基于4种典型神经网络识别路面不平度的研究*

doi:10.19562/j.chinasae.qcgc.2020.01.015

汽车工程 ›› 2020, Vol. 42 ›› Issue (1): 100-107.doi: 10.19562/j.chinasae.qcgc.2020.01.015

基于4种典型神经网络识别路面不平度的研究^*

李杰, 郭文翠, 赵旗, 谷盛丰

吉林大学,汽车仿真与控制国家重点实验室,长春 130025

收稿日期:2018-11-27 发布日期:2020-01-21
通讯作者: 李杰,教授,博士生导师,E-mail:lj@jlu.edu.cn
基金资助:
*中国汽车产业创新发展联合基金重点项目(U1564213)、国家自然科学基金国际(地区)合作与交流重点项目(61520106008)和省校共建项目(SXGJSF2017-2-1-1)资助。

Study on Road Roughness Identification Based on Four Typical Neural Networks

Li Jie, Guo Wencui, Zhao Qi, Gu Shengfeng

Jilin University, State Key Laboratory of Automotive Simulation and Control, Changchun 130025

Received:2018-11-27 Published:2020-01-21

摘要/Abstract

摘要： 为识别路面不平度,对4种典型神经网络及其应用选择、输入方案优化和评价指标进行研究,提出4种典型神经网络输入选择和输入组合优化的解决方案。建立了汽车系统振动的4自由度平面模型,通过仿真获得神经网络的输入和输出。采用正交试验设计确定了每种神经网络的32个输入方案,在常用的B级路面和车速60 km/h下得到每种神经网络输入方案的评价指标,通过方差分析选出每种神经网络的最优输入方案和4种典型神经网络中的最优神经网络。研究结果表明,4种典型神经网络中,NARX神经网络是识别路面不平度的最优神经网络,其最优输入方案的相关系数和均方根误差分别为96.75%和0.003 3。

关键词: 路面不平度识别, 典型神经网络, 最优神经网络, 输入方案优化

Abstract: To identify road roughness, four typical neural networks and their application selection, input scheme optimization and evaluation indicators are studied, and a solution scheme is proposed for the input selection and input combination optimization of four typical neural networks. A four DOF plane model for vehicle system vibration is built, with the input and output of neural networks obtained by simulation. 32 input schemes of each neural network are determined by orthogonal experimental design, the evaluation indicators for the input scheme of each neural network under common road grade B and 60 km/h driving speed are obtained, and the optimal input scheme of each neural network and the optimal neural network among four typical ones are selected through variance analysis. The results show that among four typical neural networks, NARX neural network is the optimal one in identifying road roughness with its correlation coefficient and root mean square error of optimal input scheme being 96.75% and 0.003 3 respectively

Key words: road roughness identification, typical neural network, optimal neural network, input scheme optimization

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.

Li Jie, Guo Wencui, Zhao Qi, Gu Shengfeng. Study on Road Roughness Identification Based on Four Typical Neural Networks[J]. Automotive Engineering, 2020, 42(1): 100-107.

参考文献

[1] 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.
[2] HAMMOND D S, CHAPMAN L, THORNES J E. Roughness length estimation along road transects using airborne LIDAR data[J]. Meteorological Applications,2012,19(4):420-426.
[3] CHEN W X, NI Z B, HU X H. Research on pavement roughness based on the laser triangulation[J]. Photonic Sensors,2016,6(2):177-180.
[4] 张丽霞,赵又群,徐培民,等.路面功率谱密度识别的仿真[J].农业机械学报,2007,38(5):15-18.
[5] YOUSEFZADEH M, AZADI S, SOLTANI A. Road profile estimation using neural network algorithm[J]. Journal of Mechanical Science and Technology,2010,24(3):743-754.
[6] SOLHMIRZAEI A, AZADI S, KAZEMI R. Road profile estimation using wavelet neural network and 7-DOF vehicle dynamic systems[J]. Journal of Mechanical Science and Technology,2012,26(10):3029-3036.
[7] NGWANGWA H M, HEYNS P S, LABUSCHAGNE F J, et al. An overview of the neural network based technique for monitoring of road condition via reconstructed road profiler[C]. Southern African Transport Conference,2008:312-329.
[8] NGWANGWA H M, HEYNS P S, LABUSCHAGNE F J, et al. Reconstruction of road defects and road roughness classification using vehicle responses with artificial neural networks simulation[J]. Journal of Terramechanics,2010,47(2):97-111.
[9] NGWANGWA H M, HEYNS P S, BREYTENBACH H G A, et al. Reconstruction of road defects and road roughness classification using artificial neural networks simulation and vehicle dynamic responses:application to experimental data[J]. Journal of Terramechanics,2014,53(1):1-18.
[10] 周瑜.某轿车车身姿态半主动悬架最优控制研究[D].长春:吉林大学,2016.
[11] JIANG L, WANG Y, ZHU X H. A novel vehicle dynamics identification method utilizing MIMU sensors based on support vector machine[C]. Tenth International Conference on Sensing Technology,2016, Nanjing, China.
[12] 张金.基于ARM的车辆姿态测量系统设计[D].北京:北京交通大学,2008.
[13] BORTOLIN G, BACKEN S, GUSTAFSSON T. Estimation of sideslip angles of a VOLVO A25E articulated all-wheel drive hauler based on GPS/INS measurements[C]. SAE Paper 2011-01-2156.
[14] 王朝斌,吴旭,熊吉.汽车车轮轮心与车身相对位移测量方法的探讨[J].汽车科技,2014(5):41-44.
[15] 王小川,史峰,郁磊,等.MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社,2013.
[16] 任露泉.试验优化设计与分析[M].北京:高等教育出版社,2003.
[17] 张敏,袁辉.关于标准误差应用问题的讨论[J].郑州工业大学学报,1996,17(3):95-100.
[18] 杨彦荣,戴云.基于均方根误差和相关系数评价人眼像差对视网膜像质的影响[J].光学学报,2017,37(3):382-388.
[19] 李杰,高雄,王维,等.基于UniTire模型的平顺性和操纵稳定性协同研究[J].汽车工程,2018,40(2):127-132.

基于4种典型神经网络识别路面不平度的研究^*

Study on Road Roughness Identification Based on Four Typical Neural Networks

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

Metrics

本文评价

推荐阅读 10