汽车工程 ›› 2024, Vol. 46 ›› Issue (10): 1853-1862.doi: 10.19562/j.chinasae.qcgc.2024.10.012

• • 上一篇    下一篇

人机共驾型车道保持鲁棒控制

章军辉1,2,3,4(),郭晓满2,4,刘禹希2,4,郑明强2,4,钱宇晗2,4,丁羽璇2,4   

  1. 1.常熟理工学院电气与自动化工程学院,苏州 215500
    2.无锡物联网创新中心有限公司,无锡 214029
    3.江苏省工业机器人复杂工艺智慧控制工程研究中心,苏州 215500
    4.江苏省物联网创新中心昆山分中心,苏州 215347
  • 收稿日期:2024-05-05 修回日期:2024-06-29 出版日期:2024-10-25 发布日期:2024-10-21
  • 通讯作者: 章军辉 E-mail:zjh34@mail.ustc.edu.cn
  • 基金资助:
    江苏省博士后科研资助计划(2020Z411)

Driver-Automation Shared Lane-Keeping Robust Control

Junhui Zhang1,2,3,4(),Xiaoman Guo2,4,Yuxi Liu2,4,Mingqiang Zheng2,4,Yuhan Qian2,4,Yuxuan Ding2,4   

  1. 1.School of Electrical and Automatic Engineering,Changshu Institute of Technology,Suzhou 215500
    2.Wuxi Internet of Things Innovation Center Co. ,Ltd. ,Wuxi 214029
    3.Jiangsu Engineering Research Center of Industrial Robot Complex Process Intelligent Control,Suzhou 215500
    4.Kunshan Department,Jiangsu Internet of Things Innovation Center,Suzhou 215347
  • Received:2024-05-05 Revised:2024-06-29 Online:2024-10-25 Published:2024-10-21
  • Contact: Junhui Zhang E-mail:zjh34@mail.ustc.edu.cn

摘要:

为更好地让共驾型车道保持控制系统能够预判驾驶人的转向行为,本文提出了一种间接式共驾型车道保持鲁棒控制算法。首先引入了仿驾驶人转向行为的驾驶人转向模型,并采用免疫遗传(immune genetic algorithm, IGA)算法对驾驶人转向模型参数进行离线辨识,建立了驾驶人在环的线性时变人-车-路模型;其次考虑到复杂工况下道路曲率扰动、线性模型适配不足的缺陷以及模型参数的时变特性等因素,基于T-S模糊控制理论设计了输出反馈γ次优H鲁棒控制器;再次综合考虑驾驶人转向行为、车辆横向综合偏差情况等因素,设计了一种人机控制权分配策略,用以实现驾驶控制权的平稳动态分配;最后基于驾驶人在环的集成平台对该鲁棒控制算法进行了验证与探讨。结果表明采用该人机控制权分配策略的鲁棒控制算法具有较好的扰动抑制作用,且能够有效增强共驾过程中的合作程度,提升了人机协作的友好性。

关键词: 智能汽车, 人机共驾, T-S模糊控制, 控制权分配, 驾驶行为

Abstract:

In order to enhance the ability of the intelligent control system to predict the driver's steering intention during the co-driving, an indirect shared lane-keeping robust control algorithm is thus proposed in this paper. Firstly, a driver steering model that mimics the driver’s steering behavior is introduced, with the parameters identified offline by the immune genetic algorithm (IGA). Then a linear time-varying human-vehicle-road model for derivers in the loop is established. Secondly, considering factors such as road curvature disturbance under complex working conditions, insufficient adaptation of the linear model, and time-varying characteristics of the model parameters, an output feedback γ suboptimal Hrobust controller based on T-S fuzzy control theory is designed. Then, by fully taking into account both the driver’s steering behavior and the vehicular comprehensive lateral error, a human-vehicle control allocation strategy is designed to realize dynamic smooth allocation of driving control rights. Finally, the robust control algorithm is validated and studied based on the driver in the loop integrated platform. The results show that the robust control algorithm using the human-machine control allocation strategy has good disturbance suppression effect and can effectively enhance cooperation during the co-driving process, improving the friendliness of human-machine cooperation.

Key words: intelligent vehicle, shared autonomy, T-S fuzzy control, control authority allocation, driver behavior characteristics