Austin Probe, Emergent Space Technologies, Inc.; Graham Bryan, Emergent Space Technologies, Inc.; Timothy Woodbury, Emergent Space Technologies, Inc.; Evan Novak, Emergent Space Technologies, Inc.; Shiva Iyer, The University of Texas at Austin; Apoorva Karra, The University of Texas at Austin; Moriba Jah, University of Texas at Austin
Keywords: Conjunction Assessment, Collision Avoidance, Space Traffic Management, Autonomous
Abstract:
Conjunction assessment is one of the most important components of operational satellite safety and that importance is continually growing due to the proliferation of missions and constellations in LEO. The difficulty and complexity are increased when coupled with the implementation of autonomous maneuvering for swarms or constellations, and further increased when such systems begin interacting with other autonomously maneuvering systems. With many large-scale autonomous constellations such as SpaceX Starlink, Amazon Kuiper, and other commercial providers as well as the proposed persistent LEO constellations from the SDA and MDA scheduled for deployment in the coming decade finding a scalable solution to this problem is key to Space Sustainability. The problem of conjunction assessment or conjunction analysis (CA) is currently managed by legacy centralized systems such as that managed by the 18th Space Control Squadron. These systems monitor space assets for potential collisions and alert the operators so that they can maneuver their satellites to avoid potential collisions. The current system requires inefficient communication between operators and the current turnaround time is such that unnecessary avoidance maneuvers are performed out of an abundance of caution. In many cases, as time passes and knowledge of the vehicle state improves, the conjunction risk is resolved on its own. Secondly, the current infrastructure does not have a clear path for supporting autonomous space systems that have the capacity for performing their own maneuvers. Emergent Space Technologies and the University of Texas at Austin are working to develop a prototype system to marry on-board spacecraft autonomy and navigation with a networked ground hub that autonomously manages a real-time space object catalog. This effort focuses on combining ground and space-based (edge) platforms to deliver the capability to quantify and assess on-orbit collision risk between space objects and use this space situational awareness to support decision-making processes that maximize desired outcomes. Several entities, such as SpaceXs Starlink satellites, have implemented autonomous collision avoidance maneuvers, but to date, these are executed with oftentimes flawed and outdated information and thus the collision risk is not realistic. To address this shortfall, we are working to develop the tools necessary to synchronize planned autonomous maneuvers and better models for estimating conjunction risk. Our team is building a prototype hub where spacecraft operators can submit the most recent state data and future planned maneuvers. This hub will screen the planned maneuvers to identify potential conjunction risks. To improve this process, we are also working to accurately model anthropogenic space object (ASO) motion within this system. There are five things that influence our inference of ASO motion:
The actual astrodynamics experienced by ASOs
Our models of (1) which are inaccurate and imprecise
The actual sensors observing ASOs which are inaccurate and imprecise
Our models of (3) which do not accurately or precisely capture or characterize the real inaccuracy and imprecision of our sensors
The choice of method of inference (e.g. linear regression, plethora of filters, etc.)
Our approach will be to address each of these and quantify the sensitivity to the inferred or perceived collision risk as compared to actual risk (via modeling and simulation) from uncertainties and errors in each. We will combine modeling and simulation along with real ASO data and information, as well as historical ASO population information, to achieve a holistic product that can identify the most relevant potential conjunctors for a selected space vehicle and accurately predict the potential conjunctions. Once the potential objects of concern are identified their state information and projected time of closest approach are uploaded to the space vehicle. Leveraging the CA capabilities of Emergents Autopilot and Navigator flight software and integrating this catalog of high-risk objects enables autonomous collision avoidance (COLA) based on the latest on-board vehicle state information. This system has the potential to pave the way for sustainable autonomous space traffic management as LEO grows more and more congested and a prototype being developed for evaluation as part of a NASA Space Traffic Management Experiment to be testing using the Starling satellites in 2023. This paper will discuss the capabilities of the conjunction ground hub, the methodology used for identifying potential conjunctions, the capabilities of the onboard flight software for COLA, simulation results for the system, and details of the planned flight experiment.
Date of Conference: September 27-20, 2022
Track: Conjunction/RPO