Big data-based driving information provision system and method thereof

Patent metadata

Field	Value
Publication number	US-2021182671-A1
Application number	US-202016923787-A
Country	US
Kind code	A1
Filing date	Jul 8, 2020
Priority date	Dec 11, 2019
Publication date	Jun 17, 2021
Grant date	—

Abstract

A big data-based driving information provision system may include a sensor configured to measure and collect state monitoring data of an engine, vehicle monitoring data, and vibration data; an engine electronic control unit (ECU) configured to generate a combustion characteristic index (CCI) data of the engine; and a graphic controller configured to generate a primary deep learning model which classifies the big data including the state monitoring data, the vehicle monitoring data, the vibration data, and the CCI into at least two categories.

First claim

What is claimed is:

1 . A big data-based driving information provision system, comprising:
a sensor configured to measure and collect state monitoring data of an engine, vehicle monitoring data, and vibration data; an engine electronic control unit (ECU) configured to generate a combustion characteristic index (CCI) data of the engine; and a graphic controller configured to generate a primary deep learning model, wherein the primary deep learning model is configured to classify big data including the state monitoring data, the vehicle monitoring data, the vibration data, and the CCI data into at least two data categories.

2 . The big data-based driving information provision system of claim 1 , wherein the primary deep learning model is configured to classify combinations of correlation coefficients between the big data into the at least two data categories.

3 . The big data-based driving information provision system of claim 2 , wherein the primary deep learning model is configured to classify the big data into clusters by applying a k-means algorithm to the correlation coefficients.

4 . The big data-based driving information provision system of claim 3 , wherein the primary deep learning model is configured to apply, after the k-means algorithm is applied, a Gaussian mixture model (GMM) and a deep neural network (DNN).

5 . The big data-based driving information provision system of claim 1 , wherein the ECU is configured to calculate the CCI using a crank angle, an angular velocity, and an angular acceleration of the engine.

6 . The big data-based driving information provision system of claim 5 , wherein the ECU is configured to:
calculate the CCI for each cylinder; and when a median deviation value of a median with respect to a cylinder is greater than a predetermined threshold, determine that a cylinder having a smallest minimum value is an abnormal combustion.

7 . The big data-based driving information provision system of claim 1 , wherein the graphic controller is configured to:
generate the at least two data categories for each cylinder of the engine.

8 . The big data-based driving information provision system of claim 1 , wherein the graphic controller is configured to:
determine that the at least two data categories are an irregular vibration index based on the CCI; and determine that the at least two data categories are a grade and a vibration level which are matched to a range from a maximum value to a minimum value of the irregular vibration index.

9 . The big data-based driving information provision system of claim 1 , wherein the graphic controller is configured to:
generate a secondary deep learning model that is configured to predict an irregular vibration index in an idle state of the engine based on the primary deep learning model.

10 . The big data-based driving information provision system of claim 9 , wherein the system further comprises:
a vehicle communication terminal configured to receive, from a central server, service time information indicating a time required for an inspection service by comparing prediction information based on the secondary deep learning model and a predetermined value for maintenance.

11 . A method of providing big data-based driving information, comprising:
collecting big data based on engine management system (EMS) data and controller area network (CAN) data; collecting, by an engine electronic control unit (ECU), a combustion characteristic index (CCI) representing a combustion characteristic of an engine; classifying the collected big data according to types using a primary deep learning model; and controlling a CCI of a problematic cylinder by analyzing the classified big data.

12 . The method of claim 11 , wherein classifying the big data comprises:
classifying, by the primary deep learning model, combinations of correlation coefficients between the big data into at least two data categories.

13 . The method of claim 12 , wherein classifying the big data comprises:
classifying the big data into clusters by applying a k-means algorithm to the correlation coefficients.

14 . The method of claim 13 , wherein classifying the big data comprises:
after applying the k-means algorithm, classifying the big data into the clusters by applying a Gaussian mixture model (GMM) and a deep neural network.

15 . The method of claim 11 , wherein the method comprises:
calculating the CCI using a crank angle, an angular velocity, and an angular acceleration of the engine.

16 . The method of claim 15 , wherein the method comprises:
calculating the CCI for each cylinder; and when a median deviation value of a median with respect to a cylinder is greater than a predetermined threshold, determining that a cylinder having a smallest minimum value is an abnormal combustion.

17 . The method of claim 15 , wherein the method comprises:
generating the at least two data categories for each cylinder of the engine.

18 . The method of claim 17 , wherein the method comprises:
determining that the at least two data categories are an irregular vibration index based on the CCI; and determining that the at least two data categories are a grade and a vibration level which are matched to a range from a maximum value to a minimum value of the irregular vibration index.

19 . The method of claim 11 , wherein classifying the big data comprises:
generating a secondary deep learning model that is configured to predict an irregular vibration index in an idle state of the engine based on the primary deep learning model.

20 . The method of claim 19 , wherein the method further comprises:
receiving, by a vehicle communication terminal, service time information indicating a time required for an inspection service from a central server; and comparing prediction information based on the secondary deep learning model and a predetermined value for maintenance.

Assignees

• Hyundai Motor Co Ltd
• Kia Motors Corp

Inventors

• Lee Dong-Chul
• Jung In-Soo

Classifications

• G06N3/045

  Combinations of networks · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N20/00

  Machine learning · CPC title

• G01M15/12

  by monitoring vibrations · CPC title