• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2019, Vol. 55 ›› Issue (6): 156-165.doi: 10.3901/JME.2019.06.156

• 运载工程 • 上一篇    下一篇

基于自适应扩展卡尔曼滤波的分布式驱动电动汽车状态估计

张志勇1,2, 张淑芝2, 黄彩霞3, 张刘铸2, 李博浩2   

  1. 1. 工程车辆安全性设计与可靠性技术湖南省重点实验室 长沙 410114;
    2. 长沙理工大学汽车与机械工程学院 长沙 410114;
    3. 湖南大学机械与运载工程学院 长沙 410082
  • 收稿日期:2018-03-18 修回日期:2018-10-09 出版日期:2019-03-20 发布日期:2019-03-20
  • 通讯作者: 张志勇(通信作者),男,1976年出生,博士,副教授。主要研究方向为电动汽车技术,车辆动力学及控制。E-mail:zzy04@163.com
  • 作者简介:张淑芝,女,1993年出生。主要研究方向为电动汽车技术。E-mail:1049889005@qq.com
  • 基金资助:
    国家自然科学基金(51675057)和湖南省教育厅(15B008,16C0906)资助项目。

State Estimation of Distributed Drive Electric Vehicle Based on Adaptive Extended Kalman Filter

ZHANG Zhiyong1,2, ZHANG Shuzhi2, HUANG Caixia3, ZHANG Liuzhu2, LI Bohao2   

  1. 1. Key Laboratory of Lightweight and Reliability Technology for Engineering Vehicle, Changsha 410114;
    2. College of Automobile and Mechanical Engineering, Changsha University of Science and Technology, Changsha 410114;
    3. College of Mechanical and vehicle Engineering, Hunan university, Changsha 410082
  • Received:2018-03-18 Revised:2018-10-09 Online:2019-03-20 Published:2019-03-20

摘要: 纵向车速和质心侧偏角是车辆主动安全控制系统的关键参考状态信号,通常采用卡尔曼滤波算法估计。当系统噪声和测量噪声的统计特性存在不确定性时,不仅估计精度会降低,甚至导致估计器发散。结合分布式驱动电动汽车4个车轮转矩和转速可直接测量的特点,提出一种车辆状态自适应扩展卡尔曼滤波估计方法。基于量纲一化新息平方实现车辆状态估计有效性检测,提出滑动窗口长度自适应调整规则;根据新息统计特性提出卡尔曼滤波增益和状态估计误差协方差矩阵的自适应调整策略,及基于车辆状态估计稳态误差和动态响应速度的自适应参数确定原则。数值仿真和试验证明,所提出的车辆状态估计方法,不仅估计精度较高,而且实时性和易用性较强。

关键词: 电动汽车, 分布式驱动, 扩展卡尔曼滤波, 状态估计, 自适应控制

Abstract: Longitudinal velocity and sideslip angle are the key referent state signals for vehicle active safety control system, and are usually estimated by Kalman filtering algorithm. The uncertainties of the statistical characteristics of system noise and measurement noise may cause filter to deviate or even diverge. Using the characteristics of the torques and speeds of four wheels can measurement directly in a distributed drive electric vehicle, an adaptive extended Kalman filtering method for vehicle state estimation is proposed. With normalized innovation square, the validity of vehicle state estimation is detected, and an adaptive adjustment rule of sliding window length is designed. An adaptive adjustment strategy of the gain of Kalman filter and the covariance matrix of state estimation error are proposed based on the statistical characteristics of innovation. The determination principle of adaptive parameters based on the steady-state error of vehicle state estimation and the dynamic response speed is determined. The numerical simulation and experiment can prove that the proposed algorithm of vehicle state estimation not only can improve estimation accuracy, but also has advantages of high real-time and easy to implement.

Key words: adaptive control, distributed drive, electric vehicle, extended Kalman filter, states estimation

中图分类号: