Abstract
This paper introduces a new method for estimation of expected values and standard deviations of signals in the nonlinear lateral control system of a snowplow, including the effects of disturbances acting on the vehicle. This method is based on Fourier discrete decomposition of time-varying signals and systems using an orthonormal basis of functions. Statistical linearization of the nonlinear element is combined with Spectral Methods that allow handling of time-varying systems and transient regimes of linear time-invariant (LTI) systems. This is in contrast to traditional methods of statistical linearization where one typically analyzes only steady-state regimes of LTI systems.