DE102020128497A1 - Computer-implemented method and system for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle and training method - Google Patents

Abstract

The invention relates to a computer-implemented method for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle (1). The method includes receiving (S1) natural language data (D1) describing the defect in the technical component and/or system of the vehicle (1) and classifying (S2) the natural language data (D1) into at least one of the defects in the technical component and/or system of the vehicle (1) identifying category (K) by a classification algorithm (A1). The invention also relates to a computer-implemented method for providing a classification algorithm (A1) for classifying a defect in a technical component and/or system of a vehicle (1). The invention also relates to a system for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle (1). The invention also relates to a computer program and a computer-readable data carrier.

Description

The invention relates to a computer-implemented method for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle.

The invention also relates to a system for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle.

In addition, the invention relates to a computer-implemented method for providing a classification algorithm for classifying a defect in a technical component and/or system of a vehicle.

When a defect occurs in a technical component and/or system of a vehicle, which is either displayed to the user by a vehicle information system or is determined by the user himself, it is conventionally necessary to contact a service center.

During an on-site service appointment, an employee of the service center uses an electronic test device to diagnose the vehicle systems in order to diagnose the defect. After the error diagnosis has been completed, further work processes such as ordering spare parts, etc. are carried out.

This traditional process is time inefficient for both the user and the service center. There is thus a need to provide an improved method for diagnosing a defect in a technical component and/or system of a vehicle.

The object is achieved with a computer-implemented method for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle with the features of patent claim 1.

Furthermore, the object is achieved with a computer-implemented method for providing a classification algorithm for classifying a defect in a technical component and/or system of a vehicle with the features of patent claim 12 .

In addition, the object is achieved with a system for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle with the features of patent claim 13, a computer program with the features of patent claim 14 and with a computer-readable data carrier with the features of patent claim 15 .

The present invention creates a computer-implemented method for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle.

The method includes receiving natural language data describing the defect in the technical component and/or system of the vehicle and classifying the natural language data into at least one category identifying the defect in the technical component and/or system of the vehicle using a classification algorithm.

The method also includes, based on the category identifying the defect, executing at least one remote diagnostics module operating on a vehicle-external server, which remotely diagnoses at least one onboard system of the vehicle via a wireless connection, or generating a first error code for defects that cannot be diagnosed by the remote diagnostics module .

The remote diagnosis module or diagnosis module can alternatively be operated, for example, on an in-vehicle server.

Defects that cannot be diagnosed by the onboard system can be, for example, mechanical defects such as disadvantageous noise development. This can be further determined and processed with the help of an internal microphone and/or an external audio recorder, e.g. in the form of a mobile phone with an app, and a noise identification process.

The present invention also creates a computer-implemented method for providing a classification algorithm for classifying a defect in a technical component and/or system of a vehicle.

The method includes receiving first input training data of natural language data describing the defect in the technical component and/or system of the vehicle and receiving second input training data of a category identifying the defect in the technical component and/or system of the vehicle.

In addition, the method includes training the classification algorithm using the first input training data and the second input training data by an optimie algorithm that calculates an extreme value of a loss function.

The present invention also creates a system for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle.

The system includes means for receiving natural language data describing the defect in the technical component and/or system of the vehicle and a computing device which is set up to use a classification algorithm to convert the natural language data into at least one of the defects in the technical component and/or system to classify the vehicle identifying category or to generate a first fault code for non-diagnosable by a remote diagnostic module defects.

The system also includes a vehicle-external server which is set up to execute at least one remote diagnosis module based on the category identifying the defect, which remote diagnosis module carries out a remote diagnosis of at least one on-board system of the vehicle via a wireless connection.

One idea of the present invention is to digitize the aforementioned process of detecting, diagnosing and remedying a defect in a technical component and/or system of a vehicle and thus making it more efficient in terms of time, both on the part of the user and on the part of the service center processes are automated and the overall process is therefore reduced in time.

By automatically classifying the description of the defect in the technical component and/or system of the vehicle generated by the user or the onboard system, the defect can be categorized in an advantageous manner without involving the service center, which is usually only done in a specialist workshop or a service center.

Furthermore, in a second step, based on the identified category of the defect in the technical component and/or system of the vehicle, the remote diagnosis module operating on the vehicle-external server can be called up and a remote diagnosis of relevant vehicle systems can be carried out, so that the defect can be further localized.

Advantageous embodiments and developments result from the dependent claims and from the description with reference to the figures.

