Automated data simulation tool and method based on linear regression model

Abstract

The application discloses an automatic data simulation tool and method based on a linear regression model, comprising a simulation video library module, a data conversion module, an instruction management module and a data playback module, wherein the simulation video library module, the data conversion module and the instruction management module are all associated with the data playback module, the simulation video library module comprises a video file unit and a configuration file unit; the data conversion module is a core module of the system, and comprises a data filtering unit, a data conversion unit and a data rewriting unit; the data filtering unit is used for filtering data in a parsed Log file according to an instruction before simulation starts, and target data is obtained through linear regression model conversion; and data used in the simulation process is measured by equipment in an actual road test scene, and the data has reliability, accuracy and effectiveness, and a reproduced scene is equivalent to the actual road test scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of computer software development, in particular to an automatic data simulation tool and method based on a linear regression model. BACKGROUND

[0002] Advanced Driving Assistance System is to use various sensors (millimeter wave radar, laser radar, single / dual camera and satellite navigation) installed on the car to sense the environment around the car at any time during driving, collect data, identify, detect and track static and dynamic objects, and combine navigation map data to perform system calculation and analysis, so as to let the driver aware of the possible danger in advance, effectively increase the comfort and safety of car driving. In recent years, the ADAS market has grown rapidly. Originally, such systems were limited to the high-end market, but now they are entering the mid-end market. At the same time, many low-technology applications are more common in entry-level passenger vehicles, and improved new sensor technology is also creating new opportunities and strategies for system deployment.

[0003] In the field of intelligent driving, the ADAS system contains a rich library of Corner Case scenarios, which need a large amount of experimental data to evaluate its reliability. However, due to the large number of sensors loaded on the autonomous vehicle, a large amount of data will be generated during testing, and the tester cannot accurately locate the problem. At the same time, for a large number of video libraries and data, testers usually need to repeat the configuration operation to manually simulate and verify, which is very wasteful of manpower and resources, and the simulation efficiency is low.

[0004] In view of the above defects in the prior art, the present application provides an automatic data simulation tool and method based on a linear regression model, which intercepts the data of the problem scene to simulate the actual road vehicle operation, and processes the data according to the test requirements, realizes the secondary utilization of historical scene data, and achieves the purpose of data cycle simulation. Therefore, the present application is very necessary. SUMMARY

[0005] The present application aims to provide an automatic data simulation tool and method based on a linear regression model, to solve the problem in the background art that when simulating data from the scene library in the ADAS system, the developer usually needs to upload data one by one and manually configure, and the data of different scenes cannot be reused, which is time-consuming and laborious, and the development quality is not easy to control.

[0006] In order to achieve the above object, the application provides the following technical scheme: an automatic data simulation tool based on a linear regression model, comprising a simulation video library module, a data conversion module, an instruction management module and a data playback module. The simulation video library module, the data conversion module and the instruction management module are all associated with the data playback module. The simulation video library module comprises a video file unit and a configuration file unit. The data conversion module is the core module of the system and comprises a data filtering unit, a data conversion unit and a data rewriting unit. The instruction management module comprises an instruction analysis unit and a simulation management unit. The data playback module comprises a data analysis unit, a data sending unit and an image playing unit.

[0007] Preferably, the data conversion unit processes the parsed message signal source data by using a linear regression model in combination with a configuration file and user operation to obtain target data, which is provided to the data playback module for synchronous simulation.

[0008] Preferably, the simulation management unit analyzes the instruction by using an.xlsx instruction definition standard file according to the instruction sent by an external CAN device, so as to select and simulate the scene of the simulation video library module, realize the automatic cyclic playback function of data, save the verification time of the test personnel and improve the development verification efficiency.

[0009] Preferably, the application further provides an automatic data simulation method based on a linear regression model, comprising the following steps:

[0010] S1: import.log files,.xml files and.dbc files. The system analyzes and extracts the data in the log files according to a configuration file to obtain CAN data and video image data required for playback as source data of the simulation video library module;

[0011] S2: the obtained source data is configured on a UI interface as required, and the data is processed by a linear regression model of the data conversion module to obtain target data for simulation;

[0012] S3: the target data is simulated by the data playback module as required, and the parsed CAN data is sent and the video is played back synchronously;

[0013] S4: the instruction management module is sent with a play instruction and a video selection instruction by an external CAN device as required. The system analyzes the instruction by using an.xlsx instruction definition standard file, so as to switch the source data of the simulation video library, and then return to step S2 to simulate the next scene and verify the automatic cyclic simulation of data.

