Removal of drive-by vehicle components using transfer functions with experimental investigation

In the past two decades, the application of instrumented sensing vehicles for the structural health monitoring of transport infrastructure has attracted increasing interest due to the low cost and high efficiency. This is known as an indirect method, vehicle scanning method, or drive-by monitoring. The recent advances in intelligent vehicles are accelerating the explorations and applications of drive-by monitoring for transport infrastructure assessment and adding further value of smart vehicle networks in terms of data for informed decision-making by transport infrastructure asset owners. However, the removal of undesired vehicle components from vehicle measurements remains one of the key challenges in drive-by monitoring. To realize the practical implementation of drive-by monitoring towards net zero transport infrastructure, this paper proposes novel vehicle transfer functions to isolate: (i) the bridge information, and/or (ii) contact point information, from drive-by vehicle measurements via smartphone sensing. Both theoretical, empirical, and averaged empirical transfer functions are implemented to remove unwanted vehicle information. The sensitivity to the accuracy of the proposed algorithm was investigated regarding different vehicle speeds and varying road conditions. Additionally, laboratory testing with a one-fifth scale model of Loughbrickland bridge on the A1 road (Belfast to Dublin) in Northern Ireland was carried out to verify its practical feasibility, and experimental results also demonstrate that proposed transfer functions work well in mitigating vehicle components.