KR20230132205A - System for Transslating OBD Document using AI model and Driving method thereof - Google Patents

Abstract

The present invention relates to a self-diagnosis device (OBD) document translation system and a driving method using an artificial intelligence model that translates system diagnosis contents from an automobile's self-diagnosis device (OBD) into a desired language. A diagnostic content receiver that receives the vehicle system diagnosis content from a diagnostic device that reads the system diagnosis content from (OBD) or the self-diagnosis device (OBD) inside the vehicle, and artificial intelligence that pre-learns the system diagnosis content received from the diagnostic content receiver. A translation performance unit that selects a model and performs machine translation through the selected artificial intelligence model, a translation result transmission unit that provides the translation results performed by the translation performance unit to a set user terminal, and the translation performance unit is performing machine translation. When new translation data is generated, the self-diagnosis device (OBD) document translation system using an artificial intelligence model that includes a translation model learning unit that reorganizes the new translation data into sentence pairs to learn a machine translation model is required in the automobile industry certification process. The effect of providing a self-diagnosis device (OBD) document translation device and method using an artificial intelligence model that can clearly interpret technical documents containing technical terms or new terms when submitting documents is derived.

Description

Self-diagnosis device (OBD) document translation system and driving method using artificial intelligence model {System for Transslating OBD Document using AI model and Driving method thereof}

The present invention relates to an on-diagnosis device (OBD) document translation system using an artificial intelligence model and its driving method. More specifically, the present invention relates to an artificial intelligence system that translates system diagnosis contents from an automobile's self-diagnosis device (OBD) into a desired language. This relates to an on-board diagnostic device (OBD) document translation system using a model and its operation method.

In general, an automobile's self-diagnosis device (OBD: On Board Diagnosis) is a legally prescribed sub-diagnosis/monitoring system integrated into the engine control system. The OBD system monitors all systems related to exhaust gases and evaporative gases throughout the entire driving range.

And when a failure occurs in the monitored systems, the failure details are stored in the ECU and can be viewed through a standardized interface - a 16-pin diagnostic connector. In addition to this, a malfunction message is delivered to the driver through the Malfunction Indicator Lamp (MIL) or message on the instrument panel.

The OBD system includes the functions of the catalytic converter, catalytic heater, air ratio sensor, engine misfire monitoring system, EGR-system function, fuel tank ventilation system (evaporative gas system) function, secondary air system function, and exhaust gas related functions. The component's electrical circuitry, sub-systems and sensors are monitored continuously or during each driving cycle.

The driving cycle means starting the engine, driving at a certain rotational speed and driving speed, decelerating, and stopping the engine. At this time, the engine coolant temperature must change from a minimum of 22℃ to 70℃.

Domestic ODB follows the American OBD system standard for gasoline vehicles and the European OBD system standard for diesel vehicles. This is because domestic environmental standards for environmental emissions adopt the US standard (ULEV) for gasoline vehicles and the European standard (EURO) for diesel vehicles.

The OBD system is a catalytic device for reducing exhaust gases, engine misfire, fuel evaporation gas gauge, secondary air supply system, air conditioner system, fuel system, exhaust gas recirculation (EGR), oxygen sensor, and other automobile parts related to exhaust gases. Items are checked in real time and the driver is notified when more than 1.5 times the allowable emission standard is emitted due to a breakdown. This is because the increase in vehicle emissions is caused by a variety of reasons, not just exhaust system failure.

Meanwhile, during various industrial certification procedures such as automobile Ministry of Environment certification, there are documents that must be submitted, such as the overall OBD system manual for the vehicle. However, in the case of imported vehicles, the OBD system manual is often written in English, so translation work is necessary.

KR 10-2360659 B1 KR 10-1986721 B1 KR 10-0961717 B1

The present invention was derived from this technical background, and is a self-diagnosis device using an artificial intelligence model that can clearly interpret technical documents containing technical terms or new terms when submitting documents required in the automobile industry certification process ( The purpose is to provide a document translation device and method (OBD).

In addition, when performing the necessary procedures for the automotive industry certification process, an artificial intelligence model is developed that can reduce the time and effort to prepare for the automotive industry certification process by understanding the certification criteria in advance and predicting and providing the certification results. The purpose is to provide a device and method for translating documents using an on-board diagnostic device (OBD).

