Ajou University repository

Fault localization by comparing memory updates between unit and integration testing of automotive software in an Hardware-in-the-Loop environmentoa mark
Citations

SCOPUS

0

Citation Export

Publication Year
2018-11-15
Journal
Applied Sciences (Switzerland)
Publisher
MDPI AG
Citation
Applied Sciences (Switzerland), Vol.8 No.11
Keyword
Automotive softwareElectronic control unit (ECU)Fault localizationHardware-in-the-Loop (HiL) testMemory update information
All Science Classification Codes (ASJC)
Materials Science (all)InstrumentationEngineering (all)Process Chemistry and TechnologyComputer Science ApplicationsFluid Flow and Transfer Processes
Abstract
During the inspection stage, an integration test is performed on electronic automobile parts that have passed a unit test. The faults found during this test are reported to the developer, who subsequently modifies the source code. If the tester provides the developer with memory usage information (such as functional symbol or interface signal), which works differently from normal operation in failed Hardware-in-the-Loop (HiL) testing (even when the tester has no source code), that information will be useful for debugging. In this paper, we propose a fault localization method for automotive software in an HiL environment by comparing the analysis results of updated memory between units and integration tests. Analyzing the memory usage of a normally operates unit test, makes it possible to obtain memory-updated information necessary for the operation of that particular function. By comparing this information to the memory usage when a fault occurs during an integration test, erroneously operated symbols and stored values are presented as potential root causes of the fault. We applied the proposed method to HiL testing for an OSEK/VDX-based electronic control unit (ECU). As a result of testing using fault injection, we confirmed that the fault causes can be found by checking the localized memory symbols with an average of 5.77%. In addition, when applying this methodology to a failure that occurred during a body control module (BCM) (which provides seat belt warnings) test, we could identify a suspicious symbol and find the cause of the test failure with only 8.54% of localized memory symbols.
ISSN
2076-3417
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/30463
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85056585405&origin=inward
DOI
https://doi.org/2-s2.0-85056585405
Journal URL
https://www.mdpi.com/2076-3417/8/11/2260/pdf
Type
Article
Funding
Funding: This research was supported by Next-Generation Information Computing Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF-2014M3C4A7030504); and supported by the National Research Foundation of Korea (NRF) grant funded by the MSIP (NRF-2016R1A2B1014376).This research was supported by Next-Generation Information Computing Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF-2014M3C4A7030504); and supported by the National Research Foundation of Korea (NRF) grant funded by the MSIP (NRF-2016R1A2B1014376)
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

LEE, JUNG WON Image
LEE, JUNG WON이정원
Department of Electrical and Computer Engineering
Read More

Total Views & Downloads

File Download