基于模型的车辆状态参数估计研究
发布时间：2018-11-09 18:35
【摘要】：作为汽车主动安全控制系统的关键技术之一,车辆纵向速度、质心侧偏角及横摆角速度等车辆状态参数的估计研究受到了越来越多的关注。本文研究分别基于整车动力学模型及运动学模型的状态参数估计方法,解决车辆状态参数估计的实时性与精确性问题。基于动力学模型的状态参数估计研究,采用能充分发挥动力学模型非线性模拟特性的非线性观测技术对车辆纵向速度及质心侧偏角进行估计。根据相关轮胎试验数据辨识出魔术公式轮胎模型参数,以此为基础结合车辆系统动力学建立包括纵向、侧向及横摆运动的3自由度整车模型并与车辆动力学仿真软件Carsim进行联合仿真分析。根据所建立的整车动力学模型设计非线性车辆状态观测器,所设计非线性观测器由一个非线性子观测器和一个修正模块组成。非线性子观测器采用固定增益形式保证了观测器的实时性,修正模块根据典型工况仿真结果调整增益参数保证估计的精度。基于运动学模型的车辆状态参数估计研究,采用能充分发挥运动学模型实时性好这一特点的卡尔曼滤波技术对车辆纵向速度进行估计。对轮速传感器及陀螺仪信号进行滤波处理以降低测量噪声。推导出车轮速度与车辆纵向速度关系,建立考虑纵向滑移率影响的运动学估计模型。根据估计模型应用卡尔曼滤波技术建立车辆纵向速度估计卡尔曼滤波器组,并通过加权平均融合由纵向加速度积分得到纵向速度信息估计出车辆纵向速度值。典型工况下的联合仿真结果表明,本文所提出的两种车辆状态参数估计方法具有较高的估计精度,能够满足车辆主动安全控制系统要求。
[Abstract]:As one of the key technologies of vehicle active safety control system, more and more attention has been paid to the estimation of vehicle state parameters, such as vehicle longitudinal velocity, side deflection angle of mass center and yaw angle velocity. In this paper, the methods of state parameter estimation based on vehicle dynamics model and kinematics model are studied, respectively, to solve the real-time and accuracy problems of vehicle state parameter estimation. Based on the research of state parameter estimation of dynamic model, the vehicle longitudinal velocity and side deflection angle of mass center are estimated by using nonlinear observation technology which can give full play to the nonlinear simulation characteristics of dynamic model. Based on the related tire test data, the parameters of the magic formula tire model are identified, which are combined with the vehicle system dynamics to establish the longitudinal model. The 3 DOF vehicle model of lateral and yaw motion is simulated and analyzed jointly with the vehicle dynamics simulation software Carsim. According to the established vehicle dynamics model, a nonlinear vehicle state observer is designed. The designed nonlinear observer consists of a nonlinear sub-observer and a correction module. The nonlinear subobserver adopts the fixed gain form to ensure the real-time performance of the observer, and the correction module adjusts the gain parameters to ensure the accuracy of the estimation according to the simulation results of typical operating conditions. The research of vehicle state parameter estimation based on kinematics model and the Kalman filter technique which can give full play to the real time of kinematics model is used to estimate the longitudinal velocity of vehicle. The signal of wheel speed sensor and gyroscope are filtered to reduce the measurement noise. The relationship between wheel speed and vehicle longitudinal velocity is derived, and a kinematics estimation model considering the influence of longitudinal slip ratio is established. According to the estimation model, the Kalman filter bank of vehicle longitudinal velocity estimation is established by using Kalman filter technique, and the longitudinal velocity information is obtained from the longitudinal acceleration integral by the weighted average fusion. The results of joint simulation under typical operating conditions show that the two methods proposed in this paper have high estimation accuracy and can meet the requirements of vehicle active safety control system.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.6
本文编号：2321233
[Abstract]:As one of the key technologies of vehicle active safety control system, more and more attention has been paid to the estimation of vehicle state parameters, such as vehicle longitudinal velocity, side deflection angle of mass center and yaw angle velocity. In this paper, the methods of state parameter estimation based on vehicle dynamics model and kinematics model are studied, respectively, to solve the real-time and accuracy problems of vehicle state parameter estimation. Based on the research of state parameter estimation of dynamic model, the vehicle longitudinal velocity and side deflection angle of mass center are estimated by using nonlinear observation technology which can give full play to the nonlinear simulation characteristics of dynamic model. Based on the related tire test data, the parameters of the magic formula tire model are identified, which are combined with the vehicle system dynamics to establish the longitudinal model. The 3 DOF vehicle model of lateral and yaw motion is simulated and analyzed jointly with the vehicle dynamics simulation software Carsim. According to the established vehicle dynamics model, a nonlinear vehicle state observer is designed. The designed nonlinear observer consists of a nonlinear sub-observer and a correction module. The nonlinear subobserver adopts the fixed gain form to ensure the real-time performance of the observer, and the correction module adjusts the gain parameters to ensure the accuracy of the estimation according to the simulation results of typical operating conditions. The research of vehicle state parameter estimation based on kinematics model and the Kalman filter technique which can give full play to the real time of kinematics model is used to estimate the longitudinal velocity of vehicle. The signal of wheel speed sensor and gyroscope are filtered to reduce the measurement noise. The relationship between wheel speed and vehicle longitudinal velocity is derived, and a kinematics estimation model considering the influence of longitudinal slip ratio is established. According to the estimation model, the Kalman filter bank of vehicle longitudinal velocity estimation is established by using Kalman filter technique, and the longitudinal velocity information is obtained from the longitudinal acceleration integral by the weighted average fusion. The results of joint simulation under typical operating conditions show that the two methods proposed in this paper have high estimation accuracy and can meet the requirements of vehicle active safety control system.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.6
【相似文献】
相关期刊论文 前6条
1 赵林辉;刘志远;陈虹;;一种车辆状态的非线性估计方法研究[J];系统仿真学报;2009年06期
2 郭文强;付菊;张玉杰;侯勇严;;贝叶斯网络在车辆状态远程故障诊断系统中的应用[J];陕西科技大学学报(自然科学版);2013年01期
3 王建锋;李平;;基于多信息融合的车辆状态参数估计[J];计算机仿真;2013年11期
4 赵林辉;刘志远;陈虹;;一种车辆状态滑模观测器的设计方法[J];电机与控制学报;2009年04期
5 杨志强;谭_g;;车道保持系统的车辆状态预测模型[J];农业机械学报;2008年10期
6 ;[J];;年期
相关硕士学位论文 前5条
1 黄皓;基于驾驶操作及车辆状态的疲劳驾驶行为检测研究[D];东南大学;2016年
2 佘国芹;基于模型的车辆状态参数估计研究[D];武汉科技大学;2016年
3 金未平;独立分量分析在车辆状态监测中的应用研究[D];合肥工业大学;2010年
4 屈肖蕾;基于转向操作和车辆状态的疲劳驾驶检测方法研究[D];清华大学;2012年
5 黄巍;基于CAN总线的车辆状态信息显示处理系统[D];大连理工大学;2005年
，本文编号：2321233
本文链接：https://www.wllwen.com/kejilunwen/qiche/2321233.html
