US 12,242,615 B2
Adversarial reinforcement learning system for simulating security checkpoint environments
Brian Jacob Lewis, Washington, DC (US); Jason Adam Deich, Arlington, VA (US); Stephen John Melsom, Reston, VA (US); Kara Jean Dodenhoff, Reston, VA (US); and William Tyler Niggel, Wilmington, NC (US)
Assigned to NOBLIS, INC., Reston, VA (US)
Filed by NOBLIS, INC., Reston, VA (US)
Filed on Aug. 22, 2022, as Appl. No. 17/892,228.
Application 17/892,228 is a continuation of application No. 16/864,826, filed on May 1, 2020, granted, now 11,423,157.
Claims priority of provisional application 62/847,592, filed on May 14, 2019.
Prior Publication US 2022/0414231 A1, Dec. 29, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 21/00 (2013.01); G06F 21/56 (2013.01); G06F 21/57 (2013.01)
CPC G06F 21/577 (2013.01) [G06F 21/566 (2013.01); G06F 2221/034 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A method for simulating a spatial environment, the method performed by an adversarial reinforcement learning system comprising one or more hardware processors, the method comprising:
generating, by a first model, a threat mitigation input, wherein the threat mitigation input comprises instructions for controlling one or more simulated objects in the simulation, wherein the first model is configured to minimize one or more harm outcomes of the simulation; and
generating, by a second model, a threat input, wherein the threat input comprises instructions for controlling one or more simulated objects in the simulation, wherein the second model is distinct from the first model and is configured to maximize one or more harm outcomes of the simulation; and
executing a first portion of the simulation based at least in part on the threat mitigation input and the threat input.