基于增广卡尔曼滤波并考虑车辆加速度的路面不平度识别

doi:10.19562/j.chinasae.qcgc.2022.02.012

摘要/Abstract

摘要：

为实现车辆在实际加减速行驶工况下路面不平度的准确识别，提出了一种考虑车辆加速度、基于增广卡尔曼滤波算法的路面识别方法。以车辆纵向加速度作为已知输入，车身垂向振动和俯仰振动响应作为观测向量，设计增广卡尔曼滤波观测器估计路面不平度信息；求取固定位移窗长度内的国际平整度指数，实现了对路面的等级分类。仿真结果表明在典型非匀速工况、城市运行工况和制动工况下，所提出的方法对路面不平度的识别精度和对路面等级分类的准确性，明显高于一般的增广卡尔曼滤波算法，能有效识别未知输入路面。

关键词: 路面不平度, 增广卡尔曼滤波器, 车辆加速度, 路面等级分类

Abstract:

In order to achieve the accurate identification of road unevenness under actual acceleration and deceleration driving conditions， a road surface identification method based on augmented Kalman filtering algorithm with consideration of vehicle acceleration （AKF-a） is proposed. With the longitudinal acceleration of the vehicle as the known input， and the vertical and pitching vibrations of vehicle body as the observation vectors， the augmented Kalman filter observer is designed to estimate the roughness information of road surface. The international roughness index within the fixed displacement window length is obtained to achieve the grade classification of road surface. The results of simulation show that under typical non-uniform speed conditions， urban operating conditions and braking conditions， the identification accuracy of road unevenness and the correctness of road grade classification with AKF-a algorithm proposed are apparently higher than those with general AKF algorithm and can effectively identify the unknown input road surface.

Key words: road surface roughness, augmented Kalman filter, vehicle acceleration, road grade classification

刘浪,张志飞,鲁红伟,徐中明. 基于增广卡尔曼滤波并考虑车辆加速度的路面不平度识别[J]. 汽车工程, 2022, 44(2): 247-255.

Lang Liu,Zhifei Zhang,Hongwei Lu,Zhongming Xu. Road Roughness Identification Based on Augmented Kalman Filtering with Consideration of Vehicle Acceleration[J]. Automotive Engineering, 2022, 44(2): 247-255.

图/表 23

图1

表1

车辆参数"

参数	数值	参数	数值
$m H / k g$	730	$c r ?$ / $(k N ? s ? m - 1)$	1.5
$I y / (k g ? m 2)$	1 200	$k t f ?$ / $(k N ? m - 1)$	247.03
$m f / k g$	55	$k t r$ / $(k N ? m - 1)$	247.03
$m r / k g$	52.5	$L f / m$	1.187
$k f ?$ / $(k N ? m - 1)$	28.6	$L r / m$	1.533
$k r$ / $(k N ? m - 1)$	39.6	$h / m$	0.6
$c f ?$ / $(k N ? s ? m - 1)$	1.85

表1

图2

表2

基于IRI的路面等级分类［26］"

道路等级	$G q (n 0) / 10 - 6$ m³			$I R I$ /（m·km^-1）
道路等级	下限	几何平均	上限	下限	几何平均	上限
A	8	16	32	1.658 5	2.218 2	3.137 0
B	32	64	128	3.137 0	4.436 4	6.274 0
C	128	256	512	6.274 0	8.872 8	12.548 1
D	512	1 024	2 048	12.548 1	17.745 7	25.096 2

表2

图3

图4

图5

图6

表3

测量噪声误差"

观测量	误差典型幅值	R
$x ¨ H$ /（m·s^-2）	0.005	$2.5 × 10 - 5$
$θ ˙$ /（rad·s^-1）	0.003	$9.0 × 10 - 6$
$x H$ /m	0.004 5	$2.0 × 10 - 5$
$θ$ /rad	0.001 2	$1.4 × 10 - 6$

表3

图7

表4

过程噪声误差"

