A recursive model for nonlinear spring-mass-damper estimation of a vehicle localized impact

In this paper, a recursive autoregressive model is presented to estimate modal parameters of the crash pulse recorded during a full-scale vehicle crash test. Those parameters can be further utilized to characterize a physical model, so called Maxwell model, which is composed of a serial spring-mass-damper model to simulate a vehicle crash event. Results produced by this viscoelastic system closely follow the reference vehicle's kinematics. Thanks to the application of nonlinear model's parameters estimated by RARMAX model it is possible to gain a closer insight into the nature of the vehicle collision. To evaluate the correctness of this methodology, the comparative analysis between the original vehicle's kinematics and behavior of a Maxwell model with the nonlinear parameters estimated by RARMAX model is presented.