Misbehaviour Prediction for Autonomous Driving Systems

Deep Neural Networks (DNNs) are the core component of modern autonomous driving systems. To date, it is still unrealistic that a DNN will generalize correctly to all driving conditions. Current testing techniques consist of offline solutions that identify adversarial or corner cases for improving th...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2020 IEEE/ACM 42nd International Conference on Software Engineering (ICSE) s. 359 - 371
Hlavní autori: Stocco, Andrea, Weiss, Michael, Calzana, Marco, Tonella, Paolo
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: ACM 01.10.2020
Predmet:
ISSN:1558-1225
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Deep Neural Networks (DNNs) are the core component of modern autonomous driving systems. To date, it is still unrealistic that a DNN will generalize correctly to all driving conditions. Current testing techniques consist of offline solutions that identify adversarial or corner cases for improving the training phase. In this paper, we address the problem of estimating the confidence of DNNs in response to unexpected execution contexts with the purpose of predicting potential safety-critical misbehaviours and enabling online healing of DNN-based vehicles. Our approach SelfOracle is based on a novel concept of self-assessment oracle, which monitors the DNN confidence at runtime, to predict unsupported driving scenarios in advance. SelfOracle uses autoencoder- and time series-based anomaly detection to reconstruct the driving scenarios seen by the car, and to determine the confidence boundary between normal and unsupported conditions. In our empirical assessment, we evaluated the effectiveness of different variants of SelfOracle at predicting injected anomalous driving contexts, using DNN models and simulation environment from Udacity. Results show that, overall, SelfOracle can predict 77% misbehaviours, up to six seconds in advance, outperforming the online input validation approach of DeepRoad.
AbstractList Deep Neural Networks (DNNs) are the core component of modern autonomous driving systems. To date, it is still unrealistic that a DNN will generalize correctly to all driving conditions. Current testing techniques consist of offline solutions that identify adversarial or corner cases for improving the training phase. In this paper, we address the problem of estimating the confidence of DNNs in response to unexpected execution contexts with the purpose of predicting potential safety-critical misbehaviours and enabling online healing of DNN-based vehicles. Our approach SelfOracle is based on a novel concept of self-assessment oracle, which monitors the DNN confidence at runtime, to predict unsupported driving scenarios in advance. SelfOracle uses autoencoder- and time series-based anomaly detection to reconstruct the driving scenarios seen by the car, and to determine the confidence boundary between normal and unsupported conditions. In our empirical assessment, we evaluated the effectiveness of different variants of SelfOracle at predicting injected anomalous driving contexts, using DNN models and simulation environment from Udacity. Results show that, overall, SelfOracle can predict 77% misbehaviours, up to six seconds in advance, outperforming the online input validation approach of DeepRoad.
Author Stocco, Andrea
Weiss, Michael
Tonella, Paolo
Calzana, Marco
Author_xml – sequence: 1
  givenname: Andrea
  surname: Stocco
  fullname: Stocco, Andrea
  email: andrea.stocco@usi.ch
  organization: Universitá della Svizzera italiana,Lugano,Switzerland
– sequence: 2
  givenname: Michael
  surname: Weiss
  fullname: Weiss, Michael
  email: michael.weiss@usi.ch
  organization: Universitá della Svizzera italiana,Lugano,Switzerland
– sequence: 3
  givenname: Marco
  surname: Calzana
  fullname: Calzana, Marco
  email: marco.calzana@usi.ch
  organization: Universitá della Svizzera italiana,Lugano,Switzerland
– sequence: 4
  givenname: Paolo
  surname: Tonella
  fullname: Tonella, Paolo
  email: paolo.tonella@usi.ch
  organization: Universitá della Svizzera italiana,Lugano,Switzerland
BookMark eNotjF1LwzAUQKMouE6fffClf6Az996kSR_H_ISNCerzyJJUI7aRpB3s31vQpwOHwynYWR97z9g18AWAkLdESmmABZHmJOmEFZPlpACBTtkMpNQVIMoLVuT8xTmvRdPMWL0Jee8_zSHEMZUvybtghxD7so2pXI5D7GMXx1zepXAI_Uf5esyD7_IlO2_Nd_ZX_5yz94f7t9VTtd4-Pq-W68qgUEO158a6phVS1uAJrPMtCAPoCR3XaK3iAm2tWmeFMG6iVkgKp0rXkpDm7ObvG7z3u58UOpOOuwa14KjoF1LERpM
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1145/3377811.3380353
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE/IET Electronic Library
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE/IET Electronic Library
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISBN 1450371213
9781450371216
EISSN 1558-1225
EndPage 371
ExternalDocumentID 9284027
Genre orig-research
GrantInformation_xml – fundername: H2020
  funderid: 10.13039/100010661
GroupedDBID -~X
.4S
.DC
123
23M
29O
5VS
6IE
6IF
6IH
6IK
6IL
6IM
6IN
8US
AAJGR
AAWTH
ABLEC
ADZIZ
AFFNX
ALMA_UNASSIGNED_HOLDINGS
APO
ARCSS
AVWKF
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
EDO
FEDTE
I-F
I07
IEGSK
IJVOP
IPLJI
M43
OCL
RIE
RIL
RIO
RNS
XOL
ID FETCH-LOGICAL-a247t-b0acd9f45561e31cdef14a12e32d082cc7042c67fdc44ad7fd872372f14865323
IEDL.DBID RIE
ISICitedReferencesCount 124
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000652529800030&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:32:58 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a247t-b0acd9f45561e31cdef14a12e32d082cc7042c67fdc44ad7fd872372f14865323
PageCount 13
ParticipantIDs ieee_primary_9284027
PublicationCentury 2000
PublicationDate 2020-Oct.