The present invention for achieving the above problems includes the following configuration.

In other words, the self-diagnosis device (OBD) document translation system using an artificial intelligence model according to an embodiment of the present invention reads system diagnosis contents from the self-diagnosis device (OBD) inside the vehicle or the self-diagnosis device (OBD) inside the vehicle. A diagnostic content receiving unit that receives automotive system diagnosis content from a diagnostic device, a translation performing unit that selects a pre-trained artificial intelligence model for the system diagnosis content received from the diagnostic content receiving unit and performs machine translation through the selected artificial intelligence model; A translation result transmission unit that provides the translation results performed by the translation performance unit to a set user terminal, and when new translation data is generated while the translation performance unit is performing machine translation, the new translation data is reorganized into sentence pairs to create a machine translation model. It includes a translation model learning unit for training.

Meanwhile, the method of operating the self-diagnosis device (OBD) document translation system using an artificial intelligence model is to use the self-diagnosis device (OBD) inside the vehicle or a diagnostic device that reads system diagnosis contents from the self-diagnosis device (OBD) inside the vehicle. A diagnostic content receiving step of receiving system diagnosis content, a translation performance step of selecting a pre-trained artificial intelligence model for the system diagnosis content received in the diagnostic content receiving step and performing machine translation through the selected artificial intelligence model, the translation A translation result transmission step in which the translation results performed in the performance step are provided to a set user terminal, and if new translation data is generated during machine translation in the translation performance step, the new translation data is reorganized into sentence pairs to train a machine translation model. It includes a translation model learning step.

According to the present invention, an on-board diagnostic device (OBD) document translation device and method using an artificial intelligence model is provided that can clearly interpret technical documents containing technical terms or new terms when submitting documents required for automobile industry certification procedures. The effects that can be provided are derived.

In addition, when performing the necessary procedures for the automotive industry certification process, an artificial intelligence model is developed that can reduce the time and effort to prepare for the automotive industry certification process by understanding the certification criteria in advance and predicting and providing the certification results. It is effective in providing a device and method for translating documents using an on-board diagnostic device (OBD).

Figure 1 is a block diagram showing the configuration of an on-board diagnostic device (OBD) document translation system using an artificial intelligence model according to an embodiment of the present invention.
Figure 2 is a flowchart of a method of driving an on-board diagnostic device (OBD) document translation system using an artificial intelligence model according to an embodiment of the present invention.

It should be noted that the technical terms used in the present invention are only used to describe specific embodiments and are not intended to limit the present invention. In addition, the technical terms used in the present invention, unless specifically defined in a different sense in the present invention, should be interpreted as meanings generally understood by those skilled in the art in the technical field to which the present invention pertains, and are not overly comprehensive. It should not be interpreted in a literal or excessively reduced sense.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings.

Figure 1 is a block diagram showing the configuration of an on-board diagnostic device (OBD) document translation system using an artificial intelligence model according to an embodiment of the present invention.

In the automotive industry's certification process, there are documents that must be translated and submitted. At this time, the documents are technical documents, and their usage is often different from words used in general documents. Therefore, there is a limitation in that it cannot follow the general translation industry system. Technical translation of the self-diagnosis device (OBD) document translation system 10 according to one embodiment is implemented through machine translation using an artificial intelligence model and detailed post-processing by experts.

The translation process of the OBD document translation system according to one embodiment first selects a previously trained artificial intelligence model and performs machine translation. Afterwards, the output generated from the machine translation process is provided to experts and undergoes an expert verification process.

Then, after verification by experts, the new data generated is reorganized into sentence pairs to learn a machine translation model.

Specifically, the self-diagnosis device (OBD) document translation system 10 of FIG. 1 includes a communication unit 110, a diagnosis content reception unit 120, a translation performance unit 130, a result transmission unit 140, and a translation model learning unit 150. ), a translation correction unit 160, a determination result provision unit 170, and a translation result submission unit 180.

