This research work was done in an effort to provide a tool for predicting on-center behavior of automobiles using computer simulation. It was discovered through this research effort that steering systems can have nonlinear characteristics that strongly affect the vehicle performance, especially in on-center driving. These nonlinearities must be modeled, characterized and simulated with care in order to predict vehicle performance.This research proposes a nonlinear steering system model that is based on the study of individual steering system components from both a hydraulic rack-and-pinion and a hydraulic recirculating ball power steering system. Included in the model are friction, nonlinear stiffness characeristics of the steering system and nonlinear boost curve shape. In addition, a means by which to characterize the steering system model for each steering system type is provided. Also, a methodology for the interpretation of certain static vehicle study results is developed in the context of including a steering system model in a total vehicle model.As a means to validate the proposed model, objective on-center performance of two test vehicles was measured. One vehicle had a rack-and-pinion steering system and the other a recirculating ball system. The test maneuvers included the standard Norman on-center test as well as low g swept sine input tests (random steer). For these two maneuvers, the proposed steering system model (along with a simplified linear bicycle vehicle model and simplified nonlinear tire model) is used to predict the on-center performace of the test vehicles. When driving the model with the same inputs measured from the test vehicles, the model predictions of lateral acceleration, yaw rate, steering wheel torque and even front road wheel angle are accurate. These comparisons indicate that inclusion of steering system nonlinearities is important to the prediction of on-center handling of vehicles.