A least-squares regression based method for vehicle yaw moment of inertia estimation

Abstract

Vehicle yaw moment of inertia is an important parameter for many vehicle dynamic models and control systems yet it is usually difficult to estimate. A methodology in estimating the vehicle yaw moment of inertia is presented in this article by studying the linear relationship between vehicle lateral acceleration, yaw acceleration, and rear wheel lateral tire forces. This linear relationship is derived by manipulating the equations in the single-track model such that the front wheel force disappears in the equation. Based on the linear relationship, the common global positioning system (GPS) measurement error-antenna bias angle, can be tuned based on the symmetric assumption of vehicle left and right dynamics and a least-squares regression (LSR). A lag-like lateral tire force model is applied to capture the transient dynamics of the lateral tire force. The parameter determining relaxation length of the tire model can also be tuned based on a similar LSR method. Finally, the yaw moment of inertia can be estimated from an LSR again after knowing the estimated rear wheel lateral force. Simulation results in CarSim <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> show that this proposed method is capable of generating reasonable estimations of yaw moment of inertia without knowing the front road wheel steering angle.