative to traditional sensor installations on a bridge and manual inspections. This study focuses on a drive-by method to detect permanent scour settlements at bridge supports. A novel self-calibration approach is introduced to estimate vehicle properties through an optimization process using inertial sensor data collected from the vehicle. Once these properties are identified, the inverse Newmark-Beta algorithm is applied to infer the apparent road profile, i.e., the profile experienced by the vehicle, which represents a combination of the road surface irregularities and the bridge’s deflection beneath a moving vehicle wheel. Scour damage is then identified by measuring differences in the inferred profiles between undamaged and damaged structural states. The approach is validated through simulations and field testing on a near-full-scale bridge at the European Commission’s Joint Research Centre in Ispra, Italy. Simulations are conducted to evaluate the method’s sensitivity to scour settlements as small as 2 mm, using vehicles with different properties and speeds. Field experiments demonstrate successful detection of a settlement of 4 mm across repeated runs. The apparent profile patterns consistently differ between healthy and damaged bridge conditions, with high repeatability noted for multiple of runs of the same vehicle. These findings highlight the robustness of the approach and its suitability for real-world application.