Preference learning for adaptive OTA notifications

• US 10,534,602 B2
• Filed: 03/12/2018
• Issued: 01/14/2020
• Est. Priority Date: 03/12/2018
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Miller, Jason Michael, Sangameswaran, Sangeetha, Tseng, Fling, Madrid, Daniel Joseph
• Primary Examiner(s): Chen, Qing
• Application Number:
  US15/918,337
• Publication Number:
  US 20190278581A1
• Time in Patent Office: 673 Days
• Field of Search: 717168-173
• CPC Class Codes: G06F 8/65 Updates security arrangements therefor G06F21/57
• CPC Class Codes: G06F 9/4451 User profiles; Roaming

Abstract

A vehicle includes a controller and a processor. The processor is programmed to prompt the user with a selection to install a software update to the controller responsive to a user preference value, computed from user selections whether or not to install software updates associated with contextual information that matches a software update to be installed, exceeding a threshold defined by a priority of the software update, and update the user preference value per the selection.

Claims

1. A vehicle comprising:

a controller; and

a processor programmed to;

receive a first manifest from an update server,

prompt a user to install software updates specified by the first manifest,

receive user selections of whether or not to install the software updates,

send the user selections of whether or not to install the software updates, including contextual information, to the update server, to cause the update server to determine a user preference value indicative of a user acceptance tendency for software updates, the user preference value being determined using a machine learning system trained by observing historically how users decide to accept or reject software updates in view of the contextual information, the contextual information including time, user, and route information corresponding to the user selections of whether or not to install the software updates,

receive the user preference value from the update server,

prompt for a selection of whether or not to install a controller software update to the controller responsive to the user preference value exceeding a predefined threshold defined by a priority of the controller software update, the user preference value being computed from historical selections of whether or not to install software updates that correspond to contextual information that matches current contextual information for the controller software update to be installed, and

send the user selections of whether or not to install the software updates, including the contextual information, to the update server, to cause the update server to further update the user preference value per the selection of whether or not to install the controller software update to the controller.

2. The vehicle of claim 1, wherein the contextual information includes time, day, and vehicle location.

3. The vehicle of claim 1, wherein the user preference value is an average of user preference values for a plurality of profiles matching the contextual information.

4. The vehicle of claim 1, wherein the priority of the controller software update includes one of a low priority defining a first threshold value, a medium priority defining a second threshold value lower than the first threshold value, or a high priority defining a third threshold value lower than the second threshold value.

5. The vehicle of claim 1, wherein the processor is further programmed to receive a second manifest from the update server specifying a plurality of software updates to be installed, the controller software update being included in the second manifest.

6. A method comprising:

receiving a first manifest from an update server to a vehicle;

prompting a user to install software updates specified by the first manifest;

receiving user selections of whether or not to install the software updates to the vehicle;

sending the user selections of whether or not to install the software updates to the vehicle, including contextual information, to the update server, to cause the update server to determine a user preference value indicative of a user acceptance tendency for software updates, the user preference value being determined using a machine learning system trained by observing historically how users decide to accept or reject software updates in view of the contextual information, the contextual information including time, user, and route information corresponding to the user selections of whether or not to install the software updates to the vehicle;

receiving the user preference value from the update server;

prompting for a response to install a further software update to the vehicle responsive to the user preference value exceeding a threshold defined by a priority of the further software update;

installing the further software update when the response affirms installation; and

sending the response to the update server to further train the user preference value.

7. The method of claim 6, further comprising computing the user preference value as an average of user preference values for a plurality of profiles matching the contextual information.

8. The method of claim 6, wherein the priority of the further software update includes one of a low priority defining a first threshold value, a medium priority defining a second threshold value lower than the first threshold value, or a high priority defining a third threshold value lower than the second threshold value.

9. The method of claim 6, further comprising receiving a second manifest from the update server specifying a plurality of software updates to be installed, the further software update being included in the second manifest.

10. The method of claim 6, further comprising prompting for the response using a display screen of the vehicle.

11. The method of claim 6, further comprising receiving the response using human-machine input (HMI) controls of the vehicle.

12. A system comprising:

a memory storing a plurality of user preference values; and

a processor programmed to;

receive user selections indicative of whether or not users chose to install software updates to vehicles, the user selections including contextual information regarding conditions surrounding a decision of whether or not to perform the installation of the software updates to the vehicles, the contextual information including at least two of time, user, or route information corresponding to the user selections of whether or not to install the software updates to the vehicles;

determine a user preference value indicative of a user acceptance tendency for software updates, the user preference value being determined using a machine learning system trained by observing historically how users decide to accept or reject software updates in view of the contextual information;

send, to a vehicle, the user preference value matching contextual information received in a request from the vehicle, the vehicle using the user preference value and a priority of a software update to determine whether to prompt a user to install the software update;

receive a selection from the vehicle in response to the prompt; and

further update the user preference value according to the selection using the machine learning system.

13. The system of claim 12, wherein the contextual information includes time, day, and vehicle location.

14. The system of claim 12, wherein the user preference value is an average of user preference values for a plurality of profiles matching the contextual information.

15. The system of claim 12, wherein the processor is further programmed to send, to the vehicle, a manifest specifying a plurality of software updates to be installed, the software update being included in the manifest.

16. The system of claim 12, wherein the processor is further programmed to determine the user preference value by operations including:

responsive to receiving a user selection indicating that the user has selected to apply all software updates at a location, increment a user preference value for the location by a first value, responsive to receiving a user selection indicating that the user has selected not to apply all software updates at the location, decrement the user preference value for the location by the first value,

responsive to receiving a user selection indicating that the user has selected to install a single software update, increment the user preference value for the location by a second value, the second value being smaller than the first value,

responsive to receiving a user selection indicating that the user has selected not to install a single software update, increment the user preference value for the location by the second value, and

responsive to the increment or decrement operations, clip the user preference value to be within a predefined value range.

17. The system of claim 16, wherein the processor is further programmed to set importance scores to predefined thresholds, and wherein, for a given importance score and a user preference value for current contextual information, the importance scores are configured to cause the user to be prompted to install the single software update responsive to the user preference value exceeding the given importance score.

18. The system of claim 16, wherein the processor is further programmed to initialize the user preference value to a predefined value corresponding to a priority of a software update to be installed.

19. The system of claim 16, wherein the processor is further programmed to initialize the user preference value to an importance score corresponding to a priority of a software update to be installed.

20. The system of claim 12, wherein the processor is further programmed to:

determine an overall preference value for a user as a weighted average of preference values for the user across contextual information; and

utilize the weighted average of preference values for the user as a default value for the user preference value for the user.