h1

 

I am a PhD student in the Control and Systems Center at Mines Paris , where I am supervised by Dr. Florent Di Meglio. I also collaborate with Ansys France and with the Institute of Data Science in Mechanical Engineering at RWTH Aachen, where I am co-supervised by Pr. Dr. Sebastian Trimpe.

I work on combining state estimation and dynamics learning. On the one hand, observer design can help condition the problem of learning dynamics from partial observations, in particular with Gaussian processes or neural ODEs. On the other hand, the function approximation capabilities of machine learning can make certain state estimation approaches practically applicable, such as KKL observers. The broader aim of my PhD is to learn reduced-order models of physical systems from experimental data, for example for industrial applications on the topic of digital twins. My research is motivated by industrial problems and demonstrated on use cases encountered at Ansys France.

Before starting my PhD in November 2019, I received a diploma in engineering from Mines ParisTech with a focus on applied mathematics and robotics. For my Master’s thesis, I worked on actively learning dynamics with Gaussian processes at the Max Planck Institute for Intelligent Systems under the supervision of Pr. Dr. Sebastian Trimpe.

Talks

• Invited Talk, “Recognition models to learn dynamics from partial observations with neural ODEs” at Sorbonne Université, June 2022, Paris (France)
• Invited Talk, “Towards gain tuning for numerical KKL observers” at CNAM, April 2022, Paris (France)
• "Joint state and dynamics estimation with high-gain observers and Gaussian process models” at the 2021 American Control Conference, May 2021 (virtual)
• "Actively Learning Gaussian Process Dynamics ” at the 2nd Conference on Learning for Dynamics and Control, June 2020 (virtual)

Supervised Student Projects

• "Deep Learning for Nonlinear Observer Design", 2022, Master Thesis, Lukas Bahr (RWTH Aachen University)

Links

Google Scholar

LinkedIn

Github

 

Publications

Source	Author(s)
[Preprint] Joint state and dynamics estimation with high-gain observers and Gaussian process models, 2021	Buisson-Fenet, Mona (Corresponding author) Morgenthaler, Valery Trimpe, Johann Sebastian Di Meglio, Florent
[Contribution to a book, Contribution to a conference proceedings] Joint state and dynamics estimation with high-gain observers and Gaussian process models 2021 American Control Conference (ACC 2021) : New Orleans, Louisiana, USA 25-28 May 2021, 4027-4032, 2021 [DOI: 10.23919/ACC50511.2021.9482714]	Buisson-Fenet, Mona (Corresponding author) Morgenthaler, Valery (Corresponding author) Trimpe, Johann Sebastian (Corresponding author) Di Meglio, Florent (Corresponding author)
[Journal Article] Joint state and dynamics estimation with high-gain observers and Gaussian process models IEEE control systems letters, 5 (5), 1627-1632, 2020 [DOI: 10.1109/LCSYS.2020.3042412]	Buisson-Fenet, Mona (Corresponding author) Morgenthaler, Valery Trimpe, Johann Sebastian Di Meglio, Florent
[Conference Presentation] Actively Learning Gaussian Process Dynamics 2. Conference on Learning for Dynamics and Control, 2020	Buisson-Fenet, Mona Solowjow, Friedrich Trimpe, Johann Sebastian