AIRA is a member of the consortium for Attitude Analysis of Artificial Satellites

The Centre National d’Études Spatiales (CNES) has officially awarded the consortium led by Aldoria and a major French industrial company the contract for experimentation in the field of attitude analysis within the EU SST (EU Space Surveillance and Tracking) program. The research and development initiative, scheduled for the period September 2025 – June 2026, aims to design and validate innovative algorithms for characterizing the attitude state of active and inactive space objects, through the fusion of optical, radar, and passive RF data, enhanced with machine learning techniques.

Pioneering Advances in Estimating the Attitude of NonCooperative Objects

Within EU SST, CNES seeks to fully leverage the capabilities of current and future sensors to improve the accuracy and timeliness of estimating the attitude of non-cooperative space objects. The research and development contract will focus on:

• Innovation in data fusion: combining optical images, radar echoes, and passive radio frequency signatures with advanced machine learning models to provide more robust and reliable attitude solutions.
• Enhanced observation strategies: developing new operational concepts that exploit fused multi-sensor inputs to better detect, track, and characterize satellites or space debris that are tumbling or difficult to cooperate with.
• Fast transition to service: delivering an operational attitude estimation service six months after the completion of the project, in close coordination with CNES and the EU SST community.

A Consortium with Complementary Expertise

The consortium also relies on the collaboration of European laboratories and partners:

• ONERA (Office National d’Études et de Recherches Aérospatiales) – expertise in adaptive optics, essential for improving the quality of optical data under difficult conditions.
• AIRA (Astronomical Institute of the Romanian Academy) – provision of extensive sets of optical data, required for volume and diversity in training and validating machine learning models.
• Quasar Science Resources – software development and integration for the industrialization of algorithms within CNES’s operational environment.

Expected Benefits and Roadmap

By fusing heterogeneous sensor data streams and using machine learning, the project aims:

• To enable new observation strategies that dynamically allocate sensor resources based on real-time evaluations of rotation or tumbling rates.
• To improve the responsiveness and accuracy of the service by reducing the time between the detection of an object and the delivery of the attitude estimate to users.
• To provide a credible and rapidly deployable solution, as both co-leaders of the consortium have operational experience in attitude analysis (although previously on single-sensor datasets).

"It is a recognition of the skills and value of the AIRA team participating within the consortium, an expertise that our team will aim to further expand in future activities through an academic-industrial partnership," emphasizes Mirel Bîrlan, director of the Astronomical Institute.

Published on: Oct 02, 2025