Method and device for acquiring bad road distribution, vehicle, medium and program product

Abstract

The invention relates to the technical field of traffic transportation, in particular to a bad road distribution obtaining method and device, a vehicle, a medium and a program product, and the method comprises the steps: obtaining the actual road driving data of a road where the vehicle is located in a target area; based on the actual road driving data, half-shaft vibration time-frequency characteristics corresponding to the road are extracted; and inputting the half-shaft vibration time-frequency characteristics into a pre-constructed good and bad road identification model to output a bad road identification result of the road, and obtaining bad road position distribution of the target area based on the bad road identification result. Therefore, the problems that time and labor are wasted, all roads are difficult to fully cover and some potential hidden dangers are easy to miss in the prior art due to monitoring of a smart phone are solved. In addition, the problem that the reliability of a detection result is reduced due to the fact that data are inconsistent when the model is constructed is solved.

Description

Technical Field

[0001] The present application relates to the technical field of transportation, and in particular to a method, device, vehicle, medium and program product for obtaining bad road distribution. Background Art

[0002] As the frequency of use and load of urban roads continue to increase, road damage and deformation problems are becoming increasingly serious. Bad urban roads not only affect traffic flow, but also have a significant impact on driving safety, passenger comfort and vehicle service life. Timely and accurate monitoring and evaluation of urban road conditions has become an important task in urban management and traffic planning.

[0003] In related technologies, the smartphone’s built-in sensors can be used to monitor abnormal driving conditions during vehicle driving, and the data and location information can be sent to the background server. Through location information association and data feature analysis, automatic detection and analysis of road conditions can be completed. The degree of fusion of road damage data can also be calculated based on the acquired vertical road displacement information and vehicle vibration information, thereby constructing a road damage information model.

[0004] However, in the related technologies, smartphone monitoring is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, the data inconsistency will occur when building the model, resulting in low reliability of the detection results, which urgently needs to be improved. Summary of the invention

[0005] The present application provides a method, device, vehicle, medium and program product for obtaining bad road distribution, so as to solve the problems in the related technology that smart phone monitoring is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, the data inconsistency will occur when building the model, which reduces the reliability of the detection results.

[0006] The first aspect of the present application provides a method for obtaining bad road distribution, comprising the following steps: obtaining actual road driving data of the road where the vehicle is located in the target area; based on the actual road driving data, extracting the semi-axle vibration time-frequency characteristics corresponding to the road; inputting the semi-axle vibration time-frequency characteristics into a pre-constructed good and bad road identification model to output a bad road identification result of the road, and obtaining the bad road position distribution in the target area based on the bad road identification result.

[0007] Through the above technical scheme, the corresponding half-axle vibration time-frequency characteristics can be extracted according to the actual road driving data of the vehicle on the road in the target area, and the half-axle vibration time-frequency characteristics can be input into the pre-built good and bad road identification model to output the bad road identification result of the road, and then the bad road position distribution in the target area can be obtained. The acquisition of the bad road position distribution can be completed in a relatively short time, and the half-axle vibration time-frequency characteristics are used as input without human intervention, which greatly reduces labor costs and improves work efficiency. At the same time, it also reduces the errors caused by human factors, and can also provide timely data support for road maintenance and repair, realize the monitoring of all roads in the target area, and more comprehensively understand and grasp the road conditions in the target area.

[0008] Optionally, in one embodiment of the present application, before inputting the half-axle vibration time-frequency characteristics into a pre-constructed good and bad road identification model, it also includes: obtaining the bad road half-axle vibration time-frequency characteristics corresponding to the road in the bad road scenario; obtaining the non-bad road half-axle vibration time-frequency characteristics corresponding to the road in the non-bad road scenario; and constructing the good and bad road identification model based on the bad road half-axle vibration time-frequency characteristics and the non-bad road half-axle vibration time-frequency characteristics.

[0009] Through the above technical scheme, a good and bad road identification model can be constructed according to the time-frequency characteristics of the bad road semi-axle vibration in the bad road scenario and the time-frequency characteristics of the non-bad road semi-axle vibration in the non-bad road scenario. Through the time-frequency characteristics of the semi-axle vibration in the bad road scenario and the non-bad road scenario, it can be ensured that the model training data has sufficient diversity, thereby improving the recognition accuracy of the model and enhancing the generalization ability of the model.

[0010] Optionally, in one embodiment of the present application, extracting the time-frequency characteristics of the semi-axle vibration corresponding to the road based on the actual road driving data includes: obtaining micro-segment driving data of at least one time segment corresponding to the actual road driving data based on the actual road driving data; calculating the vehicle speed per acceleration of each micro-segment driving data by a calculation formula of the vehicle speed per acceleration, and obtaining at least one vehicle speed per acceleration interval for each micro-segment driving data according to the vehicle speed per acceleration; generating a semi-axle vibration data set of the actual road driving data by combining the micro-segment driving data and the at least one vehicle speed per acceleration interval; and obtaining the semi-axle vibration time-frequency characteristics according to the semi-axle vibration data set.

