Modeling and Simulation of 2D ARMA Terrain Models for Vehicle Dynamics Applications

The modeling of the terrain/road surface is a very important and challenging task in off/on-road vehicle dynamics simulation, road construction, and stochastic road surface assessment and identification. Various studies have been focused on on-road or on off-road terrain modeling, respectively. Most of these studies developed one-dimensional terrain/road models for specific applications. The approaches used to model paved surfaces differ, usually, from those employed to simulate unprepared terrain, since they must address different uncertainty ranges, linear or non-linear profiles, with normal or non-normal statistical distribution. From the practical point of view, one also needs to be able to reconstruct a representative terrain model from a limited set of measured data.

In this study, we are interested in reconstructing a terrain/road profile for a surface with known/standard roughness or required performance, for the purpose of using the terrain/road model for on-road and off-road vehicle dynamics applications. We employ the Auto-Regressive Moving Average (ARMA) series method to model a two-dimensional terrain profile. This method has previously been used to model one-dimensional terrain profiles. Two-dimensional ARMA terrain models with different orders are developed here to match International Roughness Index (IRI)/ Ride Number (RN) terrain/road standards. The stochastic terrain surface is analyzed in the frequency domain as well.