Technical Papers
Electronic Architectures
P. Keller, N. Navet, “Approximation of Worst-Case Traversal Times in Real-Time Ethernet Networks: Exploring the Potential of Many-Objective Optimization for Simulation Aggregation“, Proc. 20th IEEE International Workshop on Factory Communication System (WFCS 2024), Toulouse, France, April 17-19, 2024. Multi-objective optimization to improve the simulation’s capability to identify corner cases.
N. Navet, “TSN timing QoS mechanisms: what did we learn over the past 10 years?“, keynote at the 13th International Workshop on Resilient Networks Design and Modeling (RNDM 2023), September 20-22, 2023.
H. H. Bengtsson (Volvo Cars), N. Navet, “Time-Predictable Communication in Service-Oriented Architecture – What are the challenges?“, Automotive Ethernet Congress, March 21-23, 2023. Practical use cases and challenging issues that OEMs need to address for timing predictability in Software-Defined Vehicles.
R. Gonzalez de Oliveira (R. Bosch GmbH), N. Navet, A. Henkel (R. Bosch GmbH), “Multi-objective optimization for safety-related available E/E architectures scoping highly automated driving vehicles“, ACM Transactions on Design Automation of Electronic Systems, Volume 28, Issue 3, Article n°41, pp 1-37, 2023. A Model-Based Systems Engineering framework with DSE for the design of the E/E architectures in automated driving vehicles.
H. H. Bengtsson (Volvo Cars), M. Hiller (Volvo Cars), J. Migge, N. Navet, “Signal-Oriented ECUs in a Centralized Service-Oriented Architecture: Scalability of the Layered Software Architecture“, Automotive Ethernet Congress, June 1-2, 2022. Transition from signal-oriented ECUs to service-oriented ECUs and scalability of next-generation zone-based E/E architecture.
C. Mauclair (Airbus Helicopters), M. Gutièrrez, J. Migge, N. Navet, “Do we really need TSN in Next-Generation Helicopters? Insights from a Case-Study“, 40th Digital Avionics Systems Conference (DASC 2021), San Antonio, Texas, October 3-7, 2021. Pdf of the slides. Which TSN timing QoS mechanisms do we actually need for next-generation helicopter’s avionics and mission systems?
J. Villanueva (Renault), N. Navet, J. Migge, “QoS-Predictable SOA on TSN: Insights from a Case-Study”, Automotive Ethernet Congress, February 9-11, 2021. Mapping SOME/IP messages to Ethernet TSN QoS mechanisms so as to meet the individual QoS requirements of the services in the FACE E/E architecture.
O. Creighton (BMW Group), N. Navet, P. Keller, J. Migge, “Towards Computer-Aided, Iterative TSN-and Ethernet based EE Architecture Design”, 2020 IEEE Standards Association (IEEE-SA) Ethernet & IP @ Automotive Technology Day, Munich, September 14-18, 2020. The 4 main current challenges at BMW in the design of E/E architectures & towards synthesizing E/E communication architectures based on design goals and constraints.
D. Pannell (NXP Semiconductor), N. Navet, “Practical Use-cases for Ethernet Redundancy”, 2020 IEEE Standards Association (IEEE-SA) Ethernet & IP @ Automotive Technology Day, Munich, September 14-18, 2020. Practical ways to use frame replication in automotive Ethernet networks and what to expect from each solution in terms of communication reliability.
N. Navet, Hoai Hoang Bengtsson (Volvo Cars), J. Migge, “Early-stage Bottleneck Identification and Removal in TSN Networks“, Automotive Ethernet Congress in Munich, February 12-13, 2020. Cost- and safety-optimizing Volvo’s centralized E/E architecture with design space exploration.
N. Navet, J. Villanueva (Renault Group), J. Migge, “Early-stage Topological and Technological Choices for TSN-based Communication Architectures“, 2019 IEEE Standards Association (IEEE-SA) Ethernet & IP @ Automotive Technology Day, Detroit, Mi, September 23-25, 2019. Quantifying and improving the “capacity” of Renault FACE service-oriented E/E architecture using synthetic data.
N. Navet, J. Migge, “Insights into the performance and configuration of TCP in Automotive Ethernet Networks“, 2018 IEEE Standards Association (IEEE-SA) Ethernet & IP @ Automotive Technology Day, London, October 8-9, 2018. Configuration and performance of TCP used for in-vehicle communication.
N. Navet, “A journey into time-triggered communication protocols with a focus on Ethernet TSN“, Digicosme: Working Group on Real-Time Systems, Paris, June 2018. An historical perspective on time-triggered (TT) protocols and a few possible misconceptions about TT communication.
