Road classification based on SVM and vehicle body vibration

Xiaoyi Fu; Yuzhuang Zhao

doi:10.1051/e3sconf/202126801074

Road classification based on SVM and vehicle body vibration

Xiaoyi Fu^*, Yuzhuang Zhao

^*Corresponding author for this work

School of Mechanical Engineering

Beijing Institute of Technology

Research output: Contribution to journal › Conference article › peer-review

Abstract

Obtaining road surface information to make the vehicle run in the best condition can not only reduce energy consumption and vehicle loss, but also improve driving safety. In this paper, specific car body information was preprocessed as root mean square value, and SVM offline training was used. The recognition rate of off-road and highway can reach 98%. Compared with traditional threshold recognition, SVM has better adaptability and robustness. On the premise of keeping easy to obtain, the discrimination accuracy of the root mean square value is obviously better than the original value and the mean value.

Original language	English
Article number	01074
Journal	E3S Web of Conferences
Volume	268
DOIs	https://doi.org/10.1051/e3sconf/202126801074
Publication status	Published - 11 Jun 2021
Event	2020 6th International Symposium on Vehicle Emission Supervision and Environment Protection, VESEP2020 - Wuhan, China Duration: 25 Nov 2020 → 27 Nov 2020

Keywords

Road surface recognition
Support vector machine
Vehicle information

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1051/e3sconf/202126801074

Cite this

@article{e1327aefafd3423e88835d4e91434e6f,

title = "Road classification based on SVM and vehicle body vibration",

abstract = "Obtaining road surface information to make the vehicle run in the best condition can not only reduce energy consumption and vehicle loss, but also improve driving safety. In this paper, specific car body information was preprocessed as root mean square value, and SVM offline training was used. The recognition rate of off-road and highway can reach 98%. Compared with traditional threshold recognition, SVM has better adaptability and robustness. On the premise of keeping easy to obtain, the discrimination accuracy of the root mean square value is obviously better than the original value and the mean value.",

keywords = "Road surface recognition, Support vector machine, Vehicle information",

author = "Xiaoyi Fu and Yuzhuang Zhao",

note = "Publisher Copyright: {\textcopyright} The Authors, published by EDP Sciences, 2021.; 2020 6th International Symposium on Vehicle Emission Supervision and Environment Protection, VESEP2020 ; Conference date: 25-11-2020 Through 27-11-2020",

year = "2021",

month = jun,

day = "11",

doi = "10.1051/e3sconf/202126801074",

language = "English",

volume = "268",

journal = "E3S Web of Conferences",

issn = "2267-1242",

publisher = "EDP Sciences",

}

TY - JOUR

T1 - Road classification based on SVM and vehicle body vibration

AU - Fu, Xiaoyi

AU - Zhao, Yuzhuang

N1 - Publisher Copyright: © The Authors, published by EDP Sciences, 2021.

PY - 2021/6/11

Y1 - 2021/6/11

N2 - Obtaining road surface information to make the vehicle run in the best condition can not only reduce energy consumption and vehicle loss, but also improve driving safety. In this paper, specific car body information was preprocessed as root mean square value, and SVM offline training was used. The recognition rate of off-road and highway can reach 98%. Compared with traditional threshold recognition, SVM has better adaptability and robustness. On the premise of keeping easy to obtain, the discrimination accuracy of the root mean square value is obviously better than the original value and the mean value.

AB - Obtaining road surface information to make the vehicle run in the best condition can not only reduce energy consumption and vehicle loss, but also improve driving safety. In this paper, specific car body information was preprocessed as root mean square value, and SVM offline training was used. The recognition rate of off-road and highway can reach 98%. Compared with traditional threshold recognition, SVM has better adaptability and robustness. On the premise of keeping easy to obtain, the discrimination accuracy of the root mean square value is obviously better than the original value and the mean value.

KW - Road surface recognition

KW - Support vector machine

KW - Vehicle information

UR - http://www.scopus.com/inward/record.url?scp=85108362457&partnerID=8YFLogxK

U2 - 10.1051/e3sconf/202126801074

DO - 10.1051/e3sconf/202126801074

M3 - Conference article

AN - SCOPUS:85108362457

SN - 2267-1242

VL - 268

JO - E3S Web of Conferences

JF - E3S Web of Conferences

M1 - 01074

T2 - 2020 6th International Symposium on Vehicle Emission Supervision and Environment Protection, VESEP2020

Y2 - 25 November 2020 through 27 November 2020

ER -

Road classification based on SVM and vehicle body vibration

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this