Machine learning updating with sensor data

Patent metadata

Field	Value
Publication number	US-11760376-B2
Application number	US-202017136093-A
Country	US
Kind code	B2
Filing date	Dec 29, 2020
Priority date	Dec 29, 2020
Publication date	Sep 19, 2023
Grant date	Sep 19, 2023

Abstract

A system includes a vehicle control module, a vehicle gateway module, and a wired vehicle communications network communicatively coupling the vehicle gateway module to the vehicle control module. The vehicle control module is programmed to receive sensor data from at least one sensor, execute a machine-learning program trained to determine whether the sensor data satisfies at least one criterion, and transmit the sensor data satisfying the at least one criterion to the vehicle gateway module. The vehicle gateway module is programmed to transmit the machine-learning program to the vehicle control module; upon receiving the sensor data from the vehicle control module, store the sensor data; and upon establishing a connection with a remote server, transmit the sensor data to the remote server.

First claim

The invention claimed is:

1. A system comprising:
a vehicle control module on board a vehicle;
a vehicle gateway module on board the vehicle; and
a wired vehicle communications network on board the vehicle that is communicatively coupling the vehicle gateway module to the vehicle control module, the vehicle gateway module configured to transmit between subnetworks of different domains of the wired vehicle communications network;
wherein the vehicle control module is programmed to
receive sensor data from at least one sensor;
execute a machine-learning program, the machine-learning program previously trained to determine whether the sensor data satisfies at least one criterion, the machine-learning program outputting a determination that the sensor data satisfies the at least one criterion; and
transmit the sensor data satisfying the at least one criterion to the vehicle gateway module; and

the vehicle gateway module is programmed to
transmit the machine-learning program to the vehicle control module;
upon receiving the sensor data from the vehicle control module, store the sensor data; and
upon establishing a connection with a remote server, transmit the sensor data to the remote server.

2. The system of claim 1 , wherein the vehicle control module is further programmed to delete the sensor data after transmitting the sensor data to the vehicle gateway module.

3. The system of claim 1 , wherein the vehicle control module is further programmed to delete the sensor data after executing the machine-learning program to determine that the sensor data fails to satisfy the at least one criterion.

4. The system of claim 1 , wherein the vehicle gateway module is further programmed to, upon receiving the sensor data from the vehicle control module, add metadata to the sensor data, and to transmit the metadata to the remote server when transmitting the sensor data to the remote server.

5. The system of claim 4 , wherein the metadata includes location data.

6. The system of claim 4 , wherein the metadata includes time data.

7. The system of claim 1 , wherein the vehicle gateway module is further programmed to transmit a notification to the vehicle control module upon detecting that a prespecified event occurred.

8. The system of claim 7 , wherein the vehicle control module is further programmed to, upon receiving the notification, transmit the sensor data from a predetermined time before the notification.

9. The system of claim 7 , wherein the prespecified event includes at least one of a triggering of an impact sensor, an airbag deployment, or a brake force above a threshold.

10. The system of claim 1 ,
wherein the vehicle control module is a first vehicle control module, the at least one sensor is at least one first sensor, the sensor data is first sensor data, the machine-learning program is a first machine-learning program, and the at least one criterion is at least one first criterion;
the system further comprising a second vehicle control module communicatively coupled to the vehicle gateway module via the wired vehicle communications network;
wherein the second vehicle control module is programmed to
receive second sensor data from at least one second sensor;
execute a second machine-learning program trained to determine whether the second sensor data satisfies at least one second criterion; and
transmit the second sensor data satisfying the at least one second criterion to the vehicle gateway module.

11. The system of claim 10 , wherein the vehicle gateway module is further programmed to, upon receiving the first or second machine-learning program from the remote server, determine to which of the first or second vehicle control module to transmit the first or second machine-learning program.

12. The system of claim 1 , further comprising the remote server, wherein the remote server is programmed to train the machine-learning program with a dataset including the sensor data received from the vehicle gateway module.

13. The system of claim 12 , wherein the remote server is further programmed to transmit the machine-learning program to the vehicle gateway module.

14. The system of claim 1 , wherein the at least one criterion includes image recognition of a prespecified type of roadway user.

15. The system of claim 14 , wherein the prespecified type of roadway user is at least one of a pedestrian, a bicycle, or a nonvehicular moving object.

16. The system of claim 1 , wherein the at least one criterion includes image recognition of a prespecified type of roadway sign.

17. The system of claim 16 , wherein the prespecified type of roadway sign is at least one of a damaged sign or an occluded sign.

18. The system of claim 1 , wherein the at least one criterion includes image recognition of a prespecified type of roadway section.

19. The system of claim 18 , wherein the prespecified type of roadway section is at least one of a roundabout, a tollway entrance, or a construction zone.

20. A method comprising:
transmitting a machine-learning program to a vehicle control module on board a vehicle over a wired vehicle communications network on board the vehicle by a vehicle gateway module on board the vehicle, wherein the vehicle gateway module is configured to transmit between subnetworks of different domains of the wired vehicle communications network, the machine-learning program is previously trained to determine whether sensor data satisfies at least one criterion, and the machine-learning program outputs a determination that the sensor data satisfies the at least one criterion;
executing the machine-learning program by the vehicle control module;
transmitting the sensor data satisfying the at least one criterion to the vehicle gateway module by the vehicle control module;
upon receiving the sensor data from the vehicle control module, storing the sensor data by the vehicle gateway module;
upon establishing a connection with a remote server, transmitting the sensor data to the remote server by the vehicle gateway module; and
training the machine-learning program with a dataset including the sensor data received from the vehicle gateway module by the remote server.

Assignees

• Ford Global Tech Llc

Inventors

• Chari Varun
• Tonshal Basavaraj
• Kondoju Panduranga

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06F18/214 Primary

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• H04L67/12 Primary

  specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks · CPC title

• B60W60/00 Primary

  Drive control systems specially adapted for autonomous road vehicles · CPC title