Byzantine fault; Machine learning; N-version system; Perception; Rejuvenation; Reliability; Data faults; Input datas; Machine-learning; N version programming; Output quality; Perception systems; Real-world; Computer Networks and Communications; Information Systems; Software; Safety, Risk, Reliability and Quality
Abstract :
[en] Machine Learning (ML) has become indispensable for real-world complex systems, such as perception systems of autonomous systems and vehicles. However, ML-based systems are sensitive to input data, faults, and malicious threats that can degrade output quality and compromise the complete system's correctness. Ensuring a reliable output of ML-based components is crucial, especially for safety-critical systems. In this paper, we investigate architectures of perception systems using N-version programming for ML to mitigate the dependence on a singular ML component and combine it with a time-based rejuvenation mechanism to maintain a healthy system over extended periods. We propose models and functions to evaluate the reliability of N-version perception systems subject to faults, malicious threats, and rejuvenation. Our numerical experiments show that a rejuvenation mechanism could benefit a multiple-version system, with a reliability improvement superior to 13%. Also, the results indicate that rejuvenation could improve output reliability when ML modules' accuracy is high.
Disciplines :
Computer science
Author, co-author :
Mendonca, Julio; University of Luxembourg, Interdisciplinary Centre for Security, Reliability and Trust (SnT), Luxembourg
MacHida, Fumio; University of Tsukuba, Department of Computer Science, Japan
VÖLP, Marcus ; University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CritiX
External co-authors :
yes
Language :
English
Title :
Enhancing the Reliability of Perception Systems using N-version Programming and Rejuvenation
Publication date :
2023
Event name :
2023 53rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)
Event place :
Porto, Prt
Event date :
27-06-2023 => 30-06-2023
Audience :
International
Main work title :
Proceedings - 53rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops Volume, DSN-W 2023
Publisher :
Institute of Electrical and Electronics Engineers Inc.
This work has been supported by the German research council (DFG) and by the Luxembourg Fond Nationale de Recherche (FNR) through the Core Inter Projects ByzRT (C19-IS-13691843) and ReSAC (C21/IS/15741419). This work is supported in part by JSPS KAKENHI Grant Numbers 22K17871.
D. Gruyer, V. Magnier, K. Hamdi, L. Claussmann, O. Orfila, and A. Rakotonirainy," Perception, information processing and modeling: Critical stages for autonomous driving applications, " Annual Reviews in Control, vol. 44, pp. 323-341, 2017.
M. A. Hanif, F. Khalid, R. V. W. Putra, S. Rehman, and M. Shafique," Robust machine learning systems: Reliability and security for deep neural networks, " in International Symposium on On-Line Testing And Robust System Design, 2018.
S. Qiu, Q. Liu, S. Zhou, and C. Wu," Review of artificial intelligence adversarial attack and defense technologies, " Applied Sciences, vol. 9, no. 5, p. 909, 2019.
National Transportation Safety Board," Final reports for 2 advanced driver assistance system crash investigations published, " 2020. [Online]. Available: https://www. ntsb. gov/news/pressreleases/Pages/NR20200319. aspx
M. Berk, O. Schubert, H.-M. Kroll, B. Buschardt, and D. Straub," Exploiting redundancy for reliability analysis of sensor perception in automated driving vehicles, " IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 12, pp. 5073-5085, 2020.
M. Qiu, P. Bazan, T. Antesberger, F. Bock, and R. German," Reliability assessment of multi-sensor perception system in automated driving functions, " in 2021 IEEE 26th Pacific Rim International Symposium on Dependable Computing (PRDC), 2021, pp. 104-112.
A. Avizienis," The n-version approach to fault-tolerant software, " IEEE Transactions on software engineering, no. 12, pp. 1491-1501, 1985.
M. Ege, M. Eyler, and M. Karakas," Reliability analysis in n-version programming with dependent failures, " in Proceedings 27th EUROMICRO Conference. 2001: A Net Odyssey. IEEE Comput. Soc, 2001, pp. 174-181.
