【作者】 李明；

【导师】 宗长富；

【作者基本信息】 吉林大学 ， 车辆工程， 2011， 硕士

【摘要】 线控技术是未来汽车技术的发展趋势,线控汽车增加了不同于传统汽车的特殊传感器。传感器的增加为估计轮胎侧偏角和路面附着系数提供了新条件。本文针对线控转向系统的特点,研究了线控转向汽车路面附着系数和轮胎侧偏角的估算算法。首先建立了线控转向系统模型,并将建立的线控转向系统模型嵌入Carsim软件中代替其中的转向系统,完成线控转向汽车模型建立;并应用卡尔曼滤波算法以转向电机电流为控制输入,前轮转角为测量输出,估算了线控转向汽车回正力矩。其次基于车辆线性模型和非线性轮胎模型,应用扩展卡尔曼滤波理论以线控转向汽车估算得到的回正力矩和侧向加速度为测量输出,分别估算了线控车辆的路面附着系数和轮胎侧偏角;最后考虑到车辆响应与路面附着状况互相影响,应用双扩展卡尔曼滤波算法同时估算了线控转向汽车的路面附着系数和轮胎侧偏角。具体工作内容如下:1、线控转向车辆动力学模型建立首先建立线控转向系统模型,包括方向盘总成模型,转向执行机构总成模型、车轮组件总成模型以及转向电机模型和容错电机模型。并将线控转向系统模型嵌入到Carsim软件中无ABS的B级车中,代替原有转向系统,建立线控转向汽车模型,通过选取典型试验工况验证线控转向系统模型的精度,为验证估算算法精度奠定基础。2、线控转向汽车回正力矩估算建立了线控转向汽车总力矩模型,包括回正力矩模型和转向电机力矩模型;同时从机械角度考虑建立了转向系统回正力矩模型,将两个模型联立,能够得出线控转向系统回正力矩与转向电机电流之间的关系。本文以转向电机电流为控制输入,前轮转角为测量输出,通过卡尔曼滤波算法估计了线控转向汽车的回正力矩,并选用典型试验工况基于Matlab/Simulink与Carsim联合仿真验证估算结果的精度。回正力矩的估算为估算线控汽车轮胎侧偏角和路面附着系数奠定基础。3、基于扩展卡尔曼滤波算法的线控转向汽车路面附着系数和轮胎侧偏角估算为了便于研究,本文研究的算法以车辆线性模型和轮胎非线性刷子模型为基础,以车辆受到的侧向加速度和前一步估算得到的回正力矩为测量输出,通过扩展卡尔曼滤波算法分别估算线控汽车路面附着系数和轮胎侧偏角,并选用典型试验工况基于Carsim与Matlab/Simulink联合仿真分别验证估算结果的精度。4、基于双扩展卡尔曼滤波算法的线控汽车轮胎侧偏角和路面附着系数联合估算为了提高估算精度,本文将两个扩展卡尔曼滤波算法平行结合,组成双扩展卡尔曼滤波算法同时将这两个变量估算出来,实现状态和参数互换信息。同时为了验证双扩展卡尔曼滤波算法的估算精度,本文以正弦工况和对接路面为例,对联合算法进行了验证,通过Matlab/Simulink与Carsim联合仿真可以看出双扩展卡尔曼滤波算法能够更精确的估算线控转向汽车的轮胎侧偏角和路面附着系数。更多还原

【Abstract】 By wire technology is the development trend of future automotive technology,By wire technology has the characteristics of increased a lot of sensors, And increase these sensors provides new conditions for estimating tire slip angle and tire-road friction coefficient.In this paper, according to characteristics of steering by wire system, study on the estimation algorithm of tire-road friction coefficient and tire slip angle. First established a model of steering by wire system, and embedded it in software of Carsim, instead of the original steering system, finished the establishment of vehicle steering by wire system. Based on kalman filtering algorithm, making the steering motor current for control input and front wheel steer angle for measurement output, estimated the aligning torque of vehicle steering by wire. Secondly based on vehicle linear model and nonlinear tire model, based on extended kalman filtering theory, making the aligning torque and lateral acceleration for the measurement outputs, estimated vehicles tire-road friction coefficient and tire slip angle respectively. Finally considering vehicle response and tire-road friction condition affecting each other, using a duel extended kalman filtering algorithm estimated the tire-road friction coefficient and tire slip angle simultaneously. Main works for this paper as followed:1、vehicle steering by wire system model establishedFirst established a model of steering by wire system, including of steering assembly model, steering actuators assembly model and wheel kit assembly model . contains steering motor model and the fault-tolerant motor model. By selecting typical test conditions validate the precision of steering by wire system model. Then embedded model in software of Carsim, instead of the original steering system, finished the establishment of vehicle steering by wire system, lay the foundation for validation algorithm accuracy.2、the aligning torque of vehicle steering by wire estimatedEstablished total torque model of vehicle steering by wire system, it contains aligning torque model and steering motor torque model; in this paper also established steering system’s mechanical torque model, simultaneousness the two model of torque, will get the relationship between aligning torque and steering motor current of steering system, Based on kalman filtering algorithm, making the steering motor current for control input and front wheel steer angle for measurement output, estimated the aligning torque of vehicle steering by wire. And choose typical test conditions based on Matlab/Simulink and Carsim simulation to verify the accuracy of the estimation results.3、Using extended kalman filtering theory estimated vehicles tire-road friction coefficient and tire slip angle respectivelyIn order to facilitate research, this paper studies the algorithm based on vehicle linear model and tire nonlinear brush model. Making the previous estimated tire slip angle and lateral acceleration for vehicle measurement output, using extended kalman filtering theory estimated vehicles tire-road friction coefficient and tire slip angle respectively, and choose typical test conditions based on Matlab/Simulink and Carsim simulation to verify the accuracy of the estimation results.4、Using duel extended kalman filtering algorithm estimated the tire-road friction coefficient and tire slip angleThis paper use two extended kalman filter parallel combinated, composition duel extended kalman filter and estimate the two variables simultaneously. Realize real-time updates between state and parameter estimation. In order to verify the accuracy of double extended kalman filter estimation, make the docking pavement for example, verified the union algorithm, Through the simulation can see duel extended kalman filter can be more accurate estimate the tire-road friction coefficient and tire slip angle.更多还原