The communication unit 110 communicates with any internal component or at least one external terminal through a wired/wireless communication network. At this time, the external arbitrary terminal is a server (not shown), a terminal (not shown), and more specifically, a self-diagnosis device 20, a diagnostic device 25, a user terminal 30, an expert terminal 40, and an automobile industry. It may include a certification authority server 50, etc. Here, wireless Internet technologies include Wireless LAN (WLAN), DLNA (Digital Living Network Alliance), Wibro (Wireless Broadband: Wibro), Wimax (World Interoperability for Microwave Access: Wimax), and HSDPA (High Speed Downlink Packet Access). ), HSUPA (High Speed Uplink Packet Access), IEEE 802.16, Long Term Evolution (LTE), LTE-A (Long Term Evolution-Advanced), Wireless Mobile Broadband Service (WMBS), etc. The communication unit 110 transmits and receives data according to at least one wireless Internet technology, including Internet technologies not listed above. In addition, short-range communication technologies include Bluetooth, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra Wideband), ZigBee, and Near Field Communication (NFC). , Ultrasound Communication (USC), Visible Light Communication (VLC), Wi-Fi, Wi-Fi Direct, etc. may be included. Additionally, wired communication technologies may include Power Line Communication (PLC), USB communication, Ethernet, serial communication, optical/coaxial cables, etc.

The self-diagnosis device (20, OBD-system) includes the functions of the catalytic converter, catalytic heater, air ratio sensor, engine misfire monitoring system, EGR-system function, fuel tank ventilation system (evaporative gas system) function, secondary Sub-systems and sensors, such as the function of the air system and the electrical circuits of exhaust-related components, are monitored continuously or during each driving cycle.

The driving cycle means starting the engine, driving at a certain rotational speed and driving speed, decelerating, and stopping the engine. At this time, the engine coolant temperature must change from a minimum of 22℃ to 70℃.

In addition, the self-diagnosis device (20, OBD-system) is a catalytic device for reducing exhaust gases, engine misfire, fuel evaporation gas meter, secondary air supply system, air conditioner system, fuel system, exhaust gas recirculation system (EGR), Automobile parts related to exhaust gas, such as oxygen detectors, are checked in real time and the driver is notified when more than 1.5 times the allowable exhaust gas standard is emitted due to a malfunction. This is because the increase in vehicle emissions is caused by a variety of reasons, not just exhaust system failure.

In the case of an engine misfire in which the fuel supplied to the cylinder does not combust properly in the flame, carbon hydrogen increases. In addition, the catalytic device that destroys the exhaust gas through a chemical reaction is damaged, resulting in a further increase in the final exhaust gas. The OBD system measures the kinetic energy imbalance between the cylinder and crankshaft due to misfire, and lights up a warning light in case of an abnormality.

Inspecting the catalytic device and recirculation device, etc. = OBD checks in real time whether the catalytic device is operating properly, and oxygen sensors attached to the front and back of the catalytic device monitor this. The amplitude of the output voltage from the oxygen sensor in front of the catalyst is large because the exhaust gas is not purified, and the amplitude of the rear sensor is small, so the amplitude ratio is compared and the OBD system is notified if the amplitude of the rear sensor is larger than the standard.

Vehicles have an exhaust gas recirculation system (EGR) that sends exhaust gases back to the cylinder for combustion, but failure of this device also causes exhaust gases to exceed the standard level. When EGR operates normally, the pressure in the engine's intake pipe increases, and a sensor measures this pressure and reports any abnormalities.

In addition, in addition to the exhaust gas coming out through the muffler, OBD also detects leakage of fuel itself, such as gasoline, from the fuel tank, and this can also be detected using sensors. Even if the fuel cap is not properly closed after refueling, the OBD system operates and the warning light turns on.

Therefore, the self-diagnosis device (20, OBD-system) monitors the sensors attached to each emission-related component, the engine control unit (ECU) that receives information from these sensors to determine whether there is a malfunction, and the dashboard warning light, etc. can do.

Automobile manufacturers must receive such self-diagnosis device (20, OBD-system) certification from the Ministry of Environment.

The diagnostic device 25 acquires the vehicle's sensor and fault code data from the self-diagnosis device (20, OBD-system) connector and outputs the vehicle's current status or faulty area in real time on the screen and in voice. In one embodiment, the diagnostic device 25 has all of the technical components to transmit the system diagnosis contents read from the self-diagnosis device (20, OBD-system) connector to the document translation system 10 according to one embodiment through various communication methods. It is interpreted to be inclusive.

