KR101565030B1 - Decision system for error of car using the data analysis and method therefor - Google Patents

Abstract

The present invention relates to an error determination system and method for an automobile, and more particularly, to an error correction system for generating an input correlation script that defines a collection specification of data generated by a sensor and an actuator of an automobile, An input processor for extracting data that can be used for generating an error judgment model or an algorithm; Receives the data extracted from the input processing unit, generates a model or algorithm for error determination, and determines whether or not the data transmitted in real time from the sensor and the actuator through the vehicle network of the vehicle according to the generated error determination model or algorithm Learning and analysis department; An output processor for transmitting data determined as an error to the vehicle network as a result of the learning and analysis unit; The present invention relates to a system and a method for determining an error of an automobile using data analysis.

Description

Technical Field [0001] The present invention relates to a system and a method for error determination of a vehicle using data analysis,

The present invention relates to an error determination system and method for an automobile, and more particularly, to an error correction system for generating an input correlation script that defines a collection specification of data generated by a sensor and an actuator of an automobile, An input processor for extracting data that can be used for generating an error judgment model or an algorithm; Receives the data extracted from the input processing unit, generates a model or algorithm for error determination, and determines whether or not the data transmitted in real time from the sensor and the actuator through the vehicle network of the vehicle according to the generated error determination model or algorithm Learning and analysis department; An output processor for transmitting data determined as an error to the vehicle network as a result of the learning and analysis unit; The present invention relates to a system and a method for determining an error of an automobile using data analysis.

In general, a communication network such as a CAN communication is used as an internal communication network used by an electronic control unit (ECU) of an automobile. Various sensor information of the vehicle, information about the state of the vehicle and control command information are transmitted and received, So that the control unit (ECU) can perform control.

However, a sensor present in a vehicle is difficult to detect an operation error, and may cause a malicious operation such as a sudden start in operation of the vehicle.

Disclosure of Invention Technical Problem [8] The present invention has been made in order to overcome the above-mentioned problems of the prior art, and it is an object of the present invention to provide an automotive vehicle which uses data analysis capable of recognizing abnormally changes The present invention provides a judgment system and a method thereof.

According to an aspect of the present invention, there is provided an error determination system and method for an automobile using data analysis, the method comprising: generating an input correlation script that defines a collection specification of data generated by a sensor and an actuator of an automobile; An input processor for extracting data that can be used to generate an error judgment model or algorithm of the data as defined in the input correlation script; Receives the data extracted from the input processing unit, generates a model or algorithm for error determination, and determines whether or not the data transmitted in real time from the sensor and the actuator through the vehicle network of the vehicle according to the generated error determination model or algorithm Learning and analysis department; An output processor for transmitting data determined as an error to the vehicle network as a result of the learning and analysis unit; And a control unit.

An automobile error determination system and method using the data analysis of the present invention having the above-described configuration can automatically detect an error due to a sensor failure or an abnormal operation in the automobile, And it has an effect of increasing the safety on the running of the vehicle and improving the fuel efficiency by detecting the operation error by judging the vehicle control state.

In addition, it is possible to variably apply the input correlation script that is individually defined for the sensor or function and controller to be detected, so that fail-safe can be further developed for the data error continuously generated after the development of the vehicle This is a very advanced invention that has the effect of reducing the maintenance cost.

1 is a block diagram of an error judgment system according to the present invention;
2 is a configuration diagram of an input processing unit of the system of the present invention,
3 is a block diagram of an analysis and learning unit of the system of the present invention,
4 is a configuration diagram of an output processing section of the system of the present invention,
FIG. 5 is a flow chart of the error determination method of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an error determination system and method of a vehicle using data analysis of the present invention will be described with reference to the drawings.

It is to be noted, however, that the disclosed drawings are provided as examples for allowing a person skilled in the art to sufficiently convey the spirit of the present invention. Accordingly, the present invention is not limited to the following drawings, but may be embodied in other forms.

In addition, unless otherwise defined, the terms used in the description of the present invention have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. In the following description and the accompanying drawings, A detailed description of known functions and configurations that may be unnecessarily blurred is omitted.

1 is a block diagram of an error judgment system according to the present invention.

Referring to the drawings, an error determination system 1 of the present invention includes an input processing unit 100, a learning and analyzing unit 200, and an output processing unit 300.

