MILLET Maxime

Ajouter à votre agenda PhD student at Sorbonne University
Team : ALSOC
    Sorbonne Université - LIP6
    Boîte courrier 169
    Couloir 24-25, Étage 4, Bureau 411
    4 place Jussieu
    75252 PARIS CEDEX 05
    FRANCE

Tel: +33 1 44 27 74 78, Maxime.Millet (at) nulllip6.fr
https://lip6.fr/Maxime.Millet

Supervision : Lionel LACASSAGNE

Co-supervision : Adrien CASSAGNE

Optimization and Time/Quality Trade-Off of an Optical Flow Algorithm on Low-Power SoC for Real-Time Meteor Detection Onboard a Nanosatellite

The detection of meteors, a luminous phenomenon resulting from the entry of extraterrestrial material into the atmosphere, is a topic of interest for astronomers. The Meteorix project aims to perform this detection from a nanosatellite, a CubeSat 3U, in order to overcome the constraints associated with ground detection. This thesis is part of this project and focuses on enriching and optimizing a space detection processing chain while adhering to two significant constraints: real-time processing (40 ms per image) on a low-power system-on-chip (SoC) limited to 10 W. This work addresses key challenges related to optimizing algorithms to ensure real-time processing and identifying possible approximations that do not compromise the detection quality.
Spatial detection has to take into account the movements of the satellite and the observed scene, rendering commonly used algorithms ineffective. For this reason, the Meteorix project application relies on an optical flow algorithm (Horn & Schunck) to estimate the various apparent motions within the scene. The complete application achieves a detection rate of 96%.
To meet execution time and energy constraints, the optical flow algorithm was parallelized (using OpenMP and SIMD) and then optimized through algorithmic transformations. These include modifications to the code semantics, two iteration pipeline schedulings, and a multi-scale approximate estimation. The optimizations achieved allow for speedups ranging from x15 to x28 while reducing energy consumption by a factor of three compared to the compiler-optimized version. Tests on embedded systems, such as the Jetson Orin Nano and the Orange Pi 5+, identified configurations that respect the 10 W energy limit while maintaining a detection rate of 96%.

Defence : 12/04/2024 - 14h - Campus Pierre et Marie Curie, salle Jacques Pitrat (25-26/105)

Jury members :

Denis Barthou, Directeur scientifique (HDR), Huawei Paris [Rapporteur]
Steven Derrien, Professeur des Universités, Lab-STICC Université de Bretagne Occidentale [Rapporteur]
François Berry, Professeur des Universités, ISPR Université Clermont Auvergne
Isabelle Bloch, Professeure des Universités, LIP6 Sorbonne Université
Michèle Gouiffes, Maîtresse de Conférences (HDR), LISN Université Paris-Saclay
Daniel Etiemble, Professeur des Universités Émérite, LRI Université Paris-Saclay
Lionel Lacassagne, Professeur des Universités, LIP6 Sorbonne Université
Adrien Cassagne, Maître de Conférences, LIP6 Sorbonne Université
Nicolas Rambaux, Maître de Conférences (HDR), IMCCE Sorbonne Université

2021-2023 Publications