Manuel Pavy, CNES; Valentin Baral, CNES; Emmanuel Delande, CNES; Yvan Gary, CNES; Pascal Richard, CNES; Raphel Ferme, CS Group
Keywords: break-up, fragmentation, cataloguing, telescopes, correlation, filter,
Abstract:
In 2024, we witnessed two major fragmentation events in geostationary orbit vicinity, a critical region for various space applications: the Atlas V Centaur on Sep 6, 2024, and the Intelsat 33E on Oct 19, 2024. Fragmentations, whether due to collisions or explosions, play a major part in the current growth of space debris in the GEO region. These events, although rare, have significant impacts on the space environment, especially at GEO altitudes where pieces of debris linger in orbit almost indefinitely.
These unfortunate events underline the critical importance of space surveillance. The timely description of the cloud of debris, and its integration to the space catalog, are essential to maintaining the safety and sustainability of space activities. Estimating the composition (size and individual states) of the cloud of debris right after its inception is paramount to the efficiency of the cataloging process, but it entails specific challenges:
1) Observing the cloud right after the fragmentation event requires specific cueing strategies of the sensor network, beyond the usual surveillance routine;
2) The measurement-to-track association is particularly ambiguous in the early stages of the cloud’s lifetime, especially so since a priori information about the composition of the cloud is scarce at best.
This paper will address these two critical steps, in the context of the French orbital catalog maintained by CNES.
We will first explain how, in the first nights, the system was able to acquire photos and garner information useful for the next nights with only three telescopes. This ability is based on the following principle:
The sooner a fragment is catalogued, the more time we have to catalogue others.
When evidence about a potential fragmentation event is collected, we program the telescope to observe according to two strategies: track the main body and detect other fragments in one hand and follow the new observed fragments in the other hand.
Image taken in the immediate aftermath of the fragmentation event, showing the detected debris
Then, we will focus on the implementation of the re-observation process, a concept introduced in [AMOS2023#35], and applying to the signal processing chain of the French orbital catalog. We will show that this agile and efficient planner plays a key role in collecting valuable information about the cloud of debris in the decisive period that follows its formation, and helped CNES in collecting critical data right after the Intelsat 33E fragmentation.
Next, we will introduce a new data association method for telescope measurements, adapted to the context of fragmentation events. Said method is based on an approximation of the underlying objects’ orbits leading to a timely association step, from which an initial description of the cloud of debris can be quickly established and integrated to the catalog.
For instance, on the following figure, we can notice the equinoctial parameters hx and hy are characteristics of each fragment represented by a different color. These characteristics help us to correlate the tracklets (and their detections) to an object or let a larger track emerge. These new tracks will be confirmed in the following days by a dedicated programming based on their Keplerian parameters.
Segregation of fragments by orbital planes using equinoctial parameters Hx/Hy
We will show how this mechanism is implemented in the signal processing chain of the French orbital catalog. Each uncorrelated tracklet is considered with objects from the neighboring hx/hy plane and analyzed on these restricted possibilities: if the plan includes a limited number of object, an orbit restitution is attempt to confirm the correlation; else, all of this is drawn and submitted to an operator.
Finally, we will illustrate these new methods to the Intelsat 33E fragmentation event. They led us to establishing the state estimates of individual pieces of debris from the second day of observation onwards, and to feeding the French orbital catalog over the following weeks to reach 80+ catalogued fragments within 40 days.
Data association process in the first weeks following the fragmentation event
The previous figure shows the dispersion of objects over time: it represents the projected longitudes on geostationary orbit as a function of time for all catalogued object.
Since the main hypothesis is based on the approximation of a circular orbit, we will present which adaptations are needed to generalize this method to a broader range of fragmentation events. We will illustrate these modifications on the break-up, of Atlas V Centaur, another fragmentation event in 2024, that occurred on a GTO orbit. We will present results from our a posteriori analysis, based on real images acquired with the re-observation method.
[AMOS2023#35] Pavy, M., Richard, P., Delande, E., and Théron, S., “Improving The Operational Signal Processing Chain for Faster Acquisition of New Objects to The French National Catalogue of Orbital Objects.”, in Proceedings of the Advanced Maui Optical and Space Surveillance (AMOS) Technologies Conference, 2023, Art. no. 35.
Date of Conference: September 16-19, 2025
Track: Space Debris