The user terminal 30 and expert terminal 40 are smart phones, portable terminals, mobile terminals, foldable terminals, and personal digital assistants (PDAs). ), PMP (Portable Multimedia Player) terminal, telematics terminal, navigation terminal, personal computer, laptop computer, Slate PC, Tablet PC, ultrabook ), wearable devices (e.g., including smartwatch, smart glass, head mounted display (HMD), etc.), Wibro terminal, IPTV (Internet Protocol Television) terminal , can be applied to various terminals such as smart TVs, digital broadcasting terminals, AVN (Audio Video Navigation) terminals, A/V (Audio/Video) systems, flexible terminals, and digital signage devices.

In one embodiment, the user terminal 30 is interpreted to encompass a terminal device possessed by a person in charge, such as the owner of the car or an agency that performs automobile authentication of the vehicle. And the expert terminal 40 is interpreted to encompass terminal devices possessed by various experts in the automobile-related field.

The automobile industry certification authority server 50 is interpreted to encompass servers of various certification authorities that request translation results provided by the document translation system 10, that is, system diagnosis contents translated into a specific language.

The diagnosis content receiving unit 120 receives vehicle system diagnosis content from the self-diagnosis device (OBD, 20) inside the vehicle or the diagnostic device (25) that reads the system diagnosis content from the self-diagnosis device (OBD, 20) inside the vehicle. .

In one embodiment, the diagnostic content receiving unit 120 includes the same technical configuration as an existing diagnostic device and is directly connected to the self-diagnosis device 20 installed in the vehicle via the communication unit 110 to immediately read the system diagnostic content. It can be implemented to include

Alternatively, it may be implemented to read the system diagnosis contents of the vehicle directly from the physically separated diagnostic device 25, which reads the system diagnosis contents from the self-diagnosis device (OBD, 20) inside the vehicle.

The translation performing unit 130 selects a pre-trained artificial intelligence model for the system diagnosis content received from the diagnostic content receiving unit 120 and performs machine translation through the selected artificial intelligence model.

In one embodiment, the translation performing unit 130 may perform translation using a machine translation (NMT: Neural Machine Translation) model using a neural network.

The translation performing unit 130 may be implemented as a machine translation module including an encoder trained to convert a source sentence into a semantic vector and a decoder trained to generate a sentence written in the target language using the semantic vector.

The translation performing unit 130 receives a random sentence written in a specific source language and language information about the specific target language into which the sentence is to be translated, and the translation performing unit that has learned the random sentence converts it into a sentence written in the target language to be translated. It can be translated.

The translation result transmission unit 140 provides the translation result performed by the translation performance unit 130 to the set user terminal 30. In other words, the translation result can be provided to the user terminal 30 possessed by the borrower or the user terminal 30 possessed by the person in charge of the certification process so that it can be printed for submission to a certification authority requiring submission of translation contents. At this time, the translation result transmission unit 140 transmits the file format or transmission method requested in advance by the user terminal 30. That is, the translation result transmission unit 140 can compress or change the translation result into a file format selected by the user.

For example, the translation result can be transmitted by sending it to an email designated by the user or by uploading it to a specific web page operated by a service platform provided by the document translation system 10 according to an embodiment. However, it is not limited to this and can be interpreted to encompass various modifications.

When new translation data is generated while the translation performing unit 130 is performing machine translation, the translation model learning unit 150 reorganizes the new translation data into sentence pairs to learn a machine translation model.

In order to show good translation performance in a machine translation (NMT: Neural Machine Translation) model using a neural network, a large parallel corpus is required.

The translation model learning unit 150 receives input of a specific source language and a specific target language for training the machine translation module. And, from among the multi-lingual parallel corpora candidates written in a plurality of source languages and a plurality of target languages, a plurality of candidate parallel corpora that are a specific source language received by the source language or a specific target language received by the target language are selected, and a plurality of candidate parallel corpora are selected. Among the sentences included in the candidate parallel corpora, a tag corresponding to the target language into which each source sentence was translated is inserted into each source sentence.

Then, we generate learning data including source sentences with inserted tags and target sentences into which the source sentences are translated, and use the learning data to learn a low-resource machine translation module so that each source sentence is translated into a target sentence according to the inserted tags. You can do it.

At this time, to generate learning data, words can be separated from the source sentences included in the candidate parallel corpora, and the separated words can be encoded as lower words using the BPE (Byte Pair Encoding) algorithm.

