A method and system for concretizing abstract autonomous driving experimental data

Abstract

The present invention discloses a method and system for concretizing and abstracting autonomous driving experimental data, which relates to the field of data analysis technology, including a data input module, a data processing module, a data visualization module, a user module and a data storage module; the method and system perform data cleaning, integration and abstraction through the data processing module, can remove noise data, correct erroneous data, fill in missing data, and improve data quality and accuracy; the data integration module can integrate data from different experimental equipment, reduce the time and error rate of manual processing, and improve data processing efficiency; the data visualization module converts the processed data into a visual form, which more intuitively reflects the meaning and association of the data; through the interactive operation of the user module, the user can set parameters and adjust the visualization method as needed to better meet personalized needs and improve the understandability of data visualization and user experience.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a method and system for concretizing abstract autonomous driving experimental data. Background Art

[0002] The methods and systems for concretizing abstract autonomous driving experimental data are a relatively new research field. It aims to transform abstract experimental data into concrete and actionable information in order to better understand and analyze the experimental results. The methods and systems for concretizing abstract autonomous driving experimental data cover multiple aspects from specific examples, chart presentation, model construction to practical experience, and are combined with advanced technical means such as digital experimental platforms and database systems. These methods and systems can effectively help people better understand and apply abstract experimental data.

[0003] The main technologies used in the current means of abstracting experimental data have some obvious shortcomings. In the preprocessing process, data loss or errors may occur, which will affect the subsequent concretization process; the data abstraction process may cause information loss, making it impossible for users to fully understand the true situation of the original data; although data visualization can help users better understand the data, sometimes charts and images may cover up the details of the data, making it impossible for users to gain a deep understanding of the true situation of the data.

[0004] Therefore, there is an urgent need for a method and system for concretizing abstract autonomous driving experimental data, which aims to solve the above problems and help users better understand and analyze experimental results by providing more accurate, comprehensive and easy-to-understand concretized experimental data methods and systems, thereby improving the efficiency and quality of the experiment. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for concretizing abstract autonomous driving experimental data to solve the above-mentioned deficiencies in the prior art.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a system for concretizing abstract autonomous driving experimental data, comprising a data input module, a data processing module, a data visualization module, a user module and a data storage module;

[0007] The data input module receives raw data output by various experimental devices capable of outputting experimental data, and transmits the received raw data to the data processing module;

[0008] The data processing module is used to process the received raw data, generate a data summary by extracting key information of the raw data, and transmit the generated data summary to the data visualization module;

[0009] The data visualization module is used to convert the processed data summary into a visual form;

[0010] The user module is used to provide user human-computer interactive operations for the data input module, data processing module and data visualization module, allowing the user to perform input settings, operation settings and result viewing for the data input module, data processing module and data visualization module respectively;

[0011] The data storage module is used to store the original data, the intermediate data in the processing process and the final generated data summary for backup.

[0012] Furthermore, the data input module includes an online input submodule and an offline input submodule, the input ends of the online input submodule and the offline input submodule are respectively connected to the output end of the user module, and the output ends of the online input submodule and the offline input submodule are respectively connected to the input end of the data storage module.

[0013] Furthermore, the data processing module includes a data cleaning submodule, a data integration submodule and a data abstraction submodule, the output end of the data integration submodule is connected to the input end of the data abstraction submodule, the output end of the data abstraction submodule is connected to the input end of the data visualization module, the output ends of the data integration submodule and the data abstraction submodule are respectively connected to the input end of the data storage module, and the input ends of the data integration submodule and the data abstraction submodule are respectively connected to the output end of the user module.

[0014] Further, the data cleaning submodule includes a noise data removal unit, an error data correction unit and a missing data filling unit, the input end of the noise data removal unit is respectively connected to the output ends of the online input submodule and the offline input submodule, the output end of the noise data removal unit is connected to the input end of the error data correction unit, the output end of the error data correction unit is connected to the input end of the missing data filling unit, the output end of the missing data filling unit is connected to the input end of the data integration submodule, the input ends of the noise data removal unit, the error data correction unit and the missing data filling unit are respectively connected to the output end of the user module, and the output ends of the noise data removal unit, the error data correction unit and the missing data filling unit are respectively connected to the input end of the data storage module.

[0015] Furthermore, the data visualization module can extract the feature vector of the data summary and adopt different visualization methods for different types of data summaries through the visualization mapping model in the data visualization module. The visualization mapping model is specifically as follows:

[0016] Let F be the feature space, where each F_i is a feature vector, and V is the visualization space;

[0017] Define a mapping function M, which accepts the feature space F and the visualization space V as input and outputs a visualization configuration C, which is expressed as:

[0018] M(F,V)→C;

[0019] C contains the chart type, axis ranges, color map, and label parameters.

[0020] A method for concretizing abstract autonomous driving experimental data, which is applicable to the system for concretizing abstract autonomous driving experimental data, specifically comprises the following steps:

[0021] First, the user module controls the data input module to receive the original data output by various experimental devices capable of outputting experimental data. The original data includes digital signals and text information, and the original data is stored in the data storage module at the same time;

[0022] Secondly, the user module controls the data processing module to clean, integrate and abstract the received raw data, extract key information and generate data summaries. The intermediate data in the processing process and the final generated data summary are stored in the data storage module;

[0023] Finally, the user module controls the data visualization module to convert the processed data summary into a visual form so that users can understand the data intuitively.

[0024] Furthermore, the data input module receives the raw data output by various experimental devices capable of outputting experimental data in an online or offline input manner.

[0025] Furthermore, the cleaning of the received raw data by the data processing module includes removing noise data, correcting erroneous data and filling missing data, processing Gaussian noise through the adaptive Kalman filter algorithm model set in the noise data removal unit, and optimizing the filtering performance by dynamically adjusting the noise covariance, adaptability to nonlinear systems and the ability to overcome filtering divergence;

[0026] The data processing module integrates the received raw data to integrate the cleaned data according to a unified standard or format for subsequent processing and analysis;

[0027] The data processing module abstracts the received raw data to extract key information from the integrated data and generate a data summary.

[0028] Furthermore, the adaptive Kalman filter algorithm model is specifically as follows:

[0029] Prediction Step

[0030] Status prediction:

[0031] Based on the state estimate x_{t-1|t-1} at the previous moment and the state transfer matrix F_t, predict the state x_{t|t-1} at the current moment;

[0032] x t∣t-1 =F t x t-1∣t-1 +B t u t ;

[0033] Among them, B_t is the control matrix, u_t is the control vector;

[0034] Covariance prediction:

[0035] Predict the estimated error covariance P_{t|t-1} at the current moment;

[0036] P t∣t-1 =F t P t-1∣t-1 F t T +Q t ;

[0037] Among them, P_{t-1|t-1} is the estimation error covariance of the previous moment, Q_t is the covariance matrix of process noise;

[0038] Update steps

[0039] Calculate the Kalman gain:

[0040] Kalman gain K_t is used to balance the weight between prediction and observation;

[0041] K t =P t∣t-1 H t T (H t P t∣t-1 H t T +R t ) -1 ;

[0042] Among them, H_t is the observation matrix, R_t is the covariance matrix of the observation noise;

[0043] Status Update:

[0044] Combine the predicted value and the observed value to obtain the current state estimate x_{t|t};

[0045] x t∣t =x t∣t-1 +K t (zt -H t x t∣t-1 );

[0046] Among them, z_t is the observation vector;

[0047] Covariance update:

[0048] Update the estimated error covariance P_{t|t} at the current moment;

[0049] P t∣t =(1-K t H t ) t∣t-1 ;

[0050] In adaptive Kalman filtering, adjustments can be made according to specific adaptive strategies.

[0051] Compared with the prior art, the present invention provides a method and system for concretizing abstract autonomous driving experimental data:

[0052] 1. By using the data cleaning, integration and abstraction functions in the data processing module, it is possible to remove noise data, correct erroneous data, and fill in missing data, thereby improving the quality and accuracy of the data;

[0053] 2. The data integration module can integrate data from different experimental equipment according to a unified standard or format, reducing the time and error rate of manual processing and improving the efficiency of data processing;

[0054] 3. The data visualization module converts the processed data into a visual form, so that users can understand the meaning and association of the data more intuitively;

[0055] 4. Through the interactive operation of the user module, users can set parameters and adjust the visualization method as needed to better meet personalized needs and improve the understandability of data visualization and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0057] Figure 1 A schematic diagram of the overall structure of the system provided by an embodiment of the present invention;

[0058] Figure 2 A first schematic diagram of a partial structure of a system provided by an embodiment of the present invention;

[0059] Figure 3 A second schematic diagram of a partial structure of a system provided by an embodiment of the present invention;

[0060] Figure 4 A third schematic diagram of a partial structure of a system provided by an embodiment of the present invention;

[0061] Figure 5 A fourth schematic diagram of a partial structure of a system provided by an embodiment of the present invention;

