Information sharing platform and method capable of providing bidirectional vehicle state information and system having information sharing platform

Patent metadata

Field	Value
Publication number	US-11775820-B2
Application number	US-202016880738-A
Country	US
Kind code	B2
Filing date	May 21, 2020
Priority date	Dec 11, 2019
Publication date	Oct 3, 2023
Grant date	Oct 3, 2023

Abstract

An information sharing platform of providing bidirectional vehicle state information between a driver and a vehicle, the information sharing platform may include a communication controller which collects measured data and vehicle Controller Area Network (CAN) information by sensors installed to components capable of diagnosing a vehicle state; and a graphic controller which provides a driver with diagnosis result output information that is generated based on a predetermined selection criterion among the components through Deep Learning based diagnosis using big data having the collected data.

First claim

What is claimed is:

1. An information sharing platform of providing bidirectional vehicle state information between a driver and a vehicle, the information sharing platform comprising:
a communication controller configured to collect data measured by sensors that are installed in components configured to diagnose a vehicle state and vehicle Controller Area Network (CAN) information; and
a graphic controller configured to provide a driver with diagnosis result output information that is generated based on a predetermined selection criterion among the components through Deep Learning based diagnosis using big data having the collected measured data and CAN information;
wherein the measured data is composed of vibration data and noise data, and the diagnosis result output information is stored to provide an abnormal prediction information; and
wherein the diagnosis result output information includes a corresponding component and an abnormal probability value.

2. The information sharing platform of claim 1 , wherein the predetermined selection criterion is set for each rank by comparing the vibration data of the measured data with noise the data of the measured data.

3. The information sharing platform of claim 1 , wherein the graphic controller is configured to:
generate driving tendency analysis information classifying the driver's tendency based on running pattern data of the driver.

4. The information sharing platform of claim 3 ,
wherein the driving tendency analysis information classifies the driver's tendency into a fuel efficiency type or a power performance type by comparing a current diagnosis result with a previous diagnosis result.

5. The information sharing platform of claim 1 , wherein the Deep Learning based diagnosis comprises:
a bidirectional Gated Recurrent Unit (GRU), a Deep Neural Network (DNN), and an Attention mechanism.

6. The information sharing platform of claim 1 , wherein the Deep Learning based diagnosis is executed when an execution instruction of the driver is input by voice or text.

7. The information sharing platform of claim 6 , wherein the graphic controller comprises:
a plurality of artificial intelligence modules, wherein the Deep Learning based diagnosis is executed based on the execution instruction.

8. The information sharing platform of claim 7 , wherein the plurality of artificial intelligence modules are configured to match with the execution instruction in advance.

9. The information sharing platform of claim 7 , wherein the execution instruction is re-input when the accuracy of confirming a specific-rank execution instruction among the execution instructions is equal to or less than a predetermined accuracy.

10. A big data based state diagnosis information providing system comprising:
an information sharing platform of providing bidirectional vehicle state information between a driver and a vehicle, the information sharing platform comprising:
a communication controller configured to collect data measured by sensors that are installed in components configured to diagnose a vehicle state and vehicle Controller Area Network (CAN) information;
a graphic controller configured to provide a driver with diagnosis result output information that is generated based on a predetermined selection criterion among the components through Deep Learning based diagnosis using big data having the collected measured data and CAN information;
wherein the measured data is composed of vibration data and noise data, and the diagnosis result is stored to provide an abnormal state prediction information; and
wherein the diagnosis result output information is includes a corresponding component and an abnormal probability value; and
a central server connected to the information sharing platform by a communication network and configured to:
store a database; and
provide answer information according to an inquiry type input by the driver; and

at least one communication terminal configured to input the answer information.

11. The big data based state diagnosis information providing system of claim 10 , wherein the inquiry type is at least one of vehicle manual related information, current vehicle state related information, or technical knowledge information.

12. An information sharing method of providing bidirectional vehicle state information between a driver and a vehicle, the information sharing method comprising:
collecting, by a communication controller, data measured by sensors of components and vehicle Controller Area Network (CAN) information;
executing, by a graphic controller, Deep Learning based diagnosis using big data composed of the collected measured data and CAN information; and
providing, by the graphic controller, a driver with diagnosis result output information generated by a predetermined selection criterion among the components by executing the Deep Learning based diagnosis;

wherein the measured data is composed of vibration data and noise data, and the diagnosis result is stored to provide an abnormal state prediction information; and
wherein the diagnosis result output information includes a corresponding component and an abnormal probability value.

13. The information sharing method of claim 12 , wherein the method further comprises:
generating, by the graphic controller, driving tendency analysis information classifying the driver's tendency based on running pattern data of the driver.

14. The information sharing method of claim 13 , wherein the generation of the driving tendency analysis information includes:
classifying, by using the driving tendency analysis information, the driver's tendency into a fuel efficiency type or a power performance type by comparing a current diagnosis result with a previous diagnosis result.

15. The information sharing method of claim 12 , wherein the Deep Learning based diagnosis comprises:
a bidirectional Gated Recurrent Unit (GRU), a Deep Neural Network (DNN), and an Attention mechanism.

16. The information sharing method of claim 12 , wherein the execution of the Deep Learning based diagnosis includes:
executing, by the graphic controller, the Deep Learning based diagnosis when an execution instruction of the driver is input by voice or text.

17. The information sharing method of claim 16 , wherein the execution of the Deep Learning based diagnosis includes:
executing, by the graphic controller, the Deep based Learning diagnosis based on the execution instruction, wherein the graphic controller comprises a plurality of artificial intelligence modules.

18. The information sharing method of claim 17 , where in the method comprises:
matching, by the plurality of artificial intelligence modules, the execution instruction in advance.

Assignees

• Hyundai Motor Co Ltd
• Kia Motors Corp

Inventors

• Jung In-Soo
• Lee Dong-Chul

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• G06F9/3836

  Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution · CPC title

• G06N3/044

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