Systems and methods for machine-assisted vehicle inspection
AssigneeState Farm Mutual Automobile Insurance Company
InventorsRyan Knuffman, Bradley A. Sliz, Lucas Allen
Life of this patent
Estimating the term…
Abstract
One or more processing elements may be trained to identify vehicle damages or vehicle damages based upon training data. A remotely-controlled (RC) and/or autonomously operated inspection device, such as a ground vehicle or drone, may capture one or more sets of imaging data indicative of at least a portion of an automotive vehicle, such as all or a portion of the undercarriage. The one or more sets of imaging data may be analyzed using the trained processing elements to identify a damage to the vehicle or defect of the vehicle.
Claims18
A computer-implemented method for identifying damage to a current vehicle based upon training sets of imaging data indicative of reference vehicles, the method comprising:
- training one or more processing elements to identify vehicle damage based upon the training sets of imaging data, the training sets of imaging data being indicative of, for each of the reference vehicles, at least a portion of the reference vehicle;
- receiving, via a communication element, an indication of a claim associated with reported damage to the current vehicle;
- determining, via the one or more trained processing elements, based upon the claim, at least a portion of the current vehicle of which to capture imaging data;
- obtaining, via the communication element, one or more current sets of imaging data captured by an imaging unit of an inspection device that is at least one of autonomously operated or remotely-controlled, each the one or more current sets of imaging data being indicative of the determined at least the portion of the current vehicle;
- and
- analyzing, via the one or more trained processing elements, the one or more current sets of imaging data to identify or verify the damage to the current vehicle.
A computer system configured to identify damage to a current vehicle based upon training sets of imaging data indicative of reference vehicles, the computer system comprising:
- one or more processing elements;
- and
- one or more memories comprising computer-executable instructions that, when executed by the one or more processing elements, cause the computer system to:
- train the one or more processing elements to identify vehicle damage based upon the training sets of imaging data, the training sets of imaging data being indicative of, for each of the reference vehicles, at least a portion of the reference vehicle;
- receive, via a communication element, an indication of a claim associated with reported damage to the current vehicle;
- determine, via the one or more trained processing elements, based upon the claim, at least a portion of the current vehicle of which to capture imaging data;
- obtain, via the communication element, one or more current sets of imaging data captured by an imaging unit of an inspection device that is at least one of autonomously operated or remotely-controlled, each the one or more current sets of imaging data being indicative of the determined at least the portion of the current vehicle;
- and
- analyze, via the one or more trained processing elements, the one or more current sets of imaging data to identify or verify the damage to the current vehicle.
Dependencies are inferred from the claim text.