[0062] Figure 6 A flow chart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0064] Embodiment 1:

[0065] See also Figure 1-Figure 5 ,A system for concretizing abstract autonomous driving experimental data, including a data input module, a data processing module, a data visualization module, a user module and a data storage module;

[0066] The data input module receives the raw data output by various experimental devices capable of outputting experimental data, and transmits the received raw data to the data processing module;

[0067] The data processing module is used to process the received raw data, generate a data summary by extracting key information of the raw data, and transmit the generated data summary to the data visualization module;

[0068] The data visualization module is used to convert the processed data summary into a visual form;

[0069] The user module is used to provide user human-computer interactive operations for the data input module, data processing module and data visualization module, allowing the user to perform input settings, operation settings and result viewing for the data input module, data processing module and data visualization module respectively;

[0070] The data storage module is used to store the original data, the intermediate data in the processing process and the final generated data summary for backup.

[0071] The data input module includes an online input submodule and an offline input submodule, the input ends of the online input submodule and the offline input submodule are respectively connected to the output end of the user module, and the output ends of the online input submodule and the offline input submodule are respectively connected to the input end of the data storage module.

[0072] The data processing module includes a data cleaning submodule, a data integration submodule and a data abstraction submodule. The output end of the data integration submodule is connected to the input end of the data abstraction submodule, the output end of the data abstraction submodule is connected to the input end of the data visualization module, the output ends of the data integration submodule and the data abstraction submodule are respectively connected to the input end of the data storage module, and the input ends of the data integration submodule and the data abstraction submodule are respectively connected to the output end of the user module.

[0073] The data cleaning submodule includes a noise data removal unit, an error data correction unit and a missing data filling unit. The input end of the noise data removal unit is connected to the output ends of the online input submodule and the offline input submodule respectively, the output end of the noise data removal unit is connected to the input end of the error data correction unit, the output end of the error data correction unit is connected to the input end of the missing data filling unit, the output end of the missing data filling unit is connected to the input end of the data integration submodule, the input ends of the noise data removal unit, the error data correction unit and the missing data filling unit are connected to the output end of the user module respectively, and the output ends of the noise data removal unit, the error data correction unit and the missing data filling unit are connected to the input end of the data storage module respectively.

[0074] The data visualization module can extract the feature vector of the data summary and use different visualization methods for different types of data summaries through the visualization mapping model in the data visualization module. The visualization mapping model is as follows:

[0075] Let F be the feature space, where each F_i is a feature vector, and V is the visualization space;

[0076] Define a mapping function M, which accepts the feature space F and the visualization space V as input and outputs a visualization configuration C, which is expressed as:

[0077] M(F,V)→C;

[0078] C contains the chart type, axis ranges, color map, and label parameters.

[0079] See also Figure 6 A method for concretizing abstract autonomous driving experimental data is applicable to a system for concretizing abstract autonomous driving experimental data, and specifically comprises the following steps:

[0080] First, the user module controls the data input module to receive the original data output by various experimental devices capable of outputting experimental data. The original data includes digital signals and text information, and the original data is stored in the data storage module at the same time;

[0081] Secondly, the user module controls the data processing module to clean, integrate and abstract the received raw data, extract key information and generate data summaries. The intermediate data in the processing process and the final generated data summary are stored in the data storage module;

[0082] Finally, the user module controls the data visualization module to convert the processed data summary into a visual form so that users can understand the data intuitively.

[0083] The data input module receives the raw data output by various experimental devices that can output experimental data, which can be input online or offline.

[0084] The data processing module cleans the received raw data by removing noise data, correcting erroneous data and filling missing data. It processes Gaussian noise through the adaptive Kalman filter algorithm model set in the noise data removal unit, and optimizes the filtering performance by dynamically adjusting the noise covariance, adaptability to nonlinear systems and the ability to overcome filtering divergence.

[0085] The data processing module integrates the received raw data to integrate the cleaned data according to a unified standard or format for subsequent processing and analysis;

[0086] The data processing module abstracts the received raw data to extract key information from the integrated data and generate a data summary.

