汽车工程 ›› 2022, Vol. 44 ›› Issue (1): 8-16.doi: 10.19562/j.chinasae.qcgc.2022.01.002

所属专题: 智能网联汽车技术专题-规划&控制2022年

• • 上一篇    下一篇

基于自适应动态滑模控制的智能汽车纵向巡航控制

赵健,杜金朋,朱冰(),王志伟,陈志成,陶晓文   

  1. 吉林大学,汽车仿真与控制国家重点实验室,长春  130022
  • 收稿日期:2021-08-23 修回日期:2021-09-28 出版日期:2022-01-25 发布日期:2022-01-21
  • 通讯作者: 朱冰 E-mail:zhubing@jlu.edu.cn
  • 基金资助:
    国家自然科学基金(51775235);吉林大学研究生创新基金项目(101832020CX130)

Longitudinal Cruise Control of Intelligent Vehicles Based on Adaptive Dynamic Sliding Mode Control

Jian Zhao,Jinpeng Du,Bing Zhu(),Zhiwei Wang,Zhicheng Chen,Xiaowen Tao   

  1. Jilin University,State Key Laboratory of Automotive Simulation and Control,Changchun  130022
  • Received:2021-08-23 Revised:2021-09-28 Online:2022-01-25 Published:2022-01-21
  • Contact: Bing Zhu E-mail:zhubing@jlu.edu.cn

摘要:

为消除参数不确定性和外部干扰等因素对智能汽车纵向巡航控制的影响,提出一种基于自适应动态滑模的纵向巡航控制方法。建立以广义纵向力导数项为控制输入的车辆纵向动力学模型,基于反步法构建保证车速与纵向加速度同时收敛于期望值的新型滑模函数;在此基础上,设计了期望广义纵向力的动态滑模控制律,并利用RBF神经网络对控制律中的未知干扰量进行自适应补偿;设计执行器选择模块将期望广义纵向力转换为执行层期望控制输入;进行仿真对比与实车测试,结果表明,所提出的智能汽车纵向巡航控制方法能有效地消除参数不确定性和外部干扰的影响,改善了传统滑模控制的抖振现象,实现对期望巡航车速的稳定准确跟踪。

关键词: 巡航控制, 动态滑模控制, 自适应控制, 反步法, RBF神经网络

Abstract:

To eliminate the effects of parameter uncertainty and external disturbances on the longitudinal cruise control of intelligent vehicles, a longitudinal cruise control method based on adaptive dynamic sliding mode is proposed. The vehicle longitudinal dynamics model is established with the derivative term of generalized longitudinal force as control input, and a novel sliding mode function is constructed based on the backsteping method to ensure that the vehicle speed and longitudinal acceleration converge to the desired value at the same time. On this basis, the dynamic sliding mode control law of the desired generalized longitudinal force is designed and the unknown disturbances in the control law are compensated adaptively by using RBF neural network. The actuator selection module is designed to convert the desired generalized longitudinal force into the desired control input in actuator layer, with the comparative simulation and real vehicle test conducted. The results show that the longitudinal cruise control method proposed for intelligent vehicles can effectively eliminate the influences of parameter uncertainty and external disturbance, improve the chattering of traditional sliding mode control, and achieve the stable and accurate tracking of the desired cruise speed.

Key words: cruise control, dynamic sliding mode control, adaptive control, backstepping method, RBF neural network