Identification of vehicle axle loads based on visual measurement

Abstract

Monitoring vehicle axle loads is very important for preventing infrastructure degradation and traffic accidents. However, developing an accurate, cost-effective, and easy-to-implement axle load identification technology still remains challenging. In this study, a vehicle axle load identification method based on visual measurement is proposed. The principle of the method is that each axle load of the vehicle is determined based on the vehicle unsprung masses, the ratio of vehicular sprung mass to unsprung mass, and the centroid position of the sprung mass. While the vehicle unsprung masses can be treated as a known with the vehicle model identified, the mass ratio can be predicted from the visually captured free damped vibrations of the vehicle as it passes over a speed bump. To be more specific, the vibration responses of a vehicle passing over a speed bump are firstly measured by a camera, and the vehicle system matrix, which is composed of the mass matrix, stiffness matrix, and damping matrix, can be obtained from the vehicle responses. The axle load can then be determined based on the element ratios in the system matrix of the vehicle whose unsprung masses are known. The performance of the proposed method was evaluated by numerical simulations and field tests. The results show that the errors of the simulation results are all less than 1% under ideal conditions. In the field test, the average error in identifying axle loads at different vehicle speeds was 7.01% when the vehicle speed was below 20 m s−1, which proved the accuracy and effectiveness of the proposed method. This study has demonstrated a novel application of computer vision technology to identify the axle loads of moving vehicles. The proposed method does not require installation of sensors on the roadway or the vehicle, making it a promising alternative for traditional weigh-in-motion systems.