[0087] The adaptive Kalman filter algorithm model is as follows:

[0088] Prediction Step

[0089] Status prediction:

[0090] Based on the state estimate x_{t-1|t-1} at the previous moment and the state transfer matrix F_t, predict the state x_{t|t-1} at the current moment;

[0091] x t∣t-1 =F t x t-1∣t-1 +B t u t ;

[0092] Among them, B_t is the control matrix, u_t is the control vector;

[0093] Covariance prediction:

[0094] Predict the estimated error covariance P_{t|t-1} at the current moment;

[0095] P t∣t-1 =F t P t-1∣t-1F t T +Q t ;

[0096] Among them, P_{t-1|t-1} is the estimation error covariance of the previous moment, Q_t is the covariance matrix of process noise;

[0097] Update steps

[0098] Calculate the Kalman gain:

[0099] Kalman gain K_t is used to balance the weight between prediction and observation;

[0100] K t =P t∣t-1 H t T (H t P t∣t-1 H t T +R t ) -1 ;

[0101] Among them, H_t is the observation matrix, R_t is the covariance matrix of the observation noise;

[0102] Status Update:

[0103] Combine the predicted value and the observed value to obtain the current state estimate x_{t|t};

[0104] x t∣t =x t∣t-1 +K t (z t -H t x t∣t-1 );

[0105] Among them, z_t is the observation vector;

[0106] Covariance update:

[0107] Update the estimated error covariance P_{t|t} at the current moment;

[0108] P t∣t =(1-K t H t ) t∣t-1 ;

[0109] In adaptive Kalman filtering, adjustments can be made according to specific adaptive strategies.

[0110] Specific implementation method: According to different experimental equipment, the user selects the online input submodule for online input through the user module or uses the data carrier and the offline input submodule for offline input. The input raw data is stored in the data storage module, and the raw data enters the noise data removal unit in the data cleaning submodule. The noise data removal unit filters out the noise data in the raw data, and the data after noise filtering is transmitted to the error data correction unit and also transmitted to the data storage module for storage. The error data correction unit processes the data and corrects the abnormal data. The data after noise removal and error correction is transmitted to the missing data filling unit and also transmitted to the data storage module for storage. The missing data filling unit fills the missing part in the data and transmits the data to the data storage module for storage. The data is stored in the data storage module. The data processed by the data cleaning submodule is transmitted to the data integration submodule. The data integration submodule performs unified processing on the data in terms of standard and format. The integrated data is transmitted to the data abstraction submodule to abstractly extract key information and generate a data summary. The integrated data and the abstracted data summary are both transmitted to the data storage module for storage. At the same time, the abstracted data is transmitted to the data visualization module for display. The data visualization module can extract the feature vector of the data summary and use different visualization methods for different types of data summaries through the visualization mapping model in the data visualization module. During the entire process of data processing, the user can issue operation instructions through the user module for control, and the visualization results are simultaneously transmitted to the data storage module for storage.

[0111] Embodiment 2:

[0112] This embodiment provides a technical solution based on the first embodiment: with the acceleration of urbanization, in the complex scenarios of urban environment experiments, especially in the scenarios where a large amount of complex experimental data needs to be processed and analyzed, the method and system for concretizing and abstracting autonomous driving experimental data involved in the above embodiments can play a significant role in experimental verification in urban environments.

[0113] The method and system are applied to the urban environment experimental verification; the composition and functions of the system are as follows:

[0114] Data input module: In the urban environment experimental verification, this module is responsible for receiving the raw data output from various urban environment monitoring equipment (such as air quality monitoring stations, traffic flow monitors, noise monitors, etc.). It supports both online and offline input methods to ensure the timeliness and integrity of the data.

[0115] Data processing module: 1) Data cleaning: remove noise data (such as outliers caused by equipment failure), correct erroneous data (such as erroneous records caused by transmission errors), fill missing data (such as temporary data loss caused by equipment failure) to ensure data accuracy and reliability. 2) Data integration: Integrate data from different monitoring devices according to a unified standard or format to facilitate subsequent analysis and comparison. 3) Data abstraction: Extract key information from the integrated data and generate data summaries to simplify the data understanding and analysis process.

[0116] Data visualization module: converts the processed data into visual forms, such as charts and images, to intuitively display the results and trends of urban environment experimental verification. It supports user-defined visualization methods, such as adjusting colors, lines, markers, etc., to meet personalized analysis needs.

[0117] User module: provides a human-computer interactive operation interface, allowing users to set input, operation and result for data input, processing and visualization. Supports users to set parameters according to experimental requirements, such as data cleaning rules, data integration standards, etc., to improve the flexibility and accuracy of experimental verification.

[0118] Data storage module: stores raw data, intermediate data during processing, and the final generated data summary for backup and subsequent analysis. Supports data export and sharing functions to facilitate data exchange and cooperation with other research institutions or government departments.

