Ryan Shepperd, Iridium
Keywords: Iridium 33, Conjunction Assessment, Collision Avoidance
Abstract:
On February 10, 2009, a collision between the operational Iridium 33 and the derelict COSMOS 2251 promoted policy changes to usher a new era of collision assessment and avoidance. At the time, only General Perturbation Two Line Element sets (TLEs) were released publicly by the U.S. Government (USG). The simplified analytical model and lack of a corresponding covariance were insufficiently accurate and precise for satellite collision avoidance. The data sufficiently capable of collision assessment of well tracked objects was the numerically integrated Special Perturbation (SP) model of the High Accuracy Catalog (HAC), which at the time was restricted from the public but used internally by the group then known as the Joint Space Operations Center (JSpOC) for conjunction detection.
The HAC alone, however, could not delineate the conjunction risks of operational and maneuverable satellites as it did not incorporate operator measurements, maneuver histories, or maneuver plans. Eight hours before the conjunction, Iridium 33 fired its hydrazine thruster twice to raise the semimajor axis by 8.3 meters for station-keeping. Iridium did not know that COSMOS 2251, as modelled in the HAC, was on the end trajectory of that maneuver. The JSpOC did not know Iridium maneuvered because no mechanism existed for willing commercial and foreign operators to inform them, and 8 hours was too short a time to uncooperatively detect the maneuver. Each party, Iridium and the JSpOC, needed the other half of the information to know a collision was probable.
The resulting collision led to immediate policy changes to correct this communication shortcoming. Recognizing the importance to space safety, Iridium worked with the JSpOC and 18th Space Control Squadron to identify minimal necessary components of SP data from the HAC while security concerns over comprehensively useful data were addressed. From relative state vectors to redacted Orbit Conjunction Messages (OCMs), known as Conjunction Summary Messages (CSMs) to external operators, this process evolved into the 95% capture screening volumes with results shared via the new CCSDS Conjunction Data Message (CDM), which effectively contains the information provided in the precursor CSM. Equally important, all Iridium maneuvers since 2009 are included in ephemerides and shared with the USG, which returns data of the secondary objects in conjunction with those ephemerides and assessed before the maneuver. With help gleaned from the NASA Conjunction Assessment Risk Analysis (CARA) team, Iridium implemented a collision assessment process to reduce the risk of future collisions.
In 2011, the JSpOC and Iridium recreated SP versus SP and SP versus Iridium ephemerides screenings, using historical data, to ascertain the effectiveness of preventing the 2009 collision using the then current data sharing of CSMs and analysis tools. That review both underscored how the data could prevent future collisions between large objects but also served as a caution for setting action thresholds and addressing covariance realism. The illuminating story has long been told by insiders and discussed at the Operator Collision Assessment Workshops in Paris and Denver but hasnt yet been recorded in an easily accessible and referenceable paper or wider forum. Considering a lack of easily accessible information about the collision could improperly prioritize future improvements or research, this paper seeks to outline what is known by those who operated Iridium 33 and what the available data later revealed. The facts of the collision and summary of the recreated screening results will be presented along with a discussion of how additional improvements since 2011 aid collision avoidance and where the gaps remain.
Date of Conference: September 19-22, 2023
Track: Conjunction/RPO