INSITU - Embedded Artificial Intelligence for in situ monitoring. Application to environmental usecases


Marielle Malfante


Marielle Malfante,

Researcher and Engineer in AI,
marielle.malfante@cea.fr



 

DESCRIPTION

We build INSITU project to contribute to embedded AI and smart sensors for environmental monitoring. The project includes developments of embedded systems for various use cases and confrontation to experimental conditions.

While the state of the art in AI was traditionally developed in either algorithmic, hardware or applicative fields, the maturity level of those fundamental components of AI allows for joint work and innovation by simultaneously considering the branches. We therefore propose in INSITU a tripartite co-conception approach, integrating algorithm, hardware, and application to create a more cohesive and efficient design of embedded AI systems (smart sensors). The chair project is positioned at the intersection of those three branches and aims at contributing to bridge the gap between those scientific communities.

Practically speaking, we structure the research program of INSITU at the intersection of two methodological axis (M1 energy constraint AI and the use of interpretability to help reduce energy consumption, and M2 reliability issue and the interest of anomaly detection to help detecting unreliable situations) when designing embedded AI and three applicative situations where the need of embedded AI arises (A1 sea ice monitoring and A2 biodiversity monitoring). We highlight the importance of the third methodological component M3, which builds on the other workstreams to structure and validate the tripartite co-conception paradigm, ensuring a rigorous integration of algorithm, hardware, and application.

In addition, and for all development, we integrate (i) a systematic confrontation of the devices to real world and experimental conditions, (ii) a systematic use of eco-conception metrics to better control and mitigate the impact of the produced devices.

ACTIVITIES

  • PhD CIFRE with Biophonia (privacy constraint): Anouk Enz, ongoing.
  • M2 Internship + PhD on embedded systems for bioacoustics (energy constraint): to begin in winter / spring 2026
  • Postdoc on embedded systems for sea ice monitoring (reliability constraint): to begin in fall 2026.
  • Several internship positions will also be open during the chair duration.


CHAIR PRESENTATION

Chair Presentation INSITU

License:
Unless otherwise stated, all documents are shared under the Creative Commons BY-NC-ND 4.0 license.
You may view and share them for non-commercial purposes, without modification, and with appropriate credit to the authors.

Published on  August 20, 2025
Updated on August 21, 2025