Road health monitoring method and system based on data frequency self-matching

Abstract

The invention provides a road health monitoring method and system based on data frequency self-matching, and relates to the technical field of service health monitoring of pavement structures, and the method comprises the steps: obtaining vehicle body vibration information and vehicle position information of a vehicle; performing frequency self-matching on the output frequency of the vehicle body vibration information and the output frequency of the vehicle position information, and extracting vibration information and position information within the same time period after frequency self-matching; respectively inputting the vibration information and the position information in the same time period into a prediction model, and predicting road elevation data and roadbed disease data; constructing a road health state model, introducing a timestamp into the road health state model, and inputting the predicted road elevation data and roadbed disease data into the road health state model; and after a time label is given, comparing the time label with road elevation data and path disease data of an initial data set in the road health state model, identifying an abnormal point of elevation change, and obtaining a potential roadbed disease risk.

Description

Technical Field

[0001] The present disclosure relates to the technical field of service health monitoring of pavement structures, and particularly to a road health monitoring method and system based on data frequency self-matching. Background Art

[0002] The statements in this part merely provide background technical information related to the present disclosure and do not necessarily constitute prior art.

[0003] Subgrade diseases, such as settlement, cracks, deformation, and soil erosion, directly affect the structural stability and bearing capacity of roads, and may lead to potential vehicle driving safety hazards and traffic jams. With the changes in vehicle loads and environmental factors, these diseases will accelerate the damage of the road surface, reduce the service life of the road, increase the maintenance cost, and even threaten driving safety. Therefore, monitoring the road health status, timely identifying, and dealing with subgrade diseases are crucial for ensuring the normal service of roads.

[0004] Currently, the methods for road health monitoring mainly include traditional visual inspection, airborne detection (such as drones and laser scanning), non-destructive testing (such as ground-penetrating radar), etc. These solutions have their own advantages and disadvantages: traditional visual inspection relies on manual work, with strong subjectivity and low efficiency; sensor monitoring may have problems such as high installation and maintenance costs; airborne detection and laser scanning require high technical levels and equipment costs; non-destructive testing techniques sometimes have difficulty obtaining comprehensive data. Overall, various methods have certain deficiencies in terms of accuracy, cost, operation complexity, and environmental adaptability. The vehicle body vibration displacement sensor has important applications in vehicle health status detection, and the Beidou system also has an important position in vehicle navigation, while the combination of the vehicle body vibration displacement sensor and the Beidou system for road health monitoring has less application. Summary of the Invention

[0005] To solve the above problems, the present disclosure proposes a road health monitoring method and system based on data frequency self-matching, which combines the vibration data of the vehicle and the vehicle positioning trajectory data, and performs data frequency self-matching alignment. Through the elevation prediction model, it predicts and analyzes the high-frequency road sections and low-frequency road sections of the vehicle's travel, and based on this, outputs the road health status and potential disease types.

[0006] According to some embodiments, the present disclosure adopts the following technical solutions:

[0007] A road health monitoring method based on data frequency self-matching, comprising:

[0008] Obtaining the vehicle body vibration information and the vehicle position information of the vehicle;

[0009] Perform frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extract the vibration information and position information within the same time period after frequency self-matching;

[0010] Input the vibration information and position information within the same time period into the prediction model respectively, introduce a feature selection method for feature processing, and predict the road elevation data and subgrade disease data respectively;

[0011] Construct a road health status model, and connect the road information with the real-time map in the road health status model;

[0012] Introduce a timestamp into the road health status model. After inputting the predicted road elevation data and subgrade disease data into the road health status model, assign time tags and compare them with the road elevation data and path disease data in the initial data set in the road health status model to predict the abnormal points of elevation change and obtain potential subgrade disease risks.

