基于卷积神经网络的路面识别及半主动悬架控制

doi:10.3969/j.issn.1000-1093.2020.08.002

兵工学报 ›› 2020, Vol. 41 ›› Issue (8): 1483-1493.doi: 10.3969/j.issn.1000-1093.2020.08.002

基于卷积神经网络的路面识别及半主动悬架控制

刘秋¹，孙晋伟²，张华³, 胡煦¹，顾亮¹

（1.北京理工大学振动与噪声控制研究所，北京 100081；2.西安航空学院车辆工程学院，陕西西安 710077；3.内蒙古一机集团宏远电器股份有限公司，内蒙古包头 014000）

收稿日期:2019-08-30 修回日期:2019-08-30 上线日期:2020-09-23
通讯作者: 顾亮（1958—），男，教授，博士生导师 E-mail:guliang@bit.edu.cn
作者简介:刘秋（1995—），男，硕士研究生。E-mail: liuqiu_bit@163.com

Road Identification and Semi-active Suspension Control Based on Convolutional Neural Network

LIU Qiu¹, SUN Jinwei², ZHANG Hua³, HU Xu¹, GU Liang¹

(1.Institute of Noise and Vibration Control, Beijing Institute of Technology, Beijing 100081, China; 2.School of Vehicle Engineering, Xi'an Aeronautical University, Xi'an 710077, Shaanxi, China;3.Hong Yuan Electric Appliances Co., Ltd., Inner Mongolia FIRMACO, Baotou 014000, Inner Mongolia, China)

Received:2019-08-30 Revised:2019-08-30 Online:2020-09-23

摘要/Abstract

摘要： 路面对车辆的平顺性、操纵稳定性有直接影响，实时获取路面信息对提升车辆性能具有重要意义。针对传统路面识别方法中难以精确识别多种路面类型的问题，采用卷积神经网络对路面类型进行识别，并根据不同路面输入下悬架系统的输出响应来调整控制器参数，使可控悬架在不同路面下均保持最优性能。建立车辆1/4半主动悬架模型；搭建卷积神经网络基本结构并通过所采集的4种典型城市和非城市路面图像对网络进行训练以及测试；采用遗传算法求取沥青路、弹石路、砂石路、水泥路4种不同路面激励下悬架的最优控制参数；根据路面识别结果及优化结果实现悬架控制参数的自适应调整。仿真结果表明：基于卷积神经网络的路面识别方法能够对多种路面进行准确识别；基于路面识别和遗传算法的半主动悬架控制系统可根据不同路面类型自适应调整悬架参数，有效提升车辆性能。

关键词: 路面识别, 半主动悬架, 卷积神经网络, 遗传算法

Abstract: Road has a direct impact on vehicle ride comfort and handling stability, so that the real-time acquisition of road information plays an important role in improving the vehicle performance. The multiple types of road are difficultly identified accurately using traditional road identification methods. The convolutional neural network is used to identify the road type, and then the identified road type is used as the basis for tuning the controller parameters of suspension system in order to make the controllable suspension system maintain the optimal performance under different road surfaces. Firstly, the quarter-vehicle semi-active suspension model is established. Secondly, the basic structure of convolutional neural network is built, and this network is trained and tested based on four typical urban and non-urban road images collected in advance. And then genetic algorithm is used to obtain the optimal control parameters of suspension system under excitations of four different roads, such as asphalt road, sandstone road, pebble road and cement road. Finally, the suspension control parameters are adaptively adjusted according to both the identified and optimized results of road surface. The simulated results show that the road identification method based on convolutional neural network can accurately identify a variety of roads; the semi-active suspension control system based on road identification and genetic algorithm can adaptively adjust the suspension parameters according to different road surfaces, thus improving the vehicle performance effectively.

Key words: roadidentification, semi-activesuspension, convolutionalneuralnetwork, geneticalgorithm

中图分类号:

TJ810.3⁺32

刘秋，孙晋伟，张华, 胡煦，顾亮. 基于卷积神经网络的路面识别及半主动悬架控制[J]. 兵工学报, 2020, 41(8): 1483-1493.

LIU Qiu, SUN Jinwei, ZHANG Hua, HU Xu, GU Liang. Road Identification and Semi-active Suspension Control Based on Convolutional Neural Network[J]. Acta Armamentarii, 2020, 41(8): 1483-1493.

