	US 12,304,668 B2
	Satellite constellation scheduling
Skylar Cox, North Logan, UT (US); Justin Whitaker, Logan, UT (US); John Humble, Logan, UT (US); and Greg Droge, Nibley, UT (US)
Filed by Utah State University Space Dynamics Laboratory, North Logan, UT (US)
Filed on Sep. 20, 2022, as Appl. No. 17/933,602.
Claims priority of provisional application 63/246,261, filed on Sep. 20, 2021.
Prior Publication US 2023/0166873 A1, Jun. 1, 2023
Int. Cl. B64G 3/00 (2006.01); B64G 1/10 (2006.01); G06F 9/48 (2006.01)

CPC B64G 3/00 (2013.01) [G06F 9/4881 (2013.01); B64G 1/1007 (2013.01); B64G 1/1021 (2013.01)]

15 Claims

1. A method comprising:

generating communication graph nodes for at least one satellite wherein a node score is assigned to each communication graph node, the communication graph nodes comprise backbone nodes that represent the at least one satellite not performing a task and connect to every target communication graph node with a time value between a given backbone node and a subsequent backbone node;

calculating communication graph edges from the communication graph nodes for the at least one satellite, wherein the communication graph nodes and the communication graph edges comprise a communication graph;

calculating node scores for each communication graph edge by assigning the node scores of each communication graph node to a corresponding communication graph edge that connects to the communication graph node as edge scores and edge costs, wherein the edge scores describe a benefit associated with the corresponding communication graph node and the edge costs represent a transition between communication graph nodes based on a slew performed, a vector x of all edge flows x_ithat represents flow along a given edge e_ifor an optimization variable i, and a memory vector y of all memory needs y_lkfor the at least one satellite at given times k;

solving the communication graph to minimize slewing the at least one satellite and prevent each element of the memory vector from exceeding a maximum memory capacity by maximizing the utility u^Tz subject to Dx=b, A_gx≤1_n_{_g}, A_dataz=b_gdata, x_i∈{0, 1}∀i, 0≤y_l,k≤y_l,max∀ l,k, and z=[x^Ty^T]^Tto yield a communication plan, wherein u is the utility of all utility u_iassociated with the optimization variable i, D is an incidence matrix, b is a right-angle hypotenuse-side congruence of a flow constant, A_gis a group constraint matrix, b_gdatais a right-angle hypotenuse-side congruence of a group constraint, l is a given satellite, and z is all optimization parameters;

generating a points-of-interest plan from the communication plan; and

controlling the at least one satellite based on the points-of-interest plan by slewing the at least one satellite between orientations for each communication graph node.