[0014] Compared with the prior art, the application has the following beneficial effects:

[0015] 1. The data used in the simulation process of this invention are all obtained from measurements taken by equipment in actual road test scenarios. The data is reliable, accurate, and effective, and the reproduced scenarios are equivalent to actual road test scenarios. Therefore, compared with frequent road test testing, this invention saves test personnel time and resource costs for on-site testing, avoids situations where unforeseen circumstances cannot be reproduced, and ensures the smooth progress of testing.

[0016] 2. This invention can simulate road test scenarios based on measurement results. During playback, different log files can be selected according to received instructions to switch simulation scenarios. It also uses a linear regression model to transform and process data based on XML and DBC configuration files. Furthermore, it provides an API call interface with a visual operation interface, offering good versatility, ease of use, and easy maintenance. By simply providing log, XML, and DBC files and configuring them on the operation interface, automated data loop playback can be achieved, enabling secondary data utilization and improving development and verification efficiency. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the operation of the present invention.

[0018] Figure 2 This is a directory diagram of the configuration files required for the simulation video library module of this invention;

[0019] Figure 3 This is a detailed diagram of the configuration files required for the simulation video library module of this invention;

[0020] Figure 4 This is a diagram showing the details of the DBC files that the data conversion module of this invention needs to convert;

[0021] Figure 5 This is a template diagram of the instruction management module receiving instructions when the present invention is applied to autonomous driving.

[0022] Figure 6 This is a reference standard diagram for the instruction management module of the present invention to receive instructions;

[0023] Figure 7 This is a functional flowchart of the data conversion module of the present invention;

[0024] Figure 8 This is the user interface of the present invention;

[0025] Figure 9 This invention relates to the image playback interface of the data playback module used in autonomous driving.

[0026] Figure 10 This is a flowchart of the instruction receiving and parsing function of the present invention.

[0027] In the figure: 1, simulation video library module; 2, data conversion module; 3, instruction management module; 4, data playback module. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0029] Referring to Figures 1-10 The present application provides an automatic data simulation tool and method based on a linear regression model, which comprises a simulation video library module 1, a data conversion module 2, an instruction management module 3 and a data playback module 4. The simulation video library module 1, the data conversion module 2 and the instruction management module 3 are all associated with the data playback module 4. The simulation video library module 1 comprises a video file unit and a configuration file unit. The data conversion module 2 comprises a data filtering unit, a data conversion unit and a data rewriting unit. The instruction management module 3 comprises an instruction analysis unit and a simulation management unit. The data playback module 4 comprises a data analysis unit, a data sending unit and an image playing unit.

[0030] In the embodiment, when used, the.log file, the.xml file and the.dbc file to be played back are first loaded into the project. According to the requirements, the data and images in the.log file are synchronously played back, and the specified message is sent to the corresponding device. At the same time, a visual operation interface with good universality, high usability and easy maintenance is provided. The user only needs to provide the file and configure on the operation interface according to the requirements, so as to realize automatic cyclic playback of data.

[0031] In the embodiment, the video file unit is a necessary configuration unit, which contains the.log binary file stored in the Corner Case scenario library recorded in advance. The subsequent operation can be performed only after the user imports.

[0032] In the embodiment, the configuration file unit is a necessary configuration unit, which contains the.xml and.dbc configuration files customized by the user. The subsequent operation can be performed only after the user imports.

[0033] In the embodiment, the simulation management unit contains the.csv file, which is a unified instruction reference standard. The CAN device instruction received by the present application must be sent according to the standard for internal analysis by the software.

[0034] The automatic data simulation method based on a linear regression model comprises the following steps:

[0035] S1: Import.log file,.xml file,.dbc file, the system parses and extracts the data in the log file according to the configuration file, obtains the required CAN data and video image data for playback as the source data of the simulation video library module;

[0036] S2: The source data is configured on the UI interface as needed, and the linear regression model of the data conversion module is used for processing the data to obtain target data for simulation;

[0037] S3: According to the need, the target data is simulated by the data playback module, and the parsed CAN data is sent and the video is played back synchronously;

[0038] S4: According to the need, the playing instruction and the video selection instruction are sent to the instruction management module through the external CAN device, the system parses the instructions through the.xlsx instruction definition standard file, so as to switch the source data of the simulation video library. Then go back to step S2, carry out the next scene simulation verification, and realize the automatic data cycle simulation.