The input processing unit 100 may be configured to receive signals from actuators 50 such as a steering wheel 40, a parking sensor, a radar, a camera, a brake, an excel or a transmission, And extracts data that can be used by the learning and analysis unit 200 to generate an error judgment model or an algorithm through the input correlation script generation module 110 to be described later.

The learning and analysis unit 200 receives the extracted data from the input processing unit 100 and learns a model or an algorithm for error determination by learning the extracted data for a predetermined period according to a mechanical learning method or an automatic code generation technique And determines whether or not the data transmitted from the sensor 40 and the actuator 50 through the vehicle network 10 of the vehicle is in error according to the generated error judgment model or algorithm.

The output processing unit 300 generates an error message for the data determined as the result of the determination by the learning and analysis unit 200 and a diagnostic message to be generated upon the determination of the error data, and outputs a fail-safe control message And generates a warning message for the error data.

An error message, a diagnosis message, a fail-safe control message, and a warning message for data determined to be generated by the output processing unit 300 are transmitted to the electronic control unit 20 of the vehicle through the network 10 for communication of the vehicle And the electronic control unit 20 displays the message through the cluster of the driver's seat.

The data input to the input processing unit 10 of the present invention may include data generated in the interior of the vehicle such as the sensor 40 and the actuator 50 and execution of an application program executed in the mobile phone 30 of the vehicle occupant Data of the result can also be input.

Hereinafter, a specific configuration of each part of the error judgment system 1 of the automobile using the data analysis of the present invention having the above-described configuration will be described with reference to FIG. 2 to FIG.

2 is a configuration diagram of an input processing unit of the system of the present invention.

The input processing unit 100 of the present invention system comprises an input correlation script generation module 110, a data collection module 120, a data extraction module 130, and a storage module 140.

The input correlation script generation module 110 is a module that generates an input correlation script and defines a collection specification for data output by the sensor 40 or the actuator 50. [

Among the data generated by the engine temperature sensor and the throttle valve sensor, which is the sensor value of the error to be judged, for example, the collecting specification such as the engine temperature and the amount of change in the engine temperature can be defined from the engine temperature sensor, Can collect data such as throttle valve opening / closing amount and throttle valve opening / closing frequency.

As shown in the figure, the script generated by the input correlation script generation module 110 has a unique ID such as a script A, a script B, a script C, and a script D depending on the sensor or the actuator, When an error occurs in the data in the analyzer 200, the error code is used for tracking the error code and generating the diagnostic code.

The collection specifications that can be included in the input correlation script of the present invention include input data types and specifications, extraction specification data types and specifications (source data specification, extraction specification), data range, data size, script ID, , Data specification (artificial intelligence algorithm specification), and the like.

Since the input correlation script as described above can be defined by changing the collection specification as needed, it is possible to collect data variably according to needs and circumstances.

The data collecting module 120 collects data generated by the sensor 40 and the actuator 50. The data extracting module 130 extracts data from the data collected by the data collecting module 120, And extracts the necessary data according to the collection specification described in the script defined in the correlation script 110. [

The data extracted by the data extraction module 130 is stored in the storage module 140 as a data set A, a data set B, a data set C, and a data set D according to the type of data.

3 is a block diagram of an analysis and learning unit of the system of the present invention.

Referring to the drawings, the learning and analysis unit 200 of the present invention includes an analysis tool generation module 210 and an error determination module 220.

The analysis tool generation module 210 receives the extracted data from the data extraction module 130, receives the learning specification from the input correlation script generation module 110, It is the part that generates analysis model or analysis algorithm for analysis.

At this time, the analysis tool generation module 210 receives a data set divided and stored as a data set A, a data set B, a data set C, and a data set D from the data extraction module 130.

The inventive analysis tool generation module 210 has a dynamic learning tool 211 and a rule based analysis tool 212 to generate the analysis model.

The dynamic learning tool 211 has a mutually complementary relationship with the rule base analysis tool 212. The dynamic learning tool 211 is a mechanical learning tool based on artificial intelligence to dynamically detect an error in a vehicle control operation that can be continuously generated. And generates an error judgment model for the data according to the learning method.

In other words, according to a conventional mechanical learning method based on artificial intelligence, the input data may be classified into data groups (eg, and generates an error judgment model or an error judgment algorithm for the data by modeling the error situation through clustering.

The rule base analyzing tool 212 has a complementary relationship with the dynamic learning tool 211. The rule base analyzing tool 212 has a basic rule for dynamically detecting an error in a vehicle control operation that can be continuously generated. And generates the error judgment algorithm based on the dynamic logic that becomes the error decision algorithm.

