Mixed H2/H∞ road feel control of EPS based on genetic algorithm

Abstract

In view of the existence of uncertainties such as system model and disturbance signal in the electric power steering (EPS) system, and the demand for system dynamic performance, the mixed H2/H∞ controller based on genetic algorithm is proposed. In order to obtain satisfactory steering feel, robust performance and steering stability, models of EPS system and a two-degree-of-freedom car are set up, then the state space model and the augmented matrixes are built. The H∞ method is introduced to minimize the effect of disturbances on the outputs, and the H2 method is applied to optimizing the system performance based on genetic algorithm. The simulation results show that the modified mixed H2/H∞ controller, which synthesizes the advantage of H2 control and H∞ control, has better robust performance and robust stability. The designed controller can attenuate the noises and disturbances caused by road random motivation, torque sensor measurement and model parameter uncertainty, enabling the driver to obtain satisfactory road feel.