In one aspect of the present invention, the translation correction unit 160 provides the machine translation result of the system diagnosis contents in the translation performance unit 130 to the expert terminal 40, and provides the machine translation result provided from the expert terminal 40. Receive a correction proposal.

At this time, the translation model learning unit 150 reorganizes the new translation data into sentence pairs with reference to the correction plan received from the translation correction unit 160 and trains the machine translation model.

In one embodiment, the OBD document is a technical document, and its usage is different from words used in general documents. Therefore, there are limits to following the general translation industry system. Therefore, after performing machine translation using an artificial intelligence model, the translation accuracy of terms, etc. for technical documents can be improved by undergoing post-editing work from experts and repeatedly learning the corrections for the post-editing work.

Additionally, the translation model learning unit 150 may further perform a verification process on training data or parallel corpus candidates from the expert terminal 40. That is, the expert terminal 40 possessed by an expert in the automobile-related field provides a multi-language parallel corpus candidate group written in a plurality of source languages and a plurality of target languages, and provides a plurality of candidate parallel corpora that are specific target languages from the expert. It may include a process of selecting one of the options. However, it is not limited to this and can be implemented to perform learning after receiving the relevant term or sentence directly from an expert and going through a verification process with at least two experts.

In other words, more accurate translation is possible through a review process by experts, and even when terms to be translated are added in the process of introducing new languages or new technologies in the automobile-related field, the effect is that the accuracy of translation can be improved by learning more accurately.

The determination result providing unit 170 determines whether at least one of the emission test, OBD test, and noise test is within the allowable range of domestic standards when proceeding with the automobile industry certification process according to the system diagnosis content received by the diagnosis content receiver 120. The determination is made and the determination result is provided to the user terminal 30 held by the borrower. Accordingly, there is an effect of being able to identify and prepare for matters that need improvement before applying for certification.

In one embodiment, the determination result providing unit 170 may receive data on the domestic standard allowable range from the automobile industry certification agency server 50 at any time.

The machine translation results from the translation performance unit 130 are provided to the user terminal 30 owned by the car owner, and the translation result submission unit 180 identifies the translation results required for certification for each automobile industry certification body, and provides the results to the user. Upon request from the terminal 30, the translation result required for authentication identified is transmitted and submitted to the automobile industry certification agency.

The translation result submission unit 180 may submit the translation result in a file format requested from the automobile industry certification authority server 50 or in a file attachment or transmission method.

In one embodiment, the translation result submission unit 180 may receive changes to condition data and submission methods related to the translation result required for authentication from the automobile industry certification authority server 50 at any time.

Figure 2 is a flowchart of a method of driving an on-board diagnostic device (OBD) document translation system using an artificial intelligence model according to an embodiment of the present invention.

A flowchart of a method of driving a self-diagnosis device (OBD) document translation system using an artificial intelligence model according to an embodiment reads system diagnosis contents from the self-diagnosis device (OBD) inside the vehicle or the self-diagnosis device (OBD) inside the vehicle. The vehicle system diagnosis information is received from the diagnostic device (S600).

In one embodiment, the diagnosis content receiving step may be implemented to include the same technical configuration as an existing diagnostic device and connect directly to a self-diagnosis device installed in the vehicle via a communication unit to immediately read system diagnosis content.

Alternatively, it may be implemented to read the system diagnosis contents of the vehicle directly from a physically separated diagnostic device that reads the system diagnosis contents from a self-diagnosis device inside the vehicle.

Then, in the stage of receiving diagnosis content, a pre-trained artificial intelligence model is selected for the received system diagnosis content and machine translation is performed through the selected artificial intelligence model (S610).

In one embodiment, the translation performance step may be performed using a machine translation (NMT: Neural Machine Translation) model using a neural network.

The translation performance step receives a random sentence written in a specific source language and language information about a specific target language to translate the sentence into, and can translate the random sentence into a sentence written in the target language to be translated in the learned translation performance step. there is.

Afterwards, the translation result performed in the translation performance step is provided to the set user terminal (S620).

In other words, the translation result can be provided to a user terminal owned by the borrower or a user terminal held by the person in charge of the certification process so that it can be printed for submission to a certification authority that requires submission of translation content. At this time, the translation result transmission step is transmitted using a file format or transmission method previously requested by the user terminal. In other words, the translation result transmission step can compress or change the translation result into a file format selected by the user.

