Abstract
A systematic approach based on stochastic optimal control and estimation theories for the optimal design of active suspension systems for a tractor-semitrailer vehicle is presented. The vehicle-road system dynamics combined with the human perception of random vibrations are cast in the standard linear quadratic Gaussian format. The time delays of the stochastic excitation process of the multi-axled vehicle are represented by a first order Padé approximation. The optimal controller/observer is derived which minimizes the mean-squared values of performance variables reflecting the human comfort, cargo safety, suspension working space, roadholding ability and control forces. The advantages of optimally controlled suspension systems over their passive counterparts are indicated.