PublicationDateYYYYMMDD 2020-10-01
PublicationDate_xml – month: 10
  year: 2020
  text: 2020-Oct.
PublicationDecade 2020
PublicationTitle 2020 IEEE/ACM 42nd International Conference on Software Engineering (ICSE)
PublicationTitleAbbrev ICSE
PublicationYear 2020
Publisher ACM
Publisher_xml – name: ACM
SSID ssj0006499
ssj0002870079
Score 2.553282
Snippet Deep Neural Networks (DNNs) are the core component of modern autonomous driving systems. To date, it is still unrealistic that a DNN will generalize correctly...
SourceID ieee
SourceType Publisher
StartPage 359
SubjectTerms anomaly detection
deep learning
misbehaviour prediction
testing
Title Misbehaviour Prediction for Autonomous Driving Systems
URI https://ieeexplore.ieee.org/document/9284027
WOSCitedRecordID wos000652529800030&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED61FQNTgRbxlgdG0iax48eIgIqFqgNI3SrHD6lLi0LC7-cuDYWBhSlR5EjRRb7X5-8-gNssSqOiTpPMoQvEeoMnVqY6KUUwttSG0o5WbELN53q5NIse3O25MCGE9vBZmNBti-X7rWuoVTY16EuxjOpDXym542rt-ykE2KWEOHVeWGIq343yyUQx5VwRp3KCFVnKSQn5l5ZKG0pmw_99xBGMfzh5bLGPNsfQC5sTGH6LMrBuj45AvhDC0HLvmwrfICCGjM8wO2X3TU0kBqz22WO1pl4C60aWj-Ft9vT68Jx04giJzYWqkzK1zpsoSN4y8Mz5EDNhszzw3GNYd07hdnRSRe-EsB6vWuVc5bhKy4Ln_BQGm-0mnAErrBE2usgdrpWisFx6lZUueI_JR9TnMCIzrN538y9WnQUu_n58CYc51aTtgbcrGNRVE67hwH3W64_qpv1pX1Aflx0
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED6VggRTgRbxJgMjaRPbsZMRAVURbdWhSN0qxw-JpUUh4fdzl4bCwMKUKHKk6CLf6_N3H8Bt7GWmfBqFsUEXiPUGD7WM0jAXLtN5mlHaUYtNqOk0XSyyWQvutlwY51x9-Mz16bbG8u3aVNQqG2ToS7GM2oHdRAgWbdha244KQXYRYU6NH5aYzDfDfGKRDDhXxKrsY00WcdJC_qWmUgeTYed_n3EIvR9WXjDbxpsjaLnVMXS-ZRmCZpd2QU4IY6jZ91WBbxAUQ-YPMD8N7quSaAxY7wePxRt1E4JmaHkPXodP84dR2MgjhJoJVYZ5pI3NvCCBS8djY52PhY6Z48xiYDdG4YY0UnlrhNAWr6liXDFclcqEM34C7dV65U4hSHQmtDeeG1wrRaK5tCrOjbMW0w-fnkGXzLB830zAWDYWOP_78Q3sj-aT8XL8PH25gANGFWp9_O0S2mVRuSvYM5_l20dxXf_AL13MmmQ
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=proceeding&rft.title=2020+IEEE%2FACM+42nd+International+Conference+on+Software+Engineering+%28ICSE%29&rft.atitle=Misbehaviour+Prediction+for+Autonomous+Driving+Systems&rft.au=Stocco%2C+Andrea&rft.au=Weiss%2C+Michael&rft.au=Calzana%2C+Marco&rft.au=Tonella%2C+Paolo&rft.date=2020-10-01&rft.pub=ACM&rft.eissn=1558-1225&rft.spage=359&rft.epage=371&rft_id=info:doi/10.1145%2F3377811.3380353&rft.externalDocID=9284027