For example, the translation result can be transmitted by sending it to an email designated by the user, or by uploading it to a specific web page operated by a service platform provided by the document translation method according to one embodiment. However, it is not limited to this and can be interpreted to encompass various modifications.

In one aspect of the present invention, the machine translation result of the system diagnosis contents in the translation performance step is provided to the expert terminal, and a correction plan for the machine translation result provided from the expert terminal is received (S630).

At this time, the translation model learning step reorganizes the new translation data into sentence pairs by referring to the correction plan received in the translation correction step and learns the machine translation model.

In one embodiment, the OBD document is a technical document, and its usage is different from words used in general documents. Therefore, there are limits to following the general translation industry system. Therefore, after performing machine translation using an artificial intelligence model, the translation accuracy of terms, etc. for technical documents can be improved by undergoing post-editing work from experts and repeatedly learning the corrections for the post-editing work.

Additionally, the translation model learning step may further perform a verification process on training data or parallel corpus candidates from the expert terminal. In other words, it provides a multi-lingual parallel corpus candidate group written in multiple source languages and multiple target languages through an expert terminal owned by experts in various fields related to automobiles, and provides one of the multiple candidate parallel corpora that is a specific target language from the expert. It may include a process of being selected.

In other words, more accurate translation is possible through a review process by experts, and even when terms to be translated are added in the process of introducing new languages or new technologies in the automobile-related field, the effect is that the accuracy of translation can be improved by learning more accurately.

And, if new translation data is generated during machine translation in the translation performance stage (S640), the new translation data is reorganized into sentence pairs to learn a machine translation model (S650).

In the translation model learning stage, when new translation data is generated during machine translation in the translation performance stage, the new translation data is reorganized into sentence pairs to learn a machine translation (NMT: Neural Machine Translation) model.

The translation model learning stage receives input of a specific source language and a specific target language for learning the machine translation module. And, from among the multi-lingual parallel corpora candidates written in a plurality of source languages and a plurality of target languages, a plurality of candidate parallel corpora that are a specific source language received by the source language or a specific target language received by the target language are selected, and a plurality of candidate parallel corpora are selected. Among the sentences included in the candidate parallel corpora, a tag corresponding to the target language into which each source sentence was translated is inserted into each source sentence.

Then, we generate learning data including source sentences with inserted tags and target sentences into which the source sentences are translated, and use the learning data to learn a low-resource machine translation module so that each source sentence is translated into a target sentence according to the inserted tags. You can do it.

At this time, to generate learning data, words can be separated from the source sentences included in the candidate parallel corpora, and the separated words can be encoded as lower words using the BPE (Byte Pair Encoding) algorithm.

In an additional aspect of the present invention, according to the system diagnosis content received in the diagnosis content reception step, when proceeding with the automobile industry certification process, it is determined whether at least one of the emission test, OBD test, and noise test is within the allowable range of the domestic standard. , the determination result is provided to the user terminal owned by the borrower (S660).

Accordingly, there is an effect of being able to identify and prepare for matters that need improvement before applying for certification.

In one embodiment, the step of providing the determination result may be provided at any time with data on the domestic standard allowable range from the automobile industry certification agency server.

In the translation performance stage, the machine translation results are provided to the user terminal owned by the car owner, and in the translation result submission stage, each automotive industry certification agency identifies the translation performance results required for certification, and provides authentication based on requests from the user terminal. Transmit and submit the necessary translation results to the automotive industry certification body.

In the translation result submission stage, translation results can be submitted in the file format requested from the automobile industry certification agency server, or in the file attachment or transmission method.

In one embodiment, in the step of submitting translation results, condition data related to translation results required for authentication and changes to the submission method may be provided at any time from the automobile industry certification agency server.

The above-described method may be implemented as an application or in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., singly or in combination.

The program instructions recorded on the computer-readable recording medium may be those specifically designed and configured for the present invention, or may be known and usable by those skilled in the computer software field.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floptical disks. media), and hardware devices specifically configured to store and perform program instructions, such as ROM, RAM, flash memory, etc.

Examples of program instructions include not only machine language code such as that created by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the invention and vice versa.