[0011] Through the above technical solution, the actual road driving data can be divided into micro-segment driving data of multiple time segments, and the vehicle driving status in each time segment can be analyzed more carefully, which helps to capture the impact of more subtle road changes on the vibration of the vehicle half-axle, and improve the precision and accuracy of feature extraction. By calculating the vehicle speed per acceleration of each micro-segment driving data, and obtaining the vehicle speed per acceleration interval based on the vehicle speed per acceleration, the vehicle driving status can be more closely linked to the road conditions. Combining the micro-segment driving data and the vehicle speed per acceleration interval to generate a half-axle vibration data set can ensure that the data set contains rich feature information closely related to the road. By extracting the time-frequency characteristics of the half-axle vibration, a more detailed assessment of the road quality can be achieved, and potential road problems such as small cracks and unevenness can be discovered, thereby providing a more accurate basis for road maintenance.

[0012] Optionally, in one embodiment of the present application, the calculation formula of the vehicle speed per acceleration may be, but is not limited to,:

[0013]

[0014] Where i represents the i-th second; v i represents the vehicle speed at the ith second; (v i+1 -v i ) / 3.24 represents the acceleration at the i-th second.

[0015] Through the above technical scheme, the vehicle's speed per acceleration can be calculated according to the calculation formula of the vehicle speed per acceleration, which can more accurately describe the changes in the speed and acceleration of the vehicle during driving, and then more accurately identify the damaged area on the road. By providing the calculation formula of the vehicle speed per acceleration, it can be applied to different types of vehicles and different road conditions, and has wider applicability and versatility in practical applications.

[0016] The second aspect of the present application provides a device for acquiring bad road distribution, including: a first acquisition module, used to acquire actual road driving data of the road where the vehicle is located in the target area; an extraction module, used to extract the semi-axle vibration time-frequency characteristics corresponding to the road based on the actual road driving data; a second acquisition module, used to input the semi-axle vibration time-frequency characteristics into a pre-constructed good and bad road identification model to output the bad road identification result of the road, and acquire the bad road position distribution in the target area based on the bad road identification result.

[0017] Through the above technical scheme, the corresponding half-axle vibration time-frequency characteristics can be extracted according to the actual road driving data of the vehicle on the road in the target area, and the half-axle vibration time-frequency characteristics can be input into the pre-built good and bad road identification model to output the bad road identification result of the road, and then the bad road position distribution in the target area can be obtained. The acquisition of the bad road position distribution can be completed in a relatively short time, and the half-axle vibration time-frequency characteristics are used as input without human intervention, which greatly reduces labor costs and improves work efficiency. At the same time, it also reduces the errors caused by human factors, and can also provide timely data support for road maintenance and repair, realize the monitoring of all roads in the target area, and more comprehensively understand and grasp the road conditions in the target area.

[0018] Optionally, in one embodiment of the present application, it also includes: a third acquisition module, used to obtain the bad road semi-axle vibration time-frequency characteristics corresponding to the road in the bad road scene before inputting the semi-axle vibration time-frequency characteristics into the pre-constructed good and bad road identification model; a fourth acquisition module, used to obtain the non-bad road semi-axle vibration time-frequency characteristics corresponding to the road in the non-bad road scene; a construction module, used to construct the good and bad road identification model based on the bad road semi-axle vibration time-frequency characteristics and the non-bad road semi-axle vibration time-frequency characteristics.

[0019] Through the above technical scheme, a good and bad road identification model can be constructed according to the time-frequency characteristics of the bad road semi-axle vibration in the bad road scenario and the time-frequency characteristics of the non-bad road semi-axle vibration in the non-bad road scenario. Through the time-frequency characteristics of the semi-axle vibration in the bad road scenario and the non-bad road scenario, it can be ensured that the model training data has sufficient diversity, thereby improving the recognition accuracy of the model and enhancing the generalization ability of the model.

[0020] Optionally, in one embodiment of the present application, the extraction module includes: an acquisition unit, used to acquire micro-segment driving data of at least one time segment corresponding to the actual road driving data based on the actual road driving data; a calculation unit, used to calculate the vehicle speed per acceleration of each micro-segment driving data by using the calculation formula of the vehicle speed per acceleration, and obtain at least one vehicle speed per acceleration interval of each micro-segment driving data according to the vehicle speed per acceleration; a first generation unit, used to combine the micro-segment driving data and the at least one vehicle speed per acceleration interval to generate a semi-axle vibration data set of the actual road driving data; and a second generation unit, used to obtain the semi-axle vibration time-frequency characteristics according to the semi-axle vibration data set.

[0021] Through the above technical solution, the actual road driving data can be divided into micro-segment driving data of multiple time segments, and the vehicle driving status in each time segment can be analyzed more carefully, which helps to capture the impact of more subtle road changes on the vibration of the vehicle half-axle, and improve the precision and accuracy of feature extraction. By calculating the vehicle speed per acceleration of each micro-segment driving data, and obtaining the vehicle speed per acceleration interval based on the vehicle speed per acceleration, the vehicle driving status can be more closely linked to the road conditions. Combining the micro-segment driving data and the vehicle speed per acceleration interval to generate a half-axle vibration data set can ensure that the data set contains rich feature information closely related to the road. By extracting the time-frequency characteristics of the half-axle vibration, a more detailed assessment of the road quality can be achieved, and potential road problems such as small cracks and unevenness can be discovered, thereby providing a more accurate basis for road maintenance.

[0022] Optionally, in one embodiment of the present application, the calculation formula of the vehicle speed per acceleration may be, but is not limited to,:

[0023]

[0024] Where i represents the i-th second; v i represents the vehicle speed at the ith second; (v i+1 -v i ) / 3.24 represents the acceleration at the i-th second.

