Vibration Feature Fusion and Machine Learning Based Road Surface Identification Method

Yuyan Zhu

doi:10.54097/smabv419

Authors

Yuyan Zhu

DOI:

https://doi.org/10.54097/smabv419

Keywords:

Road Unevenness, Vibration Signal Analysis, Feature Extraction, Machine Learning, Real-time Classification

Abstract

To address the demand for real-time road condition monitoring in intelligent transportation systems, a pavement roughness classification method based on multi-feature fusion and machine learning is proposed. By constructing vibration datasets for four typical types of road surfaces, 17-dimensional feature vectors including time-domain, frequency-domain, and statistical features are extracted. Feature importance evaluation and classification are performed using Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and other algorithms. Experimental results show that the proposed method achieves a classification accuracy of over 95.2% on the test set. Random Forest outperforms comparative methods in both training efficiency and feature interpretability, providing reliable technical support for real-time road condition monitoring and automated classification.

Downloads

Download data is not yet available.

References

[1] Wu D .Ride Comfort Prediction on Urban Road Using Discrete Pavement Roughness Index[J].Applied Sciences, 2024, 14(7): 3108-.DOI:10.3390/APP14073108.

[2] Patricia A P ,Henri G .A recent overview of the effect of road surface properties on road safety, environment, and how to monitor them.[J].Environmental science and pollution research international,2022,29(44):65993-66009. DOI: 10.1007/ S1135 6-022-21847-X.

[3] Linlin Z ,Jasper M ,Bill W , et al.Installation Quality Inspection for High Formwork Using Terrestrial Laser Scanning Technology [J].Symmetry,2022,14(2):377-377.DOI:10.3390/ SYM14020377.

[4] Jian L ,Yusen W ,Hongzheng L , et al.Pavement surface defect recognition method based on vehicle system vibration data and feedforward neural network[J].International Journal of Pavement Engineering, 2023,24(1): DOI:10.1080/ 10298436. 2023. 2188594.

[5] Wenzhi Y ,Qun Y .A Novel Method for Pavement Transverse Crack Detection Based on 2D Reconstruction of Vehicle Vibration Signal[J].KSCE Journal of Civil Engineering, 2023, 27 (7):2868-2881.DOI:10.1007/S12205-023-1972-1.

[6] Guido R ,Ferrisi S ,Lofaro D , et al.An Overview on the Advancements of Support Vector Machine Models in Healthcare Applications: A Review[J]. Information, 2024, 15 (4): 235-.DOI:10.3390/INFO15040235.

[7] Zhang W ,Yan S ,Tang H , et al.An improved shuffled frog leaping algorithm based on support vector machine for hybrid flow shop rescheduling with disturbance prediction[J].Swarm and Evolutionary Computation,2025,93101843-101843. DOI: 10.1016/J.SWEVO.2025.101843.

[8] Hui W, Guizhi W .Improving random forest algorithm by Lasso method[J].Journal of Statistical Computation and Simulation, 2021,91(2):353-367.DOI:10. 1080/ 00949655. 2020. 1814776.

[9] Wang Q ,Chen H .Optimization of parallel random forest algorithm based on distance weight[J].Journal of Intelligent & Fuzzy Systems,2020,39(2):1-13.DOI:10.3233/jifs-179965.

[10] Xingyu F ,Yingyue C ,Jingru Y , et al.BGRF: A broad granular random forest algorithm[J].Journal of Intelligent & Fuzzy Systems,2023,44(5):8103-8117.DOI:10.3233/JIFS-223960.

[11] Zhao L ,Zhang Z .A improved pooling method for convolutional neural networks.[J].Scientific reports,2024, 14 (1): 1589-1589.DOI:10.1038/S41598-024-51258-6.

[12] Chaoyang H ,Fengjiao W ,Chi A , et al.An Improved New Convolutional Neural Network Method for Inverting the Pore Pressure in Oil Reservoir by Surface Vertical Deformation [J]. LITHOSPHERE,2021,2021.DOI:10.2113/2021/5597238.

[13] Legnaro E ,Guastavino S ,Piana M , et al.Deep Learning for Active Region Classification: A Systematic Study from Convolutional Neural Networks to Vision Transformers[J].The Astrophysical Journal,2025,981(2):157-157.DOI:10. 3847/ 1538-4357/ADB41A.

[14] Abolfazl A ,Biswajeet P ,M. A A .An ensemble architecture of deep convolutional Segnet and Unet networks for building semantic segmentation from high-resolution aerial images[J]. Geocarto International,2022,37(12):3355-3370.DOI: 10.1080/ 10106049.2020.1856199.