According to a preferred refinement, it is provided that at least one second error code is assigned to the at least one category for defects that can be diagnosed by the remote diagnosis module, the category identifying the defect in the technical component and/or system of the vehicle, the remote diagnosis module using the second error code to carry out the remote diagnosis of at least one of the vehicle's onboard system associated with the second fault code.

Thus, an efficient diagnosis of the defects that can be diagnosed by the remote diagnosis module can be carried out in an advantageous manner.

Another preferred development provides that the natural language data describing the defect in the technical component and/or system of the vehicle is text data and/or audio data, with the classification algorithm being set up to process audio data and/or text data.

The classification algorithm is thus advantageously able to process natural language data of different types and sources.

According to a further preferred development, it is provided that the classification algorithm is rule-based, in particular using a knowledge graph, or using a trained machine learning algorithm, in particular an artificial neural network. The classification algorithm can thus be embodied by various alternative types of algorithms.

According to a further preferred development, it is provided that the category of defect in the technical component and/or system of the vehicle identified by the classification algorithm is evaluated by an evaluation algorithm with a confidence score, in particular a numerical score value, wherein if the determined confidence -Score exceeds or falls below a predetermined threshold value, an active learning algorithm asks a user for an identification of the category assigned to the natural language data and/or audio data of a noise behavior of a vehicle component describing the defect in the technical component and/or system of the vehicle .

The categories of defects identified by the classification algorithm, which have a low confidence score, can thus advantageously be used by a user be labeled or tagged to improve the accuracy of the classification algorithm.

According to a further preferred development, it is provided that the natural language data of the description of the defect in the technical component and/or system of the vehicle by a user, in particular through communication with a digital voice assistant, creation of a text message and/or an error description via menu navigation Mobile phone app, or generated by the vehicle's onboard system. The description of the defect can thus advantageously be generated by different sources.

According to a further preferred development, it is provided that the remote diagnosis module creates a remote diagnosis plan based on the identified category and/or the at least one error code of the defect in the technical component and/or system of the vehicle, which contains a list of tests to be carried out and onboard systems to be tested in the vehicle Vehicle has.

The categorization of the damage by the classification algorithm can thus be used advantageously to create the remote diagnosis plan.

According to a further preferred development, it is provided that when the description of the defect is automatically generated by the onboard system of the vehicle due to a failure of the technical component and/or system of the vehicle, an extended description of the defect is generated using a data memory in the vehicle and /or outside the vehicle, i.e. operating instructions stored on a vehicle-external server is retrieved.

As a result, the user is advantageously supplied with a description of the operating manual that corresponds to the classified damage or that corresponds to it, without having to consult an onboard operating manual. This enables the user to better assess whether the system limitation is only temporary or how urgent a service requirement is.

According to a further preferred development, it is provided that a control device arranged in the vehicle scans vehicle electrical system signals to carry out anomaly detection and checks whether measured values determined are within a target range, with the vehicle-external server carrying out a remote diagnosis at least if the measured values determined are outside of the target range one of the vehicle's onboard systems associated with the anomaly.

Vehicles in a vehicle fleet can thus advantageously be monitored to ensure that parameters of the on-board systems are always operating within a target range.

According to a further preferred development, it is provided that technical parameters of the vehicle, in particular an engine speed, a torque, a number of gear shifts and/or a number of actuations of electrical components, are recorded continuously or at intervals Defective in electronic communication with a service center, an appointment proposal for a service appointment of the vehicle is generated.

This results in a significantly more efficient execution of the service process, from diagnosis to making an appointment for maintenance or repairs.

According to a further preferred development, after the remote diagnosis of at least one onboard system of the vehicle has been carried out or the first error code has been generated for defects that cannot be diagnosed by the remote diagnosis module, at least one spare part assigned to the diagnosed defect is pre-ordered in electronic communication with a service center.

The automated pre-ordering of spare parts enables a further gain in time efficiency, since this process can usually only be carried out after the vehicle has arrived at the service center.

The configurations and developments described can be combined with one another as desired.

Further possible configurations, developments and implementations of the invention also include combinations of features of the invention described above or below with regard to the exemplary embodiments that are not explicitly mentioned.

character list

The accompanying drawings are provided to provide a further understanding of embodiments of the invention. They illustrate embodiments and, together with the description, serve to explain principles and concepts of the invention.

Other embodiments and many of the foregoing advantages will become apparent by reference to the drawings. The illustrated elements of the drawings are not necessarily shown to scale with respect to one another.

