Fast calibration for vibration-based pavement roughness measurement based on model updating of vehicle dynamics

Abstract

ABSTRACT The vibration-based pavement roughness measurement method has garnered significant attention in recent years due to the development of low-cost sensors. The vehicle calibration process is necessary before field measurement due to the effect of changeable vehicle dynamics, restricting the large-scale implementation of pavement roughness measurement. This paper proposes a novel calibration method for vibration-based pavement roughness measurement. It can quickly calibrate vehicle dynamic parameters through a few controlled impact tests. The model updating technique is employed to estimate vehicle dynamic parameters, resulting in the development of a full-car numerical model that accurately simulates the vibration of the test vehicle. A road-vehicle coupled simulation system is introduced, incorporating randomly generated road profiles and the calibrated full-car model. This system can simulate vibration responses under various conditions. A linear relationship between IRI (International Roughness Index) and vehicle vibration indices (e.g., RMS – Root Mean Square of vertical acceleration) is estimated based on simulated results. The efficiency and accuracy of the calibration method are validated on actual roads in Shanghai, China. The average IRI difference between the ground truth and test data reached 0.5 m/km in a 100 m road section, demonstrating the reliability and practicality of this fast calibration method.