Design and evaluation of an observer-based disturbance rejection controller for electric power steering systems

Abstract

The goal of this paper is to develop an observer-based disturbance rejection electric power steering (EPS) controller to provide steering assistance and improve the driver’s steering feel. For the purpose of control design, a control-oriented model of a vehicle with a column-assist EPS system is developed and verified against a high-fidelity multibody dynamics model of the vehicle. The high-fidelity model is used to mimic vehicle dynamics to study controller performance in realistic driving conditions. Then, a linear quadratic Gaussian approach is used to design an EPS optimal controller, in which a Kalman filter estimates the unmeasured steering system’s states and external disturbance. A new formulation for the linear quadratic regulator objective function is proposed to take advantages of the known information about the system dynamics to attenuate the disturbance and magnify the driver’s torque. Finally, the EPS controller is applied to the high-fidelity vehicle model in a software-in-the-loop simulation to evaluate its robustness and performance under realistic conditions. The results show that the proposed controller can effectively reduce the disturbance induced in the steering rack, and simultaneously magnify the driver’s steering torque by use of a bi-linear EPS characteristic curve. Then, to show the disturbance rejection properties of this EPS controller, its performance is compared with H2/ H∞ and PID control designs using time and frequency domain analysis.