Jehyun Cha, Hanyang University; Joonghyun Ryu, Hanyang University; Chanyoung Song, Hanyang University; Misoon Mah, Air Force Office of Scientific Research (AFOSR); Roberto Furfaro, University of Arizona; Deok-Soo Kim, Hanyang University
Keywords: Conjunction prediction, Optimal maneuver, Constellation, Starlink, Webserver, Dynamic Voronoi diagram, Spatial reasoning, Moving objects
Abstract:
Conjunction management (CM) involves the prediction and avoidance of a possible conjunction among orbital objects in a foreseeable future and is one of the most critical issues for space traffic management (STM). As witnessed by the Starlink project of SpaceX, the number of valuable orbital objects is rapidly growing. Hence, STM is getting more important and so is the accurate and efficient decision-making for CM.
However, it remains a mathematical and computational challenge. Despite many prior works, a unified theoretical and computational framework of CM has not been uncovered yet. One of the main challenges has been and will be the combinatorial explosion among the objects to be considered for a closer investigation during a prediction time window: We want to identify all potential conjunctions for a many-to-many case rather than one-to-many for the entire prediction time window. To be more concrete, we want to locate a conjunction as soon as and as accurately as possible and find the best, not a better, maneuver plan which will be safe over the entire prediction time window.
In the two prior AMOS reports, we have presented a paradigm-shifting approach based on the dynamic Voronoi diagram (DVD) of moving orbital objects [REF1]. The first one in AMOS2017 reported that DVD can be used to predict a conjunction accurately and efficiently [REF2] and the second in AMOS2018 reported that the same DVD can be used to find the best maneuver plan by quickly enumerating and evaluating many alternatives during the entire time window [REF3].
In the current study, we report a novel webserver COOP (Conjunctive Orbital Objects Predictor and Planner) which implements the two algorithms above together with realistic and practical astrodynamics considerations. All challenges are beautifully solved within a single mathematical and computational framework of the DVD of spherical 3D balls moving through the linear approximation of orbits. COOP guarantees no missing case of conjunction prediction, computes a solution very quickly, and can be used to efficiently solve diverse application problems involving many objects including the conjunction prediction problem.
COOP produces an event sequence file, called EVENTSEQ, with a very compact size of O(NW) where N represents the number of objects and W the length of prediction time window. By tracking the events in the EVENTSEQ file, COOP can replay the transition of DVD accurately and efficiently. Once the EVENTSEQ file is available on COOP, anyone can download and run his or her own application program to get an answer to many queries very efficiently by scanning the EVENTSEQ file. COOP will be a key engine of constellations, e.g. Starlink, etc. and diverse applications are expected to be found.
COOP downloads the TLE data from Space Catalogue and preprocesses to get the EVENTSEQ data for a prescribed prediction time window. Upon a users request, COOP currently provides answers for the following three types of queries:
Query 1. Find all proximal events in time horizon;
Query 2. Find the proximal events for an object-of-interest;
Query 3. Find the optimal maneuver plan for an object-of-interest.
Notes.
Object-of-interest (OOI): The object under investigation, e.g. a particular satellite.
Proximal event: An event that two objects (including OOI) are closer than a predefined threshold distance.
Optimal maneuver plan: The best maneuver path (or orbit) among many alternatives of the current path of OOI.
COOP is planned to be ported to the VerSSA system (leverging the CyVerse system[ REF4]) which has been developed and maintained by University of Arizona so that users can have convenient accesses to COOP.
References
[REF1] Chanyoung Song. Jehyun Cha, Mokwon Lee, Deok-Soo Kim, Dynamic Voronoi Diagram for Moving Disks, IEEE Transactions on Visualization and Computer Graphics, DOI: 10.1109/TVCG.2019.2959321, 2019.
[REF2] Jehyun Cha, Drew McNeely, Joonghyun Ryu, Misoon Mah, Moriba Jah, Deok-Soo Kim, DVD-COOP: Innovative Conjunction Prediction using Voronoi-filter based on the Dynamic Voronoi Diagram of 3D Spheres, Advanced Maui Optical and Space Surveillance Technologies Conference(AMOS), September 19-22, Wailea Marriot, Maui, Hawaii, USA, 2017.
[REF3] Jehyun Cha, Joonghyun Ryu, Mokwon Lee, Chanyoung Song, Youngsong Cho, Paul Schumacher, Misoon Mah, Deok-Soo Kim, DVD-COOP for Optimal Design of Maneuver Path for Conjunctive Objects, Advanced Maui Optical and Space Surveillance Technologies Conference(AMOS), September 12-14, Wailea Marriot, Maui, Hawaii, USA, 2018.
[REF4] Devisetty, U.K., Kennedy, K., Sarando, P., Merchant, N. and Lyons, E., 2016. Bringing your tools to CyVerse discovery environment using Docker.?F1000Research,?5.
Date of Conference: September 15-18, 2020
Track: Astrodynamics