[0013] According to some embodiments, the present disclosure adopts the following technical solutions:

[0014] A road health monitoring system based on data frequency self-matching, comprising:

[0015] A data acquisition module for acquiring the vehicle body vibration information and the vehicle position information of the vehicle;

[0016] A frequency self-matching module for performing frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extracting the vibration information and position information within the same time period after frequency self-matching;

[0017] A prediction module for inputting the vibration information and position information within the same time period into the prediction model respectively, introducing a feature selection method for feature processing, and predicting the road elevation data and subgrade disease data respectively;

[0018] A comparison and analysis module for constructing a road health status model, and connecting the road information with the real-time map in the road health status model; introducing a timestamp into the road health status model, inputting the predicted road elevation data and subgrade disease data into the road health status model, assigning time tags and comparing them with the road elevation data and path disease data in the initial data set in the road health status model to predict the abnormal points of elevation change and obtain potential subgrade disease risks.

[0019] According to some embodiments, the present disclosure adopts the following technical solutions:

[0020] A computer program product includes a computer program which, when executed by a processor, implements the road health monitoring method based on data frequency self-matching as described above.

[0021] According to some embodiments, the present disclosure adopts the following technical solutions:

[0022] A non-transitory computer-readable storage medium is used to store computer instructions which, when executed by a processor, implement the road health monitoring method based on data frequency self-matching as described above.

[0023] According to some embodiments, the present disclosure adopts the following technical solutions:

[0024] An electronic device includes: a processor, a memory, and a computer program; wherein, the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device runs, the processor executes the computer program stored in the memory so that the electronic device implements the road health monitoring method based on data frequency self-matching.

[0025] Compared with the prior art, the beneficial effects of the present disclosure are:

[0026] The road health monitoring method based on data frequency self-matching of the present disclosure predicts the elevation data of the road by collecting the vibration data of the driving vehicle and the position trajectory data based on Beidou, and then predicts potential subgrade diseases; by processing and collecting the data, it can predict the elevation and potential diseases of the road, and can output information such as the disease location, disease degree and the evolution trend of the road elevation. The road health monitoring method based on data frequency self-matching can effectively predict the elevation change of the road and potential subgrade diseases by collecting the vibration data of the driving vehicle and the position information based on Beidou. This method has high accuracy in data processing and analysis, can accurately output the location, degree of road diseases and the evolution trend of elevation, so as to provide a scientific basis for road maintenance and repair, optimize road management and maintenance decisions, and improve road safety and service life.

[0027] The road health monitoring method based on data frequency self-matching of the present disclosure utilizes the GOOSE-LSTM prediction algorithm in deep learning. By first training and learning the experimental data and then establishing a prediction model, it realizes predicting the road elevation by inputting the vibration data and position data of the vehicle, predicting and analyzing the high-frequency road end and low-frequency road end of the vehicle's driving by inputting the trajectory data of the vehicle, and based on the output, the road health degree and potential disease types.

[0028] The disclosed road health monitoring method based on data frequency self-matching uses the GOOSE-LSTM prediction algorithm in deep learning to bring significant benefits to road health monitoring. First, the model can not only accurately predict the elevation change of the road through training and learning of experimental data, but also deeply analyze the response of high-frequency road ends and low-frequency road ends during vehicle driving, so as to comprehensively evaluate the health of the road. This intelligent data analysis capability improves the accuracy and efficiency of monitoring and greatly reduces the cost and workload of manual detection. In terms of model structure, GOOSE-LSTM combines the time series prediction capability and adaptive weight adjustment mechanism of the long short-term memory network (LSTM), which can effectively mine complex patterns and long-term dependencies in the data. This enables the model to converge quickly and produce high-quality prediction results after inputting dynamic vehicle vibration data and position data. In addition, the output of this method can not only provide a quantitative assessment of road health, but also identify potential disease types and locations, provide important references for maintenance decisions of relevant departments, and ultimately realize intelligent and precise road management. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure.