That is, the input data can be dynamically configured by an algorithm capable of identifying an error situation according to logic specifications and parameter specifications defined in an input correlation script, and can be performed using an ordinary code automatic generation technique.

The error determination module 220 receives the data generated by the analysis tool generation module 210 from the sensor 40 and the actuator 50 through the vehicle network 10 using the error determination model and the error determination algorithm, To determine the error of the data set to be transmitted.

The error determination module 220 includes an error analysis model A, an error analysis model B, and an error analysis model C, which correspond to the data set A, data set B, data set C, and data set D received from the data extraction module 130, , And an error analysis model D. The data set received in real time is analyzed for errors by the corresponding error analysis model.

The error determination module 220 determines that the data set in which the error has not occurred is normal and the data set in which the error has occurred is determined to be abnormal and transmits the data set to the output processing unit 300 to be described later.

4 is a configuration diagram of an output processing section of the system of the present invention.

Referring to the drawings, the output processing unit 300 of the present invention includes an error analysis result generation module 310 and an error display and error control module 320.

The error analysis result generation module 310 treats the data set determined by the error determination module 220 as normal and treats the data set in which the error has occurred as abnormal.

Then, the data set normally handled is sent back to the vehicle network 10, and the data set treated as abnormal sends out the abnormal result analysis value to the original data to the vehicle network 10 , The data set sent to the vehicle network 10 is transmitted to the electronic control unit 20 and used for automobile control.

The error display and error control module 320 generates a diagnostic message for the error message and the error data for the data set in which the error has occurred from the error analysis result generation module 310 and generates a failure message regarding the error data -safe) Generates a control message and generates a warning message about the error data.

An error message, a diagnosis message, a fail-safe control message, and a warning message for data determined to be generated by the output processing unit 300 are transmitted to the electronic control unit 20 of the vehicle through the network 10 for communication of the vehicle And the electronic control unit 20 notifies the driver of the occurrence of the error through the cluster of the vehicle's driver's seat and informs the driver of the occurrence of the error and other controllers using the data refer to the result of the error data to perform safe vehicle control I will help you.

FIG. 5 is a flow chart of the error determination method of the present invention.

Referring to the drawings, an error determination method of the present invention is performed in an error determination system including an input processing unit 100, a learning and analysis unit 200, and an output processing unit 300.

First, the input processing unit 100 loads an input correlation script that defines a collection specification of data output from a sensor or an actuator of an automobile (S 10).

As described above, the collection specification of the input correlation script includes a collection specification such as an engine temperature and a change amount of the engine temperature from the engine temperature sensor, which is the sensor value of the error judgment target, and data generated by the throttle valve sensor And the throttle valve sensor data includes data such as throttle valve opening / closing amount and throttle valve opening / closing frequency.

The input processing unit 100 collects data generated by the sensor and the actuator (S 20), and extracts necessary data according to the collection specification described in the script defined in the input correlation script among the collected data ( S 30).

The learning and analysis unit 200 receives the extracted data from the input processing unit 100 and learns the extracted data according to a mechanical learning method or an automatic code generation technique for a predetermined period of time, And generates an algorithm (S40).

The learning and analysis unit 200 determines whether the model or algorithm for error determination is sufficient to determine an error of data transmitted in real time through a vehicle network of the vehicle (S50) .

If it is determined in step S50 that a sufficient model or algorithm has been generated for the error determination, the learning and analysis unit 200 loads data transmitted from the sensor and the actuator in real time through the vehicle network (S60).

Then, the learning and analysis unit 200 determines an error of the loaded data using an error determination model and an error determination algorithm (S70).

Then, the learning and analysis unit 200 determines whether the numerical range of the data has continuity and continuity using an error judgment model and an error judgment algorithm (S 80).

In addition, the learning and analysis unit 200 determines whether the value of the data maintains a logical value with respect to the running state of the vehicle using the error judgment model and the error judgment algorithm (S90).

If there is no data error at step S70, step S80, step S90, the output processing unit 300 transmits the original data that has been judged to be error to the vehicle network at step S140. However, if it is determined that the data is an error in these steps, the output processing unit 300 generates an error code for the data (S100), generates an error message for the data (S110) (S 120), generates a warning message for the error data (S 130), tags the abnormal result analysis value to the errored original data to generate a fail-safe control message To the network.

Thereafter, the data in which the error has occurred is transmitted to the electronic control unit 20 and used for automobile control.