[0039] Specific use, the realization of data simulation:

[0040] (1) Left-click the "File" option in the menu bar at the top left corner of the UI interface ( Figure 8 ), select the "Load LOG file" function;

[0041] (2) After the pop-up window appears, select the directory where the log file to be simulated is stored ( Figure 2 ), and all folders under the log folder must be named in a standard manner, and each folder must contain.log,.xml and.dbc files ( Figure 3 );

[0042] (3) After loading, the UI interface left side drop-down box will display all the selectable log file paths, and the table in the left side will display the messages and signals in the DBC file belonging to the current log file;

[0043] (4) Left-click the "File" option in the menu bar at the top left corner of the UI interface ( Figure 8 ), select the "Load DBC file" function;

[0044] (5) After the pop-up window appears, select the directory where the.dbc file to be converted is stored ( Figure 4 );

[0045] (6) After loading, the UI interface right side drop-down box will display all the selectable.dbc files, and the table in the right side will display the messages and signals in the currently selected DBC file;

[0046] (7) Manually click the signals in the tables on the left and right sides of the UI interface to associate or disassociate the signals Figure 8 );

[0047] (8) If the signal association is updated, left-click the menu bar "File" option in the upper left corner of the UI interface Figure 8 ) before exiting the program, select the "Save XML file" function, and the system will automatically update the current log file belonging to the.xml configuration file. If there is no update, this step is ignored.

[0048] (9) After determining that there is no error, click the "Start playback" button to perform data simulation. During this period, the application will parse and filter the data in the.log file, and convert the source data Figure 7 ), and finally send it to the external CAN device and play the scene video synchronously Figure 9 ).

[0049] (10) The system monitors the external CAN device in real time, and if a command is received, the command is first parsed and processed Figure 10 ) according to the command standard defined by the application Figure 5 、 Figure 6 ). At the same time, the command content is also displayed on the UI interface Figure 8 ) for the user to check. The system switches the log file required for simulation according to the result and repeats the above steps to realize automatic simulation data recycling.

[0050] (11) Click the "Terminate playback" button on the UI interface to manually stop the current data playback process Figure 8 ).

[0051] Although embodiments of the application have been shown and described, it is to be understood that various modifications, substitutions, replacements and changes can be made to these embodiments without departing from the principles and spirit of the application, and the scope of the application is defined by the appended claims and their equivalents.

Claims

1. An automated data simulation tool based on a linear regression model, comprising a simulation video library module (1), a data conversion module (2), an instruction management module (3), and a data playback module (4), characterized in that, The simulation video library module (1), data conversion module (2), and instruction management module (3) are all associated with the data playback module (4). The simulation video library module (1) includes a video file unit and a configuration file unit. The data conversion module (2) is the core module. The data conversion module (2) includes a data filtering unit, a data conversion unit, and a data rewriting unit. The data filtering unit is used to filter the data in the parsed Log file according to the instructions before the simulation starts, and to convert the data to the target data through a linear regression model. The instruction management module (3) includes an instruction parsing unit and a simulation management unit. The simulation management unit receives the instructions sent by the external CAN device, and first uses the .xlsx instruction to define a standard file to parse the instructions, thereby switching the simulation video source data required for the next simulation. It selects and processes the source data according to the data conversion instructions, and finally sends the target CAN data and plays the target video image. The data playback module (4) includes a data parsing unit, a data sending unit, and an image playback unit. The data parsing unit parses and processes the data in the Log file before the playback starts, and performs synchronous processing on the processed target video data and CAN data. The synchronous processing includes synchronously sending CAN data and playing back the image video.

2. The automated data simulation tool based on a linear regression model according to claim 1, characterized in that: The video file unit and configuration file unit contain a pre-recorded .log binary file stored in the Corner Case scene library, as well as custom .xml and .dbc configuration files.

3. The automated data simulation tool based on a linear regression model according to claim 1, characterized in that: The instruction management module (3) controls the system simulation process by sending instructions from an external CAN device. It contains a custom .xlsx configuration file, which serves as a reference manual for the instructions.

4. The simulation method of the automated data simulation tool based on a linear regression model according to any one of claims 1-3, characterized in that: Including the following steps: S1: Import .log file, .xml file, and .dbc file. The system parses and extracts the data in the log file according to the configuration file to obtain the CAN data and video image data to be played back as the source data of the simulation video library module (1). S2: The source data obtained is configured in the UI interface as needed, and the data is processed by the linear regression model of the data conversion module (2) to obtain the target data for simulation. S3: As needed, the target data is simulated and processed by the data playback module (4), the parsed CAN data is sent and the video is played back synchronously; S4: As needed, send playback instructions and video selection instructions to the instruction management module (3) through an external CAN device. The system parses the instructions through the .xlsx instruction definition standard file, thereby switching the simulation video library source data; then return to step S2 to perform the next scenario simulation verification and realize automated data loop simulation.

Images