基于融合特征的无人履带车辆道路类型识别方法

doi:10.12382/bgxb.2022.0038

摘要/Abstract

摘要：

无人履带车辆行驶路况复杂,将道路类型作为无人履带车辆悬架控制、自动换挡决策、路径规划等任务的先验信息,有利于提升车辆性能。针对使用单一信号识别道路类型环境适应性差或准确率低的问题,提出一种基于融合特征的道路类型识别方法,将图像的深度特征和悬置质量垂向加速度时域、频域、功率谱密度信号的统计特征相结合,利用机器学习分类算法实现道路类型识别。对单一特征和融合特征进行对比发现:融合特征实现了图像特征和悬置质量垂向加速度特征的互补,提高了道路类型识别的准确率和环境适应能力;融合特征方法与仅使用图像特征的方法实时性相差极小。对多种机器学习分类算法进行对比,试验结果表明:支持向量机和随机森林在准确性和实时性方面都表现优越,总体准确率均可以达到90%以上,识别速度可以达到14帧/s。

关键词: 无人履带车辆, 融合特征, 机器学习, 卷积神经网络, 道路类型识别

Abstract:

Unmanned tracked vehicles often navigate challenging terrain, and incorporating road types as priori information for tasks such as suspension control, automatic gear decision, and path planning can enhance their performance. However, methods based on single-class features have limitations in accuracy and environmental adaptability. To overcome this, a road-type identification method based on fusion features is proposed, combining deep image features with the statistical features of vertical acceleration in time, frequency, and power spectral density domains. Machine learning classification algorithms are used to identify the road types. Compared to using single class of features, the proposed method using fusion features enriches image features and vertical acceleration features, and improves the accuracy and environmental adaptability. The response speed of the method based on fusion features is similar to that of the image-based methods. Five machine learning classification algorithms are compared. The results show that support vector machine and random forest are the most accurate and fastest classification algorithms, achieving over 90% accuracy with a speed of 14 frames per second.

Key words: unmanned tracked vehicles, fusion features, machine learning, convolutional neural network, road types identification

刘佳, 刘海鸥, 陈慧岩, 毛飞鸿. 基于融合特征的无人履带车辆道路类型识别方法[J]. 兵工学报, 2023, 44(5): 1267-1276.

LIU Jia, LIU Hai’ou, CHEN Huiyan, MAO Feihong. Road Types Identification Method of Unmanned Tracked Vehicles Based on Fusion Features[J]. Acta Armamentarii, 2023, 44(5): 1267-1276.

图/表 10

图1 道路类型识别方法整体框架图

Fig.1 Overview of the identification method for proposed road types

图2 卷积神经网络结构图

Fig.2 Convolutional neural network structure

图3 履带式数据采集平台

Fig.3 Tracked data collection platform

图4 4种道路图像数据与垂向加速度数据

Fig.4 Images and vertical acceleration data of four road types

图5 训练过程中的准确率与损失值变化

Fig.5 Changes in accuracy and loss during training

图6 使用不同特征的道路类型识别方法准确率

Fig.6 Accuracy of road types identification methods based on different features

图7 使用不同特征的道路类型识别方法混淆矩阵

Fig.7 Confusion matrixes of road-type identification methods based on different features

图8 使用不同特征的方法的识别结果

Fig.8 Identification results of methods based on different features

图9 不同机器学习分类算法的道路类型识别准确率

Fig.9 Accuracy of road-type identification methods with different machine learning classification algorithms

表1 不同机器学习分类算法单次识别用时

Table 1 Time taken for identification using different machine learning classification algorithms

