US 12,456,384 B2
System and methods for implementing an unmanned aircraft tracking system
Alex Brofos, Hopkinton, NH (US); Brett C. Bishop, Waltham, MA (US); and Craig P. Vandervest, Lutherville Timonium, MD (US)
Assigned to The MITRE Corporation, McLean, VA (US)
Filed by The MITRE Corporation, McLean, VA (US)
Filed on Apr. 7, 2020, as Appl. No. 16/842,563.
Claims priority of provisional application 62/830,973, filed on Apr. 8, 2019.
Prior Publication US 2024/0331553 A1, Oct. 3, 2024
Int. Cl. G08G 5/72 (2025.01); G01S 13/91 (2006.01); G01S 13/933 (2020.01); G08G 5/22 (2025.01)
CPC G08G 5/727 (2025.01) [G01S 13/91 (2013.01); G01S 13/933 (2020.01); G08G 5/22 (2025.01)] 30 Claims
OG exemplary drawing
 
1. A method for classifying flying objects in an airspace, the method comprising:
generating a persistent clutter mask, comprising:
mapping training data to a three-dimensional grid corresponding to a field of view of a radar system, and determining one or more cells of the grid that have a coefficient of variation below a predetermined threshold, wherein the persistent clutter mask comprises the one or more cells of the grid that have a coefficient of variation below the predetermined threshold;
receiving plot data from the radar system, wherein the plot data indicates a location of one or more flying objects encountered by the radar system;
masking a portion of the received plot data based on the persistent clutter mask;
identifying one or more values of attributes belonging to the one or more flying objects based on portions of the received plot data from the radar system that are not masked;
classifying each identified flying objects into one or more categories based on their one or more values of one or more attributes, wherein classifying the identified flying objects into one or more categories comprises:
generating a hypercube with a plurality of dimensions, wherein each dimension of the hypercube is based on an attribute of the one or more attributes;
mapping each identified flying object's one or more attributes to a data point within the hypercube;
determining one or more geometric proximities of the data point to one or more previously mapped data points within the hypercube; and
placing the identified flying object into one or more categories based on the one or more geometric proximities to the one or more previously mapped data points within the hypercube.