[0025] Through the above technical scheme, the vehicle's speed per acceleration can be calculated according to the calculation formula of the vehicle speed per acceleration, which can more accurately describe the changes in the speed and acceleration of the vehicle during driving, and then more accurately identify the damaged area on the road. By providing the calculation formula of the vehicle speed per acceleration, it can be applied to different types of vehicles and different road conditions, and has wider applicability and versatility in practical applications.

[0026] The third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for obtaining bad road distribution as described in the above embodiment.

[0027] The fourth aspect of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above method for obtaining the bad road distribution.

[0028] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed, implements the method for obtaining the bad road distribution as described above.

[0029] The embodiment of the present application can form the corresponding semi-axle vibration time-frequency characteristics through the actual road driving data of the road where the vehicle is located in the target area, and input the semi-axle vibration time-frequency characteristics into the pre-constructed good and bad road identification model to obtain the bad road identification result of the road, and then generate the bad road position distribution in the target area. By associating the semi-axle vibration time-frequency characteristics with the road conditions, not only the detection efficiency is improved, but also the detection cost is reduced. By analyzing the semi-axle vibration time-frequency characteristics, the damage of the road can be accurately judged, providing an important basis for road maintenance, thereby improving the safety and comfort of the road, and realizing more intelligent and accurate road monitoring and management. Thus, it solves the problem that in the related technology, smart phone monitoring is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, the construction of the model will result in inconsistent data, which reduces the reliability of the detection results.

[0030] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0032] Figure 1 A flowchart of a method for obtaining bad road distribution according to an embodiment of the present application;

[0033] Figure 2 A flow chart of generating time-frequency characteristics of semi-axle vibration according to an embodiment of the present application;

[0034] Figure 3 A flowchart of constructing a bad road identification model according to an embodiment of the present application;

[0035] Figure 4 A schematic block diagram of a bad road scene with potholes, cracks and other road surface damage provided according to an embodiment of the present application;

[0036] Figure 5 A schematic diagram of the distribution of bad roads in a XX area according to an embodiment of the present application;

[0037] Figure 6 A schematic block diagram of a method for obtaining bad road distribution according to an embodiment of the present application;

[0038] Figure 7 A block diagram of a device for acquiring bad road distribution according to an embodiment of the present application;

[0039] Figure 8It is a schematic diagram of the structure of a vehicle provided according to an embodiment of the present application.

[0040] Reference numerals:

[0041] Among them, 10 is a device for acquiring bad road distribution; 100 is a first acquisition module, 200 is an extraction module, 300 is a second acquisition module; 801 is a memory, 802 is a processor, and 803 is a communication interface. DETAILED DESCRIPTION

[0042] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0043] The following describes the acquisition method, device, vehicle, medium and program product of the bad road distribution of the embodiment of the present application with reference to the accompanying drawings. The monitoring of the smart phone mentioned in the above background technology is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, the data inconsistency will occur when building the model, which reduces the reliability of the detection result. The present application provides a method for acquiring the bad road distribution of the road. In this method, the corresponding semi-axle vibration time-frequency characteristics can be formed by the actual road driving data of the road where the vehicle is located in the target area, and the semi-axle vibration time-frequency characteristics are input into the pre-built good and bad road identification model to obtain the bad road identification result of the road, and then generate the bad road position distribution of the target area. By associating the semi-axle vibration time-frequency characteristics with the road conditions, not only the detection efficiency is improved, but also the detection cost is reduced. By analyzing the semi-axle vibration time-frequency characteristics, the damage of the road can be accurately judged, providing an important basis for road maintenance, thereby improving the safety and comfort of the road, and realizing more intelligent and accurate road monitoring and management. Thus, it solves the problem that the monitoring of the smart phone in the related technology is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, building the model may result in data inconsistency, reducing the reliability of the test results, and other issues.

[0044] Specifically, Figure 1 The present invention is a flowchart of a method for obtaining bad road distribution according to an embodiment of the present application.

[0045] like Figure 1 As shown, the method for obtaining the bad road distribution of the road includes the following steps:

[0046] In step S101, actual road driving data of the road where the vehicle is located in the target area is obtained.

[0047] It is understandable that the actual road driving data may include road driving data including half-axle vibration collected by the vehicle over the years, or may include road driving data including half-axle vibration collected by the vehicle in real time. Among them, half-axle vibration refers to the vibration signal at the half-axle of the vehicle, which may be collected and obtained by a vibration acceleration sensor installed at the half-axle of the vehicle, or may be obtained by other methods, and this application does not impose any specific restrictions.

[0048] In addition, it should be noted that the actual road driving data of the embodiment of the present application may also include parameters such as time, vehicle speed, acceleration, longitude and latitude, and half-axle vibration, which can be specifically set by technicians in this field according to actual conditions, and this application does not impose any specific restrictions.

[0049] As a possible implementation method, the embodiments of the present application can obtain actual road driving data of vehicles on roads in the target area through GNSS (Global Navigation Satellite System), GPS (Global Positioning System), Internet of Things, target area traffic information network, camera sensors, radar sensors, etc.

[0050] For example, the embodiment of the present application obtains actual road driving data of about 300,000 kilometers in the past three years in the XX region through the XX region traffic information network.

