Florian Grivet

PhD Student in Satellite Telemetry Anomaly Detection
CNES · LAAS-CNRS · INSA Toulouse

I am a first-year PhD student at the French National Centre for Space Studies (CNES) and the Laboratory for Analysis and Architecture of Systems (LAAS-CNRS), affiliated with INSA Toulouse.

My research is centered on Anomaly Detection in Satellite Telemetry, with a specific focus on developing an explainable method based on the Christoffel function to detect anomalies in multivariate time series. This work has direct applications in enhancing the reliability and safety of satellite operations.

My research is supervised by Louise Travé-Massuyès, Didier Henrion, and Jean-Bernard Lasserre from LAAS-CNRS. Additionally, I am supported by my industrial co-applicants, Lionel Vintenat and Clément Hinderer from CNES.

CV Scholar ORCID GitHub LinkedIn Email

Publications

2026

Cost trade-offs in matrix inversion updates for streaming outlier detection

Florian Grivet, Louise Travé-Massuyès

Array · 2026

F. Grivet and L. Travé-Massuyès, Cost trade-offs in matrix inversion updates for streaming outlier
                  detection. Array (2026), doi:
                  https://doi.org/10.1016/j.array.2026.100737.

PDF Code BibTex

2026

Christoffel Function based Anomaly Detection for Satellite Telemetry

Florian Grivet, Louise Travé-Massuyès, Didier Henrion, Jean-Bernard Lasserre, Lionel Vintenat, Clément Hinderer

ANITI Days · Toulouse, France · 2026

F. Grivet, L. Travé-Massuyès, D. Henrion, J.-B. Lasserre, L. Vintenat, and C. Hinderer, Christoffel
                  Function based Anomaly Detection for Satellite Telemetry. ANITI Days 2026, Feb 2026, Toulouse,
                  France.(hal-05521204).

HAL Poster BibTex

Experience & Education

CV.pdf

2025 - present

Anomaly Detection PhD Student

CNES · LAAS-CNRS · INSA Toulouse

Topic: Christoffel Function based Anomaly Detection for Satellite Telemetry.
Supervisor: Louise Travé-Massuyès. Co-supervisor: Didier Henrion.
Advisor: Jean-Bernard Lasserre. Industry advisors: Lionel Vintenat and Clément Hinderer.

2022 — 2025

AI/ML apprentice for satellite monitoring

CNES · Toulouse · France

Apprenticeship under the supervision of Lionel Vintenat.
• Improvement of a tool for detecting univariate anomalies in satellite telemetry based on a OCSVM (IHM, performances, large scale deployment).
• Development and integration of a detection method that takes user feedback into account: semi-supervised OCSVM.
• Development of a multivariate anomaly detection algorithm using a CNN autoencoder and an OCSVM with algorithm explainability.

2022 — 2025

Double degree in Modeling and Artificial Intelligence (ModIA)

INSA Toulouse · INP ENSEEIHT · Toulouse · France

Double degree through work-study program.
• INSA degree specializing in Applied Mathematics.
• INP-ENSEEIHT degree specializing in Computer Science and Telecommunications

January 2022 — May 2022

Semester abroad exchange program

Université Laval · Québec · Canada

2020 — 6 weeks

Worker Internship in the Connectivity of Connected Objects & IoT department

SNCF · Lyon · France

• Set up the new management and billing system
• Ensure compliance and consistency between the database (Excel), the new management system (Mytem), and the operator portals (Orange, SFR, Bouygues Telecom)

2019 — 2022

Preparatory Cycle

INSA Toulouse · Toulouse · France

Integrated preparatory course followed by specialization in the third year in Computer Modeling and Communication (MIC)

Tools & Languages

Python C C++ R ADA LaTeX Git Linux PyTorch Tensorflow Dash Scikit-Learn Pandas Plotly Dash OCSVM Autoencoder CNN Supervised Learning Unsupervised Learning Deep Learning Clustering Signal Processing Image Processing Statistics Optimization Data Assimilation