算法	单次识别用时/ms
SVM	2.48
ANN	0.14
DT	0.08
Bagging+SVM	44.54
RF	3.65

参考文献 25

[1]	赵丰. 基于路面感知的车辆智能悬架控制策略研究[D]. 北京: 北京理工大学, 2016.
	ZHAO F. Controlstrategy of intelligent vehicle suspension system based on the perception of the road[D]. Beijing: Beijing Institute of Technology, 2016. (in Chinese)
[2]	郝胜强. 复杂道路工况下重型越野车辆AMT换挡策略优化[D]. 北京: 北京理工大学, 2018.
	HAO S Q. The shift strategy optimization of AMT heavy off-road vehicles under complex road conditions[D]. Beijing: Beijing Institute of Technology, 2018. (in Chinese)
[3]	邹家祥. 野外机器人的地形识别与自然场景分割[D]. 济南: 山东大学, 2019.
	ZHOU J X. Terrain classification and natural scene segmentation in field robot[D]. Jinan: Shandong University, 2019. (in Chinese)
[4]	王振宇. 基于计算机视觉的半主动悬架预瞄控制研究[D]. 北京: 北京理工大学, 2019.
	WANG Z Y. Research on semi-active suspension system preview control based on computer vision[D]. Beijing: Beijing Institute of Technology, 2019. (in Chinese)
[5]	DEWANGAN D K, SAHU S P. RCNet: road classification convolutional neural networks for intelligent vehicle system[J]. Intelligent Service Robotics, 2021, 14(2):199-214. doi: 10.1007/s11370-020-00343-6
[6]	ŠABANOVIC E, ZURAULIS V, PRENTKOVSKIS O, et al. Identification of road-surface type using deep neural networks for friction coefficient estimation[J]. Sensors, 2020, 20(3):612. doi: 10.3390/s20030612 URL
[7]	WANG W L, ZHANG B T, WU K Q, et al. A visual terrain classification method for mobile robots’ navigation based on convolutional neural network and support vector machine[J]. Transactions of the Institute of Measurement and Control, 2022, 44(4):744-753. doi: 10.1177/0142331220987917 URL
[8]	王志红, 王少博, 颜莉蓉, 等. 基于语义分割模型的路面类型识别技术研究[J]. 公路交通科技, 2021, 38(1):128-134. doi: 10.3969/j.issn.1002-0268.2021.01.016
	WANG Z H, WANG S B, YAN L R, et al. Pavement type recognition based on semantic segmentation model[J]. Journal of Highway and Transportation Research and Development, 2021, 38(1):128-134. (in Chinese)
[9]	LIU J, WANG B Y, LIU H O, et al. Slip Estimation for autonomous tracked vehicles via machine learning[C]∥Proceedings of 2021 IEEE Intelligent Vehicles Symposium(IV).Nagoya, Japan:IEEE, 2021:1-98.
[10]	WANG S F, KODAGODA S, SHI L, et al. Road-terrain classification for land vehicles:employing an acceleration-based approach[J]. IEEE Vehicular Technology Magazine, 2017, 12(3):34-41. doi: 10.1109/MVT.2017.2656949 URL
[11]	BAI C C, GUO J F, ZHENG H X. Three-dimensional vibration-based terrain classification for mobile robots[J]. IEEE Access, 2019, 7: 63485-63492. doi: 10.1109/Access.6287639 URL
[12]	BAI C C, GUO J F, GUO L L, et al. Deep multi-layer perception based terrain classification for planetary exploration rovers[J]. Sensors, 2019, 19(14): 3102. doi: 10.3390/s19143102 URL
[13]	WANG M M, YE L, SUN X Y. Adaptive online terrain classification method for mobile robot based on vibration signals[J]. International Journal of Advanced Robotic Systems, 2021, 18(6):17298814211062035.
[14]	SKOCZYLAS A, STACHOWIAK M, STEFANIAK P, et al. Terrain classification using neural network based on inertial sensors for wheeled robot[C]∥Proceedings of Asian Conference on Intelligent Information and Database Systems.Phuket, Thailand:Springer, 2021:429-440.
[15]	CHENG C, CHANG J, LÜ W, et al. Frequency-temporal disagreement adaptation for robotic terrain classification via vibration in a dynamic environment[J]. Sensors, 2020, 20(22): 6550. doi: 10.3390/s20226550 URL
[16]	SHI W, LI Z, LÜ W, et al. Laplacian support vector machine for vibration-based robotic terrain classification[J]. Electronics, 2020, 9(3):513. doi: 10.3390/electronics9030513 URL
[17]	ANDRADES I S, CASTILLO AGUILAR J J, GARCíA J M V, et al. Low-cost road-surface classification system based on self-organizing maps[J]. Sensors, 2020, 20(21): 6009. doi: 10.3390/s20216009 URL
[18]	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. doi: 10.3969/j.issn.1000-1093.2017.08.023
	WANG S F, DOU 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)
[19]	DIMASTROGIOVANNI M, CORDES F, REINA G. Terrain estimation for planetary exploration robots[J]. Applied Sciences, 2020, 10(17):6044. doi: 10.3390/app10176044 URL
[20]	BONFITTO A, FERACO S, TONOLI A, et al. Combined regression and classification artificial neural networks for sideslip angle estimation and road condition identification[J]. Vehicle System Dynamics, 2020, 58(11):1766-1787. doi: 10.1080/00423114.2019.1645860 URL
[21]	SANDLER M, HOWARD A, ZHU M L, et al. MobileNetV2:inverted residuals and linear bottlenecks[C]∥Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City,UT,US:IEEE, 2018:4510-4520.
[22]	MARSLAND S. Machine learning: an algorithmic perspective[M]. New York,NY, US: Chapman and Hall/CRC, 2011:119-130.
[23]	BREIMAN L, FRIEDMAN J H, OLSHEN R A, et al. Classification and regression trees[M].Boca Raton,FL, US:Routledge, 2017:23-48.
[24]	BREIMAN L. Bagging predictors[J]. Machine Learning, 1996, 24(2):123-140.
[25]	BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1):5-32. doi: 10.1023/A:1010933404324 URL