Max Geissbuhler, Slingshot Aerospace; Nicholas Noyes, Slingshot Aerospace; Jesse Gutknecht, Slingshot Aerospace; Tyler Mancini, Slingshot Aerospace; Jeffrey Shaddix, Slingshot Aerospace; Belinda Marchand, Slingshot Aerospace
Keywords: Launch tracking, LEOP, framework, rideshare, multi-payload launch, characterization, cataloguing, insights, optical fence
Abstract:
The cost of launching individual payloads to low Earth orbit (LEO) has drastically decreased in the past decade for a multitude of reasons. One major contributor to the decreased cost is the emergence of ride-sharing launch vehicles. Ride-shares distribute the cost of launch across many payloads, offering a low-cost alternative to traditional single-payload launches. While the cost to individual operators may be reduced by rideshares, one major implication of rideshare launches involves the burden of tracking, characterization, and cataloguing of such launches, which may contain over 100 distinct payloads. This is particularly important for uncooperative launches, which may disguise several nefarious payloads amongst the deployment of innocuous payloads. At present, it can take over 30 days for objects from multi-payload launches to be officially catalogued by space-track.org. During this time, the payloads may present a conjunction risk to other objects on-orbit or may perform nefarious actions that may be unnoticed. Thus, rapid cataloguing of such payloads is necessary to maintain proper space situational and domain awareness.
There are several different data sources used for performing space object cataloging, each with their unique benefits and drawbacks. This paper will focus on describing a method for tracking multi-payload launches in LEO utilizing data derived from electro-optical (EO) systems. EO systems provide a combination of astrometric and photometric insights, both of which can be combined for discriminating between payloads on multi-payload missions. We utilize both cued telescope and uncued optical EO systems; the former being important for closely spaced object discrimination, and the latter having the advantage of collecting data on launches without tasking or prior launch nominal information. With these data sources in mind, the method for tracking multi-payload launches is summarized below.
First, a nominal estimate of the deployment orbit is generated to identify data collected in the vicinity of the deployment orbit. The nominal orbit is generated using publicly available information about the launch, such as launch site, anticipated launch time, and temporary flight restriction (TFR) locations for launch vehicle drop-zones. If the launch will contain payloads for a known constellation, historic information such as inclination and deployment altitude are used to refine the nominal launch profile.
Next, the nominal launch profile is updated using the exact time of the launch. This is accomplished by shifting the right ascension of the ascending node (RAAN) to accommodate for any discrepancies in the launch time. This is a necessary step, as some launches may have windows on the order of hours, and the nominal must be accurate to within seconds of the actual launch profile to be useful for cueing gimbaled systems. If the launch time is inaccurate, the rotation of the Earth will drastically change the RAAN component of the profile and will result in inaccuracies exceeding hundreds of arcminutes with respect to a ground observer.
Then, an initial orbital determination (IOD) is performed against all space surveillance data identified as likely originating from the launch using a batch least squares method. This data originates from a global network of uncued EO optical fence systems. From the IOD candidates, analyst objects are generated for initial cataloguing (i.e. the identity of the candidates are unknown, but the orbital states can be catalogued for future identification). Then, a conjunction assessment is performed to identify the neighborhood of the payloads and to identify any risks to high value assets. A stability assessment is then performed to categorize the candidates (i.e. 3-axis stable versus tumbling). Finally, the analyst objects are promoted to the full catalog once a NORAD ID is assigned by space-track.org.
After describing a process for rapidly cataloguing multi-payload launches, this paper will apply the process to a real-world launch using data collected from a combination of gimbaled and optical fence EO systems. With this example, the authors will demonstrate that the framework is capable of discriminating between the upper stage rocket body and over 10 payloads, all of which is accomplished within 72 hours of the launch.
Date of Conference: September 16-19, 2025
Track: Space Domain Awareness