Rack force fault tolerance estimation of steer-by-wire system under different resolver faults based on sensor flows

Abstract

In the context of vehicle intelligence, steer-by-wire (SbW) has great potential for development and has become a current research hotspot. The key challenge for SbW systems is to obtain the nonlinear rack force generated by tire-road interaction, which is the source of the driver's real steering feel and the main disturbance of steering angle tracking control. Existing rack force estimators either are only suitable for non-aggressive steering or assume that the measured signals are reliable. This paper proposes a high-precision rack force estimation algorithm that is robust to primary sensor faults of a SbW system. Further, a rack force estimator based on the established high-order SbW mechanism model is designed, and a detailed stability proof and feedback gain selection guidance are given. The resolver sensor model is provided and analyzed, and a fault detection method based on discrete Fourier transform (DFT) and data fusion is developed, which has good real-time performance and high accuracy. The performance of rack force estimator was verified and compared by simulations and hardware-in-loop (HIL) experiments with various steering maneuvers and different resolver faults. The experiment results show that the proposed rack force estimation method can not only effectively improve the estimation accuracy even in aggressive steering condition and uneven road profiles, but also achieve accurate estimation of the rack force in the case of resolver faults. The proposed method obtains high-precision and high-reliability rack force estimation results while being easily embedded applied in engineering practice.