Vehicle Parameter Estimation Based on Full-Car Dynamic Testing

Yu Zhang,Lifu Wang,Bangji Zhang,Nong Zhang
DOI: https://doi.org/10.4271/2015-01-0636
2015-01-01
Abstract:Effectively obtaining physical parameters for vehicle dynamic model is the key to successfully performing any computer-based dynamic analysis, control strategy development or optimization. For a spring and lump mass vehicle model, which is a type of vehicle model widely used, its physical parameters include sprung mass, unsprung mass, inertial properties of the sprung mass, stiffness and damping coefficient of suspension and tire, etc. To minimize error, the paper proposes a method to estimate these parameters from vehicle modal parameters which are in turn obtained through full-car dynamic testing. To verify its effectiveness, a visual vehicle with a set of given parameters, build in the Adams(Automatic Dynamic Analysis of Mechanical Systems)/Car environment, is used to perform the dynamic testing and provide the testing data for the parameter estimation. Two parameter estimation methods based on half-car and full-car models are illustrated and compared with the parameters given in the Adams. Furthermore, the full-car parameter estimation method is applied on a bus, and the simulation results using the estimated parameters are compared with the data collected from real vehicle tests.
What problem does this paper attempt to address?