Road Roughness Modelling by Using Spectral Factorization Methods

Abstract

In this paper, road roughness model for the right and the left tracks are constructed by utilizing two methods. In the first method, from time-domain measurements the empirical auto and cross-power spectral densities of the left and the right tracks are estimated by using the Welch method. Next, the road roughness is decomposed into three stochastically independent random processes and the linear-shape filter parameters of the right and the left tracks are estimated via a recently developed subspace-based identification algorithm. In the second method, a multi-input/multi-output subspacebased identification algorithm is directly used, without employing spectral decomposition to derive the spatial-domain models of the right and the left tracks from the Welch power spectral density estimates. Profile measurements for a specific road section from the Long Term Pavement Performance archive of the University of Michigan Transportation Research Institute is used to demonstrate the successfull application of the subspace approach.