[0119] By deploying air quality monitoring stations, air quality data in various areas of the city are collected in real time.

[0120] The method and system of concretizing and abstracting autonomous driving experimental data are used to clean, integrate and visualize the data, revealing the temporal and spatial distribution patterns and changing trends of air quality, and providing scientific basis and decision-making support for urban environmental management and air quality improvement.

[0121] At the same time, specific optimization of urban environment experimental verification is achieved by combining the following aspects, as follows:

[0122] Traffic flow monitoring: By installing traffic flow monitors, the traffic flow data of various roads in the city can be recorded in real time. The data can be processed and analyzed to identify traffic congestion points and bottleneck sections, optimize traffic light control and road layout, and improve the efficiency of urban traffic operation and the travel experience of travelers.

[0123] Noise pollution monitoring: Use noise monitors to collect noise data from various areas of the city. Integrate and analyze the data to identify noise pollution sources and noise pollution areas. Provide a basis for the formulation of noise pollution prevention and control measures and plans. Improve data quality and accuracy: Through the data cleaning and correction process, remove noise data and erroneous data, fill in missing data, and ensure the reliability and accuracy of experimental verification results.

[0124] Improve data processing efficiency: Data integration and abstraction functions can reduce the time and error rate of manual processing and improve the efficiency and quality of data processing.

[0125] Enhanced data visualization: The data visualization module can intuitively display experimental verification results and trends, making it easier for users to understand and analyze data.

[0126] Support user customization requirements: The user module provides an interactive operation interface and customized parameter setting functions to meet the user's personalized experimental verification needs.

[0127] Embodiment three:

[0128] See also Figure 1-Figure 6 On the basis of the first embodiment, a technical solution is provided, in which a method for concretizing abstract autonomous driving experimental data and a system for concretizing abstract autonomous driving experimental data are combined to concretize the radar detection target information.

[0129] After the radar measures the target's distance, angle (azimuth and elevation), speed (radial speed and tangential speed), shape and size and other basic information, the user inputs the above information online into the data processing module through the online input submodule. Due to the influence of factors such as weather, distance, earth curvature and complex electromagnetic environment, there is a certain amount of noise in the original data. The noise data removal unit in the data cleaning submodule uses an adaptive Kalman filter algorithm model to remove the noise data. The data after the noise data is removed may have some abnormal points and burrs. At this time, the error data correction unit corrects the above abnormal data. The error-corrected data may be missing during the measurement, filtering and correction process. At this time, the missing data filling unit is used to fill in the gaps and fill in the gaps to form complete data transmission to the data integration submodule. The data integration submodule The different specifications of data measured by different radars are processed uniformly to facilitate the exchange, comparison and storage of data between different points. The integrated data is abstracted by the data abstraction submodule using statistical methods, data mining technology or machine learning algorithms to extract key information and generate data summaries. The original data, intermediate data in the processing process and the final generated data summary are all transmitted to the data storage module for storage. At the same time, the data summary is transmitted to the data visualization module for display. The data visualization module can extract the feature vector of the data summary and use different visualization methods for different types of data summaries through the visualization mapping model in the data visualization module. The data characteristics of the radar detection target information can be expressed by two-dimensional or three-dimensional images. The user can use the user module to issue operation instructions throughout the process.

[0130] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A system for concretizing abstract autonomous driving experimental data, characterized in that: It includes a data input module, a data processing module, a data visualization module, a user module and a data storage module; The data input module receives raw data output by various experimental devices capable of outputting experimental data, and transmits the received raw data to the data processing module; The data processing module is used to process the received raw data, generate a data summary by extracting key information of the raw data, and transmit the generated data summary to the data visualization module; The data visualization module is used to convert the processed data summary into a visual form; The user module is used to provide user human-computer interactive operations for the data input module, data processing module and data visualization module, allowing the user to perform input settings, operation settings and result viewing for the data input module, data processing module and data visualization module respectively; The data storage module is used to store the original data, the intermediate data in the processing process and the final generated data summary for backup.

2. The system for concretizing abstract autonomous driving experimental data according to claim 1, characterized in that: The data input module includes an online input submodule and an offline input submodule, the input ends of the online input submodule and the offline input submodule are respectively connected to the output end of the user module, and the output ends of the online input submodule and the offline input submodule are respectively connected to the input end of the data storage module.