[0051] In step S102, based on the actual road driving data, the time-frequency characteristics of the semi-axle vibration corresponding to the road are extracted.

[0052] It can be understood that the time-frequency characteristics of the semi-axle vibration in the embodiment of the present application refer to the performance characteristics of the semi-axle vibration in the time domain and frequency domain during the vehicle driving process. Among them, the time-frequency characteristics can include but are not limited to parameters such as variance, root mean square value, peak factor, kurtosis factor, center of gravity frequency and root mean square frequency, and the present application does not make specific restrictions.

[0053] As a possible implementation method, the embodiment of the present application can perform data processing and unsupervised cluster analysis on the actual road driving data obtained, and then obtain the time-frequency characteristics of the half-axle vibration corresponding to the road. The specific content will be described below and will not be repeated here.

[0054] Optionally, in one embodiment of the present application, based on the actual road driving data, the semi-axle vibration time-frequency characteristics corresponding to the road are extracted, including: based on the actual road driving data, obtaining micro-segment driving data of at least one time segment corresponding to the actual road driving data; calculating the vehicle speed per acceleration of each micro-segment driving data by the calculation formula of the vehicle speed per acceleration, and obtaining at least one vehicle speed per acceleration interval of each micro-segment driving data according to the vehicle speed per acceleration; combining the micro-segment driving data and at least one vehicle speed per acceleration interval to generate a semi-axle vibration data set of the actual road driving data; obtaining the semi-axle vibration time-frequency characteristics according to the semi-axle vibration data set. Among them, the calculation formula of the vehicle speed per acceleration can be but is not limited to:

[0055]

[0056] Where i represents the i-th second; v i represents the vehicle speed at the ith second; (v i+1 -v i ) / 3.24 represents the acceleration at the i-th second.

[0057] In some embodiments, the flowchart of the embodiment of the present application for generating the time-frequency characteristics of the semi-axis vibration is as follows: Figure 2 As shown, the main contents include:

[0058] Step S201: Acquire micro-segment driving data.

[0059] Among them, the embodiment of the present application segments the actual road driving data to obtain micro-segment driving data with a fixed duration of k seconds (k can be 5 seconds, and this application does not impose any specific limitation).

[0060] Step S202: Calculate vehicle speed per acceleration.

[0061] In the embodiment of the present application, the vehicle speed per acceleration of each micro-segment driving data is calculated by the calculation formula of the vehicle speed per acceleration, wherein the calculation formula of the vehicle speed per acceleration may be but is not limited to:

[0062]

[0063] Where i represents the i-th second; v i represents the vehicle speed at the ith second; (v i+1 -v i ) / 3.24 represents the acceleration at the i-th second.

[0064] Step S203: Divide the vehicle speed into acceleration intervals.

[0065] The embodiment of the present application divides the micro-segment driving data into (-∞,-3) m according to the vehicle speed per acceleration. 2 / s 3、[-3,-1)m 2 / s 3 、[-1,-0)m 2 / s 3 , [0,1)m 2 / s 3 、[1,3)m 2 / s 3 、[3,5)m 2 / s 3 , [5,10)m 2 / s 3 、[10,15)m 2 / s 3 、[15,+∞)m 2 / s 3 And so on for 9 vehicle speeds and acceleration intervals.

[0066] Step S204: Generate a semi-axis vibration data set.

[0067] Among them, the embodiment of the present application performs low-pass filtering on the half-shaft vibration in all micro-segment driving data in each vehicle speed and acceleration interval, and calculates the time-frequency characteristics of the half-shaft vibration after filtering, thereby obtaining a half-shaft vibration data set that takes into account the vehicle's operating status.

[0068] Step S205: extracting the time-frequency characteristics of the semi-axis vibration.

[0069] Among them, the embodiment of the present application performs unsupervised cluster analysis based on the semi-axle vibration data set in each vehicle speed and acceleration interval to obtain the semi-axle vibration time-frequency characteristics. Specifically, in each vehicle speed and acceleration interval, the embodiment of the present application uses the time-frequency characteristics of the semi-axle vibration as input, uses the AP (Affinity Propagation) clustering algorithm to obtain the cluster center, and integrates the cluster centers of all vehicle speed and acceleration intervals, divides the micro-segment driving data into the corresponding cluster centers, and then obtains the semi-axle vibration time-frequency characteristics.

[0070] For example, the present application processes the actual road driving data of the past three years in the XX region to obtain the semi-axle vibration data set of the past three years in the XX region. Specifically, first, the present application embodiment cuts the actual road driving data into segments to obtain micro-segment driving data with a fixed time segment length of 5 seconds; then, the present application embodiment calculates the vehicle speed per acceleration of the micro-segment driving data according to the calculation formula of the vehicle speed per acceleration, and divides the micro-segment driving data into (-∞,-3)m according to the vehicle speed per acceleration. 2 / s 3 、[-3,-1)m 2 / s 3 、[-1,-0)m 2 / s3 , [0,1)m 2 / s 3 、[1,3)m 2 / s 3 、[3,5)m 2 / s 3 , [5,10)m 2 / s 3 、[10,15)m 2 / s 3 、[15,+∞)m 2 / s 3 Finally, in each vehicle speed per acceleration interval, the embodiment of the present application performs a low-pass filter with a cutoff frequency of 30 Hz on the semi-axle vibration of all micro-segment driving data, and calculates the time-frequency characteristics of the semi-axle vibration after filtering, thereby forming a semi-axle vibration data set that takes into account the vehicle running state.

