Systems and methods for identifying a vehicle platform using machine learning on vehicle bus data

Patent metadata

Field	Value
Publication number	US-2023401444-A1
Application number	US-202318450346-A
Country	US
Kind code	A1
Filing date	Aug 15, 2023
Priority date	Feb 27, 2020
Publication date	Dec 14, 2023
Grant date	—

Abstract

Embodiments of the invention include a vehicle telematics system that obtains vehicle bus data for a time period, determines identification information regarding a vehicle platform using a machine learning process on the vehicle bus data, and obtains a set of communication data for communicating with at least one vehicle module on the vehicle bus based on the identified vehicle platform.

First claim

1 . A server for determining configuration settings for a telematics device to facilitate communications with vehicle modules of a vehicle, the server comprising:
a communications interface; one or more processors; and a memory having stored therein a plurality of instructions that, when executed by the one or more processors, cause the server to:
receive, via the communications interface and from a telematics device of each of a plurality of vehicles, an electronic signature of each corresponding vehicle, wherein the electronic signature comprises vehicle bus data of the corresponding vehicle captured by the corresponding telematics device during a time period;
generate a machine learning model using the electronic signatures received from the telematics devices, wherein the machine learning model is configured to classify a subsequently received electronic signature to one of a plurality of vehicle platforms; and
identify, for each one of the plurality of vehicle platforms, a set of configuration settings usable by a telematics device to communicate with one or more vehicle modules of the corresponding vehicle.

2 . The server of claim 1 , wherein the electronic signature includes identification data that identifies the corresponding vehicle or a vehicle module of the corresponding vehicle.

3 . The server of claim 1 , wherein to generate the machine learning model comprises to analyze a pattern, a frequency, or a content of data included in each electronic signature.

4 . The server of claim 1 , wherein to generate the machine learning model comprises to analyze a frequency of a vehicle module ID included in each electronic signature, wherein the frequency comprises a number of occurrences of the vehicle module ID in the electronic signature.

5 . The server of claim 1 , wherein to generate the machine learning model comprises to:
receive, for each electronic signature, a labeled electronic signature, wherein the labeled electronic signature comprises data added to the electronic signature that identifies characteristics of the electronic signature; and train a supervised learning classifier using the labeled electronic signature.

6 . The server of claim 1 , wherein to generate the machine learning model comprises to establish an unsupervised machine learning classifier to perform, on each electronic signature, cluster analysis to identify one or more relationships between data included in the electronic signatures.

7 . The server of claim 1 , wherein the plurality of instructions, when executed by the one or more processors, further cause the server to transmit the machine learning model to each of the telematics devices.

8 . The server of claim 1 , wherein the plurality of instructions, when executed by the one or more processors, further cause the server to transmit the machine learning model to a telematics device different from the telematics devices from which the electronic signatures were received.

9 . The server of claim 1 , wherein the set of configuration settings comprises On-Board Diagnostic Parameter IDs (OBD-II PIDs).

10 . The server of claim 1 , wherein the electronic signature comprises Controller Area Network (CAN) vehicle bus data of the corresponding vehicle captured by the corresponding telematics device during the time period.

11 . A server for determining configuration settings for a telematics device to facilitate communications with vehicle modules of a vehicle, the server comprising:
a communications interface; one or more processors; and a memory having stored therein a plurality of instructions that, when executed by the one or more processors, cause the server to: receive, via the communication interface and from the telematics device of the vehicle, an electronic signature of the vehicle, wherein the electronic signature comprises vehicle bus data of the corresponding vehicle captured by the corresponding telematics device during a time period; determine, using a machine learning model, a vehicle platform of the vehicle based on the electronic signature; identify, based on the determined vehicle platform, a set of configuration settings usable by the telematics device to communicate with the vehicle modules of the vehicle; and transmit, via the communication interface, the set of configuration settings to the telematics device.

12 . The server of claim 11 , wherein the electronic signature includes identification data that identifies the vehicle or one or more of the vehicle modules of the vehicle.

13 . The server of claim 11 , wherein the time period is dynamically determined.

14 . The server of claim 11 , wherein the initiation of the time period is based on a vehicle event of the vehicle.

15 . The server of claim 11 , wherein to determine the vehicle platform based on the electronic signature comprises to analyze a pattern, a frequency, or a content of data included in the electronic signature.

16 . The server of claim 11 , to determine the vehicle platform based on the electronic signature comprises to analyze a frequency of a vehicle module ID included in the electronic signature, wherein the frequency comprises a number of occurrences of the vehicle module ID in the electronic signature.

17 . The server of claim 11 , wherein to determine the vehicle platform based on the electronic signature comprises to determine the vehicle platform using, on the electronic signature, a supervised learning classifier that has been trained on labeled electronic signatures received from other telematics devices of other vehicles.

18 . The server of claim 11 , wherein to determine the vehicle platform based on the electronic signature comprises to determine the vehicle platform using, on the electronic signature, an unsupervised machine learning classifier that has previously performed cluster analysis on electronic signatures received from other telematics devices of other vehicles to identify one or more relationships between data included in the electronic signatures received from the other telematics devices.

19 . The server of claim 1 , wherein the set of configuration settings comprises On-Board Diagnostic Parameter IDs (OBD-II PIDs).

20 . The server of claim 1 , wherein the electronic signature comprises Controller Area Network (CAN) vehicle bus data of the corresponding vehicle captured by the the telematics device during the time period.

Assignees

• Calamp Corp

Inventors

• Fu Brian
• Abhishek
• Rau Amrit Krishna

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• G06F17/18

  for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G07C5/0808 Primary

  Diagnosing performance data (testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005) · CPC title