Parameter estimation in analytical models of automotive vehicles and fault diagnosis

This paper describes a “global” model based method for fault detection and physical diagnostics via parameter identification in dynamical ground vehicle problems. The procedure has been applied to specific absorbers and to the dynamical behaviour of lateral motion comprising roll-yaw-front, rear and center of mass skid dynamics, in which all the physical parameters involved and to be estimated have then a physical significance. Changes in the model process and in its outputs and signal parameters are used for fault detection through identification of changes in their values in subsets of characteristic ranges. Here nine and twenty physical parameters are respectively considered in the first and second