[0071] Furthermore, the embodiment of the present application performs unsupervised cluster analysis in each vehicle speed and acceleration interval based on the half-axle vibration data set, thereby obtaining the time-frequency characteristics of the half-axle vibration.

[0072] Optionally, in one embodiment of the present application, before the half-axle vibration time-frequency characteristics are input into a pre-constructed good and bad road identification model, it also includes: obtaining the bad road half-axle vibration time-frequency characteristics corresponding to the road in the bad road scene; obtaining the non-bad road half-axle vibration time-frequency characteristics corresponding to the road in the non-bad road scene; and constructing the good and bad road identification model based on the bad road half-axle vibration time-frequency characteristics and the non-bad road half-axle vibration time-frequency characteristics.

[0073] It can be understood that the bad road scenario in the embodiment of the present application can be understood as a scenario in which a vehicle drives through potholes, cracks and other damaged road surfaces. In addition, in the embodiment of the present application, the actual road driving data in the bad road scenario can be acquired through a data acquisition test. The actual road driving data includes road driving data including half-axle vibration, and also includes parameters such as time, vehicle speed, acceleration, longitude and latitude, and half-axle vibration. This application does not impose any specific restrictions.

[0074] As a possible implementation method, the embodiment of the present application obtains actual road driving data under a bad road scenario, obtains the time-frequency characteristics of the semi-axle vibration under the bad road scenario through data processing, and marks the clustering centers of all vehicle speeds and each acceleration interval, thereby obtaining the time-frequency characteristics of the semi-axle vibration of the bad road scenario and the time-frequency characteristics of the semi-axle vibration of the non-bad road scenario. According to the time-frequency characteristics of the semi-axle vibration of the bad road scenario and the time-frequency characteristics of the semi-axle vibration of the non-bad road scenario, a good and bad road identification model is constructed using a machine learning method.

[0075] For example, the flowchart of constructing the bad road identification model in the embodiment of the present application is as follows: Figure 3 As shown, the main contents include:

[0076] Step S301: Processing actual road driving data under a bad road scenario to obtain the time-frequency characteristics of half-axle vibration under the bad road scenario.

[0077] Among them, first, the embodiment of the present application segments the actual road driving data in the bad road scenario to obtain micro-segment driving data with a fixed time segment of k seconds (the value of k here is consistent with step S201). Then, the embodiment of the present application calculates the vehicle speed per acceleration of the micro-segment driving data through the calculation formula of the vehicle speed per acceleration, and divides the micro-segment driving data into (-∞,-3)m according to the vehicle speed per acceleration. 2 / s 3 、[-3,-1)m 2 / s 3 、[-1,-0)m 2 / s 3 , [0,1)m 2 / s 3 、[1,3)m 2 / s 3 、[3,5)m 2 / s 3 , [5,10)m 2 / s 3 、[10,15)m 2 / s 3 、[15,+∞)m 2 / s 3 Finally, in each vehicle speed acceleration interval, the embodiment of the present application performs low-pass filtering on the semi-axle vibration of all micro-segment driving data, and calculates the time-frequency characteristics of the semi-axle vibration after filtering (the parameters of the time-frequency characteristics are selected in the same way as in step S204), and simultaneously calculates the average value of the time-frequency characteristics of the semi-axle vibration, that is, the average value of each parameter in the time-frequency characteristics, as the time-frequency characteristics of the semi-axle vibration in the bad road scenario.

[0078] Step S302: Mark the cluster centers of all vehicle speeds and acceleration intervals according to the time-frequency characteristics of the semi-axle vibration in the bad road scenario.

[0079] Among them, in the embodiment of the present application, within each vehicle speed and acceleration interval, based on the time-frequency characteristics of the semi-axle vibration under the bad road scenario, the cluster center with the smallest Euclidean distance is found, and the cluster center is marked as the bad road scenario cluster center under the vehicle speed and acceleration interval, and the remaining cluster centers are marked as non-bad road scenario cluster centers under the vehicle speed and acceleration interval. In this way, the cluster centers of the bad road scenarios and the non-bad road scenarios under all vehicle speed and acceleration intervals can be obtained, and then the clustering results of the actual road driving data for the bad road scenarios and non-bad road scenarios can be obtained.

[0080] Step S303: construct a good and bad road identification model using a machine learning method according to the time-frequency characteristics of the semi-axle vibration in the bad road scenario and the non-bad road scenario.

[0081] Among them, the embodiment of the present application obtains the half-axle vibration time-frequency characteristics of bad road scenarios and non-bad road scenarios at each vehicle speed and acceleration interval based on the clustering results of actual road driving data of bad road scenarios and non-bad road scenarios. Furthermore, the embodiment of the present application uses the half-axle vibration time-frequency characteristics as input in each vehicle speed and acceleration interval, and records the bad road scene as 1 and the non-bad road scene as 0, and takes 1 (bad road scenario) and 0 (non-bad road scenario) as output, and uses the random forest algorithm in machine learning to construct a good and bad road identification model.

[0082] For example, Figure 4 As shown, the embodiment of the present application provides a schematic diagram of a bad road scene with potholes, cracks and other damaged road surfaces, wherein: Figure 4 (a) is a schematic diagram of a bad road scene with potholes and damaged road surface provided according to an embodiment of the present application; Figure 4 (b) A schematic diagram of a bad road scene with cracked pavement provided by an embodiment of the present application; and through experiments, nearly 300 kilometers of actual road driving data under bad road scenes were collected in XX area, according to Figure 3 The bad road identification model is constructed as shown in the figure, and its main contents include:

[0083] First, the embodiment of the present application processes the actual road driving data under the bad road scenario according to step S301 to obtain the time-frequency characteristics of the semi-axle vibration under the bad road scenario. The main contents include: First, the embodiment of the present application segments the actual road driving data under the bad road scenario to obtain micro-segment driving data with a fixed duration of 5 seconds. Then, the embodiment of the present application calculates the vehicle speed per acceleration of the micro-segment driving data through the calculation formula of the vehicle speed per acceleration, and divides the micro-segment driving data into (-∞,-3) m according to the vehicle speed per acceleration. 2 / s 3 、[-3,-1)m 2 / s 3 、[-1,-0)m 2 / s 3 , [0,1)m 2 / s 3 、[1,3)m 2 / s 3 、[3,5)m 2 / s 3 , [5,10)m 2 / s 3 、[10,15)m 2 / s 3 、[15,+∞)m 2 / s3 Finally, in each vehicle speed acceleration interval, the embodiment of the present application performs low-pass filtering on the semi-axle vibration of all micro-segment driving data, calculates the time-frequency characteristics of the semi-axle vibration after filtering, and simultaneously calculates the average value of the time-frequency characteristics of the semi-axle vibration, that is, the average value of each parameter in the time-frequency characteristics, as the time-frequency characteristics of the semi-axle vibration in the bad road scenario.

[0084] Next, the embodiment of the present application marks the cluster centers of all vehicle speeds per acceleration interval according to the time-frequency characteristics of the semi-axle vibration in the bad road scenario in step S302. The main contents include: First, in each vehicle speed per acceleration interval, the embodiment of the present application finds the cluster center with the smallest Euclidean distance according to the time-frequency characteristics of the semi-axle vibration in the bad road scenario, marks the cluster center as the bad road scenario cluster center in the vehicle speed per acceleration interval, and marks the remaining cluster centers as the non-bad road scenario cluster centers in the vehicle speed per acceleration interval. Then, the cluster centers of bad road scenarios and non-bad road scenarios in all vehicle speed per acceleration intervals can be obtained, and finally the clustering results of actual road driving data belonging to bad road scenarios and non-bad road scenarios are obtained, as shown in Table 1. The embodiment of the present application provides a schematic table of bad road scenario cluster centers. Among them, Table 1 is a box schematic table of bad road scenario cluster centers provided according to an embodiment of the present application.

[0085] Table 1

[0086]

[0087] Finally, the embodiment of the present application uses a machine learning method to construct a good and bad road identification model according to the clustering results of the actual road driving data of the bad road scene and the non-bad road scene in step S303. The main contents include: the embodiment of the present application obtains the half-axle vibration time-frequency characteristics in the bad road scene and the non-bad road scene in each vehicle speed and each acceleration interval according to the clustering results of the actual road driving data of the bad road scene and the non-bad road scene, and uses the half-axle vibration time-frequency characteristics as input in each vehicle speed and each acceleration interval, and records the bad road scene as 1 and the non-bad road scene as 0, and uses 1 (bad road scene) and 0 (non-bad road scene) as output, and uses the random forest algorithm in machine learning to construct a good and bad road identification model.

[0088] In step S103, the half-axle vibration time-frequency characteristics are input into a pre-built good and bad road identification model to output a bad road identification result of the road, and the bad road position distribution in the target area is obtained based on the bad road identification result.

[0089] It can be understood that the good and bad road identification model constructed in the embodiment of the present application can be used to obtain the bad road location distribution of the target map.

[0090] It will be understood by those skilled in the art that the embodiments of the present application can input the time-frequency characteristics of the semi-axle vibration obtained in the target area into a pre-built good and bad road identification model, output the bad road identification result, and obtain the bad road location distribution in the target area based on the latitude and longitude information of the micro-segment driving data corresponding to the bad road identification result.

[0091] For example, the embodiment of the present application processes the actual road driving data of about 300,000 kilometers in the XX region over the past three years, and obtains the time-frequency characteristics of the half-axle vibration in 10 vehicle speed acceleration intervals.

[0092] Furthermore, the embodiment of the present application inputs the half-axle vibration time-frequency characteristics into the pre-built good and bad road identification model to obtain the bad road identification result, and obtains the bad road location distribution in the XX area based on the latitude and longitude information of the micro-segment driving data corresponding to the identification result. Figure 5 As shown in FIG. 1 , a bad road distribution map of the XX region is given, and the location of the black solid line is the bad road distribution location in the XX region.

[0093] Combine the following Figure 6 As shown, the working principle of the method for obtaining bad road distribution proposed in the embodiment of the present application is introduced in detail with a specific embodiment.

[0094] Step S601: Acquire actual road driving data, and calculate and obtain the clustering result of the vehicle's actual road driving data through data processing and unsupervised cluster analysis.

[0095] Step S602: Acquire actual road driving data under bad road scenarios, calculate the time-frequency characteristics of half-axle vibration under bad road scenarios through data processing, and mark the clustering centers of all vehicle speeds and acceleration intervals, thereby obtaining clustering results of actual road driving data belonging to bad road scenarios and non-bad road scenarios.