3. The system for concretizing abstract autonomous driving experimental data according to claim 1, characterized in that: The data processing module includes a data cleaning submodule, a data integration submodule and a data abstraction submodule. The output end of the data integration submodule is connected to the input end of the data abstraction submodule, the output end of the data abstraction submodule is connected to the input end of the data visualization module, the output ends of the data integration submodule and the data abstraction submodule are respectively connected to the input end of the data storage module, and the input ends of the data integration submodule and the data abstraction submodule are respectively connected to the output end of the user module.

4. The system for concretizing abstract autonomous driving experimental data according to claim 3, characterized in that: The data cleaning submodule includes a noise data removal unit, an error data correction unit and a missing data filling unit. The input end of the noise data removal unit is connected to the output ends of the online input submodule and the offline input submodule respectively, the output end of the noise data removal unit is connected to the input end of the error data correction unit, the output end of the error data correction unit is connected to the input end of the missing data filling unit, the output end of the missing data filling unit is connected to the input end of the data integration submodule, the input ends of the noise data removal unit, the error data correction unit and the missing data filling unit are connected to the output end of the user module respectively, and the output ends of the noise data removal unit, the error data correction unit and the missing data filling unit are connected to the input end of the data storage module respectively.

5. The system for concretizing abstract autonomous driving experimental data according to claim 1, characterized in that: The data visualization module can extract the feature vector of the data summary and adopt different visualization methods for different types of data summaries through the visualization mapping model in the data visualization module. The visualization mapping model is as follows: Let F be the feature space, where each F_i is a feature vector, and V is the visualization space; Define a mapping function M, which accepts the feature space F and the visualization space V as input and outputs a visualization configuration C, which is expressed as: M(F,V)→C; C contains the chart type, axis ranges, color map, and label parameters.

6. A method for concretizing abstract autonomous driving experimental data, characterized in that: It is applicable to a system for concretizing abstract autonomous driving experimental data as described in any one of claims 1 to 5, and specifically comprises the following steps: First, the user module controls the data input module to receive the original data output by various experimental devices capable of outputting experimental data. The original data includes digital signals and text information, and the original data is stored in the data storage module at the same time; Secondly, the user module controls the data processing module to clean, integrate and abstract the received raw data, extract key information and generate data summaries. The intermediate data in the processing process and the final generated data summary are stored in the data storage module; Finally, the user module controls the data visualization module to convert the processed data summary into a visual form so that users can understand the data intuitively.

7. The method for concretizing abstract autonomous driving experimental data according to claim 6, characterized in that: The data input module receives the raw data output by various experimental devices capable of outputting experimental data, and can input the data online or offline.

8. The method for concretizing abstract autonomous driving experimental data according to claim 6, characterized in that: The data processing module cleans the received raw data by removing noise data, correcting erroneous data and filling missing data, processes Gaussian noise through the adaptive Kalman filter algorithm model set in the noise data removal unit, and optimizes the filtering performance by dynamically adjusting the noise covariance, adaptability to nonlinear systems and the ability to overcome filtering divergence; The data processing module integrates the received raw data to integrate the cleaned data according to a unified standard or format for subsequent processing and analysis; The data processing module abstracts the received raw data to extract key information from the integrated data and generate a data summary.

9. The method for concretizing abstract autonomous driving experimental data according to claim 8, characterized in that: The adaptive Kalman filter algorithm model is specifically as follows: Prediction Step Status prediction: Based on the state estimate x_{t-1|t-1} at the previous moment and the state transfer matrix F_t, predict the state x_{t|t-1} at the current moment; x t∣t-1 =F t x t-1∣t-1 +B t u t ; Among them, B_t is the control matrix, u_t is the control vector; Covariance prediction: Predict the estimated error covariance P_{t|t-1} at the current moment; P t∣t-1 =F t P t-1∣t-1 F t T +Q t ; Among them, P_{t-1|t-1} is the estimation error covariance of the previous moment, Q_t is the covariance matrix of process noise; Update steps Calculate the Kalman gain: Kalman gain K_t is used to balance the weight between prediction and observation; K t =P t∣t-1 H t T (H t P t∣t-1 H t T +R t ) -1 ; Among them, H_t is the observation matrix, R_t is the covariance matrix of the observation noise; Status Update: Combine the predicted value and the observed value to obtain the current state estimate x_{t|t}; x t∣t =x t∣t-1 +K t (z t -H t x t∣t-1 ); Among them, z_t is the observation vector; Covariance update: Update the estimated error covariance P_{t|t} at the current moment; P t∣t =(1-K t H t )P t∣t-1 ; In adaptive Kalman filtering, adjustments can be made according to specific adaptive strategies.

Images