Space Asset Management

Project lead: SERC

Researchers: James Bennett, Daniel Kucharski, Michael Lachut, Marek Möckel, Sven Flegel, Alex Pollard, David Kooymans, James Allworth

Participants: SERC, Optus Satellite Systems, EOS Space Systems

Non Participants: Graz Observatory, Zimmerwald Observatory

The core function of this RP was to develop a SOC database containing space objects of interest.

Optus space object ephemeris data, range data and manoeuvre data is automatically retrieved and stored in the SOC. The ephemeris data is used in the conjunction assessment process allowing SERC to provide conjunction assessments which include Optus’ satellite manoeuvre plans. Currently Optus use data provided by CSpOC, however, testing of the SERC system is underway.

EOSSS tracking data is automatically stored in the SOC. When sufficient data has been stored, automatic ephemeris generation is triggered which performs an orbit determination using numerical integration of a complete force model. A complete force model provides the best accuracy. This produces a state and covariance for use in scheduling, conjunction assessments (or refinements) and conjunction probability estimation. The states and covariances are also used to estimate the time when the orbital state error becomes non-Gaussian (ie: not a normal distribution). The connection between the database and the CATW system has been made so the state and covariance can be sent automatically for conjunction assessment.

The conjunction information is then returned to the database for storage and viewing. The conjunction information is autonomously provided to Optus by automatic email transfer.

The catalogue pushes the object states and covariances to the conjunction assessment system for automated conjunction analysis. The resulting conjunction data message is returned to the database and an interface has been set up to monitor the conjunctions.

Innovative research and technologies have resulted from the object characterisation work that demonstrate the application of complex methods of science, hardware and software:

• Photon pressure force on space debris TOPEX/Poseidon (T/P) measured by satellite SLR and photometry. The combined photometric and SLR data collected over the 11-year time span indicate that T/P is continuously gaining rotational energy since being decommissioned in January 2006. The physical analysis of the satellite’s rotational dynamics makes it possible to determine the amount of the solar photon pressure force that is exerted on the object and perturbs its orbital motion. The results of this research have been published in the American Geographical Union Earth and Space Science, 2017: http://dx.doi.org/10.1002/2017EA000329
• Development of the High-rate photon counter: extremely low noise of 45 dark counts-per-second, max count rate of 50 million counts-per-second, data storing rate: 50,000 count rate samples per second.
• Voltus – post-processing software for charged-coupled device (CCD) imagery analysis, light curve extraction and spin determination.

The following techniques using machine learning have been developed to improve the correlation of tracked objects, designed for reliable object identification or use in the laser manoeuvre campaign:

• Methods for corrected object magnitude generated from light curves.
• Improved object identification that is independent of flat and dark frame collection.
• Improved multi-object tracking with astrometric and photometric correction enabled for all detected objects in the image.
• Includes streaking Resident Space Objects (RSO).
• Magnitudes down to 17.8 have been identified in GEO. This is an ongoing area to detect and identify faint objects.
• RSOs in all orbital regimes are able to be tracked in burst-mode with the SERC GEO tracker telescope Zyla camera which gives fast frame imagery for object characterisation.
• Improved astrometric correction (within pipeline, rather than during the post-processing stage).
• Significantly more stars are mapped per track (order of 1-3 magnitudes) more than existing methods.
• Increased astrometric correction robustness for larger mount model errors.
• Astrometric correction has been verified using Consolidated Prediction Formats (CPF) for calibration objects.
• A visualisation tool has been developed to demonstrate the object tracking in the post-processing stage.

The pipeline has been tested and refined by:

• Parallelising the image processing pipeline to increase the processing speed.
• Investigating the detection limitations and determining the uncertainty in the characterisation of faint objects.
• Comparing the astrometric correction capabilities which are now in use with the correction performed at the observation post-processing stage.

Project lead: SERC

Researchers: James Bennett, Marek Möckel, Sven Flegel, Daniel Kucharski, Michael Lachut, Jeffrey Wardman, Andrew Edwards

Participants: SERC, Optus Satellite Systems, EOS Space Systems

SERC has developed a conjunction analysis system for all objects (all-on-all conjunction analysis) in the SOC which can provide actionable conjunction alerts to subscribers. The SERC conjunction system also distributes priority tasking to sensors in the SOC network to provide increased tracking and updating for potential collisions which are considered serious.

An Information-Gain based scheduler, which distributes the tasking of sensors by maximising the information gain, has been developed and optimised in C++ to allow for a larger network of sensors. The software has been tested and the scheduler used in the lead up to the on-sky experiments. 

The conjunction assessment system has been validated in test scenarios running on SERC’s own catalogued objects as well as on special perturbations, and ephemeris data provided by Optus. Results have been validated and are in agreement with CSpOC conjunction threat warnings. 

Currently the conjunction system allows for: 

• Whole catalogue all-on-all TLE conjunction assessments 
• TLE vs TLE 
• Multi-TLE versus multi-TLE 
• Multi-SP versus Multi-SP, eg CSpOC 
• SERC SP state vector versus SERC SP state vector 
• SERC SP versus operator ephemeris 
• Combinations of the above, eg CSpOC SP vs SERC SP 

The CATW system can fuse data from the EOSSS sensor network to improve collision predictions. State error information is generated during the SERC ephemeris generation allowing the probability of conjunction, and the time until the error becomes non-Gaussian, to be calculated. This information is unique to SERC and not provided by CSpOC for the TLEs or Special Perturbations (SP).

The CATW system has been set up as a server so multiple conjunction assessments tasks can be queued for execution. A client front-end interface has also been developed to allow the user to run and view the conjunction assessments.

Several historical close approach scenarios have been used to validate the results of the SERC CATW system. One example of this is the close approach between Optus 10 and a Block DM-SL rocket body on 26th November 2018.

This is a particularly interesting case due to the highly eccentric orbit of the piece of space debris. 

The SERC CATW collision prediction and the JSpOC (JSpOC has been renamed CSpOC) predictions were in agreement.

The SERC SP vs Optus ephemeris results compare well with the JSpOC results in this historical close approach and is part of the validation process of the SERC SP vs Optus ephemeris generation and conjunction assessment method.

CATW trials undertaken at Optus demonstrated the SERC CATW System produces actionable results to satellite operators.