汽车工程 ›› 2025, Vol. 47 ›› Issue (11): 2238-2249.doi: 10.19562/j.chinasae.qcgc.2025.11.017

• • 上一篇    

考虑轮胎侧偏特性的汽车线控转向系统前轮转角估计方法

孙晓强(),丁佳伟,汤浩然,蔡英凤,陈龙   

  1. 江苏大学汽车工程研究院,镇江 212013
  • 收稿日期:2025-02-14 修回日期:2025-03-22 出版日期:2025-11-25 发布日期:2025-11-28
  • 通讯作者: 孙晓强 E-mail:sxq@ujs.edu.cn
  • 基金资助:
    国家重点研发计划项目(2023YFB2504502)

Estimation of Front Wheel Steering Angle for Vehicle Steer-by-Wire System Considering Tire Cornering Characteristics

Xiaoqiang Sun(),Jiawei Ding,Haoran Tang,Yingfeng Cai,Long Chen   

  1. Institute of Automotive Engineering,Jiangsu University,Zhenjiang 212013
  • Received:2025-02-14 Revised:2025-03-22 Online:2025-11-25 Published:2025-11-28
  • Contact: Xiaoqiang Sun E-mail:sxq@ujs.edu.cn

摘要:

前轮转角是汽车线控转向系统(SBW)的主要跟踪控制目标,其精准估计是线控转向系统冗余设计的重要内容。针对传统方法无法保证汽车大范围行驶工况下前轮转角估计精度的问题,本文提出了一种考虑轮胎非线性侧偏特性的SBW前轮转角估计方法。首先,构建了2自由度横摆-侧倾车辆动力学模型和SBW执行模型,并完成了轮胎非线性侧偏特性的分段仿射(PWA)辨识;随后,推导了PWA系统状态方程,并采用最大相关熵平方根容积卡尔曼滤波(MCSCKF)算法,设计了汽车SBW前轮转角估计策略,以提升系统大范围状态切换时的估计精度;最后,基于CarSim和Simulink建立了汽车SBW前轮转角估计性能联合仿真验证平台,结合两种典型工况,对前轮转角估计效果进行了验证。结果表明,正弦转向工况下,MCSCKF算法相较于EKF、MCEKF算法,估计误差最大降幅分别为66.3%和41.1%;双移线转向工况下,MCSCKF算法相较于EKF、MCEKF算法,估计误差最大降幅分别为64.3%和38.2%,验证了所提出方法能够有效提升汽车大范围行驶工况下的前轮转角估计精度。

关键词: 线控转向系统, 前轮转角估计, 轮胎侧偏特性, 分段仿射辨识, MCSCKF算法

Abstract:

The steering angle of the front wheels is the primary tracking control objective in vehicle steer-by-wire (SBW) system, and its accurate estimation is a critical aspect of the redundancy design in such systems. For the problem that traditional methods fail to ensure the estimation accuracy of the front wheel steering angle across a wide range of driving conditions, an estimation method for the front wheel steering angle in SBW systems is proposed, which takes into account of the nonlinear characteristics of tire lateral deflection. Firstly, a two-degree-of-freedom yaw-roll vehicle dynamics model and a SBW system model are constructed, followed by the completion of piecewise affine (PWA) identification for the tire nonlinear cornering characteristics. Subsequently, the state equation of the PWA system is derived, and a front wheel steering angle estimation strategy for the vehicle SBW system is designed using the maximum correlation square root cubature Kalman filter (MCSCKF) algorithm to enhance the estimation accuracy during extensive state transitions. Finally, a co-simulation validation platform for the performance estimation of the front wheel steering angle in vehicle SBW systems is established based on CarSim and Simulink. The effectiveness of the front wheel steering angle estimation is verified in conjunction with two typical operational conditions. The results show that under sinusoidal steering conditions, the MCSCKF algorithm has a maximum reduction in estimation error of 66.3% and 41.1% compared to EKF and MCEKF algorithms, respectively. Under the dual lane steering condition, the MCSCKF algorithm has a maximum reduction in estimation error of 64.3% and 38.2% compared to the EKF and MCEKF algorithms, respectively, which verifies that the proposed method can effectively improve the accuracy of front wheel steering angle estimation under a wide range of driving conditions for automobiles.

Key words: steer-by-wire system, front wheel steering angle, tire cornering characteristics, piecewise affine identification, maximum correlation entropy square root volume Kalman filter