ASTRID - Accompagnement spécifique des travaux de recherches et d’innovation défense

Artificial Intelligence for adaptation to detection & identification of thermal targets – IAADICT

Submission summary

The aim of the project proposed by Bertin Technologies, CEA and ISL is to develop a system to assist in the detection, recognition and identification of threats in the visible and infrared domains. The algorithm developed during this project is intended to be embedded in optronic systems used in the battlefield. It will therefore be subject to material and energy frugality constraints. The proposed approach will rely on artificial intelligence solutions based on deep neural networks. This technology is very promising for detection, recognition and identification applications. However, it requires supervised learning with a lot of annotated data representative of the usecases, which is difficult to obtain for military applications, as well as a high computing power during use. ISL, CEA and Bertin Technologies will bring their expertise to bear in data acquisition, AI training and optimisation, hardware integration and optronics to overcome the three technical hurdles of learning from a frugal dataset, joint exploitation of visible and infrared data, and frugality in computing power. A demonstration model based on Bertin Technologies' existing optronic solutions will be produced in order to validate the solution developed.
This project is in line with French and European strategies to pursue research efforts in the field of AI, which has major economic and strategic implications. From a military point of view, it addresses strong tactical and operational challenges, in particular for surveillance, reconnaissance, identification and intelligence gathering applications. Embedded in optronic systems, artificial intelligence can facilitate the processing of information transmitted by sensors, limit the risks of human error, facilitate rapid decision-making and optimize the consumption of sensor batteries. Threat detection, recognition and identification functions are also of interest in civilian areas, particularly in the surveillance of sensitive areas (ports, borders) or industrial sites, as well as self driving cars.

Project coordination

Vincent Martin (Bertin Technologies)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

CEA Commissariat à l'énergie atomique et aux énergies alternatives
ISL Institut franco-allemand de recherches de Saint-Louis
BER Bertin Technologies

Help of the ANR 280,347 euros
Beginning and duration of the scientific project: - 24 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter