Parameter Estimation of Coupled Road-Vehicle Systems

Abstract

Using a novel gray-box identification paradigm we can reconstruct both vehicle and road model parameters with one set of measurements. In the paper we present a non-linear parameter estimation method to determine the parameters of a full-car suspension system. Since there are non-measured variables that are necessary for the identification, special numerical techniques need to be applied, such as a numerical integration of the measured signals. It is also shown that the selection of the sampling time might be critical in this type of application. Based on the results of this identification procedure the road disturbance can be reconstructed. The road roughness estimation is based on this signal using autoregressive model.