Data Cleaning of Speed Monitoring Based on Driving Behavior Characteristics for Commercial Vehicle

International Conference on Manufacturing Technology, Materials and Chemical Engineering

机械制造;材料科学;化学工业

Sun, Chuan^1,2,3 ; Wang, Jianyu^1,2,3 ; Xie, Lian^1,2,3 ; Chu, Duanfeng^1,2,3 ; Liu, Liqun^1,2,3

College of Transportation, Huanggang Normal University, Huanggang, Hubei

438000, China^1

School Architecture and Transportation Engineering, Guilin University of Electronic Technology, Guilin, Guangxi

541004, China^2

Intelligent Transportation Systems Research Center, Wuhan University of Technology, Wuhan, Hubei

430063, China^3

关键词: Acceleration/deceleration;    Driving behavior;    Noise jamming;    Real-world drivings;    Road condition;    Time information;    Vehicle monitoring;    Vehicle Monitoring System;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/392/6/062156/pdf
DOI  :  10.1088/1757-899X/392/6/062156

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

There are lots of speed monitoring data in the commercial vehicle monitoring system, and the accuracy of the data depends on accuracy of the analysis for driving behavior characteristics. In this paper, the driving behavior characteristics for commercial vehicles is analyzed so that six common types of errors for vehicle monitoring data are obtained, such as false date for time information, outlying high/low speed values, zero-speed signal drift, false zero-speed records, outlying acceleration/deceleration values, noise jamming data. Then data cleaning rules are made. Finally, we compare data cleaning effect through typical data source case. Results show that the data cleaning method is used for fitting the data, and correct the error signal from the vehicle driving curves more correspond with the real-world driving and road conditions.