Model-based feedforward control for suppressing torque oscillation of electric power steering system

Abstract

Oscillation suppression is essential for the stability design of electric power steering (EPS) systems. The stability controller module in EPS controller is the key to solve the stability control problem of EPS system. This paper proposes a new method of stability analysis and stability controller module design for EPS systems. Furthermore, the dynamic characteristics of the EPS system are analyzed, and two critical factors on the resulting EPS stability, that is, large assist and variable assist gain are investigated experimentally. The transfer function from steering torque to sensor torque is redefined. A new transfer function is proposed for measuring the effect of variable assist gain on system performance. Based on the above factors and transfer functions, constraints on the stability controller design are proposed. Then the optimal parameters in the controller are obtained by maximizing an objective function including phase margin, gain margin, and crossover frequency. It is concluded from simulations and bench tests that the proposed stability controller can significantly reduce the torque oscillation of the EPS system.