状态变量	Q	状态变量	Q
$x H$	$2.4 × 10 - 7$	$x ˙ H$	$2.9 × 10 - 5$
$θ$	$3.1 × 10 - 7$	$θ ˙$	$3.0 × 10 - 5$
$x f$	$1.8 × 10 - 7$	$x ˙ f$	$7.8 × 10 - 4$
$x r$	$6.8 × 10 - 7$	$x ˙ r$	$3.6 × 10 - 3$

表4

图8

图9

图10

表5

表6

图11

图12

图13

图14

图15

图16

表7

冲击路面识别效果"

识别方法	$ρ$	RSME/mm
AKF	0.351 9	4.568
AKF-a	0.948 5	0.973 6

表7

参考文献 26

1	WANG R C， LIU W， DING R K， et al. Switching control of semi-active suspension based on road profile estimation ［J］. Vehicle System Dynamics， 2021. DOI： 10.1080/00423114.2021.1889621
2	KWON B S， KANG D， YI K. Wheelbase preview control of an active suspension with a disturbance-decoupled observer to improve vehicle ride comfort ［J］. Proceedings of the Institution of Mechanical Engineers， Part D： Journal of Automobile Engineering， 2019， 234（6）： 1725-1745.
3	GOHRLE C， SCHINDLER A， WAGNER A， et al. Road profile estimation and preview control for low-bandwidth active suspension systems ［J］. IEEE-ASME Trans Mechatron， 2015， 20（5）： 2299-2310.
4	吴秉军，刘东海，孙源泽，等. 基于路面高程自动测量的全断面平整度计算方法［J］. 中国公路学报， 2016， 29（11）： 10-17.
	WU B J， LIU D H， SUN Y Z， et al. Pavement roughness calculation of road surface based on automatic road elevation measuring［J］. China Journal of Highway and Transport， 2016， 29（11）：10-17.
5	段虎明，石峰，谢飞，等. 路面不平度研究综述［J］. 振动与冲击， 2009， 28（9）： 95-101，216.
	DUAN H M， SHI F， XIE F， et al. A survey of road roughness study［J］. Journal of Vibration and Shock， 2009， 28（9）： 95-101，216.
6	IMINE H， DELANNE Y， M'SIRDI N K. Road profile input estimation in vehicle dynamics simulation ［J］. Vehicle System Dynamics， 2006， 44（4）： 285-303.
7	KAVIANIPOUR O， MONTAZERI-GH M， MOAZAMIZADEH M. Road profile measurement using the two degrees of freedom response-type mechanism ［J］. Proceedings of the Institution of Mechanical Engineers， Part C： Journal of Mechanical Engineering Science， 2014， 229（6）： 1074-1087.
8	PREM H. A laser-based highway-speed road profile measuring system ［J］. Vehicle System Dynamics， 1988， 17（s1）： 300-304.
9	LARS B C， RICHARD M M. Model-based estimation of off-highway road geometry using single-axis LADAR and inertial sensing ［J］. IEEE International Conference on Robotics and Automation， 2006.
10	KUMAR P， LEWIS P， MCELHINNEY C P， et al. An algorithm for automated estimation of road roughness from mobile laser scanning data ［J］. Photogramm Rec， 2015， 30（149）： 30-45.
11	NGUYEN T， LECHNER B， WONG Y D. Response-based methods to measure road surface irregularity： a state-of-the-art review ［J］. European Transport Research Review， 2019， 11（1）.
12	QIN Y C， LANGARI R， WANG Z F， et al. Road excitation classification for semi-active suspension system with deep neural networks ［J］. Journal of Intelligent & Fuzzy Systems， 2017， 33（3）： 1907-1918.
13	QIN Y C， XIANG C， WANG Z F， et al. Road excitation classification for semi-active suspension system based on system response ［J］. Journal of Vibration and Control， 2017， 24（13）： 2732-2748.
