Space Surveillance

We design and develop processing chains dedicated to space surveillance, covering all functions relating to the detection, tracking and location of objects in orbit.

Advanced processing algorithms

We develop high-performance algorithms for:

• the detection of low-contrast objects in noisy environments
• multi-object tracking and trajectory estimation
• location in celestial reference frame

These developments are based on a proven software foundation that is continually enhanced to meet the growing demands of the SSA sector.

Contributions of artificial intelligence

To complement traditional approaches, we incorporate machine learning methods to:

• improve detection performance under degraded conditions
• perform advanced noise reduction and image restoration
• automate certain processing steps

These approaches are implemented in a controlled manner in combination with physical models to ensure robustness.

Multi-sensor compatibility

Our solutions are designed to be compatible with a wide range of optical sensors:

• ground-based telescopes
• satellite-borne sensors
• observation systems dedicated to space surveillance

This flexibility allows for easy integration into existing or future architectures.

Simulation and validation

We develop high-fidelity optical space scene simulators, enabling us to:

• generate representative data to test and validate algorithms
• assess system performance
• train artificial intelligence models
• train operators

These tools are a key driver for accelerating development and ensuring performance under operational conditions.