• 1 ein Ablaufdiagramm eines computerimplementierten Verfahrens zur dialogunterstützten Ferndiagnose eines Defekts an einem technischen Bauteil und/oder System eines Fahrzeugs gemäß einer bevorzugten Ausführungsform der Erfindung;
• 2 ein Ablaufdiagramm des computerimplementierten Verfahrens zur dialogunterstützten Ferndiagnose des Defekts an dem technischen Bauteil und/oder System eines Fahrzeugs gemäß des bevorzugten Ausführungsform der Erfindung;
• 3 ein Ablaufdiagramm eines computerimplementierten Verfahrens zum Bereitstellen eines Klassifikationsalgorithmus zum Klassifizieren eines Defekts an einem technischen Bauteil und/oder System eines Fahrzeugs; und
• 4 eine schematische Darstellung eines Systems zur dialogunterstützten Ferndiagnose des Defekts an dem technischen Bauteil und/oder System eines Fahrzeugs gemäß des bevorzugten Ausführungsform der Erfindung.

Show it:

• 1 a flowchart of a computer-implemented method for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle according to a preferred embodiment of the invention;
• 2 a flowchart of the computer-implemented method for dialog-supported remote diagnosis of the defect in the technical component and/or system of a vehicle according to the preferred embodiment of the invention;
• 3 a flowchart of a computer-implemented method for providing a classification algorithm for classifying a defect in a technical component and/or system of a vehicle; and
• 4 a schematic representation of a system for dialog-supported remote diagnosis of the defect in the technical component and/or system of a vehicle according to the preferred embodiment of the invention.

In the figures of the drawings, the same reference symbols designate the same or functionally identical elements, parts or components, unless otherwise stated.

This in 1 The method shown includes receiving S1 natural language data D1 or a recorded noise, e.g. creaking/rubbing of the brakes, a description of the defect in the technical component and/or system of the vehicle and classifying S2 the natural language data D1 or functional audio data of the noise in at least one category K identifying the defect in the technical component and/or system of the vehicle 1 by a classification algorithm A1.

In addition, the method includes, based on the category K identifying the defect, executing S3a at least one remote diagnosis module 12 operating on a vehicle-external server 10, which carries out a remote diagnosis of at least one onboard system 14 of the vehicle 1 via a wireless connection, or generating S3b a first Error codes for defects that cannot be diagnosed by the remote diagnostic module 12.

At least one second error code is assigned to the at least one category K identified as the defect in the technical component and/or system of the vehicle 1 for defects that can be diagnosed by the remote diagnosis module 12 . Using the second error code, the remote diagnosis module 12 carries out the remote diagnosis of at least one onboard system 14 of the vehicle 1 associated with the second error code.

The natural language data D1 describing the defect in the technical component and/or system of the vehicle 1 is text data and/or audio data. The classification algorithm A1 is set up here to process audio data and/or text data.

The classification algorithm A1 is preferably embodied by a trained machine learning algorithm, in particular an artificial neural network. Alternatively, the classification algorithm A1 can be rule-based, for example, and in particular can be designed using a knowledge graph.

The category K of the defect in the technical component and/or system of the vehicle 1 identified by the classification algorithm A1 is evaluated by an evaluation algorithm A2 with a confidence score 16, in particular a numerical score value.

If the determined confidence score 16 exceeds or falls below a predetermined threshold value, an active learning algorithm A3 asks a user for an identification 18 of the natural language data D1 describing the defect in the technical component and/or system of the vehicle 1 Category K off.

The natural language data D1 describing the defect in the technical component and/or system of the vehicle 1 is transmitted by a user, in particular by communicating with a digital language assistant 20a or creating a text message 20b and/or an error description via menu navigation in a mobile phone app. or generated by the onboard system 14 of the vehicle 1 .

Based on the identified category K and/or the at least one error code of the defect in the technical component and/or system of the vehicle 1, the remote diagnosis module 12 creates a remote diagnosis plan which has a list of tests to be carried out and onboard systems of the vehicle 1 to be tested.

After performing the remote diagnosis of at least one onboard system 14 of the vehicle 1 or generating S3b the first error code for defects that cannot be diagnosed by the remote diagnosis module 12 in electronic communication 24 with a service center 25, at least one spare part assigned to the diagnosed defect is pre-ordered.

As part of the dialogue with the service center 25 and/or with a digital personal assistant, after the remote diagnosis has been carried out by the remote diagnosis module 12, a possible error can be further localized by the driver's dialogue with the digital personal assistant and/or an employee of the service center 25.

The digital personal assistant is implemented using a machine learning algorithm that has been trained on the basis of dialogs that have already been conducted.

If the digital personal assistant is unable to narrow down the possible error based on its training data, i.e. if a confidence in the prediction of the error by the digital personal assistant falls below a predetermined threshold value, an employee of the service center 25 takes over the further dialogue. The machine learning algorithm can then be retrained based on the detected error to increase its accuracy.

