Propagation of Error Applied to Linear Vehicle Dynamics

This paper applies basic statistics to the simulation of vehicle dynamics in the time domain and the frequency domain. The methods presented here yield an expectation for the variation of the computed results as a function of variation in input parameters. Applications include important steady state measures of vehicle performance such as understeer gradient and yaw rate gain. Follow up analysis includes measures of transient response in the time domain and of amplitude and phase relationships in the frequency domain. There are several potential applications for this methodology, perhaps most important, an understanding of the consequences of parameter uncertainty on the credibility of simulated results.