Enhancement of Vehicle Lateral Stability by Non-linear Optimal Control of Yaw Dynamics

Non-linear characteristic of tire forces is the main cause of vehicle lateral dynamics instability, while direct yaw moment control is an effective method to recover the vehicle stability. In this work, an optimal non-linear controller for yaw dynamics to track a linear model, limited by road friction coefficient, is developed. The yaw rate response of an extended 2DOF non-linear model is first predicted by Taylor series expansion and then a control law is introduced by minimizing the local differences between the predicted and the desired responses. Here, the main properties of the proposed control law and its advantages over the other conventional control methods are discussed. The derived control law has an analytical form which is easy to apply and handles the control input saturation explicitly. The effectiveness of the designed controller is investigated through simulations of severe lane change maneuvers, using a developed non-linear full vehicle dynamic model. The simulation results show that the vehicle stability can be remarkably improved when the optimal non-linear controller is applied.