Smartphone-based vehicle-road contact-point response identification with unknown vehicle parameters

Drive-by sensing has the potential to become a city-level monitoring tool, aiding in the management of ageing transportation infrastructure, such as roads and bridges. However, drive-by measurements collected from sensors installed on moving vehicles are predominantly influenced by vehicle vibration components, which significantly hinder the identifiability of road and bridge characteristic information. Recognizing that the vehicle-road contact-point response is unaffected by vehicle vibration components, this paper proposes a vehicle-road contact-point response identification method with unknown vehicle parameters. The identification of vehicle-road contact-point responses and vehicle parameters is formulated as an optimization problem, incorporating a physics-informed constraint, namely, the front and rear tires experience the same road profile when traveling in a straight path. This constraint helps regularize the optimization process. The feasibility and advantages of the proposed approach are validated numerically and through field road testing. Results demonstrate that the developed method can efficiently identify the contact-point response with an acceptable accuracy and is robust to varying vehicle speeds, arbitrary smartphone placements on the vehicle, and different vehicle models. This study introduces two key innovations: (i) a practical signal pre-processing pipeline, including a data-driven sensor re-orientation method, to improve drive-by measurement data quality and enable its application in real-world field environments, and (ii) the integration of physical information into a data-driven optimization model to estimate vehicle–road contact-point responses with unknown vehicle parameters. These advancements address fundamental challenges in drive-by sensing and significantly enhance its feasibility for city-level transportation infrastructure health monitoring. • This paper proposes a novel approach for contact-point response identification. • Smartphone measurement and optimization with physics-informed constraint are used. • Vehicle parameters are not required for this drive-by sensing based SHM approach. • The proposed approach is validated through numerical and field road tests. • The contact-point responses can be identified with a very good accuracy.