Although the above has been described with reference to embodiments, those skilled in the art will understand that various modifications and changes can be made to the present invention without departing from the spirit and scope of the present invention as set forth in the following patent claims. You will be able to.

10: Document translation system 20: Self-diagnosis device
25: diagnostic device 30: user terminal
40: Expert terminal 50: Automotive industry certification body server
110: Communication unit 120: Diagnosis content receiving unit
130: Translation performance unit 140: Result transmission unit
150: Translation model learning unit 160: Translation correction unit
170: Determination result provision unit 180: Translation result submission unit

Claims (8)

A diagnostic content receiver that receives vehicle system diagnostic content from an on-board diagnostic device (OBD) inside the vehicle or a diagnostic device that reads the system diagnostic content from an on-board diagnostic device (OBD) inside the vehicle;
a translation performing unit that selects a pre-learned artificial intelligence model for the system diagnosis contents received from the diagnostic contents receiving unit and performs machine translation through the selected artificial intelligence model;
a translation result transmission unit that provides the translation result performed by the translation performance unit to a set user terminal; and
When new translation data is generated while performing machine translation in the translation performance unit, a translation model learning unit that reorganizes the new translation data into sentence pairs to train a machine translation model; self-diagnosis device (OBD) using an artificial intelligence model, including ) Document translation system.
According to claim 1,
It further includes a translation correction unit that provides a machine translation result of the system diagnosis contents of the translation performance unit to an expert terminal and receives a correction plan for the machine translation result provided from the expert terminal,
The translation model learning unit,
An on-board diagnostic device (OBD) document translation system using an artificial intelligence model that reorganizes new translation data into sentence pairs and learns a machine translation model by referring to the correction plan received from the translation correction unit.
According to claim 2,
According to the system diagnosis content received through the above diagnostic content receiver, when proceeding with the automobile industry certification process, it is determined whether at least one of the emission test, OBD test, and noise test is within the allowable range of domestic standards, and the determination result is sent to the car owner. A self-diagnosis device (OBD) document translation system using an artificial intelligence model, further comprising a determination result providing unit provided to the user terminal.
According to claim 2,
Providing machine translation results from the translation performance unit to a user terminal owned by the borrower,
It further includes a translation result submission unit that identifies the translation results required for certification for each automotive industry certification body and transmits and submits the translation results required for authentication identified in response to a request from the user terminal to the automotive industry certification body. , Self-diagnosis device (OBD) document translation system using an artificial intelligence model.
A diagnostic content receiving step of receiving vehicle system diagnostic content from an on-board diagnostic device (OBD) inside the vehicle or a diagnostic device that reads system diagnostic content from an on-board diagnostic device (OBD) inside the vehicle;
A translation performance step of selecting a pre-trained artificial intelligence model for the system diagnosis content received in the diagnostic content receiving step and performing machine translation through the selected artificial intelligence model;
A translation result transmission step of providing the translation result performed in the translation performance step to a set user terminal; and
In the translation performance step, when new translation data is generated during machine translation, a translation model learning step of reorganizing the new translation data into sentence pairs to train a machine translation model; self-diagnosis device (OBD) using an artificial intelligence model, including ) How to operate the document translation system.
According to claim 5,
It further includes a translation correction step of providing a machine translation result of the system diagnosis contents in the translation performance step to an expert terminal and receiving a correction plan for the machine translation result provided from the expert terminal,
The translation model learning step is,
A method of driving an on-board diagnostic device (OBD) document translation system using an artificial intelligence model, which rearranges new translation data into sentence pairs and learns a machine translation model by referring to the correction plan received in the translation correction step.
According to claim 6,
According to the system diagnosis content received in the above diagnosis content reception step, when proceeding with the automobile industry certification process, it is determined whether at least one of the emission test, OBD test, and noise test is within the allowable range of domestic standards, and the result of the determination is held by the car owner. A method of driving a self-diagnosis device (OBD) document translation system using an artificial intelligence model, further comprising providing a determination result provided to a user terminal.
According to claim 6,
Providing machine translation results from the translation performance step to a user terminal owned by the borrower,
A translation result submission step of identifying the translation results required for certification for each automotive industry certification body and transmitting and submitting the translation results required for authentication identified in response to a request from the user terminal to the automotive industry certification body; further comprising: , Method of operating an on-board diagnostic device (OBD) document translation system using an artificial intelligence model.