2 shows a flow chart of the computer-implemented method for dialog-supported remote diagnosis of the defect in the technical component and/or system of a vehicle according to the preferred embodiment of the invention.

When the description of the defect is automatically generated 15 due to a failure of the technical component and/or system of the vehicle 1, the onboard system 14 of the vehicle 1 creates an extended description of the defect using a data memory in the vehicle 1 or on a vehicle-external Server stored manual retrieved.

A control unit 21 arranged in the vehicle 1 scans vehicle electrical system signals to carry out anomaly detection 22 and checks whether the measured values determined are within a target range or have unexpected patterns, with the vehicle-external server 10 performing a remote diagnosis if the measured values determined are outside the target range at least one onboard system 14 of the vehicle 1 assigned to the anomaly.

A continuous or intermittent detection of technical parameters of the vehicle 1, in particular an engine speed, a torque, a number of gear shifts and/or a number of actuations of electrical components, is carried out, with diagnosis of a defect or impending defect in electronic communication 27 with a service center 25 an appointment proposal for a service appointment of the vehicle 1 is generated.

3 shows a flowchart of a computer-implemented method for providing a classification algorithm for classifying a defect in a technical component and/or system of a vehicle.

The method includes receiving S1' first input training data of natural language data D1 of a description of the defect in the technical component and/or system of the vehicle 1.

The method also includes receiving S2' second input training data of a category K identifying the defect in the technical component and/or system of vehicle 1 and training S3' of the classification algorithm A1 using the first input training data and the second input training data by an optimization algorithm A4, which Extreme value of a loss function calculated.

4 shows a schematic representation of a system for dialog-supported remote diagnosis of the defect in the technical component and/or system of a vehicle according to the preferred embodiment of the invention.

The system includes means 26 for receiving S1 natural language data D1 describing the defect in the technical component and/or system of vehicle 1.

The system also includes a computing device 28, which is set up to classify the natural language data D1 into at least one category K identifying the defect in the technical component and/or system of the vehicle 1, using a classification algorithm A1, or into a first error code for a remote diagnosis module 12 non-diagnosable defects to generate.

The system also includes a vehicle-external server 10, which is set up to run at least one remote diagnosis module 12 based on the category K that identifies the defect binding carries out a remote diagnosis of at least one onboard system 14 of the vehicle 1 .

The term vehicle includes cars, trucks, buses, mobile homes, motorcycles, etc., which are used to transport people, goods, etc.

In particular, the term includes motor vehicles for passenger transport. Additionally or alternatively, a hybrid or electric vehicle according to embodiments may be a pure electric vehicle (BEV) or a plug-in hybrid vehicle (PHEV). However, other forms of propulsion can also be used, for example in the form of a diesel or petrol-powered vehicle. The vehicle can also be in the form of a rail vehicle.

Although the invention has been illustrated and explained in more detail by means of preferred exemplary embodiments, the invention is not restricted by the disclosed examples and other variations can be derived therefrom by a person skilled in the art without departing from the protective scope of the invention.

It is therefore clear that a large number of possible variations exist. Exemplary embodiments are only examples and should not be construed as limiting in any way the scope, applications, or configuration of the invention.

Rather, the preceding description and the description of the figures enable the person skilled in the art to concretely implement the exemplary embodiments, whereby the person skilled in the art, knowing the disclosed inventive concept, can make a variety of changes, for example with regard to the function or the arrangement of individual elements mentioned in an exemplary embodiment, without having to Leaving the scope of protection defined by the claims and their legal equivalents, such as further explanations in the description.

In the present embodiment, the classification algorithm A1 is operated on the vehicle-external server. Alternatively, the classification algorithm A1 can be operated on a control unit of the vehicle, for example.

Claims (15)

Computer-implemented method for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle (1), with the steps: Receiving (S1) natural language data (D1) describing the defect in the technical component and/or system of the vehicle (1); Classifying (S2) the natural language data (D1) into at least one category (K) identifying the defect in the technical component and/or system of the vehicle (1) using a classification algorithm (A1); and based on the category (K) identifying the defect, executing (S3a) at least one remote diagnosis module (12) operating on a server (10) external to the vehicle, which carries out remote diagnosis of at least one onboard system (14) of the vehicle (1) via a wireless connection or generating (S3b) a first error code for defects that cannot be diagnosed by the remote diagnosis module (12). Computer-implemented method claim 1 , wherein at least one second error code is assigned to the at least one category (K) identified for the defect in the technical component and/or system of the vehicle (1) for defects that can be diagnosed by the remote diagnosis module (12), the remote diagnosis module (12) using the second Error code remote diagnosis of at least one of the second error code associated onboard system (14) of the vehicle (1) performs. Computer-implemented method claim 1 or 2 , wherein the natural language data (D1) describing the defect in the technical component and/or system of the vehicle (1) is text data and/or audio data, the classification algorithm (A1) being set up to process audio data and/or text data. Computer-implemented method according to one of the preceding claims, wherein the classification algorithm (A1) is rule-based, in particular by a knowledge graph, or by a trained machine learning algorithm, in particular an artificial neural network. Computer-implemented method according to one of the preceding claims, wherein the category (K) of the defect in the technical component and/or system of the vehicle (1) identified by the classification algorithm (A1) is evaluated by an evaluation algorithm (A2) with a confidence score (16), in particular a numerical score value, is evaluated, in which case if the determined confidence score (16) exceeds or falls below a predetermined threshold value, an active learning algorithm (A3) searches a user for a label (18) of the natural language data ( D1) the description of the defect queries the category (K) assigned to the technical component and/or system of the vehicle (1). Computer-implemented method according to one of the preceding claims, wherein the natural language data (D1) of the description of the defect in the technical component and / or system of the vehicle (1) by a user, in particular by communication with a digital language assistant (20a) or creating a text message (20b) and/or an error description via menu navigation of a mobile phone app, or by the onboard system (14) of the vehicle (1). Computer-implemented method according to one of the preceding claims, wherein the remote diagnosis module (12) based on the identified category (K) and / or the at least one error code of the defect in the technical component and / or system of the vehicle (1) creates a remote diagnosis plan, which a list tests to be carried out and onboard systems to be tested of the vehicle (1). Computer-implemented method according to one of the preceding claims, wherein when the technical component and/or system of the vehicle (1) fails, the description of the defect is automatically generated (15) by the onboard system (14) of the vehicle (1) with an extended Description of the defect using an operating manual stored in a data memory of the vehicle (1) is retrieved. Computer-implemented method claim 8 , wherein a control unit (21) arranged in the vehicle (1) for carrying out anomaly detection (22) scans vehicle electrical system signals and checks whether measured values determined are within a target range, wherein if the measured values determined are outside the target range, the vehicle-external server ( 10) carries out a remote diagnosis of at least one onboard system (14) of the vehicle (1) associated with the anomaly. Computer-implemented method claim 8 or 9 , wherein a continuous or interval-like detection of technical parameters of the vehicle (1), in particular an engine speed, a torque, a number of gear shifts and / or a number of actuations of electrical components, is carried out, wherein when diagnosing a defect or imminent defect in electronic communication (27) with a service center (25) an appointment proposal for a service appointment of the vehicle (1) is generated. Computer-implemented method according to one of the preceding claims, wherein after performing the remote diagnosis of at least one onboard system (14) of the vehicle (1) or generating (S3b) the first error code for defects in electronic communication (24) that cannot be diagnosed by the remote diagnosis module (12) at least one spare part assigned to the diagnosed defect is pre-ordered with a service center (25). Computer-implemented method for providing a classification algorithm (A1) for classifying a defect in a technical component and/or system of a vehicle (1), with the steps: Receiving (S1') first input training data of natural language data (D1) of a description of the defect in the technical component and/or system of the vehicle (1); receiving (S2') second input training data of a category (K) identifying the defect in the technical component and/or system of the vehicle (1); and Training (S3') the classification algorithm (A1) using the first input training data and the second input training data by an optimization algorithm (A4) which calculates an extreme value of a loss function. System for dialog-supported remote diagnosis of a defect in a technical component and/or system of a vehicle (1), comprising: Means (26) for receiving (S1) natural language data (D1) describing the defect in the technical component and/or system of the vehicle (1); and a computing device (28) which is set up to classify the natural language data (D1) into at least one category (K) identifying the defect in the technical component and/or system of the vehicle (1) using a classification algorithm (A1), or generate a first error code for defects that cannot be diagnosed by a remote diagnostic module (12); and a vehicle-external server (10), which is set up to execute at least one remote diagnosis module (12) based on the category (K) identifying the defect, which remote diagnosis of at least one onboard system (14) of the vehicle (1) via a wireless connection performs. Computer program with program code for at least one of the methods according to one of Claims 1 until 12 to be performed when the computer program is run on a computer. Computer-readable data carrier with program code of a computer program to at least one of the methods according to one of Claims 1 until 12 to be performed when the computer program is run on a computer.

Images