ABSTRACT
The paper deals with the identification of physical parameters in vehicle dynamics from measurement data. The dynamical behaviour of the vehicle is modeled by a finite-dimensional, linear, time-invariant, time-continuous, mechanical system. The identification procedure has respect to several aspects such as the inclusion of hydraulic or electrical substructures to the mechanical system, the efficient use of the available sensor configuration, the necessity of external feedback to stabilize the system, and the finiteness and the noise corruption of the data set. An instrumental variable identification algorithm is constructed for the solution of a related time-discrete parameter estimation problem which leads by subsequent backtransformation and reconstruction techniques to the desired physical parameters of the original time-continuous mechanical system.
The new efficient algorithm of parameter estimation is applied to a magnetically levitated vehicle. The related suspension control system is a representative of a big class of vehicle dynamic problems where the proposed identification method can be successfully applied because all typical difficulties appear: different types of sensors, stabilizing feedback, noise, and a dynamic excitation mechanism model according to the dynamic behaviour of electro-magnets. This example illustrates the efficiency of the new method of parameter estimation in vehicle dynamics.
