基于多约束自适应模型预测控制的智能车路径跟踪与稳定性集成控制  

Integrated Path Tracking and Stability Control for Intelligent Vehicles Based on Multiple-constraint Adaptive Model Predictive Control

作　　者：唐爽 付锐[1] 孙秦豫 刘文晓 周伟[2] TANG Shuang;FU Rui;SUN Qin-yu;LIU Wen-xiao;ZHOU Wei(School of Automobile,Chang'an University,Xi'an 710064,Shaanxi,China;College of Mechanical and Vehicle Engineering,Chongqing University,Chongqing 400044,China)

机构地区：[1]长安大学汽车学院,陕西西安710064 [2]重庆大学机械与运载工程学院,重庆400044

出　　处：《中国公路学报》2025年第3期65-81,共17页China Journal of Highway and Transport

基　　金：国家自然科学基金项目(52102451,52272412);中央高校基本科研业务费专项资金项目(300102224501,300102224302)。

摘　　要：为了解决智能车路径跟踪精度下降与稳定性变差的问题,提出一种基于多约束自适应模型预测控制(Multiple Constraint Adaptive Model Predictive Control,CAMPC)的路径跟踪与稳定性集成控制方法。基于车辆动力学与预瞄误差模型建立预测模型,分析道路附着系数对轮胎侧向力的非线性特性和轮胎侧偏刚度的影响,采用魔术公式设计一种轮胎侧偏刚度修正系数来实时修正预测模型中的侧偏刚度。基于相平面法对车辆稳定性进行分析,得到车辆横摆角速度和质心侧偏角极限值,用于构造车辆稳定性包络线约束;随后根据车辆相轨迹到包络线边界的距离设计一种稳定性指标来表征车辆稳定性程度,基于此指标设计一种权重自适应调节机制;通过添加稳定性包络线、道路环境等多约束,并结合权重自适应机制提出一种CAMPC控制方法来实现对车辆的路径跟踪与稳定性集成控制。基于MATLAB/Simulink与CarSim的联合仿真平台,对CAMPC控制方法的有效性进行验证。研究结果表明:侧偏刚度修正系数可以改善附着系数变化引起的模型失配问题,提高路径跟踪性能;在积雪路面上,CAMPC相对于传统MPC,将横摆角速度最大值和质心侧偏角最大值分别降低了10.8%和59%,改善了车辆的稳定性;在冰雪路面上,CAMPC相对于传统MPC,将横摆角速度最大值和质心侧偏角最大值分别降低了59.6%和71.5%,改善了车辆稳定性并提高了车辆路径跟踪精度;在附着系数突变路面上,传统包络线约束效果不明显时,CAMPC可以有效降低质心侧偏角,改善车辆侧滑程度;在变车速和变附着系数路面工况下,与滑模控制(SMC)、线性二次型调节器(LQR)以及Stanley控制方法相比,所提控制方法可以提高路径跟踪精度以及车辆稳定性。所提控制方法可为自动驾驶控制技术研究提供一种思路。To solve the declining path-tracking accuracy and stability of intelligent vehicles,an integrated path-tracking and stability-control method based on multiple-constraint adaptive model predictive control(CAMPC)was proposed.A predictive model was established based on vehicle-dynamics and preview-error models,and the effects of the road adhesion coefficient on the nonlinear characteristics of the tire lateral force and cornering stiffness were analyzed.A corrective coefficient for tire cornering stiffness based on the Magic Formula was designed to correct the cornering stiffness of the predictive model in real time.Based on the phase-plane method,vehicle stability was analyzed to obtain the limit values of the yaw rate and sideslip angle for constructing the envelope constraints of vehicle stability.Subsequently,a stability index was designed based on the distance from the vehicle phase trajectory to the envelope boundaries to represent the degree of vehicle stability.A weight-adaptive mechanism was designed based on the stability index.By adding multiple constraints,such as the envelope constraints of vehicle stability and road environment,and then combining them with the weight-adaptive mechanism,a CAMPC control method was proposed to realize integrated path tracking and stability control.The effectiveness of the CAMPC control method was verified using joint simulation platforms MATLAB/Simulink and CarSim.The results show that the corrective coefficient for the tire cornering stiffness can improve the model mismatch caused by a change in the adhesion coefficient and improve the path-tracking performance.On roads covered by snow,compared with the conventional model predictive control(MPC),the CAMPC can reduce the maximum yaw rate and maximum sideslip angle by 10.8%and 59%,respectively,whereas it can reduce them by 59.6%and 71.5%,respectively,on roads covered by ice and snow,thus improving the vehicle stability and path-tracking accuracy.When the adhesion coefficient changes abruptly and the conventional envelope-c

关 键 词：汽车工程 路径跟踪控制 多约束自适应模型预测控制 稳定性指标 权重自适应机制 自动驾驶安全 

分 类 号：U461.1[机械工程—车辆工程]