主办:陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊:1976年

汽车实用技术 ›› 2024, Vol. 49 ›› Issue (7): 45-53.DOI: 10.16638/j.cnki.1671-7988.2024.007.008

• 智能网联汽车 • 上一篇    

基于纵横向协同控制的路径跟踪策略研究

李永恒,刘瑞军*,毕 昭,崔恩安,杨金龙   

  1. 山东理工大学 交通与车辆工程学院
  • 发布日期:2024-04-10
  • 通讯作者: 刘瑞军
  • 作者简介:李永恒(1998-),男,硕士研究生,研究方向为智能车无人驾驶,E-mail:kycj333@163.com。 通信作者:刘瑞军(1965-),男,博士,副教授,研究方向为汽车电气与电子技术、无人驾驶,E-mail:liuruijun66@ sdut.edu.cn。

Research on Path Tracking Ttrategy Based on Vertical and Horizontal Cooperative Control

LI Yongheng, LIU Ruijun* , BI Zhao, CUI En'an, YANG Jinlong   

  1. School of Transportation and Vehicle Engineering, Shandong University of Technology
  • Published:2024-04-10
  • Contact: LIU Ruijun

摘要: 为了提高智能车的路径跟踪精度和行驶稳定性,针对智能车路径跟踪控制提出了一种 考虑车辆纵横向协同的跟踪策略。从车辆整体系统出发,对纵向运动和横向运动进行解耦, 采用分层控制的结构,上层控制器利用基于径向基函数(RBF)神经网络的自适应滑模变结 构控制对车辆运动学耦合进行解耦,并用 RBF 神经网络对模型不确定性造成的系统扰动实时 追踪;下层控制器以轮胎利用附着系数作为优化目标,将轮胎力约束在附着椭圆内。基于纵 横向协同控制对纵横向轮胎力进行优化分配,从而提高极限工况下车辆路径跟踪的精确度和 稳定性。

关键词: 路径跟踪控制;纵横向协同控制;滑模变结构;RBF 神经网络

Abstract: In order to improve the path tracking accuracy and driving stability of intelligent vehicle, a tracking strategy considering the vertical and horizontal direction coordination is proposed for intelligent vehicle path tracking control. From the perspective of the vehicle system as a whole, vertical motion and horizontal motion are decoupted, and a hierarchical control structure is adopted. The upper controller decoups the vehicle kinematic coupling through the adaptive sliding mode variable structure control based on radial basis function (RBF) neural network, and uses the RBF neural network to track the system disturbance caused by the model uncertainty in real time;The lower controller is the optimal tire force distributor based on the vertical and horizontal cooperative control, and takes the tire adhesion coefficient as the optimization objective. The tire force is confined within the attachment ellipse, and the vertical and horizontal tire force is optimized for distribution, so as to improve the accuracy and stability of vehicle path tracking in extreme working conditions.

Key words: Path tracking control; Vertical and horizontal cooperative control; Sliding mode variable structure; RBF neural network