US 11,836,684 B2
Automated vehicle repair estimation by preferential ensembling of multiple artificial intelligence functions
Abhijeet Gulati, San Diego, CA (US)
Assigned to Mitchell International, Inc., San Diego, CA (US)
Filed by Mitchell International, Inc., San Diego, CA (US)
Filed on Sep. 30, 2020, as Appl. No. 17/039,287.
Claims priority of provisional application 62/908,354, filed on Sep. 30, 2019.
Claims priority of provisional application 62/908,348, filed on Sep. 30, 2019.
Claims priority of provisional application 62/908,361, filed on Sep. 30, 2019.
Prior Publication US 2021/0097505 A1, Apr. 1, 2021
Int. Cl. G06Q 10/20 (2023.01); G06Q 40/08 (2012.01); G06N 20/20 (2019.01); G06N 5/04 (2023.01); G06Q 10/10 (2023.01); G06T 7/00 (2017.01); G06N 20/00 (2019.01); G06F 18/214 (2023.01)
CPC G06Q 10/20 (2013.01) [G06F 18/2148 (2023.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01); G06N 20/20 (2019.01); G06Q 10/10 (2013.01); G06Q 40/08 (2013.01); G06T 7/0002 (2013.01); G06T 7/0004 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30248 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system, comprising:
a hardware processor; and
a non-transitory machine-readable storage medium encoded with instructions executable by the hardware processor to perform a method comprising:
receiving, from each source of a plurality of sources, a respective vehicle repair recommendation set for a damaged vehicle, wherein each vehicle repair recommendation in the vehicle repair recommendation set identifies a recommended vehicle repair operation of a plurality of vehicle repair operations for the damaged vehicle and includes one or more images of the damaged vehicle, wherein each source of the plurality of sources is a respective trained artificial intelligence function with an initial weight corresponding to the recommended vehicle repair operation;
determining a respective source rank of each source from a plurality of the source ranks;
generating a composite vehicle repair recommendation set that identifies the recommended vehicle repair operation of the plurality of vehicle repair operations in an order determined according to the source ranks of the respective sources by providing the vehicle repair recommendation sets to an ensembling artificial intelligence function, wherein the ensembling artificial intelligence function generates the composite vehicle repair recommendation set based on the vehicle repair recommendation sets provided to the ensembling artificial intelligence function, wherein the ensembling artificial intelligence function is trained using a plurality of first vehicle repair training sets, and wherein each first vehicle repair training set comprises:
one or more images of a further damaged vehicle, and
a composite vehicle repair recommendation set for the further damaged vehicle;
providing the composite vehicle repair recommendation set to one or more vehicle repair insurance claims management systems;
determining, via the ensembling artificial intelligence function, whether a predetermined retraining trigger event occurs, wherein the predetermined retraining trigger event comprises an evaluation metric falling below a predefined metric;
requesting one or more of the artificial intelligence functions be retrained when the predetermined retraining trigger event occurs; and
retraining the artificial intelligence functions by adjusting the weights corresponding to the recommended vehicle repair operations.