[Portrait 02] Olivier Liandrat, computer science engineer in image processing and machine learning

IRT Saint Exupéry pursues a series of portraits devoted to the men and women who best represent the institute: its researchers. Their high-level skills and wealth of experience contribute hugely to IRT Saint Exupéry’s performance and unique position, which is so crucial for its members and partners.

Olivier Liandrat, age 30, is a computer science engineer specialising in image processing and machine learning. He is seconded to IRT Saint Exupéry by the SME Reuniwatt. From solar rays with photovoltaic to laser beams with optical communications, Olivier keeps his head in the clouds!

Can you tell us about your career?

I did an intensive foundation degree in Le Havre (where I come from) and Rouen before joining Ensimag[1] in Grenoble, majoring in image processing. It’s an area I’ve always liked because when you process an image, you see the results straightaway, and it’s quite concrete. Then I went on an Erasmus[2] exchange to Valencia,[3] where I could choose and customise my courses on computer vision systems and machine learning.
After my end-of-course internship at Reuniwatt, a start-up based on Réunion Island, I was the first person they hired. My work focused on analysing images of the sky recorded from the ground with observation cameras in order to deduce changes in cloud cover. After three and a half years, I wanted to return to mainland France, just when Reuniwatt had started having closer links with IRT Saint Exupéry, and I was asked to join the ALBS project in 2015.

What area of expertise @Reuniwatt was IRT Saint Exupery interested in?

Reuniwatt was founded by Nicolas Schmutz in 2009, and the start-up initially positioned itself in the photovoltaic sector: Réunion, as an island, is a laboratory in terms of the renewable energy mix. Reuniwatt’s core business is predicting the evolution of clouds on different spatial and temporal scales. For local, short-term forecasts, we rely on ground-based cameras and satellite images, which offer much greater precision than the grids used in conventional weather models. This expertise was highly relevant and transferable for the optical communications in the ALBS project, where the goal is to use a laser to enable a ground station to communicate with a satellite.



Could you tell us a little more about ALBS and Reuniwatt’s role?

A laser beam is like the sun’s rays in that it can’t travel through clouds that are too thick – which means it’s helpful to forecast the cloud cover so that communication can continue without interruption. Sky Insight, an instrument patented by Reuniwatt, is equipped with an observation camera that works using thermal infra-red 24 hours a day. This observation band gives a very good contrast between clear sky emissions and cloud emissions, since clouds are warmer. One of the main challenges is to qualify optical thickness accurately, which measures the level of transparency in the middle so you can anticipate how the cloud will impact on the link. An optical system consists of a network of stations distributed over the region, all equipped with a laser and camera showing the arrival of cloud fronts. Using these forecasts, the laser link can be switched to another station in a matter of a few minutes.

What noteworthy results have been obtained by IRT Saint Exupéry’s members?

Our solution now meets the specifications of optical systems much better. We have one year left to improve the product, validating the optical thicknesses of our instruments located on our two measuring stations. The first was set up at the Côte d’Azur Observatory in Nice, and the second at Airbus Defence & Space near the B612 building in Toulouse. A third site is planned at the end of the project for a data campaign by LIDAR[4], a precision instrument that precisely analyses the composition of the atmosphere, as found in some weather laboratories such, as SIRTA[5].

What do you like at IRT Saint Exupéry?

IRT Saint Exupéry has opened up new perspectives for me with the discovery of the world of space, which is very demanding, and which I didn’t know anything about. I’ve switched from solar rays to laser beams but I’ve still got my head in the clouds!

What I like about IRT Saint Exupéry is that we’ve got more time to work on the specifications, and to think about the methodology and really refine our product. With the resources allocated to the project, we have achieved things together that our SME could never have done by itself. IRT has played the role of facilitator, providing access to major customers by factoring in the risk-taking that is linked to R&D work. Being housed alongside Airbus Defence & Space and Thales Alenia Space employees means we can understand their challenges and needs better. It’s a very enriching experience.

There are lots of opportunities with the ALBS project: for example, CNES invited me to present our solution at a CCSDS[6] meeting in October 2018. This international committee deliberates on the weather measurements to be made on an optical reception station to contribute to the drafting of a volume of recommendations.

Thanks to this cooperation as part of ALBS, Reuniwatt now has an official office in Toulouse in the B612 building within the District, a new co-working space managed by the Aerospace Valley cluster.


Cloud cover forecast from a ground-based all sky infrared thermal cameraO. Liandrat, S. Cros, A. Braun, L. Saint-Antonin, J. Decroix, N. Schmutz. In Remote Sensing of Clouds and the Atmosphere XXII, Vol. 10424, p. 104240B, Oct. 2017, International Society for Optics and Photonics

Day and night cloud occurrence climatology for space-to-ground optical communication feasibility: the advantages of a thermal infrared sky imager.
O. Liandrat, C. Bertin, S. Cros, L. Saint-Antonin. In AGU Fall Meeting Abstracts (Dec. 2016)

[1] École Nationale Supérieure d’Informatique et de Mathématiques Appliquées, Grenoble INP.
[2] EuRopean Action Scheme for the Mobility of University Students: an exchange programme for students and teachers between universities, European grandes écoles and educational institutions across the world.

[3] Valencia Polytechnique University (Spain).

[4] Laser Imaging Detection and Ranging.

[5] A ground-based atmospheric observatory for cloud and aerosol research, the atmospheric research laboratory belonging to the Institut Pierre Simon Laplace (IPSL), Paris-Saclay.

[6] Consultative Committee for Space Data Systems.

[Portrait 02] Olivier Liandrat, computer science engineer in image processing and machine learning
Scroll to top