参考文献［1］

［1］

WARD C C，IAGNEMMA K. Speed-independent vibration-based terrain classification for passenger vehicles［J］.Vehicle System Dynamics，2009，47(9):1095-1113.［2］ QIN Y, DONG M, ZHAO F. Road profile classification for vehicle semi-active suspension system based on adaptive neuro-fuzzy inference system［C］∥Proceedings of IEEE Conference on Decision and Control．Osaka，Japan: IEEE，2015:1533-1538．［3］ NGWANGWA H M, HEYNS P S，BREYTENBACH H G A. 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.［4］ BEKHTI M A，KOBAYASHI Y． Prediction of vibrations as a measure of terrain travers ability in outdoor structured and natural environments［C］∥Proceedings of the 7th Pacific-Rim Symposium on Image and Video Technology. Auckland，New Zealand: Springer，2015: 282-294.［5］ DECATUR S E．Application of neural networks to terrain classification［C］∥Proceedings of International Joint Conference on Neural Networks．San Diego，CA，US:IEEE，1989: 283-288.［6］王世峰,都凯悦,孟颖，等.基于机器学习的车辆路面类型识别技术研究［J］.兵工学报, 2017,38(8): 1642-1648. WANG S F, DU K Y, MENG Y，et al. Machine learning-based road terrain recognition for land vehicles［J］. Acta Armamentarii,2017,38(8): 1642-1648. (in Chinese)［7］ KWOK N M, HA Q P, NGUYEN T H, et al. A novel hysteretic model for magnetorheological fluid dampers and parameter identification using particle swarm optimization［J］. Sensors and Actuators A (Physical), 2006, 132(2):441-451.［8］赵济海.路面不平度的测量分析与应用［M］.北京：北京理工大学出版社，2000.ZHAO J H. Measurement analysis and application of road surface roughness［M］. Beijing: Beijing Institute of Technology Press,2000. (in Chinese)［9］余志生.汽车理论［M］.北京：机械工业出版社，2009.YU Z S. Automobile theory［M］. Beijing: China Machine Press,2009. (in Chinese)［10］ GOODFELLOW I, BENGIO Y, COURVILE A. Deep learning ［M］．Cambridge，MA，US：The MIT Press，2016.［11］ BRUNA J, SZLAM A, LECUN Y. Signal recovery from pooling representations［C］∥Proceedings of the 31st International Conference on Machine Learning. Beijing, China: International Machine Learning Society,2014: 1585-1598. ［12］ IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift［C］∥Proceedings of the 32nd International Conference on Machine Learning. Lille, France: International Machine Learning Society,2015: 448-456.［13］叶韵.深度学习与计算机视觉［M］.北京：机械工业出版社，2017.YE Y. Deep learning and computer vision［M］. Beijing: China Machine Press,2017. (in Chinese)［14］ MARMANIS D, DATCU M, ESCH T, et al. Deep learning earth observation classification using ImageNet pretrained networks［J］. IEEE Geoscience and Remote Sensing Letters, 2016，13（1）： 105-109.［15］孙晋伟，秦也辰，顾亮.基于路面识别的非线性悬架系统自适应控制［J］. 东北大学学报(自然科学版)，2018,39(9):1299-1303.SUN J W, QIN Y C, GU L. Adaptive control of the nonlinear suspension system based on road estimation［J］. Journal of Northeastern University(Natural Science) ，2018,39(9):1299-1303. (in Chinese)［16］ RAJAMANI R. Vehicle dynamics and control［M］．New York，NY, US: Springer，2006．

[1]	马也，范文慧，常天庆. 基于智能算法的无人集群防御作战方案优化方法[J]. 兵工学报, 2022, 43(6): 1415-1425.
[2]	袁航，罗迎，李开明，陈怡君，张群. 基于涡旋电磁波雷达的人体目标步态精细识别[J]. 兵工学报, 2022, 43(5): 1167-1174.
[3]	汤润之, 班利明, 钱乐, 雷争军, 宋卫星. 资源受限条件下车辆维修工序调度优化[J]. 兵工学报, 2021, 42(9): 2032-2039.
[4]	普运伟，刘涛涛，郭江，吴海潇. 基于卷积神经网络和模糊函数主脊坐标变换的雷达辐射源信号识别[J]. 兵工学报, 2021, 42(8): 1680-1689.
[5]	司骥跃，庞兆君，由锰，冯广斌，杜忠华. 空间绳网结构设计及优化方法[J]. 兵工学报, 2021, 42(5): 1065-1073.
[6]	冯睽睽，张发平，王武宏，张文杰，张田会. 基于非支配排序遗传算法的涡轮发动机转子系统装配参数优化[J]. 兵工学报, 2021, 42(5): 1092-1100.
[7]	高子茵，杜明刚，李慎龙，李晋. 基于遗传算法优化和模糊控制动态优化的自动变速器换挡规律设计[J]. 兵工学报, 2021, 42(4): 684-696.
[8]	马月红，孔梦瑶. 基于改进快速区域卷积网络的目标检测轻量化算法[J]. 兵工学报, 2021, 42(12): 2664-2674.
[9]	陈路明，廖自力，张征. 基于路面自适应的多轮轮毂电机驱动车辆驱动防滑控制[J]. 兵工学报, 2021, 42(10): 2278-2290.
[10]	王少华，吕会强，董原生，张远. 装备伴随抢修任务分配决策方法[J]. 兵工学报, 2021, 42(1): 192-198.
[11]	张宏鹏，黄长强，唐上钦，轩永波. 基于卷积神经网络的无人作战飞机飞行轨迹实时预测[J]. 兵工学报, 2020, 41(9): 1894-1903.
[12]	叶海潮，范武林，秦国华，林锋，郑许，杨扬. 基于翘曲度阈值约束的铝合金板材热轧工艺优化与实验验证[J]. 兵工学报, 2020, 41(8): 1623-1632.
[13]	杨管金子，李建辰，黄海，国琳娜. 基于最优解析的陀螺现场免转台标校方法[J]. 兵工学报, 2020, 41(3): 577-584.
[14]	赵星宇，白春华，姚箭，孙彬峰. 燃料空气炸药爆轰产物JWL状态方程参数计算[J]. 兵工学报, 2020, 41(10): 1921-1929.
[15]	张富毅，吴钦，赵晓阳，刘影，王国玉. 基于响应面方法的喷水推进器进水流道多目标优化[J]. 兵工学报, 2020, 41(10): 2071-2080.

基于卷积神经网络的路面识别及半主动悬架控制

Road Identification and Semi-active Suspension Control Based on Convolutional Neural Network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献［1］

相关文章 15

编辑推荐

Metrics

本文评价

基于卷积神经网络的路面识别及半主动悬架控制

Road Identification and Semi-active Suspension Control Based on Convolutional Neural Network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 ［1］

相关文章 15

编辑推荐

Metrics

本文评价

参考文献［1］