[1]查云飞,刘鑫烨,马芳武,等.基于EKF和RBF的路面附着系数估计[J].福建工程学院学报,2022,20(01):1-6+34.[doi:10.3969/j.issn.1672-4348.2022.01.001]
　ZHA Yunfei,LIU Xinye,MA Fangwu,et al.Estimation of road adhesion coefficient based on EKF and RBF[J].Journal of FuJian University of Technology,2022,20(01):1-6+34.[doi:10.3969/j.issn.1672-4348.2022.01.001]
点击复制

基于EKF和RBF的路面附着系数估计()
分享到：

《福建工程学院学报》[ISSN:2097-3853/CN:35-1351/Z]

卷:
第20卷
期数:
2022年01期
页码:
1-6+34
栏目:
出版日期:
2022-02-25

文章信息/Info

Title:
Estimation of road adhesion coefficient based on EKF and RBF
作者:
查云飞; 刘鑫烨; 马芳武; 吕小龙
福建工程学院福建省汽车电子与电驱动技术重点实验室
Author(s):
ZHA Yunfei;  LIU Xinye;  MA Fangwu;  LYV Xiaolong
Fujian Provincial Key Laboratory of Automotive Electronics and Electric Drive Technology, Fujian University of Technology
关键词:
路面附着系数; 算法融合; 扩展卡尔曼滤波; 径向基神经网络; 决定系数优化
Keywords:
road adhesion coefficient;  algorithm fusion;  extended Kalman filter;  radial basis function neural network;  determination coefficient optimization
分类号:
U463.5
DOI:
10.3969/j.issn.1672-4348.2022.01.001
文献标志码:
A
摘要:
针对车辆行驶下的路面附着系数估计问题，提出了扩展卡尔曼滤波算法(EKF，Extended Kalman Filter)与径向基神经网络(RBF，Radial Basis Functionneural network)相融合。通过扩展卡尔曼滤波算法得出路面附着系数估计所需要的车辆状态参数，结合轮速等直接数据采用径向基神经网络对路面附着系数进行估计。神经网络的训练样本通过Carsim/Simulink收集不同行驶工况，并采用差值寻优的方法对径向基神经网络算法中的决定系数进行优化。基于双移线工况验证了该算法在路面附着系数估计上具有较高的精准度。
Abstract:
Aiming at the problem of road adhesion coefficient estimation under vehicle driving, a fusion algorithm is proposed integrating extended Kalman filter and radial basis function neural network. The vehicle state parameters needed to estimate the road adhesion coefficient were obtained by the extended Kalman filter algorithm. Combined with the direct data such as wheel speed, the road adhesion coefficient was estimated by radial basis function neural network. Training samples of neural networks were collected by Carsim/Simulink in different driving conditions, and the decision coefficients of radial basis neural network algorithm were optimized by the method of difference optimization. Finally, the accuracy of the algorithm was verified based on double lane change conditions.

参考文献/References:

[1] 王少帅, 李登峰. 基于无迹卡尔曼滤波的路面附着系数估计[J]. 电子设计工程, 2020, 28（1）: 27-31.[2] 吕梦男, 孙涛, 李洁. 基于扩展卡尔曼滤波的车辆状态参数估计[J]. 农业装备与车辆工程, 2018, 56（5）: 77-80.[3] 王健, 杨君, 于蓬, 等. 不同路况下路面附着系数实时估计[J]. 河北科技大学学报, 2020, 41(2): 172-180.[4] ENISZ K, SZALAY I, KOHLRUSZ G, et al. Tyre-road friction coefficient estimation based on the discrete-time extended Kalman filter[J]. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2015, 229（9）: 1158-1168.[5] RATH J J, VELUVOLU K C, ZHANG D, et al. Estimation of road adhesion coefficient using higher-order sliding mode observer for torsional tyre model[C]//Intelligent Robotics and Applications, 2013: 202-213. [6] 余志生. 汽车理论[M]. 5版. 北京: 机械工业出版社, 2009: 108-109.[7] 宋涛. 轮毂驱动电动车的路面附着系数估计方法研究[D]. 哈尔滨: 哈尔滨工业大学, 2017.[8] 周磊, 张向文. 基于Dugoff轮胎模型的爆胎车辆运动学仿真[J]. 计算机仿真, 2012, 29（6）: 308-311, 385.[9] 樊旭, 黄颖, 冷文楠, 等. 基于径向基神经网络的地基微波辐射计反演算法研究[J]. 气象与环境学报, 2020, 36（2）: 62-69.

更新日期/Last Update: 2022-02-25