Model Predictive Control for Vehicle Yaw Stability

1. Abstract Yaw stability of an automotive vehicle in a turn is critical to the overall stability of the vehicle. In this paper, we present a method of vehicle stability control (VSC) based on Model Predictive Control (MPC). Conventional VSCs work passively as they detect excessive yaw rate or slip angle of a vehicle. However, in many cases, when excessive yaw rate or slip angle of a vehicle is detected, the vehicle is already in unstable states. Using the MPC scheme, the proposed controller can actuate brakes to generate correction moment in advance of vehicle being unstable by predicting vehicle movement of several hundred milliseconds ahead. The differences of desired vehicle states from the bicycle model and the estimated vehicle state are minimized by applying the MPC scheme. The performance of the proposed method is evaluated using the vehicle dynamics software CARSIM.