[0096] Step S603: According to the clustering results of actual road driving data of bad road scenarios and non-bad road scenarios, the time-frequency characteristics of the bad road half-axle vibration of the bad road scenario and the time-frequency characteristics of the non-bad road half-axle vibration of the non-bad road scenario are obtained, and a good and bad road identification model is constructed using a machine learning method.

[0097] Step S604: input the half-axle vibration time-frequency characteristics into a pre-built good and bad road identification model, output the bad road identification result of the road, and obtain the bad road position distribution in the target area according to the bad road identification result.

[0098] According to the method for obtaining the bad road distribution of roads proposed in the embodiment of the present application, the corresponding semi-axle vibration time-frequency characteristics can be formed through the actual road driving data of the road where the vehicle is located in the target area, and the semi-axle vibration time-frequency characteristics can be input into the pre-constructed good and bad road identification model to obtain the bad road identification results of the road, and then generate the bad road position distribution in the target area. By associating the semi-axle vibration time-frequency characteristics with the road conditions, not only the detection efficiency is improved, but also the detection cost is reduced. By analyzing the semi-axle vibration time-frequency characteristics, the damage of the road can be accurately judged, providing an important basis for road maintenance, thereby improving the safety and comfort of the road, and realizing more intelligent and accurate road monitoring and management. Thus, it solves the problem in the related technology that smartphone monitoring is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, the construction of the model will result in data inconsistency, reducing the reliability of the detection results, and other problems.

[0099] Next, a device for acquiring bad road distribution according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0100] Figure 7 A block diagram of a device for acquiring bad road distribution according to an embodiment of the present application.

[0101] like Figure 7 As shown, the device 10 for acquiring bad road distribution includes: a first acquisition module 100 , an extraction module 200 and a second acquisition module 300 .

[0102] The first acquisition module 100 is used to acquire actual road driving data of the road where the vehicle is located in the target area.

[0103] The extraction module 200 is used to extract the time-frequency characteristics of the semi-axle vibration corresponding to the road based on the actual road driving data.

[0104] The second acquisition module 300 is used to input the half-axle vibration time-frequency characteristics into a pre-built good and bad road identification model to output a bad road identification result of the road, and obtain the bad road position distribution in the target area based on the bad road identification result.

[0105] Optionally, in one embodiment of the present application, it further includes: a third acquisition module, a fourth acquisition module and a construction module.

[0106] The third acquisition module is used to obtain the bad road half-axle vibration time-frequency characteristics corresponding to the road in the bad road scene before inputting the half-axle vibration time-frequency characteristics into the pre-built good and bad road identification model.

[0107] The fourth acquisition module is used to obtain the time-frequency characteristics of the non-bad road semi-axle vibration corresponding to the road in the non-bad road scene.

[0108] A building module is used to build a good and bad road identification model according to the time-frequency characteristics of the vibration of the bad road half shaft and the time-frequency characteristics of the vibration of the non-bad road half shaft.

[0109] Optionally, in one embodiment of the present application, the extraction module 200 includes: an acquisition unit, a calculation unit, a first generation unit and a second generation unit.

[0110] The acquisition unit is used to acquire micro-segment driving data of at least one duration segment corresponding to the actual road driving data based on the actual road driving data.

[0111] The calculation unit is used to calculate the vehicle speed per acceleration of each micro-segment driving data by using the calculation formula of the vehicle speed per acceleration, and obtain at least one vehicle speed per acceleration interval of each micro-segment driving data according to the vehicle speed per acceleration.

[0112] The first generating unit is used to generate a half-axle vibration data set of actual road driving data by combining the micro-segment driving data and at least one vehicle speed per acceleration interval.

[0113] The second generating unit is used to obtain the semi-axis vibration time-frequency characteristics according to the semi-axis vibration data set.

[0114] Optionally, in one embodiment of the present application, the calculation formula of vehicle speed per acceleration may be, but is not limited to,:

[0115]

[0116] Where i represents the i-th second; v i represents the vehicle speed at the ith second; (v i+1 -v i ) / 3.24 represents the acceleration at the i-th second.

[0117] It should be noted that the above explanation of the embodiment of the method for obtaining bad road distribution is also applicable to the device for obtaining bad road distribution of this embodiment, and will not be repeated here.

[0118] According to the device for acquiring the bad road distribution of roads proposed in the embodiment of the present application, the corresponding semi-axle vibration time-frequency characteristics can be formed through the actual road driving data of the road where the vehicle is located in the target area, and the semi-axle vibration time-frequency characteristics can be input into the pre-constructed good and bad road identification model to obtain the bad road identification results of the road, and then generate the bad road position distribution in the target area. By associating the semi-axle vibration time-frequency characteristics with the road conditions, not only the detection efficiency is improved, but also the detection cost is reduced. By analyzing the semi-axle vibration time-frequency characteristics, the damage of the road can be accurately judged, providing an important basis for road maintenance, thereby improving the safety and comfort of the road, and realizing more intelligent and accurate road monitoring and management. Thus, it solves the problem in the related technology that smartphone monitoring is not only time-consuming and laborious, but also difficult to fully cover all roads, and it is easy to miss some potential hidden dangers. In addition, when building a model, there will be problems such as inconsistent data and reduced reliability of the detection results.

