基于前馈+预测LQR的智能车循迹控制器设计  

Intelligent Vehicle Tracking Controller Design Based on Feedforward+Prediction LQR

作　　者：崔凯晨 高松[1] 王鹏伟[1] 周恒恒 张宇龙 CUI Kai-chen;GAO Song;WANG Peng-wei;ZHOU Heng-heng;ZHANG Yu-long(Shandong University of Technology,School of Transportation and Vehicle Engineering,Zibo 25500,China)

机构地区：[1]山东理工大学交通与车辆工程学院,淄博255000

出　　处：《科学技术与工程》2024年第10期4287-4299,共13页Science Technology and Engineering

基　　金：国家自然科学基金(52102465);山东省自然科学基金面上项目(ZR2022MF230);山东省高校学科人才引进与培养计划:智能网联汽车技术研究与创新团队;中国博士后科学基金(2020M680091)。

摘　　要：为提升智能车辆循迹性能,基于线性二次调节器(linear quadratic regulator,LQR)理论和滑模理论,提出了一种兼顾横纵向跟踪精度与转向稳定性的横纵向控制器。首先,构建了基于二自由度横向动力学模型的前馈LQR控制器。针对模型线性化后前馈LQR控制器转向稳定性降低的问题,结合恒定转弯率和速度(constant turn rate and velocity,CTRV)模型设计预测控制器,建立了基于实时车速-曲率模糊自适应预测时间的前馈LQR控制器。此外为提升纵向车速跟踪稳定性和跟踪精度,提出了一种基于滑模控制理论的纵向跟踪方法。并进行了联合仿真和硬件在环实验验证。结果表明:文中提出的横纵向控制器有效解决了跟踪精度与稳定性两者难以兼顾的问题,提升了智能车辆循迹性能。To improve the tracking performance of intelligent vehicle,a lateral controller based on LQR theory and a longitudinal controller based on sliding mode theory were proposed to meet the requirements of tracking accuracy and stability in this paper.Firstly,a feedforward LQR controller was established based on 2-DOF(two degree of freedom)dynamics model.To solve the problem of feedforward LQR controller stability reduce caused by model linearization,a real-time velocity-road curvature fuzzy adaptive prediction LQR controller combined with CRTV model was established.In addition,to improve the stability and tracking accuracy of longitudinal velocity,a longitudinal tracking controller was proposed based on SMC(sliding mode control)theory.To verify the proposed controller,co-simulation and HIL(hardware in loop)experiment were conducted.The results show that the proposed controller combines tracking accuracy and stability.The tracking performance is significantly improved.

关 键 词：自动驾驶车辆 前馈LQR 预测控制器 滑模控制 硬件在环仿真 

分 类 号：U463.42[机械工程—车辆工程]