F. Machida," N-version machine learning models for safety critical systems, " in 2019 49th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W). IEEE, 6 2019, pp. 48-51.
-," Using Diversities to Model the Reliability of Nversion Machine Learning System, " 2021. [Online]. Available: https://doi. org/10. 36227/techrxiv. 16435656. v1
Q. Wen and F. Machida," Reliability models and analysis for triplemodel with triple-input machine learning systems, " in IEEE Conference on Dependable and Secure Computing (DSC), 2022.
S. Latifi, B. Zamirai, and S. Mahlke," Polygraphmr: Enhancing the reliability and dependability of cnns, " in IEEE/IFIP International Conference on Dependable Systems and Networks (DSN), 2020.
A. Gujarati, S. Gopalakrishnan, and K. Pattabiraman," New wine in an old bottle: N-version programming for machine learning components. " Institute of Electrical and Electronics Engineers Inc., 10 2020, pp. 283-286.
H. Xu, Z. Chen, W. Wu, Z. Jin, S.-y. Kuo, and M. Lyu," Nv-dnn: Towards fault-tolerant dnn systems with n-version programming, " in IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W), 2019.
M. Voelp and P. Esteves-Verissimo," Intrusion-tolerant autonomous driving, " in IEEE International Symposium on Real-Time Computing, 2018.
J. Bai, X. Chang, G. Ning, Z. Zhang, and K. S. Trivedi," Service availability analysis in a virtualized system: A markov regenerative model approach, " IEEE Transactions on Cloud Computing, vol. 10, no. 3, pp. 2118-2130, 2020.
A. M. M. Paing," Analysis of availability model based on software aging in sdn controllers with rejuvenation, " in IEEE Conference on Computer Applications (ICCA), 2020.
K. S. Trivedi and K. Vaidyanathan," Software reliability and rejuvenation: Modeling and analysis, " in Performance Evaluation of Complex Systems: Techniques and Tools, M. C. Calzarossa and S. Tucci, Eds., 2002, pp. 318-345.
M. Castro and B. Liskov," Practical byzantine fault tolerance, " in Proceedings of the Third Symposium on Operating Systems Design and Implementation, ser. OSDI '99, 1999, p. 173-186.
D. S. Silva, R. Graczyk, J. Decouchant, M. Volp, and P. Esteves-Verissimo," Threat adaptive byzantine fault tolerant state-machine replication, " in 2021 40th International Symposium on Reliable Distributed Systems (SRDS), 2021, pp. 78-87.
P. Sousa, N. F. Neves, and P. Verissimo," Proactive resilience through architectural hybridization, " in Proceedings of the 2006 ACM Symposium on Applied Computing, ser. SAC '06, 2006, p. 686-690.
M. Marsan and G. Chiola," On petri nets with deterministic and exponentially distributed firing times, " in Advances in Petri Nets 1987, 1987, vol. 266, pp. 132-145.
N. Pitropakis, E. Panaousis, T. Giannetsos, E. Anastasiadis, and G. Loukas," A taxonomy and survey of attacks against machine learning, " Computer Science Review, vol. 34, p. 100199, 2019.
A. Zimmermann," Modelling and performance evaluation with timenet 4. 4, " in Quantitative Evaluation of Systems, N. Bertrand and L. Bortolussi, Eds. Cham: Springer International Publishing, 2017, pp. 300-303.
J. Stallkamp, M. Schlipsing, J. Salmen, and C. Igel," The german traffic sign recognition benchmark: A multi-class classification competition, " in The 2011 International Joint Conference on Neural Networks, 2011.
F. Oboril, C. Buerkle, A. Sussmann, S. Bitton, and S. Fabris," Mtbf model for avs-from perception errors to vehicle-level failures, " in IEEE Intelligent Vehicles Symposium (IV), 2022.