Description of the Related Art [0002]
One; In the present invention,
100; The input processor
110; Input Correlation Script Generation Module
120; Data acquisition module
130; Data extraction module
140; Storage module
200; Analysis and Learning Department
210; Analysis tool generation module
220; Error determination module
300; The output processor
310; Error analysis result generation module
320; Error indication and error control module

Claims (8)

1. An error judgment system for data of an automobile,
To generate an input correlation script that defines a collection specification of data generated by the sensor 40 of the vehicle and the actuator 50 and to generate an error decision model or algorithm of the data as defined in the input correlation script An input processing unit 100 for extracting usable data;
The controller 40 receives the data extracted from the input processing unit 100 and generates a model or an algorithm for error judgment and generates a model or an algorithm based on the sensor 40 and the actuator 50 A learning and analysis unit 200 for determining whether data transmitted in real-time from the mobile terminal 100 is erroneous or not;
An output processing unit (300) for transmitting data determined as an error to the vehicle network (10) as a result of the determination by the learning and analysis unit (200); And,
The input processing unit (100)
An input correlation script generation module 110 for generating the above input correlation script;
A data collection module 120 for collecting data generated by the sensor 40 and the actuator 50;
A data extraction module (130) for extracting necessary data according to a collection specification described in a script defined in the input correlation script (110) among data collected by the data collection module (120);
And an error determination system for data of an automobile.
delete 2. The apparatus of claim 1, wherein the learning and analysis module (200)
A dynamic learning tool 211 for generating an error judgment model for the data by a mechanical learning method based on artificial intelligence and a rule base analyzing tool 211 for generating an error judgment algorithm based on a dynamic logic An analysis tool generation module 210 having a tool 212;
An error of data transmitted in real time from the sensor 40 and the actuator 50 through the vehicle network 10 is determined using the error judgment model and the error judgment algorithm to the data generated by the analysis tool generation module 210 An error determination module 220;
And an error determination system for data of an automobile.
The apparatus of claim 1, wherein the output processor (300)
The data that the learning and analysis unit 200 determines that there is no error sends the original data back to the vehicle network 10 and the errored data is tagged with the original resultant data for analyzing the abnormal result, 10), an error analysis result generation module 310;
Generating an error message for the errored data set and a diagnostic message for the error data, generating a fail-safe control message on the error data, generating an error message for the error data, And an error control module (320).
And an error determination system for data of an automobile.
The error judgment system according to claim 1, characterized in that data of the execution result of the application program executed in the mobile phone (30) is input to the input processing unit (100).
A method for determining an error of an automobile performed in an error determination system including an input processing unit (100), a learning and analysis unit (200), and an output processing unit (300)
(S10) of loading an input correlation script that defines a collection specification for data output by a sensor or an actuator of the vehicle;
The input processing unit 100 collects data generated by the sensor and the actuator (S 20), and extracts necessary data according to the collection specification described in the script defined in the input correlation script among the collected data (S 30);
The learning and analysis unit 200 receives the extracted data from the input processing unit 100 and learns a model or an algorithm for error determination by learning the extracted data for a predetermined period according to a mechanical learning method or automatic code generation technique (S40);
A step (S50) of determining whether the learning and analysis unit (200) is sufficient to apply a model or an algorithm for error determination to an error of data received in real time via a vehicle network of the vehicle;
(S60) loading data delivered from the sensors and actuators in real time via the vehicle network when the learning and analysis unit 200 determines that a sufficient model or algorithm has been generated for the error determination;
The learning and analysis unit 200 may determine an error of the loaded data using an error judgment model and an error judgment algorithm (S 70);
As a result of the determination, if there is no error in the data, the output processing unit 300 transmits the original data,
As a result of the determination, when it is determined that an error has occurred in the data, the output processing unit 300 generates an error code for the data (S100), tags the abnormal result analysis value to the error- (S 140);
The learning and analysis unit 200 determines whether the numerical range of the data has continuity and continuity using an error judgment model and an error judgment algorithm (S 80);
The learning and analysis unit 200 determines whether the value of the data maintains a logical value with respect to a running state of the vehicle using an error judgment model and an error judgment algorithm; And determining the error of the data of the vehicle.
delete 7. The apparatus according to claim 6, wherein the output processing unit (300)
If it is determined that an error has occurred in the data, an error message for the data is generated (S110), a fail-safe control message for the error data is generated (S120), and a warning message (S 130). ≪ Desc / Clms Page number 19 >

Images