Programmatically identifying a personality of an autonomous vehicle
AssigneeMicron Technology, Inc.
InventorRobert Richard Noel Bielby
Life of this patent
Estimating the term…
Abstract
Systems and methods for assigning personalities to autonomous vehicles are disclosed. In one embodiment, a method is disclosed comprising receiving data from an autonomous vehicle; generating a vector representing the autonomous vehicle based on the data; classifying the vector into one or more personalities; receiving a search query from a user; identifying one or more autonomous vehicles responsive to the search query based on personalities assigned to the one or more autonomous vehicles, the one or more autonomous vehicles including the autonomous vehicle; and transmitting the one or more autonomous vehicles to the user.
Claims14
A method comprising:
- initializing, by a central system, statistics associated with autonomous vehicles;
- updating, by the central system, the statistics based on operational data received from the autonomous vehicles;
- clustering, by the central system, the autonomous vehicles into a set of clusters;
- assigning, by the central system, each cluster in the set of clusters a personality identifier using a natural language processing (NLP) algorithm, the NLP algorithm comprising:
- retrieving a set of text descriptions associated with one of a make, model, or year of the autonomous vehicles in a given cluster,
- generating a plurality of sentiments for the given cluster based on the text descriptions,
- ranking the plurality of sentiments based on a type of statistics associated with the given cluster, the ranking performed by comparing a similarity of the type of statistics with the plurality of sentiments, and
- selecting a subset of the plurality of sentiments as the personality identifier;
- training, by the central system, a model using the clusters and the personality identifiers;
- and
- predicting, by the central system, one of a future fault or repair date for an autonomous vehicle by inputting an object representing the autonomous vehicle into the model.
A non-transitory computer-readable storage medium for tangibly storing computer program instructions capable of being executed by a computer processor, the computer program instructions defining steps of:
- initializing statistics associated with autonomous vehicles;
- updating the statistics based on operational data received from the autonomous vehicles;
- clustering the autonomous vehicles into a set of clusters;
- assigning each cluster in the set of clusters a personality identifier using a natural language processing (NLP) algorithm, the NLP algorithm comprising:
- retrieving a set of text descriptions associated with one of a make, model, or year of the autonomous vehicles in a given cluster,
- generating a plurality of sentiments for the given cluster based on the text descriptions,
- ranking the plurality of sentiments based on a type of statistics associated with the given cluster, the ranking performed by comparing a similarity of the type of statistics with the plurality of sentiments, and
- selecting a subset of the plurality of sentiments as the personality identifier;
- training a model using the clusters and the personality identifiers;
- and
- predicting one of a future fault or repair date for an autonomous vehicle by inputting an object representing the autonomous vehicle into the model.
A device comprising:
- a processor;
- and
- a storage medium for tangibly storing thereon program logic for execution by the processor, the stored program logic, when executed by the processor, causing the processor to perform steps of:
- initializing statistics associated with autonomous vehicles;
- updating the statistics based on operational data received from the autonomous vehicles;
- clustering the autonomous vehicles into a set of clusters;
- assigning each cluster in the set of clusters a personality identifier using a natural language processing (NLP) algorithm, the NLP algorithm comprising:
- retrieving a set of text descriptions associated with one of a make, model, or year of the autonomous vehicles in a given cluster,
- generating a plurality of sentiments for the given cluster based on the text descriptions,
- ranking the plurality of sentiments based on a type of statistics associated with the given cluster, the ranking performed by comparing a similarity of the type of statistics with the plurality of sentiments, and
- selecting a subset of the plurality of sentiments as the personality identifier;
- training a model using the clusters and the personality identifiers;
- and
- predicting one of a future fault or repair date for an autonomous vehicle by inputting an object representing the autonomous vehicle into the model.
Dependencies are inferred from the claim text.