US 12,293,304 B1
Decision recommendation via causal feature display
Aruna Jammalamadaka, Camarillo, CA (US); Rajan Bhattacharyya, Sherman Oaks, CA (US); Samuel D. Johnson, Santa Monica, CA (US); and Rashmi N. Sundareswara, Topanga, CA (US)
Assigned to HRL LABORATORIES, LLC, Malibu, CA (US)
Filed by HRL Laboratories, LLC, Malibu, CA (US)
Filed on May 19, 2021, as Appl. No. 17/325,082.
Claims priority of provisional application 63/105,665, filed on Oct. 26, 2020.
Int. Cl. G06N 5/04 (2023.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01)
CPC G06N 5/04 (2013.01) [G06N 7/01 (2023.01); G06N 20/00 (2019.01)] 17 Claims
OG exemplary drawing
 
1. A system for generating and displaying decision recommendations, the system comprising:
an autonomous vehicle; and
one or more processors and a non-transitory computer-readable medium associated with the autonomous vehicle, the non-transitory computer-readable medium having executable instructions encoded thereon such that when executed, the one or more processors perform operations of:
receiving a set of historical observations X and a set of navigational courses of action Y corresponding to the set of historical observations X;
estimating a conditional density P(Y|X) between the set of historical observations X and the set of navigational courses of action Y;
clustering the set of historical observations X according to the P(Y) X), thereby obtaining a plurality of observational clusters;
merging the observational clusters to determine a causal partition having corresponding causal features which map to various navigational courses of action;
mapping a new observation to the causal partition and computing a ranking of navigational courses of action for the new observation;
providing a visualization of the causal features and the ranking of navigational courses of actions corresponding to the new observation to a display; and
causing the autonomous vehicle to perform a physical maneuver corresponding to a highest rank navigational course of action.