Vehicle overload seriously affects the traffic safety, damages the road infrastructure, and the road service life is reduced. There are many shortcomings in the current detection methods for overloaded vehicles. Traditional static weighing stations are relatively fixed and affect the road traffic efficiency; the cost of weighing in motion station is high, while the precision is not enough; on-board weighing electronic products require the modification of vehicles, which is difficult to promote.
In this paper, a non-contact overload detection method based on body vibration is proposed. After the analyzation of the body vibration response of dynamic vehicles under specific vibrational excitation, the load value data can be obtained combined with standard parameters and mathematical calculation model. Firstly, the body vibration response model under specific excitation is established. Then, roadbed facilities are arranged according to specific requirements, cameras calibration are finished, and the identification environment is built. Machine vision technology is used to identify the vibration track of the characteristic point on the vehicle body in the vertical direction. The vibration response characteristic parameters are extracted using the established response model. Finally, the vehicle load value data can be obtained by resolve the characteristic parameters. Compared with the rated load data in the database, the overload judgment of the vehicle is obtained.
In the experiment part, the road speed-control hump was used as the vibration excitation source. The vehicle experiments were carried out with Dongfeng Aeolus S30 and Yuejin Shangjun X500. The results show that the load identification error can be controlled within 20%–30%. This method above can detect overload vehicles without affecting the traffic efficiency and also has certain guiding significance for the development of intelligent vehicles.