Road Roughness Estimation Method Based On Vibration Test Of Vehicle-Road Coupled System

Pavement roughness is an important index to evaluate the quality of pavement,which directly affects the performance of the road.The existence of pavement roughness will cause vehicle vibration,which will affect the stability and comfort of the vehicle.At the same time,the vibration of the vehicle will act on the pavement,which will make the pavement subjected to greater load and accelerate the damage of the pavement.Compared with traditional instrument detection,the road roughness estimation method based on vehicle dynamic response and non-linear filtering algorithm has the advantages of simple operation,low cost and high accuracy.However,the current research on vehicle-road model focuses on vehicle dynamics or road dynamics alone,which is difficult to truly reflect the actual vehicle-road dynamic behavior.Vehicle suspension shock absorber,spring,tire and other non-linear factors will also cause unacceptable errors in analysis and calculation.Therefore,the paper establishes a vehicle-road coupling system model considering vehicle nonlinearity,and uses the improved unscented Kalman filter(UKF)method to study the estimation of road roughness.The specific research contents are as follows:Firstly,various simulation methods of pavement roughness are summarized.The time-domain signal conversion power spectral density and Fourier inverse transformation method are used to simulate the pavement roughness by using MATLAB software.The simulation of pavement roughness of different pavement grades is realized.Secondly,based on the improved UKF method which is used to update the non-linear finite element model by Astroza et al.,a new improved UKF method,NOWUKF method,is proposed.The traditional UKF method and two improvements under different window lengths,different observation noises and different models are compared and analyzed.The result of simulation shows that NOW-UKF has better recognition effect when it loses a certain time cost.Thirdly,a vehicle-road coupled system model considering vehicle nonlinearity is established.The pavement roughness is identified by UKF method and NOW-UKF method,respectively.An improved algorithm for identifying the pavement roughness by using the response information of multiple degrees of freedom is proposed,which improves the recognition accuracy.