Embedded Sytems Architecture, Real-Time Operating Systems, and Model Based Software Engineering for the improvement of Embedded Software reliability and costs efficiency.
A Benchmark on Artificial Neural Networks and Embedded Targets Couples Adequacy, Presented in IEEE 15th International Conference on Ubiquitous and Future Networks, Budapest (HU), 2024
Study of Architectural Models for Code Generation in Real-Time, Embedded, Electrical Metering and Protection Domain, PhD Thesis, defended in November 2022
Components based approach for Distributed Electrical Metering Architectures Simulation and Implementation in Mathworks Simulink, joint with Roland Bohrer, Benoit Leprettre, Frédéric Pétrot, and Frédéric Rousseau. Presented in IEEE 4th International Conference on Industrial Cyber-Physical Systems, Victoria (CA), May 2021
Abstract: Component-based software engineering is appealing for industrial embedded systems design for two main reasons: these systems benefit from a strong legacy from previous iterations of the design reducing overall costs and time; and they target platform-independent development to minimize hardware dependent costs. However, such approaches are not yet developed for asynchronous algorithms, limiting the possibilities and performances of code generation. In this paper we propose a framework to describe, simulate, and test asynchronous processing algorithms via a Model Based Design (MBD) approach implemented in MathWorks Simulink. Our approach also facilitates code generation for such algorithms. We demonstrate our approach by designing and validating in simulation a reference architecture for electrical measurement and protection applications.
Multi-Triggered Embedded Software Code Generation for Electrical Metering and Protection Applications, joint with Roland Bohrer, Benoit Leprettre, Frédéric Pétrot, and Frédéric Rousseau. Presented in Embedded Systems Week, 30th International Workshop on Rapid System Prototyping, New York (USA), Oct. 2019
Abstract: Code generation can be an effective way to improve quality of a product and reduce development time. But applied naively to multitriggered applications such as electrical metering and protection algorithms, performance can be drastically degraded. In this paper we present a solution to model a multi-triggered application in Mathworks Simulink for efficient code generation using a buffering mechanism specified in the model, with little overhead in usage of resources. The performance of the presented solution is compared with different modeling approaches and with a reference manually-coded implementation.
Full text available here. ACM Digital Library
Spring 2019, Spring 2020
Initiation to embedded C language: Grenoble Institute of Technology
Embedded C language for microcontroller: Polytech Grenoble Engineering School
Matlab Advisory Board - Member
Schneider Electric Edison Technical Expert program - Group Expert candidate
Roland Bohrer : roland.bohrer (at) se.com
Benoit Leprettre, PhD: benoit.leprettre (at) se.com
Pr. Frédéric Pétrot: frederic.petrot (at) univ-grenoble-alpes.fr
Pr. Frédéric Rousseau: frederic.rousseau (at) univ-grenoble-alpes.fr