Method of processing and analysing vehicle driving big data and system thereof

• US 9,600,541 B2
• Filed: 02/06/2015
• Issued: 03/21/2017
• Est. Priority Date: 05/02/2014
• Status: Active Grant

• Current Assignee: Kookmin University
• Original Assignee: Kookmin University Industry Academy Cooperation Foundation
• Inventors: Cho, Won Hee, Choi, Eun Mi
• Primary Examiner(s): Anwari, Maceeh
• Application Number:
  US14/616,633
• Publication Number:
  US 20150317844A1
• Time in Patent Office: 774 Days
• Field of Search: 709/29.3
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G07C 5/008 communicating information to a remotely located station transmission systems for measured values G08C

Abstract

Provided is a method of processing and analyzing vehicle driving big data, the method including: refining vehicle driving data of raw data; acquiring statistical data based on the refined vehicle driving data; and performing mining analysis based on at least one of the refined vehicle driving data and the acquired statistical data.

Claims

1. A method of processing vehicle driving data by a big data processing apparatus, the method comprising:

detecting a first outlier in each field of sensing data included in the vehicle driving data, the first outlier representing field values out of a predetermined range;

generating first refined data by correcting or removing the first outlier from the sensing data;

detecting a second outlier in the first refined data, the second outlier representing field values having a deviation greater than a predetermined threshold between consecutive field values of a same field or between different fields correlated to each other; and

generating second refined data by removing the second outlier from the first refined data;

generating statistical data based on the second refined data; and

performing mining analysis based on at least one of the second refined data and the statistical data.

2. The method of claim 1, wherein the vehicle driving data further includes identification data,

wherein the identification data indicates at least one of a vehicle identification number, a vehicle type, a vehicle registration number, a transportation business operator registration number, and a driver identification code, shipping document information, name of shipper and commodity, and

wherein the each field of the sensing data indicates at least one of a travel distance, a driving time, starting time, a data acquisition period, a data acquisition time, a speed, an Revolution Per Minute (RPM), a break signal, a position, an azimuth, and acceleration, location of the vehicle, On Board Diagnostics II (OBD-II) information.

3. The method of claim 1, wherein the generating first refined data comprises changing the field values of the first outlier to new field values within the predetermined range.

4. The method of claim 1, wherein the statistical data includes driving statistical data and tendency statistical data,

wherein the driving statistical data indicates at least one of a daily travel distance, a total driving time, an average speed, a maximum speed, and a maximum RPM, mileage, fuel consumption, energy consumption, Exhaust emissions, driving area, distance between cars, and

wherein the tendency statistical data indicates at least one of a number of overspeed, a number of dangerous overspeed, a number of long-term overspeed, a number of sudden acceleration, a number of sudden deceleration, a number of sudden start, a number of sudden stop, a number of overtaking, a number of sudden course changing, a number of sudden turn, a number of idling, and a number of high RPM.

5. The method of claim 1, further comprising:

detecting a field value for each field of the second refined data that is repetitively counted within a predetermined time in generating the statistical data; and

correcting the statistical data to count the detected field value once during the predetermined time.

6. The method of claim 1, wherein the performing of the mining analysis includes:

classifying drivers to a plurality of driving tendencies based on the statistical data; and

indexing the drivers with a degree of a corresponding driving tendency.

7. The method of claim 6, wherein the performing of the mining analysis further includes:

generating index information indicating a result of the indexing; and

analyzing a correlation between at least one of fields of the statistical data and the index information.

8. The method of claim 6, wherein the performing of the mining analysis further includes:

generating index information showing distribution of a result of the indexing with respect to transportation business operators, based on a transportation business operator registration number field of the second refined data.

9. The method of claim 6, wherein the performing of the mining analysis further includes:

generating index information showing a result of the indexing;

detecting a change of the index information, the second refined vehicle driving data, and the statistical data after a driver recognizes the index information; and

analyzing a change of a driving tendency of the driver based on a result of the detecting.

10. The method of claim 1, wherein the performing of the mining analysis includes:

detecting a repeated pattern occurring during a specific time period in the second refined data and the statistical data; and

analyzing a driving type of drivers during the specific time period based on the pattern.

11. The method of claim 1, wherein the statistical data includes history data indicating an accident history field and a maintenance history field, and

the performing of the mining analysis includes:

generating a learning model for predicting a change in at least one of the accident history field and the maintenance history field, based on the second refined data and the statistical data; and

analyzing at least one of accident potential and a maintenance schedule by using the learning model.

12. The method of claim 11, wherein the performing of the mining analysis further includes generating index information indicating a degree of the accident potential and how much time is left before next maintenance according to the maintenance schedule, based on a result of the analyzing.

13. The method of claim 1, wherein the performing of the mining analysis includes:

extracting location information of vehicles based on the second refined data and the statistical data; and

determining a location having high accident potential or high air pollution potential based on the extracted location information and driving tendencies that are determined based on the statistical data.

14. The method of claim 1, wherein the performing of the mining analysis includes:

extracting location information of vehicles based on the second refined data and the statistical data; and

analyzing driving tendencies of drivers for each administrative district, based on the extracted location information, the statistical data, and map data.

15. The method of claim 1, further comprising:

collecting the vehicle driving data from vehicle driving recorders or sensors of a plurality of vehicles.

16. The method of claim 1, wherein the statistical data includes statistics showing at least one of fuel consumption and exhaust emissions with respect to at least one of drivers, transportation business operators, regions, and roads, and

the performing of the mining analysis includes generating at least one of eco driving information and overloading information based on the statistics.

17. The method of claim 1, wherein the statistical data includes statistics showing diagnostic code of On Board Diagnostics II (OBD-II) with respect to at least one of drivers, transportation business operators, vehicle models, vehicle years, and

the performing of the mining analysis includes generating at least one of prediction information regarding abnormality of vehicle, accident potential, and vehicle maintenance schedule based on the statistics.

18. A system for processing vehicle driving data, the system comprising:

a storing unit configured to store the vehicle driving data; and

a processing unit configured to;

detect a first outlier in each field of sensing data included in the vehicle driving data, the first outlier representing field values out of a predetermined range;

generate first refined data by correcting or removing the first outlier from the sensing data;

detect a second outlier in the first refined data, the second outlier representing field values having a deviation greater than a predetermined threshold between consecutive field values of a same field or between different fields correlated to each other; and

generate second refined data by removing the second outlier from the first refined data;

generate statistical data based on the second refined data, and

perform mining analysis based on at least one of the second refined data and the statistical data.