Modeling and identification of the vehicle suspension characteristics using local linear model trees

An automotive MacPherson suspension unit is identified and modeled using a radial basis function network with local linear weighting functions (LOLIMOT). The output of the network represents the nonlinear force characteristic of the spring-and-damper unit in dependency of the measured spring travels at the front axle and the corresponding spring travel velocities. The network parameters are interpreted physically by comparing the five local linear models identified with the network to a physical second-order model of the wheel suspension unit. Implemented in an overall vehicle model the network or its look-up table representation represents an adaptable model of the nonlinear vehicle suspension characteristics.