Skip to main navigation Skip to main content
  • E-Submission

JKSPE : Journal of the Korean Society for Precision Engineering

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS
REGULAR

기계학습을 이용한 지능형 타이어의 마모도 추정

Wear Estimation of an Intelligent Tire Using Machine Learning

Journal of the Korean Society for Precision Engineering 2023;40(2):113-121.
Published online: February 1, 2023

1 부산대학교 기계공학부

2 부경대학교 스마트로봇융합응용교육연구단

1 School of Mechanical Engineering, Pusan National University

2 Smart Robot Convergence and Applications Research Center, Pukyong National University

#E-mail: slee@pnu.edu, TEL: +82-51-510-2320
• Received: September 2, 2022   • Revised: November 25, 2022   • Accepted: November 28, 2022

Copyright © The Korean Society for Precision Engineering

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 7 Views
  • 0 Download
  • 1 Crossref
  • 1 Scopus
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • A Study on Wheel Member Condition Recognition Using 1D–CNN
    Jin-Han Lee, Jun-Hee Lee, Chang-Jae Lee, Seung-Lok Lee, Jin-Pyung Kim, Jae-Hoon Jeong
    Sensors.2023; 23(23): 9501.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Wear Estimation of an Intelligent Tire Using Machine Learning
J. Korean Soc. Precis. Eng.. 2023;40(2):113-121.   Published online February 1, 2023
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
Wear Estimation of an Intelligent Tire Using Machine Learning
J. Korean Soc. Precis. Eng.. 2023;40(2):113-121.   Published online February 1, 2023
Close

Figure

  • 0
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
Wear Estimation of an Intelligent Tire Using Machine Learning
Image Image Image Image Image Image Image Image Image
Fig. 1 Acceleration-based intelligent tire
Fig. 2 Structure of tire conditions monitoring system
Fig. 3 Equipment module installed in tire: (a) Equipment module, (b) Battery
Fig. 4 Analysis of acceleration signal in time domain; (a) 7 mm tread depth, (b) 5.6 mm tread depth, (c) 4.2 mm tread depth, (d) 1.4 mm tread depth
Fig. 5 Analysis of acceleration signal in frequency domain; (a) 7 mm tread depth, (b) 5.6 mm tread depth, (c) 4.2 mm tread depth, (d) 1.4 mm tread depth
Fig. 6 Distribution of features by tire wears
Fig. 7 Structure of tire tread depth estimation algorithm
Fig. 8 Comparison of machine learning ideas
Fig. 9 Classification accuracy for different k values for K-NN
Wear Estimation of an Intelligent Tire Using Machine Learning

Number of training datasets

Velocity
[km/h]
Tire tread depth Total
7 mm 5.6 mm 4.2 mm 1.4 mm
40 72 72 144 144 432
50 72 72 72 72 288
60 72 72 72 72 288
Total 216 216 288 288 1,008

Number of test datasets

Velocity
[km/h]
Tire tread depth Total
7 mm 5.6 mm 4.2 mm 1.4 mm
40 144 72 144 144 504
50 144 72 72 144 432
60 72 72 72 72 288
Total 360 216 288 360 1,224

Accuracy of MLR for classify the tire tread depth

[%]

Velocity
[km/h]
Tire tread depth Total
7 mm 5.6 mm 4.2 mm 1.4 mm
40 0.0 93.1 66.7 5.6 33.9
50 1.4 87.5 55.6 30.6 34.5
60 8.3 62.5 52.8 30.6 38.5
Total 2.2 81.0 60.4 20.6 35.2

Accuracy of SVM for classify the tire tread depth

[%]

Velocity
[km/h]
Tire tread depth Total
7 mm 5.6 mm 4.2 mm 1.4mm
40 91.0 77.8 97.9 98.6 93.3
50 50.0 81.9 100.0 94.4 78.5
60 88.9 87.5 98.6 97.2 93.1
Total 74.2 82.4 98.6 96.7 88.0

Accuracy of K-NN for classify the tire tread depth

[%]

Velocity
[km/h]
Tire tread depth Total
7 mm 5.6 mm 4.2 mm 1.4mm
40 98.6 95.8 98.6 98.6 98.2
50 95.8 87.5 98.6 98.6 95.8
60 97.2 88.9 100.0 98.6 96.2
Total 97.2 90.7 99.0 98.6 96.9

Runtime and training time of machine learning models to classify tire tread depth

MLR SVM K-NN
Training time [ms] 6.984 34.612 14.842
Runtime [ms] 1.209 1.530 1.958
Table 1 Number of training datasets
Table 2 Number of test datasets
Table 3 Accuracy of MLR for classify the tire tread depth [%]
Table 4 Accuracy of SVM for classify the tire tread depth [%]
Table 5 Accuracy of K-NN for classify the tire tread depth [%]
Table 6 Runtime and training time of machine learning models to classify tire tread depth