N. Navet, J. Migge, J. Villanueva (Renault Group), M. Boyer, “Pre-shaping bursty transmissions under IEEE802.1Q as a simple and efficient QoS mechanism”, oral presentation at the WCX18: SAE World Congress Experience, Detroit, Mi, April 2018. A shaping strategy for Ethernet TSN not requiring dedicated HW unlike CBS and TAS. Pdf of the slides. Written version published in SAE International Journal of Passenger Cars—Electronic and Electrical Systems, 11(3):2018, https://doi.org/10.4271/2018-01-0756. Authors’ preprint available here.
L. Fejoz, B. Régnier (CNES), P. Miramont (CNES), N. Navet, “Simulation-Based Fault Injection as a Verification Oracle for the Engineering of Time-Triggered Ethernet networks”, Proc. Embedded Real-Time Software and Systems (ERTS 2018), Toulouse, France, January 31-February 2, 2018. Best Paper Award. Experimental assessment of TTE clock synchronization precision. Pdf of the slides.
J. Migge, J. Villanueva (Renault Group), N. Navet, M. Boyer (Onera), “Insights on the performance and configuration of AVB and TSN in automotive networks”, Proc. Embedded Real-Time Software and Systems (ERTS 2018), Toulouse, France, January 31-February 2, 2018. Applying TSN protocols on a QoS demanding automotive case-study.
Machine Learning
T.L. Mai, N. Navet, “Improvements to Deep-Learning-based Feasibility Prediction of Switched Ethernet Network Configurations“, Proc. 29th International Conference on Real-Time Networks and Systems (RTNS’2021), Nantes, France, April 7-9, 2021. Improvemements in prediction accuracy to the base Graph Neural Network (GNN) model for feasibility prediction.
T.L. Mai, N. Navet, “Deep Learning to Predict the Feasibility of Priority-Based Ethernet Network Configurations“, ACM Trans. on Cyber-Physical Systems (TCPS), Special Issue on Artificial Intelligence and Cyber-Physical Systems, Volume 5, Issue 4, October 2021. This first deep-learning model to predict the feasibility of Ethernet networks provides a large speed-up compared to conventional schedulability analysis, which unlocks new possibilities for design-space exploration such as near-interactive design tools.
T.L. Mai, N. Navet, J. Migge, “On the use of supervised machine learning for assessing schedulability: application to Ethernet TSN“, Proc. 27th International Conference on Real-Time Networks and Systems (RTNS 2019), Toulouse, France, November 6-8, 2019. The use of supervised machine learning algorithms to speed up the verification of in-vehicle networks.
T.L. Mai, N. Navet, J. Migge, “A Hybrid Machine Learning and Schedulability Method for the Verification of TSN Networks“, Proc. 15th IEEE International Workshop on Factory Communication System (WFCS2019), Sundsvall, Sweden, May 27-29, 2019. Determining a confidence level for ML prediction to possibly switch back to conventional mathematical analysis below a user-chosen threshold.
N. Navet, T.L. Mai, J. Migge, “Using Machine Learning to Speed Up the Design Space Exploration of Ethernet TSN networks“, Technical Report, January 2019. Exploring the use of supervised and unsupervised learning for determining the feasibility of Ethernet TSN configurations.
Model-Driven Engineering for Critical Systems
T. Hu, I. Cibrario Bertolotti, N. Navet, L. Havet, “Automated Fault Tolerance Augmentation in Model-Driven Engineering for CPS“, Computer Standards & Interfaces, Elsevier, in press, 2020. Automated “augmentation” of control system models with dependability mechanisms – illustration with N-Version Programming.
S. M. Sundharam, N. Navet, S. Altmeyer, L. Havet, “A Model-Driven Co-design Framework for Fusing Control and Scheduling Viewpoints“, special issue “Design and Implementation of Future CPS”, Sensors, 18(2), MDPI, 2018. A design framework to address the gaps between control and real-time software engineering.
G. Brau, N. Navet and J. Hugues, “Towards the Systematic Analysis of Non-Functional Properties in Model-Based Engineering for Real-Time Embedded Systems“, Science of Computer Programming, Elsevier, volume 156, pages 1-20, 1 May 2018. A framework to systematize and automate the analysis of non-functional properties in Model-Based Systems Engineering.
T. Hu, I. Cibrario Bertolotti, N. Navet, “Towards Seamless Integration of N-Version Programming in Model-Based Design“, 22nd IEEE International Conference on Emerging Technologies And Factory Automation (ETFA’2017), Limassol, Cyprus, September 12-15, 2017. Sofware patterns for implementing NVP.
N. Navet, T. Hu, I. Cibrario Bertolotti, “Software Patterns for Fault-Injection in CPS Engineering“, WiP paper at the 22nd IEEE International Conference on Emerging Technologies And Factory Automation (ETFA’2017), Limassol, Cyprus, September 12-15, 2017. A step towards automated software fault-injection in DSLs for CPS.