[0030] Figure 1 is an overall flow chart of the method of the embodiment of the present disclosure;

[0031] Figure 2 A schematic diagram of data frequency self-matching and time data alignment according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

[0033] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0034] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0035] Example 1

[0036] In an embodiment of the present disclosure, a road health monitoring method based on data frequency self-matching is provided, including:

[0037] Step 1: Obtain the vehicle body vibration information and vehicle position information of the vehicle;

[0038] Step 2: Perform frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extract the vibration information and position information within the same time period after frequency self-matching;

[0039] Step 3: Input the vibration information and position information within the same time period into the prediction model respectively, introduce a feature selection method for feature processing, and predict the road elevation data and subgrade disease data;

[0040] Step 4: Construct a road health status model, and connect the road information with the real-time map in the road health status model;

[0041] Step 5: Introduce a timestamp into the road health status model. After inputting the predicted road elevation data and subgrade disease data into the road health status model, assign time tags to them and compare them with the road elevation data and path disease data in the initial data set in the road health status model to identify abnormal points of elevation change and obtain potential subgrade disease risks.

[0042] As an embodiment, the road health monitoring method based on data frequency self-matching of the present disclosure is specifically implemented as follows:

[0043] Step 1: Obtain the vehicle body vibration information and vehicle position information of the vehicle;

[0044] Specifically, install vehicle body vibration displacement sensors on the four tire bearings of the vehicle, and use the Beidou system to locate vehicle information, specifically:

[0045] S1: Install the vehicle body vibration displacement sensors on the four tire bearings of the vehicle so that it can collect the vehicle's vibration data (vibration amplitude, vibration magnitude) and steering (turning) data;

[0046] S2: Set the data output system of the Beidou system, and output the vehicle's position data (position, orientation) and trajectory data after locating the vehicle.

[0047] Step 2: Perform frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extract the vibration information and position information within the same time period after frequency self-matching;

[0048] Specifically, design the frequencies of the vehicle body vibration displacement sensor and the Beidou system data output. The data output frequency of the vehicle body vibration displacement sensor is designed to be 5 Hz, and the data output frequency of the Beidou system is 1 Hz, so that they can complete data coincidence and achieve data alignment through frequency self-matching.

[0049] Among them, as Figure 2 shown, the specific process of frequency self-matching is as follows:

[0050]

[0051] Among them, n m-1 represents the normalized frequency data, m represents the currently input candidate state, T1 represents the data collected by the vehicle body vibration displacement sensor, and T2 represents the data collected by the Beidou system.

[0052] After frequency self-matching, the data at the same time point is output, and the data is output according to the frequency of the data group with the lower acquisition frequency. As Figure 2 shown, the acquisition frequencies are 1 Hz and 5 Hz respectively, that is, the data volume of the Beidou system is equivalent to five times the data volume of the vehicle body vibration displacement sensor, which is convenient for the next step of data processing.

[0053] Step 3: Input the vibration information and position information within the same time period into the prediction model respectively, introduce a feature selection method for feature processing, and predict the road elevation data and subgrade disease data;

[0054] Specifically, first, the prediction model adopts a GOOSE-LSTM neural network structure, and the construction method of the prediction model with the GOOSE-LSTM neural network structure is as follows:

[0055] S1: Preprocess the GOOSE message data, extract key information such as timestamps, state changes, event types, etc., and organize them into a time series format. The goal of this processing is to make the data adaptable to the subsequent LSTM model and be able to capture time-dependent relationships and temporal features;

[0056] S2: Design the network structure of the LSTM, determine the number of nodes in the input layer, which usually corresponds to the number of features of the input time series. Design the LSTM hidden layer, which may include one or more LSTM layers, and each layer can set different numbers of units (nodes) to capture different levels of time features;

[0057] In the network structure of the LSTM, first, the input layer contains 10 nodes, corresponding to 10 features respectively (such as the vibration data and position information of the vehicle), which are used to receive time series data. Next, the first hidden layer is set to 64 LSTM units, mainly extracting the short-term time features of the data and capturing instantaneous changes. The second hidden layer is set to 32 LSTM units, further processing the output of the first layer and focusing on more complex time relationships and long-term dependence features. Then there is a fully connected layer with 3 nodes, which is used to map the extracted features to the final output. Finally, the output layer is also set to 3 nodes, outputting the predicted road elevation, road health, and potential disease types.

[0058] S3: Select a suitable training set for model training to ensure that the training set data is representative. Then, set the loss function. Mean Squared Error (MSE) is usually a common choice, which is used to evaluate the difference between the predicted value and the actual value of the model under specific inputs. After training is completed, use the validation set to evaluate the model effect and adjust the parameters;

[0059] S4: Analyze the number of nodes and layers of the selected input layer, hidden layer, and output layer, and test different network structures through cross-validation or grid search to find the best combination of nodes and layers;

[0060] S5: Determine the time window of the input data (vibration data, steering data, trajectory data, and position data), and analyze how these inputs affect the output. Determine the output data (road elevation, potential disease types, etc.). Introduce a feature selection method. Through the Lasso regression method, after model training, the algorithm will generate feature importance scores. By analyzing these scores, select the features that contribute the most to the prediction results, remove redundant and unnecessary features, and improve the performance and efficiency of the model.

[0061] Furthermore, input the vibration information and position information into the prediction model respectively, introduce a feature selection method for feature processing, and predict the road elevation data and subgrade disease data;

[0062] Among them, the road elevation data is predicted through the vibration data at the vibration end. The elevation change trend of the subsequent elevation is predicted based on the historical data elevation. Based on the vehicle flow trajectory data, such as speed and turning, when the vehicle encounters a diseased road end, such as a pothole, it will detour, and when it encounters a high elevation end, it will decelerate. Then, based on these features, the disease location is predicted.

[0063] Step 4: Build a road health status model, connect the road information with the real-time map in the road health status model, and introduce timestamps in the road health status model. Specifically, it includes:

[0064] S1: Input relevant map road information in the interface of the control platform of the road health status model, including the location, name, category, grade, start and end points, important traffic nodes, etc. of the road. Combine with the high-precision geographical location information provided by the Beidou system to establish a connection between the input road information and the real-time map data of the Beidou system;

[0065] S2: Establish a timestamp system, which will assign time tags to all input data for data synchronization and collation. When road data (vehicle position, vibration data, road elevation and trajectory data) is input into the system, each record should be appended with an accurate timestamp. These timestamps help us align the data from different data sources in time to ensure that all data information can be compared within the same time frame;

[0066] S3: Input certain basic parameters when collecting road-related data. These parameters include the width of the road, engineering grade (such as first-class highway, second-class highway, etc.), original elevation (initial height of the road surface), etc. Combine these parameters with the timestamp data to form an initial dataset for describing the basic characteristics of the road in a healthy service state. Collect the current road condition data through sensors or inspection records to establish a preliminary road health status model;

[0067] S4: Identify the trend of road elevation change by analyzing and comparing the collected data. Adopt deep learning methods, combine parameters such as road width and engineering grade, and identify the occurrence probability and development trend of different types of road diseases (such as cracks, potholes, etc.) through existing deep learning-based prediction models. Improve the prediction accuracy by comparing different categories of data groups, and thus provide a scientific basis for road maintenance and management.

[0068] Step 5: After inputting the predicted road elevation data and subgrade disease data into the road health status model, assign time tags and then compare them with the road elevation data and path disease data in the initial dataset in the road health status model to identify abnormal points of elevation change and obtain potential subgrade disease risks.

[0069] S1: Obtain the vibration information generated by the vehicle body vibration displacement sensor. This vibration displacement sensor predicts the elevation data related to the road surface by monitoring the vibration generated during vehicle driving. By performing spatial interpolation or model fitting on these elevation information, the overall elevation state of the highway can be inferred. By using geographic information system tools, the inferred elevation data can be compared with the existing geographic data to form a comprehensive road surface elevation map;

[0070] Specifically, using the elevation data of known geographical locations, the elevation distribution of the entire road or area is estimated and inferred through a raster model. This process can generate a continuous elevation model, reflecting the overall elevation state of the road.

[0071] S2: Set a time comparison domain to analyze the relationship between current elevation data and historical elevation data. By comparing, the abnormal points of elevation change are predicted, and then the potential risks of subgrade diseases, such as subgrade settlement, uplift or deformation, etc., are speculated;

[0072] Specifically, to set a time comparison domain to analyze the relationship between current elevation data and historical elevation data, first, the time frame for comparison needs to be defined. For example, set the elevation data of the most recent year to be compared with that of the past five years. Next, collect the elevation data during this period and perform preprocessing to ensure the accuracy and consistency of the data. Data analysis can calculate the difference between the current elevation and the historical elevation, and use the Z-score to identify significant changes and mark the abnormal points. Then, for these abnormal points, combined with time series analysis methods (moving average, trend analysis), explore their change trends, so as to speculate on potential subgrade disease risks, such as settlement, uplift or deformation, etc. Finally, establish a visualization tool to intuitively display the elevation change and the disease risk area.

[0073] S3: Analyze traffic flow data. By collecting the track data of driving vehicles through sensors and the Beidou system, the lanes with high and low driving frequencies can be identified. Utilize the fact that vehicles may have avoidance behaviors in disease sections, such as reducing speed or changing the driving trajectory, resulting in a significant decrease in the driving frequency on certain lanes. By analyzing this data, the specific locations where diseases may exist are speculated. In addition, combined with historical inspection records and accident occurrence data, evaluate the severity of diseases at these locations, and combine the potential disease locations with other available data (such as elevation change, vibration characteristics) to form a more comprehensive analysis;

[0074] S4: Integrate and output the road health assessment data, risk analysis data, and potential disease location data obtained from the previous analysis. This output can take various forms, such as generating a report, creating a visualization dashboard, or directly updating it into the management system. The output data should include information such as health score, disease type, and risk level, and mark the sections that need to be maintained preferentially.

[0075] Specifically, the methods for predicting and statistically analyzing the elevation and potential diseases of the road through time comparison, elevation comparison, and trajectory data comparison include the following steps:

[0076] S1: Speculate the overall elevation information of the road based on the elevation information predicted by the vehicle body vibration displacement device;

[0077] S2: Set up a time comparison time domain to obtain the current potential subgrade disease risks by comparing the previous elevation data;

[0078] S3: Identify the high-frequency and low-frequency driving lanes by reading the track data, and infer the disease occurrence location based on the disease avoidance characteristics of the vehicle owners;

[0079] S4: Output the road health evaluation data, risk data, and location data to facilitate future maintenance by the staff.

[0080] This disclosure is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the disclosure. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce a means for realizing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, such that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process, so that the instructions executed on the computer or other programmable devices provide steps for realizing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0082] Although the specific implementation manners of the present disclosure have been described above in conjunction with the accompanying drawings, it is not a limitation to the protection scope of the present disclosure. Those skilled in the art should understand that based on the technical solutions of the present disclosure, various modifications or deformations that can be made by those skilled in the art without creative efforts are still within the protection scope of the present disclosure.

Claims

1. A road health monitoring method based on self-matching of data frequencies, characterized in that, Including: Obtain the vehicle body vibration information and vehicle position information of the vehicle; Perform frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extract the vibration information and trajectory position information within the same time period after frequency self-matching; Respectively input the vibration information and position information within the same time period into the prediction model, introduce a feature selection method for feature processing, and respectively predict the road elevation data and subgrade disease data; Construct a road health status model, and connect the road information with the real-time map in the road health status model; Introduce a time stamp in the road health status model. After inputting the predicted road elevation data and subgrade disease data into the road health status model, assign a time label to them and compare them with the road elevation data and path disease data in the initial data set in the road health status model to predict the abnormal points of elevation change and obtain potential subgrade disease risks.

2. The road health monitoring method based on data frequency self-matching according to claim 1, wherein The prediction model adopts a GOOSE-LSTM neural network structure, converts the input vibration information and position information into a time series format, extracts time series features, designs the node structures of the input layer, LSTM hidden layer and output layer, trains the LSTM model using the training set, and adjusts the parameters of the model by minimizing the loss function; determine the number of layers of the structural nodes of the final GOOSE-LSTM network, and determine the parameters of the number of input layer, the number of LSTM hidden layers and the number of nodes in the output layer.

3. The road health monitoring method based on data frequency self-matching according to claim 1, characterized in that, Obtain the vehicle body vibration information and vehicle trajectory position information of the vehicle, including installing vehicle body vibration displacement sensors on the four tire bearings of the vehicle to obtain the vehicle body vibration information, and using the Beidou system to locate the vehicle position information. The vehicle body vibration information includes vibration amplitude, vibration magnitude and turning data, and the position information includes position data, azimuth information and trajectory data.

4. The road health monitoring method based on data frequency self-matching according to claim 1, wherein Perform frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extract the vibration information and position information within the same time period after frequency self-matching. Specifically: Design the output frequencies of the vehicle body vibration displacement sensor and the Beidou system. Design the data output of the vehicle body vibration displacement sensor to be 5Hz, and the data output frequency of the Beidou system to be 1Hz, so that they can complete data coincidence and achieve data frequency alignment; output the data within the same time period. The amount of data of the Beidou system is equivalent to five times the amount of data of the vehicle body vibration displacement sensor.

5. The road health monitoring method based on data frequency self-matching according to claim 1, characterized in that Construct a road health status model, and connect the road information with the real-time map in the road health status model, including: input relevant map road information, including the location, name, category, grade, starting and ending points and important traffic nodes of the road, combine the geographical location information provided by the Beidou system, establish a connection between the input road information and the real-time map data, establish a time stamp to assign a time label to all input data, and attach a time stamp to each record when the vehicle position, vibration information, road elevation and trajectory data are input into the road health status model to ensure that all data information is compared within the same time frame.

6. The road health monitoring method based on data frequency self-matching according to claim 5, characterized in that Input basic parameters when constructing the road health status model. The basic parameters include the width of the road, the engineering grade, and the original elevation. Combine these parameters with the timestamp data to form an initial dataset, which describes the basic characteristics of the road in a healthy service state.

7. A road health monitoring system based on data frequency self-matching, characterized in that, including: A data acquisition module for acquiring the vehicle body vibration information and vehicle position information of the vehicle; A frequency self-matching module for performing frequency self-matching on the output frequencies of the vehicle body vibration information and the vehicle position information, and extracting the vibration information and position information within the same time period after frequency self-matching; A prediction module for respectively inputting the vibration information and position information within the same time period into the prediction model, introducing a feature selection method for feature processing, and predicting the road elevation data and subgrade disease data; A comparison and analysis module for constructing a road health status model and connecting the road information with the real-time map in the road health status model; introducing a timestamp in the road health status model, after inputting the predicted road elevation data and subgrade disease data into the road health status model, giving them time tags and comparing them with the road elevation data and path disease data in the initial dataset in the road health status model, predicting the abnormal points of the elevation change, and obtaining potential subgrade disease risks.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the road health monitoring method based on data frequency self-matching according to any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by the processor, it implements the road health monitoring method based on data frequency self-matching according to any one of claims 1-6.

10. An electronic device, characterized in that, including: A processor, a memory, and a computer program; wherein, the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device runs, the processor executes the computer program stored in the memory, so that the electronic device executes and implements the road health monitoring method based on data frequency self-matching according to any one of claims 1-6.

Images