• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2019, Vol. 55 ›› Issue (22): 93-102.doi: 10.3901/JME.2019.22.093

• 状态与参数估计 • 上一篇    下一篇

扫码分享

分布式驱动电动汽车双无迹卡尔曼滤波状态参数联合观测

金贤建1,2, 杨俊朋1, 殷国栋2,3, 王金湘3, 陈南3, 卢彦博3   

  1. 1. 上海大学机电工程与自动化学院 上海 200072;
    2. 吉林大学汽车仿真与控制国家重点实验室 长春 130025;
    3. 东南大学机械工程学院 南京 211189
  • 收稿日期:2019-08-23 修回日期:2019-11-07 出版日期:2019-11-20 发布日期:2020-02-29
  • 通讯作者: 殷国栋(通信作者),男,1976年出生,博士,教授,博士研究生导师。主要研究方向为车辆动力学及其控制、先进电动汽车、智能无人汽车等。发表SCI/Ei论文150余篇。E-mail:ygd@seu.edu.cn
  • 作者简介:金贤建,男,1986年出生,博士,讲师,硕士研究生导师。主要研究方向为车辆动力学及其控制、先进电动汽车、自动驾驶等。发表SCI/Ei论文40余篇。E-mail:jinxianjian@yeah.net
  • 基金资助:
    国家自然科学基金(51905329,51575103,U1664258)、省地联合招标成果转化(BA2018023)和汽车仿真与控制国家重点实验室开放基金(20181112)资助项目。

Combined State and Parameter Observation of Distributed Drive Electric Vehicle via Dual Unscented Kalman Filter

JIN Xianjian1,2, YANG Junpeng1, YIN Guodong2,3, WANG Jinxiang3, CHEN Nan3, LU Yanbo3   

  1. 1. School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072;
    2. State Key Laboratory of Automotive Simulation and Control, Jilin University, Changchun 130025;
    3. School of Mechanical Engineering, Southeast University, Nanjing 211189
  • Received:2019-08-23 Revised:2019-11-07 Online:2019-11-20 Published:2020-02-29

摘要: 为实时观测分布式驱动电动汽车行驶过程中的车身质心侧偏角等状态及车辆惯性参数如车辆质量、横摆转动惯量等信息,针对车辆惯性参数估计易敏感于载荷参数变化的挑战,发展面向载荷参数不确定(乘客或货物加载)的纵向、侧向、横摆运动的四轮车辆非线性动力学系统估计模型,在融合轮毂转矩等车载多传感器信息的基础上,将能适应强非线性系统的无迹卡尔曼滤波引入到车辆惯性参数估计中,设计车辆并联双无迹卡尔曼滤波状态参数联合观测系统,其中一个无迹卡尔曼滤波器观测车辆速度、车身质心侧偏角等状态,而另一个无迹卡尔曼滤波器观测车辆惯性参数。在CarSim/Matlab高保真环境中使用双移线、正弦工况对观测器在不同的载荷加载条件的可行性和有效性进行仿真验证,结果表明:该观测系统能实时观测车辆运行的状态及车辆惯性参数,即使在重载荷加载条件下仍具有较高的观测精度。

关键词: 分布式驱动, 电动汽车, 状态观测, 惯性参数, 双无迹卡尔曼滤波

Abstract: In order to estimate vehicle different states such as vehicle sideslip angle and vehicle inertia parameters such as vehicle mass and yaw moment of inertia in distributed drive electric vehicle, the uncertain load parameters (passenger or cargo loaded)-based four-wheeled vehicle nonlinear dynamics estimation model including longitudinal, lateral, and yaw motions is developed to deal with sensitive challenges of vehicle inertia parameters estimation due to load parameter changes. Based on vehicle multi-sensor data fusion from the hub torque and other measurements, the unscented Kalman filter that can adapt to the strong nonlinear system is introduced, and then the nonlinear observer with the dual unscented Kalman filter is designed, where the first unscented Kalman filter observes vehicle different states and the other parallel unscented Kalman filter is used to observe the vehicle inertia parameters. Simulations for double lane change and sine manoeuvres are carried out to evaluate the feasibility and effectiveness of observer under different load parameters in CarSim/Matlab environment. The results show that the proposed observation system can observe vehicle states and inertia parameters in real time, and it has high observation accuracy even under large load condition.

Key words: distributed drive, electric vehicle, state observation, inertia parameters, dual unscented Kalman filter

中图分类号: