Method and device for detecting vehicle-mounted controller

Abstract

The application discloses a kind of detection method, device and storage medium of vehicle controller, belong to vehicle safety inspection technical field, the method is applied to detection equipment, the detection equipment is connected with the CAN bus of vehicle, the CAN bus of vehicle is connected with the vehicle controller in vehicle, and the method comprises: obtaining the signal characteristic curve of target signal of the vehicle controller to be detected in the CAN bus in preset time period;The signal characteristic curve of the vehicle controller to be detected in preset time period is substituted into target neural network model, to obtain the signal indication state of the target signal, wherein the input of the target neural network model includes signal characteristic curve, and the output includes signal indication state, wherein signal indication state includes normal state and abnormal state.The method can improve the detection efficiency of each vehicle controller in vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle safety inspection, and in particular to a detection method and device for a vehicle-mounted controller. BACKGROUND

[0002] Due to a short development cycle, a vehicle used for research and development and testing is less mature than a formally marketed vehicle, and therefore all vehicle-mounted controllers in the vehicle need to be detected before use to ensure that each vehicle-mounted controller can be normally used, thereby ensuring normal use of the vehicle.

[0003] In the related art, for a vehicle-mounted controller in a vehicle, an operator needs to connect a detection device to the vehicle-mounted controller and then use the detection device to detect the vehicle-mounted controller. However, this detection method requires the operator to connect the detection device to each vehicle-mounted controller one by one for detection, which is not only time-consuming and laborious but also low in detection efficiency. SUMMARY

[0004] In view of this, the present application provides a detection method and device for a vehicle-mounted controller, which can improve the detection efficiency of each vehicle-mounted controller in a vehicle.

[0005] Specifically, the technical solutions include the following.

[0006] In a first aspect, an embodiment of the present application provides a detection method for a vehicle-mounted controller, applied to a detection device, the detection device being connected to a CAN bus of a vehicle, the CAN bus of the vehicle being connected to vehicle-mounted controllers in the vehicle, and the method comprising:

[0007] obtaining a signal feature curve of a target signal of a vehicle-mounted controller to be detected in the CAN bus in a preset time period;

[0008] substituting the signal feature curve of the vehicle-mounted controller to be detected in the preset time period into a target neural network model to obtain a signal indication state of the target signal, wherein the input of the target neural network model includes the signal feature curve, and the output includes the signal indication state, and wherein the signal indication state includes a normal state and an abnormal state.

[0009] In some embodiments, the input of the target neural network model further includes vehicle basic running parameters, wherein the vehicle basic running parameters include an average vehicle speed, a maximum vehicle speed, a number of braking times, an acceleration time length, and a driving road condition of the vehicle in a preset time length.

[0010] In some embodiments, the driving road condition includes that the vehicle is driving on a flat road, the vehicle is driving on an uphill road, the vehicle is driving on a downhill road, the vehicle is driving on a gravel road, and the vehicle is driving on a mountain road.

[0011] In some embodiments, before the obtaining the target neural network model and the feature curve of the target signal of the vehicle controller to be detected in the CAN bus in a preset time period, the method further comprises:

[0012] Obtaining the basic running parameters of the vehicle.

[0013] In some embodiments, before the obtaining the target neural network model and the feature curve of the target signal of the vehicle controller to be detected in the CAN bus in a preset time period, the method further comprises:

[0014] Establishing the target neural network model.

[0015] In some embodiments, the establishing the target neural network model comprises:

[0016] Obtaining a plurality of sample signal feature curves of each vehicle controller in a plurality of preset time periods and corresponding calibrated signal indication states thereof;

[0017] Training a preset neural network model based on the plurality of sample signal feature curves of each vehicle controller in the plurality of preset time periods and the corresponding calibrated signal indication states thereof to obtain the target neural network model.

[0018] In some embodiments, the obtaining a plurality of sample signal feature curves of each vehicle controller in a plurality of preset time periods comprises:

[0019] Obtaining a diagnostic file of each vehicle controller from the CAN bus;

[0020] For the diagnostic file of each vehicle controller, parsing the diagnostic file of the vehicle controller to obtain a parsed diagnostic file of the vehicle controller;

[0021] Based on the parsed diagnostic file of the vehicle controller, obtaining a plurality of sample signal feature curves of the vehicle controller in a plurality of preset time periods.

[0022] In some embodiments, before the obtaining a plurality of sample signal feature curves of the vehicle controller in a plurality of preset time periods based on the parsed diagnostic file of the vehicle controller, the method further comprises:

[0023] Performing preliminary detection on the parsed diagnostic file of the vehicle controller;

[0024] In response to the difference between the maximum peak value of the sample signal feature curve in a preset time period and the peak value threshold not being within the target range, calibrating the vehicle controller.

[0025] In some embodiments, after the signal feature curve of the to-be-detected vehicle-mounted controller in the preset time period is substituted into the target neural network model to obtain the signal indication state of the target signal, the method further comprises:

[0026] In response to the signal indication state of the target signal being an abnormal state, generating and displaying prompt information.

[0027] In a second aspect, the embodiments of the present application also provide a detection device of a vehicle-mounted controller, the device comprising:

[0028] The acquisition module is configured to acquire a signal feature curve of a target signal of a to-be-detected vehicle-mounted controller in a CAN bus in a preset time period;

[0029] The signal indication state obtaining module is configured to substitute the signal feature curve of the to-be-detected vehicle-mounted controller in the preset time period into a target neural network model to obtain a signal indication state of the target signal, wherein the input of the target neural network model comprises the signal feature curve, and the output comprises the signal indication state, and wherein the signal indication state comprises a normal state and an abnormal state.

[0030] The technical solutions provided by the embodiments of the present application have at least the following beneficial effects:

[0031] The detection method of the vehicle-mounted controller provided by the embodiments of the present application, when detecting a to-be-detected vehicle-mounted controller in a vehicle, since the detection device is connected to the vehicle-mounted controller in the vehicle through the CAN bus, the signal feature curve of the target signal of the to-be-detected vehicle-mounted controller in a preset time period can be acquired through the CAN bus, and the feature curve of the signal is substituted into the target neural network model to obtain the signal indication state of the target signal, so as to determine whether the target signal is in a normal state or an abnormal state. The method is connected to the CAN bus through the detection device, and the feature curve of the target signal of any to-be-detected vehicle-mounted controller in the vehicle is acquired through the CAN bus, which replaces the mode of connecting the vehicle-mounted controller and the detection device one by one in the related art, and improves the detection efficiency of the vehicle-mounted controller. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0033] Figure 1 A flowchart of a detection method of a vehicle-mounted controller provided by the embodiments of the present application;

[0034] Figure 2 A flowchart of another method for detecting a vehicle-mounted controller according to an embodiment of the present application is provided.

[0035] Figure 3 A flowchart of a method for establishing a target neural network model in a method for detecting a vehicle-mounted controller according to an embodiment of the present application is provided.

[0036] Figure 4 A flowchart of a method for obtaining a signal feature curve of a target signal of a vehicle-mounted controller to be detected in a CAN bus in a predetermined time period in a method for detecting a vehicle-mounted controller according to an embodiment of the present application is provided.

[0037] Figure 5 A structural schematic diagram of a detection device for a vehicle-mounted controller according to an embodiment of the present application is provided.

[0038] The above-described figures have shown the specific embodiments of the present application, which will be described in more detail hereinafter. These figures and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the figures in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative work fall within the scope of protection of the present application.

[0040] Unless otherwise defined, all the technical terms used in the embodiments of the present application have the same meanings as commonly understood by those skilled in the art. Some technical terms appearing in the embodiments of the present application will be described below.

[0041] In the embodiments of the present application, the term “DBC file” generally refers to an XML format file used to define a CAN bus communication protocol in a CAN bus diagnosis tool.

[0042] In order to make the technical solutions and advantages of the present application clearer, the embodiments of the present application will be described in further detail below in combination with the figures.

[0043] Before a new vehicle is officially put on sale, the vehicle manufacturer usually produces some vehicles for research and testing. Due to the short development cycle, the vehicles for research and testing are less mature than the officially sold vehicles. In addition, after being tested for many times, the vehicle controllers will also be damaged to different degrees. Therefore, before use, all vehicle controllers in the vehicle need to be detected to ensure that the vehicle controllers can be used normally, thereby ensuring the normal use of the vehicle and avoiding affecting the test results and the personal safety of related personnel.

[0044] In the related art, for each vehicle controller in the vehicle, an operator needs to connect the detection equipment with the vehicle controller and then use the detection equipment to detect the vehicle controller. However, this detection method needs the operator to connect the detection equipment with each vehicle controller one by one for detection, which is not only time-consuming and laborious, but also low in detection efficiency.

[0045] To solve the technical problems in the related art, the embodiments of the present application provide a detection method of a vehicle controller, which can improve the detection efficiency of each vehicle controller in the vehicle.

[0046] The embodiments of the present application provide a detection device. The detection device has a processor and a memory connected to each other. The detection device can be at least one of a smart phone, a laptop computer, a portable computer, a tablet computer, and the like. The detection device is connected to the CAN bus of the vehicle, and the CAN bus of the vehicle is connected to the vehicle controller in the vehicle.

[0047] Figure 1 A flowchart of a detection method of a vehicle controller provided by the embodiments of the present application is shown in FIG. 1. Figure 1 The method is applied to the above-mentioned detection device, and the method includes the following steps.

[0048] In step 101, a signal feature curve of a target signal of a vehicle controller to be detected in a CAN bus in a preset time period is obtained.

[0049] In step 102, the signal feature curve of the vehicle controller to be detected in the preset time period is substituted into a target neural network model to obtain a signal indication state of the target signal, wherein the input of the target neural network model includes the signal feature curve, and the output includes the signal indication state. The signal indication state includes a normal state and an abnormal state.

[0050] The vehicle controller detection method provided by the embodiments of the present application can obtain the signal characteristic curve of the target signal of the vehicle controller to be detected in the CAN bus in a preset time period, and the signal characteristic curve of the target signal is substituted into the target neural network model to obtain the signal indication state of the target signal, so as to determine whether the target signal is in a normal state or an abnormal state. The method is connected with the CAN bus through the detection device, and the characteristic curve of the target signal of any detection vehicle controller in the vehicle is obtained through the CAN bus, which replaces the mode of connecting the vehicle controller and the detection device one by one in the related art, and improves the detection efficiency of the vehicle controller.

[0051] In some embodiments, the input of the target neural network model further includes vehicle basic running parameters, wherein the vehicle basic running parameters include average vehicle speed, maximum vehicle speed, braking times, acceleration time length and driving road conditions of the vehicle in a preset time length.

[0052] In some embodiments, the driving road conditions include flat road driving, uphill road driving, downhill road driving, gravel road driving and mountain road driving of the vehicle.

[0053] In some embodiments, before obtaining the target neural network model and the characteristic curve of the target signal of the vehicle controller to be detected in the CAN bus in a preset time period, the vehicle controller detection method further includes:

[0054] Obtaining vehicle basic running parameters.

[0055] In some embodiments, before obtaining the target neural network model and the characteristic curve of the target signal of the vehicle controller to be detected in the CAN bus in a preset time period, the vehicle controller detection method further includes:

[0056] Establishing a target neural network model.

[0057] In some embodiments, establishing a target neural network model includes:

[0058] Obtaining sample signal characteristic curves and corresponding calibrated signal indication states of the vehicle in a plurality of preset time lengths of each vehicle controller;

[0059] Training the preset neural network model based on the characteristic curves and corresponding calibrated signal indication states of each vehicle controller in a preset time length to obtain a target neural network model.

[0060] In some embodiments, obtaining the signal characteristic curve of the target signal of the vehicle controller to be detected in the CAN bus in a preset time period includes:

[0061] obtaining a diagnostic file of each vehicle-mounted controller from the CAN bus;

[0062] For the diagnostic file of each vehicle-mounted controller, parsing the diagnostic file of the vehicle-mounted controller to obtain a parsed diagnostic file of the vehicle-mounted controller;

[0063] Based on the parsed diagnostic file of the vehicle-mounted controller, obtaining a plurality of sample signal feature curves within a preset time length of the vehicle-mounted controller.

[0064] In some embodiments, before obtaining a plurality of sample signal feature curves within a preset time length of the vehicle-mounted controller based on the parsed diagnostic file of the vehicle-mounted controller, the method further comprises:

[0065] performing preliminary detection on the parsed diagnostic file of the vehicle-mounted controller;

[0066] In response to the difference between the maximum peak value of the sample signal feature curve within the preset time length and the peak value threshold not being within the target range, calibrating the vehicle-mounted controller.

[0067] In some embodiments, after substituting the signal feature curve of the vehicle-mounted controller to be detected within a preset time period into the target neural network model to obtain the signal indication state of the target signal, the method further comprises:

[0068] In response to the signal indication state of the target signal being an abnormal state, generating and displaying prompt information.

[0069] Figure 2 Another flowchart of a vehicle-mounted controller detection method provided by the embodiments of the present application is provided, see Figure 2 The method is applied to the detection device mentioned above, and the method further comprises the following steps:

[0070] Step 201: Obtain vehicle basic running parameters.

[0071] By obtaining the vehicle basic running parameters, the subsequent training and learning of the target neural network model can be facilitated.

[0072] The vehicle basic running parameters include the average speed, the maximum speed, the number of braking times, the acceleration time length, and the driving road conditions of the vehicle within a preset time length.

[0073] It can be understood that the vehicle basic running parameters reflect the running conditions of the vehicle, and when they are used as the input of the target neural network model, the target neural network can be better learned based on the specific running conditions of the vehicle.

[0074] In some embodiments, the driving road conditions include that the vehicle is driving on flat road, the vehicle is driving on uphill road, the vehicle is driving on downhill road, the vehicle is driving on gravel road, and the vehicle is driving on mountain road.

[0075] Here, different driving road conditions indicate that the vehicle is in different driving environments, and distinguishing the driving road conditions can facilitate the target neural network to learn more finely.

[0076] Step 202, establishing a target neural network model.

[0077] By establishing the target neural network model, a signal indication state of a target signal of a to-be-detected vehicle-mounted controller can be obtained based on the target neural network model.

[0078] The input of the target neural network model includes a signal feature curve and a basic running parameter of the vehicle, and the output includes the signal indication state, where the signal indication state includes a normal state and an abnormal state.

[0079] In some embodiments, referring to Figure 3 , step 202 includes the following steps:

[0080] Step 2021, obtaining a sample signal feature curve of each vehicle-mounted controller in a plurality of preset time lengths and a corresponding calibrated signal indication state thereof.

[0081] The signal feature can be mean, variance, skewness, kurtosis, root mean square, root amplitude, waveform factor, peak factor, etc., so the sample signal feature can be at least two of the above signal features. Correspondingly, the signal feature curve refers to the change curve of the value of the signal feature with time.

[0082] In the embodiments of the present application, different vehicle-mounted controllers can correspond to different signal features, that is, the sample signal feature type of each vehicle-mounted controller in the preset time length can be different.

[0083] In the embodiments of the present application, the sample signal feature curve corresponding to the calibrated signal indication state is obtained by human calibration.

[0084] In some embodiments, referring to Figure 4 , the obtaining of the sample signal feature curve of each vehicle-mounted controller in the plurality of preset time lengths in step 2021 specifically includes the following steps:

[0085] Step 20211, obtaining a diagnostic file of each vehicle-mounted controller from the CAN bus.

[0086] In some embodiments, the diagnostic file can be a DBC file.

[0087] The DBC file contains CAN communication signals, message IDs, data lengths, and periodicity information, and the CAN communication signals contain signals of each vehicle-mounted controller.

[0088] At step 20212, for each vehicle-mounted controller diagnosis file, the vehicle-mounted controller diagnosis file is parsed to obtain a parsed vehicle-mounted controller diagnosis file.

[0089] Since the signal characteristic curve cannot be directly obtained directly from the diagnosis file, the diagnosis file needs to be parsed to facilitate obtaining the signal characteristic curve based on the parsed diagnosis file.

[0090] At step 20213, the parsed vehicle-mounted controller diagnosis file is preliminarily detected.

[0091] The parsed vehicle-mounted controller diagnosis file is preliminarily detected to preliminarily determine whether the signal of the vehicle-mounted controller has a problem.

[0092] At step 20214, in response to a difference between a maximum peak value of a sample signal characteristic curve within a preset time length and a peak value threshold not being within a target range, the vehicle-mounted controller is calibrated.

[0093] That is, if the difference between the maximum peak value of the sample signal characteristic curve within the preset time length and the peak value threshold is not within the target range, it indicates that the vehicle-mounted controller may have a problem, and it can be prioritized for calibration, and subsequent attention is focused on the vehicle-mounted controller.

[0094] At step 20215, based on the parsed vehicle-mounted controller diagnosis file, a plurality of sample signal characteristic curves of the vehicle-mounted controller within a preset time length are obtained.

[0095] The parsed vehicle-mounted controller diagnosis file contains a plurality of sample signal characteristic curves of the vehicle-mounted controller within a preset time length, so the plurality of sample signal characteristic curves of the vehicle-mounted controller within the preset time length can be directly obtained from the parsed vehicle-mounted controller diagnosis file.

[0096] At step 2022, based on the plurality of sample signal characteristic curves of each vehicle-mounted controller within a preset time length and the corresponding calibrated signal indication state, a preset neural network model is trained to obtain a target neural network model.

[0097] By inputting the plurality of sample signal characteristic curves of each vehicle-mounted controller within a preset time length and the corresponding calibrated signal indication state into the preset neural network model, the preset neural network model is trained and learned, and the target neural network model can be obtained.

[0098] In some embodiments, the preset neural network model can be a convolutional neural network model.

[0099] In step 203, a signal characteristic curve of the target signal of the vehicle-mounted controller to be detected in the preset time period is acquired.

[0100] In some embodiments, the preset time period is usually 10 seconds.

[0101] In step 204, the signal characteristic curve of the vehicle-mounted controller to be detected in the preset time period and the basic running parameter of the vehicle are substituted into the target neural network model to obtain a signal indication state of the target signal.

[0102] Since the input of the target neural network model includes the signal characteristic curve and the basic running parameter of the vehicle, and the output includes the signal indication state, after the signal characteristic curve of the vehicle-mounted controller to be detected in the preset time period is substituted into the target neural network model, the target neural network model can output the signal indication state of the target signal. Since the signal indication state includes the normal state and the abnormal state, the communication condition of the vehicle-mounted controller can be determined based on the signal indication state, and the detection of the vehicle-mounted controller to be detected is realized.

[0103] In step 205, in response to the signal indication state of the target signal being the abnormal state, prompt information is generated and displayed.

[0104] In the case where the signal indication state of the target signal is the abnormal state, the prompt information is generated and displayed to prompt the user to pay attention to the communication abnormality of the vehicle-mounted controller to be detected, which needs to be repaired.

[0105] In some embodiments, the prompt information includes prompt text displayed on the display screen of the detection device and / or the signal characteristic curve in the preset time period.

[0106] In some embodiments, in addition to generating and displaying the prompt information, the detection device also controls the vehicle to give a prompt. For example, the vehicle light is controlled to flash.

[0107] Therefore, the detection method of the vehicle-mounted controller provided in the embodiments of the present application, when detecting the vehicle-mounted controller to be detected in the vehicle, since the detection device is connected with the vehicle-mounted controller in the vehicle through the CAN bus, the signal characteristic curve of the target signal of the vehicle-mounted controller to be detected in the preset time period can be acquired through the CAN bus, and the characteristic curve of the signal is substituted into the target neural network model to obtain the signal indication state of the target signal to determine whether the target signal is in the normal state or the abnormal state. The method connects the detection device with the CAN bus, and acquires the characteristic curve of the target signal of any vehicle-mounted controller to be detected in the vehicle through the CAN bus, which replaces the mode of connecting the vehicle-mounted controller with the detection device one by one in the related art, and improves the detection efficiency of the vehicle-mounted controller.

[0108] Figure 5 A structural schematic diagram of a detection device of a vehicle-mounted controller provided by an embodiment of the present application is shown in FIG. 1. Figure 5 The device 500 is located in a detection equipment and includes:

[0109] The acquisition module 501 is configured to acquire a signal characteristic curve of a target signal of a vehicle-mounted controller to be detected in a preset time period in a CAN bus.

[0110] The signal indication state obtaining module 502 is configured to input the signal characteristic curve of the vehicle-mounted controller to be detected in the preset time period into a target neural network model to obtain a signal indication state of the target signal, wherein the input of the target neural network model includes the signal characteristic curve, and the output includes the signal indication state, and wherein the signal indication state includes a normal state and an abnormal state.

[0111] In some embodiments, the input of the target neural network model further includes vehicle basic running parameters, wherein the vehicle basic running parameters include an average vehicle speed, a maximum vehicle speed, a number of braking times, an acceleration time length, and a driving road condition of the vehicle in a preset time length.

[0112] In some embodiments, the driving road condition includes that the vehicle is driving on a flat road, the vehicle is driving on an uphill road, the vehicle is driving on a downhill road, the vehicle is driving on a gravel road, and the vehicle is driving on a mountain road.

[0113] In some embodiments, the detection device of the vehicle-mounted controller further includes:

[0114] The parameter acquisition module is configured to acquire the vehicle basic running parameters.

[0115] In some embodiments, the detection device of the vehicle-mounted controller further includes:

[0116] The modeling module is configured to establish the target neural network model.

[0117] In some embodiments, the modeling module includes:

[0118] The first acquisition submodule is configured to acquire a plurality of sample signal characteristic curves in a preset time length and corresponding calibrated signal indication states of each vehicle-mounted controller.

[0119] The first obtaining submodule is configured to train a preset neural network model based on the plurality of sample signal characteristic curves in the preset time length and the corresponding calibrated signal indication states of each vehicle-mounted controller to obtain the target neural network model.

[0120] In some embodiments, the acquisition module includes:

[0121] The second acquisition submodule is configured to acquire a diagnostic file of each vehicle-mounted controller from the CAN bus.

[0122] a second obtaining sub-module, configured to, for each diagnostic file of the vehicle-mounted controller, parse the diagnostic file of the vehicle-mounted controller to obtain a parsed diagnostic file of the vehicle-mounted controller;

[0123] a third obtaining sub-module, configured to, based on the parsed diagnostic file of the vehicle-mounted controller, obtain a plurality of sample signal feature curves within a preset time length of the vehicle-mounted controller.

[0124] In some embodiments, the apparatus further includes:

[0125] a detecting module, configured to perform preliminary detection on the parsed diagnostic file of the vehicle-mounted controller;

[0126] a calibrating module, configured to, in response to a difference between a maximum peak value of the sample signal feature curve within the preset time length and the peak value threshold not being within the target range, calibrate the vehicle-mounted controller.

[0127] In some embodiments, the vehicle-mounted controller detection apparatus further includes:

[0128] a display generating module, configured to, in response to the signal indication state of the target signal being the abnormal state, generate and display prompt information.

[0129] Therefore, the vehicle-mounted controller detection apparatus provided by the embodiments of the present application replaces the manner of connecting the vehicle-mounted controllers with the detection device one by one in the related art by connecting with the CAN bus and obtaining the feature curve of the target signal of the vehicle-mounted controller in the vehicle through the CAN bus, thereby improving the detection efficiency of the vehicle-mounted controller.

[0130] In the present application, the terms "first" and "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance. The term "a plurality of" refers to two or more, unless otherwise explicitly limited.

[0131] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. The specification and examples given are considered exemplary only, and the scope of the application is not to be considered limited by such. The application is intended to cover any variations, uses or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within the known and customary practice of the art to which the application pertains. The specification and examples are to be regarded as illustrative only.

[0132] It should be understood that the present application is not limited to the precise construction that has been described and illustrated herein and that various modifications and changes can be made therein without departing from the scope of the present application. The scope of the present application is limited only by the claims appended hereto.

Claims

1. A method of detecting a vehicle-mounted controller, characterized by, The method is applied to a detection device connected with a CAN bus of a vehicle, the CAN bus of the vehicle is connected with vehicle-mounted controllers in the vehicle, and the method comprises the following steps: establishing a target neural network model; obtaining a signal characteristic curve of a target signal of a to-be-detected vehicle-mounted controller in the CAN bus in a preset time period, the signal characteristics including mean, variance, skewness, kurtosis, root mean square, root amplitude, waveform factor and peak factor, and the signal characteristic curve being a curve of the value of the signal characteristic changing with time; substituting the signal characteristic curve of the to-be-detected vehicle-mounted controller in the preset time period into the target neural network model to obtain a signal indication state of the target signal, wherein the input of the target neural network model includes the signal characteristic curve, and the output includes the signal indication state, and the signal indication state includes a normal state and an abnormal state; wherein the establishment of the target neural network model comprises: obtaining a plurality of sample signal characteristic curves of each vehicle-mounted controller in a preset time period and a corresponding calibrated signal indication state; training a preset neural network model based on the plurality of sample signal characteristic curves of each vehicle-mounted controller in the preset time period and the corresponding calibrated signal indication state to obtain the target neural network model, the preset neural network model being a convolutional neural network model; the obtaining of the plurality of sample signal characteristic curves of each vehicle-mounted controller in the preset time period comprises: obtaining a diagnostic file of each vehicle-mounted controller from the CAN bus; for the diagnostic file of each vehicle-mounted controller, analyzing the diagnostic file of the vehicle-mounted controller to obtain an analyzed diagnostic file of the vehicle-mounted controller; based on the analyzed diagnostic file of the vehicle-mounted controller, obtaining the plurality of sample signal characteristic curves of the vehicle-mounted controller in the preset time period.

2. The method of detecting an in-vehicle controller according to claim 1, characterized by, The input of the target neural network model further includes vehicle basic running parameters, wherein the vehicle basic running parameters include average vehicle speed, maximum vehicle speed, braking frequency, acceleration time length and driving road condition of the vehicle in a preset time period.

3. The method of claim 2, wherein The driving road condition includes flat road driving, uphill road driving, downhill road driving, gravel road driving and mountain road driving.

4. The method of claim 2, wherein Before obtaining the target neural network model and the characteristic curve of the target signal of the to-be-detected vehicle-mounted controller in the CAN bus in the preset time period, the method further comprises: obtaining the vehicle basic running parameters.

5. The method of detecting an in-vehicle controller according to claim 1, wherein Before obtaining the plurality of sample signal characteristic curves of the vehicle-mounted controller in the preset time period based on the analyzed diagnostic file of the vehicle-mounted controller, the method further comprises: performing preliminary detection on the analyzed diagnostic file of the vehicle-mounted controller; in response to a difference between a maximum peak value of a sample signal characteristic curve in a preset time period and a peak value threshold not being within a target range, calibrating the vehicle-mounted controller.

6. The method of detecting an in-vehicle controller according to claim 1, wherein After substituting the signal characteristic curve of the to-be-detected vehicle-mounted controller in the preset time period into the target neural network model to obtain the signal indication state of the target signal, the method further comprises: In response to the signal indication state of the target signal being an abnormal state, prompt information is generated and displayed.

7. A detection device of an in-vehicle controller characterized by comprising: The device comprises: a modeling module for establishing a target neural network model; an acquisition module for acquiring a signal characteristic curve of a target signal of a to-be-detected vehicle-mounted controller in a CAN bus in a preset time period, the signal characteristics including mean, variance, skewness, kurtosis, root mean square, root amplitude, waveform factor, and peak factor, and the signal characteristic curve being a curve of the value of the signal characteristics changing with time; a signal indication state obtaining module for obtaining a signal indication state of the target signal by substituting the signal characteristic curve of the to-be-detected vehicle-mounted controller in the preset time period into the target neural network model, wherein the input of the target neural network model includes the signal characteristic curve, and the output includes the signal indication state, and wherein the signal indication state includes a normal state and an abnormal state; wherein the modeling module comprises: a first acquisition submodule for acquiring a plurality of sample signal characteristic curves of each vehicle-mounted controller in a preset time period and their corresponding calibrated signal indication states; a first obtaining submodule for training a preset neural network model based on the plurality of sample signal characteristic curves of each vehicle-mounted controller in a preset time period and their corresponding calibrated signal indication states to obtain the target neural network model, wherein the preset neural network model is a convolutional neural network model; the acquisition module comprises: a second acquisition submodule for acquiring a diagnostic file of each vehicle-mounted controller from the CAN bus; a second obtaining submodule for, for each diagnostic file of a vehicle-mounted controller, parsing the diagnostic file of the vehicle-mounted controller to obtain a parsed diagnostic file of the vehicle-mounted controller; a third obtaining submodule for obtaining a plurality of sample signal characteristic curves of the vehicle-mounted controller in a preset time period based on the parsed diagnostic file of the vehicle-mounted controller.

Images