基于意图辨识的线控汽车紧急转向控制方法  

Emergency Steering Control Based on Driver Steering Intention Recognition for Steer-by-Wire Vehicle

作　　者：何磊[1] 马伯祥 宗长富[1] 郑宏宇[1] 

机构地区：[1]吉林大学汽车仿真与控制国家重点实验室,吉林长春130025 [2]中国第一汽车股份有限公司技术中心,吉林长春130011

出　　处：《湖南大学学报（自然科学版）》2014年第1期81-86,共6页Journal of Hunan University:Natural Sciences

基　　金：国家自然科学基金资助项目(50775096)

摘　　要：针对线控转向汽车在紧急转向时,按理想转向传动比控制得到的横摆角速度动态响应慢、超调量大、稳定时间长的问题,提出了一种基于驾驶员转向意图辨识的横摆角速度反馈控制方法.该方法在正常转向时,车辆按照理想转向传动比控制;在紧急转向时,在理想转向传动比控制基础上,叠加横摆角速度反馈控制.车辆紧急转向引入驾驶员转向意图辨识环节,以判定何时叠加横摆角速度反馈控制.转向意图辨识利用多维高斯隐马尔可夫模型建模,通过离线训练参数、在线辨识识别的方式实现.实验验证结果表明:该方法能够有效降低线控汽车瞬态转向响应的超调量、减少稳定时间.To address the slow dynamic response,large overshoot and long stability time of yaw rate of a steer-by-wire vehicle in an emergency steering by the ideal steering ratio control,this paper proposed a yaw rate feedback control method on the basis of driver steering intention recognition.The proposed method focuses on normal steering and emergency steering.The vehicle was controlled according to the ideal steering ratio control in normal steering. And it was controlled according to the yaw rate feedback control superimposed on ideal steering ratio control in emer-gency steering.The method adopts driver steering intention recognition to determine in superimposing yaw rate feed-back control.The driver steering intention recognition was modeled and implemented by using multi-dimensional Gauss Hidden Markov Model,offline parameter training and online recognition.The results have indicated that the driver steering intention recognition model has good precision and the method can effectively improve the transient steering response of the steer-by-wire vehicle.

关 键 词：汽车 线控转向 意图辨识 隐马尔可夫模型 反馈控制 

分 类 号：U463.4[机械工程—车辆工程]