基于模糊MPC的智能车队纵向跟随控制  

Longitudinal following control of intelligent vehicle fleet based on fuzzy MPC

作　　者：雷利利[1] 张通 LEI Lili;ZHANG Tong(School of Automotive and Traffic Engineering,Jiangsu University,Zhenjiang,Jiangsu 212013,China;Guangde Branch of SAIC General Motors Co.,Ltd.,Xuancheng,Anhui 242000,China)

机构地区：[1]江苏大学汽车与交通工程学院,江苏镇江212013 [2]上汽通用汽车有限公司广德分公司,安徽宣城242000

出　　处：《江苏大学学报（自然科学版）》2022年第4期394-399,430,共7页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金资助项目(51906089)。

摘　　要：针对车队纵向跟随控制存在跟随效率低且系统不稳定的问题,提出一种基于模糊模型预测控制(model predictive control,MPC)方法的智能车队纵向跟随模型.根据车队动力学方程推导出包含跟随性和乘坐舒适性的MPC目标函数,在MPC控制器的基础上引入模糊策略,实时调整MPC目标函数中的跟随性权重系数,输出符合驾驶场景需求的理想加速度.结合纵向逆动力学模型和比例、积分、微分(proportional integral derivative,PID)控制建立下层控制器,将期望加速度转化为节气门开度或制动压力.搭建Carsim和Simulink联合仿真平台,模拟智能车队高速行驶工况,并分别与PID方法和传统的MPC方法对比.结果表明:模糊MPC控制器在高速工况下,动力学参数均满足约束条件,间距误差控制在8 m以内,相较于PID控制和传统MPC控制最大速度误差分别减小了6.6、2.5 m·s^(-1),且在紧急制动场景中车速变化更加平稳,车队的乘坐舒适性得到了提高.To solve the problems of low efficiency and unstable system in the traditional control,a longitudinal following model of the intelligent vehicle fleet was proposed based on the fuzzy model predictive control(MPC)method.According to the longitudinal following model of the fleet,the objective function of MPC including following performance and riding comfort was derived,and the fuzzy strategy was introduced based on the MPC controller.The followability weight coefficient in the MPC objective function was adjusted in real time,and the ideal acceleration was outputted to meet the needs of the driving scene.The lower⁃level controller was established by combining the longitudinal inverse dynamics model and PID control to convert the expected acceleration into throttle opening or brake pressure.The joint simulation platform of Carsim and Simulink was established to simulate the high⁃speed driving conditions of the intelligent vehicle fleet,and the simulation result was compared with those of PID method and traditional MPC method,respectively.The results show that the dynamic parameters of the fuzzy MPC controller under high⁃speed conditions can meet the constraints with the distance error less than 8 m.Compared with PID control and traditional MPC control,the maximum speed error is reduced by 6.6 and 2.5 m·s^(-1),respectively.In the emergency braking scenario,the speed change is more stable,and the riding comfort of the vehicle fleet is improved.

关 键 词：智能车队 模型预测控制 模糊控制 车辆跟随 CARSIM 

分 类 号：U461.1[机械工程—车辆工程]