[0119] Figure 8 This is a schematic diagram of the structure of a vehicle provided according to an embodiment of the present application. The vehicle may include:

[0120] A memory 801 , a processor 802 , and a computer program stored in the memory 801 and executable on the processor 802 .

[0121] When the processor 802 executes the program, the method for obtaining the bad road distribution provided in the above embodiment is implemented.

[0122] Furthermore, the vehicle also includes:

[0123] The communication interface 803 is used for communication between the memory 801 and the processor 802 .

[0124] The memory 801 is used to store computer programs that can be executed on the processor 802 .

[0125] The memory 801 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0126] If the memory 801, the processor 802 and the communication interface 803 are implemented independently, the communication interface 803, the memory 801 and the processor 802 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0127] Optionally, in a specific implementation, if the memory 801, the processor 802 and the communication interface 803 are integrated on a chip, the memory 801, the processor 802 and the communication interface 803 can communicate with each other through an internal interface.

[0128] The processor 802 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0129] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above method for obtaining bad road distribution.

[0130] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, implements the above method for obtaining bad road distribution.

[0131] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0132] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0133] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or vehicle (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or vehicle and execute the instructions), or in combination with these instruction execution systems, devices or vehicles. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or vehicle, or in combination with these instruction execution systems, devices or vehicles. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0135] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of multiple of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0136] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0137] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0138] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for obtaining bad road distribution, characterized in that: The following steps are involved: Obtaining actual road travel data of the roads where vehicles are located in the target area; Based on the actual road driving data, extracting the time-frequency characteristics of the semi-axle vibration corresponding to the road; The semi-axle vibration time-frequency characteristics are input into a pre-constructed good and bad road identification model to output a bad road identification result of the road, and the bad road position distribution in the target area is obtained based on the bad road identification result.

2. The method according to claim 1, characterized in that Before the half-shaft vibration time-frequency characteristics are input into the pre-built good and bad road identification model, the method further includes: Obtaining the bad road semi-axle vibration time-frequency characteristics corresponding to the road in the bad road scenario; Obtaining the time-frequency characteristics of non-bad road semi-axle vibration corresponding to the road in the non-bad road scenario; The good and bad road identification model is constructed according to the bad road semi-axle vibration time-frequency characteristics and the non-bad road semi-axle vibration time-frequency characteristics.

3. The method according to claim 1, characterized in that The step of extracting the time-frequency characteristics of the semi-axle vibration corresponding to the road based on the actual road driving data includes: Acquire micro-segment driving data of at least one duration segment corresponding to the actual road driving data based on the actual road driving data; Calculating the vehicle speed per acceleration of each micro-segment driving data by using a calculation formula of the vehicle speed per acceleration, and obtaining at least one vehicle speed per acceleration interval of each micro-segment driving data according to the vehicle speed per acceleration; combining the micro-segment driving data and the at least one vehicle speed per acceleration interval to generate a semi-axle vibration data set of the actual road driving data; The semi-axis vibration time-frequency characteristics are obtained according to the semi-axis vibration data set.

4. The method according to claim 3, characterized in that The calculation formula of the vehicle speed per acceleration is: Where i represents the i-th second; v i represents the vehicle speed at the ith second; (v i+1 -v i ) / 3.24 represents the acceleration at the i-th second.

5. A device for acquiring bad road distribution, characterized in that: include: A first acquisition module is used to acquire actual road driving data of the road where the vehicle is located in the target area; An extraction module, used for extracting the time-frequency characteristics of the semi-axle vibration corresponding to the road based on the actual road driving data; The second acquisition module is used to input the half-axle vibration time-frequency characteristics into a pre-built good and bad road identification model to output a bad road identification result of the road, and obtain the bad road position distribution in the target area based on the bad road identification result.

6. The device according to claim 5, characterized in that Also includes: A third acquisition module is used to obtain the bad road semi-axle vibration time-frequency characteristics corresponding to the road in the bad road scene before inputting the semi-axle vibration time-frequency characteristics into the pre-built good and bad road identification model; A fourth acquisition module is used to acquire the time-frequency characteristics of the non-bad road semi-axle vibration corresponding to the road in the non-bad road scenario; A construction module is used to construct the good and bad road identification model according to the time-frequency characteristics of the bad road semi-axle vibration and the time-frequency characteristics of the non-bad road semi-axle vibration.

7. The device according to claim 5, characterized in that The extraction module comprises: an acquisition unit, configured to acquire, based on the actual road driving data, micro-segment driving data of at least one duration segment corresponding to the actual road driving data; a calculation unit, configured to calculate the vehicle speed per acceleration of each micro-segment driving data by using a calculation formula of the vehicle speed per acceleration, and obtain at least one vehicle speed per acceleration interval of each micro-segment driving data according to the vehicle speed per acceleration; A first generating unit, configured to generate a semi-axle vibration data set of the actual road driving data by combining the micro-segment driving data and the at least one vehicle speed per acceleration interval; The second generating unit is used to obtain the semi-axis vibration time-frequency characteristics according to the semi-axis vibration data set.

8. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for obtaining bad road distribution as described in any one of claims 1 to 4.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for obtaining bad road distribution as described in any one of claims 1 to 4.

10. A computer program product, characterized in that It comprises a computer program, which, when executed, is used to implement the method for obtaining bad road distribution as described in any one of claims 1-4.

Images