14	QIN Y C， DONG M M， LANGARI R， et al. Adaptive hybrid control of vehicle semiactive suspension based on road profile estimation ［J］. Shock Vib， 2015. 636739.
15	ZHANG Z. Road profile reconstruction using connected vehicle responses and wavelet analysis ［J］. Journal of Terramechanics， 2018，80：21-30.
16	TUDON-MARTINEZ J C， FERGANI S， SENAME O， et al. Adaptive road profile estimation in semiactive car suspensions ［J］. IEEE Transactions on Control Systems Technology， 2015， 23（6）： 2293-2305.
17	DOUMIATI M， MARTINEZ J， SENAME O， et al. Road profile estimation using an adaptive Youla-Kucera parametric observer： comparison to real profilers ［J］. Control Eng Practice， 2017， 61： 270-278.
18	LIU W， WANG R C， DING R K， et al. On-line estimation of road profile in semi-active suspension based on unsprung mass acceleration ［J］. Mechanical Systems and Signal Processing， 2020， 135. 106370.
19	QIN Y C， WANG Z F， XIANG C L， et al. Speed independent road classification strategy based on vehicle response： theory and experimental validation ［J］. Mechanical Systems and Signal Processing， 2019， 117： 653-666.
20	KANG S W， KIM J S， KIM G W. Road roughness estimation based on discrete Kalman filter with unknown input ［J］. Vehicle System Dynamics， 2019， 57（10）： 1530-1544.
21	KIM G W， KANG S W， KIM J S， et al. Simultaneous estimation of state and unknown road roughness input for vehicle suspension control system based on discrete Kalman filter ［J］. Proc IMechE， Part D： J Automobile Engineering， 2020， 234（6）： 1610-1622.
22	丁仁凯，蒋俞，汪若尘，等. 考虑未知输入的主动悬架路面高程与等级识别研究［J］. 汽车工程， 2021，43（2）：278-286.
	DING R K， JIANG Y， WANG R C， et al. Research on road elevation and grade identification of active suspension considering unknown inputs［J］. Automotive Engineering， 2021，43（2）：278-286.
23	YU W， ZHANG X， GUO K， et al. Adaptive real-time estimation on road disturbances properties considering load variation via vehicle vertical dynamics ［J］. Mathematical Problems in Engineering， 2013. 283528.
24	FAURIAT W， MATTRAND C， GAYTON N， et al. Estimation of road profile variability from measured vehicle responses ［J］. Vehicle System Dynamics， 2016， 54（5）： 585-605.
25	XUE K， NAGAYAMA T， ZHAO B Y. Road profile estimation and half-car model identification through the automated processing of smartphone data ［J］. Mechanical Systems and Signal Processing， 2020，142. 106722.
26	张振伟. 基于国际不平度指数和车辆振动响应的路面统计特性幂函数模型识别［D］. 长春：吉林大学， 2019.
	ZHANG Z W. Power function model identification of road statistical characteristics based on international roughness index and vehicle vibration response ［D］. Changchun： Jilin University， 2019.

识别方法	工况一	工况二	工况三	工况四
AKF	2.495	4.263	6.174	8.154
AKF-a	1.638	1.852	2.117	2.443

识别方法	工况一	工况二	工况三	工况四
AKF	0.779 8	0.564 0	0.398 9	0.281 5
AKF-a	0.891 9	0.864 0	0.823 5	0.766 8

[1]	丁仁凯,蒋俞,汪若尘,刘伟,孟祥鹏,孙泽宇. 考虑未知输入的主动悬架路面高程与等级识别研究[J]. 汽车工程, 2021, 43(2): 278-286.
[2]	李杰, 郭文翠, 赵旗, 谷盛丰. 基于4种典型神经网络识别路面不平度的研究^*[J]. 汽车工程, 2020, 42(1): 100-107.
[3]	李杰, 郭文翠, 谷盛丰, 赵旗. 基于NARX神经网络的路面不平度识别[J]. 汽车工程, 2019, 41(7): 807-814.