Road surface condition monitoring method, device, vehicle and storage medium

Abstract

The application discloses a road surface condition monitoring method and device, a vehicle and a storage medium. The road surface condition monitoring method comprises the following steps: obtaining a target road section identifier of a target road surface based on a preset first map, wherein the target road section identifier is obtained based on pre-marking of the first map; matching a corresponding road surface clustering section according to the target road section identifier, wherein the road surface clustering section is obtained based on feature extraction and clustering of the corresponding road surface by different vehicles; and obtaining a current condition of the target road surface according to the road surface clustering section. The road surface rectangular frame is obtained based on feature extraction and clustering of the corresponding road surface by different vehicles, so that the road surface condition can be effectively monitored in vehicle driving, and the driving comfort and safety of the vehicle are improved.

Description

Technical Field

[0001] This application relates to the field of vehicle driving technology, and in particular to a road condition monitoring method, device, vehicle, and storage medium. Background Technology

[0002] In recent years, with the improvement of economic strength, people's living standards have also been greatly improved, and more convenient transportation has greatly improved people's life and work efficiency. With the rapid development of the automotive industry, vehicle driving comfort and safety have become key issues that need to be focused on.

[0003] However, during daily driving, there are often bumpy road conditions. For example, many roads may have large potholes due to road excavation or other reasons; or there may be frequent speed bumps on certain sections of the road; or there may be steep slopes. Bumpy road conditions not only affect the comfort of the user, but may also scrape the vehicle's chassis. Summary of the Invention

[0004] The main objective of this application is to provide a road condition monitoring method, device, vehicle, and storage medium, which aims to effectively monitor road conditions during vehicle driving, thereby improving driving comfort and safety.

[0005] To achieve the above objectives, this application provides a road surface condition monitoring method, the road surface condition monitoring method comprising:

[0006] Based on a preset first map, the target road segment identifier is obtained, wherein the target road segment identifier is obtained based on the pre-marked first map;

[0007] Match the corresponding road surface clustering road segment according to the target road segment identifier, wherein the road surface clustering road segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface;

[0008] Based on the road surface clustering, the current condition of the target road surface is obtained.

[0009] Optionally, the road surface clustering segment includes a road surface rectangle, and before the step of matching the corresponding road surface clustering segment according to the target road segment identifier, the method further includes:

[0010] Acquire road surface data collected by different vehicles on corresponding road surfaces, and obtain the driving trajectories of different vehicles on corresponding road surfaces;

[0011] Based on the road surface data, the road surface attributes of the corresponding road surface are obtained;

[0012] Based on the road surface attributes, features are extracted and clustered from the driving trajectory to obtain the corresponding road surface bounding boxes.

[0013] Optionally, the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface bounding boxes includes:

[0014] The driving trajectory is broken into a sequence of straight line segments;

[0015] Feature extraction is performed on the sequence of line segments to obtain the corresponding trajectory geometric features;

[0016] Based on the road surface attributes, the trajectory geometric features are merged to obtain the corresponding cluster groups;

[0017] Feature extraction is performed on the clustered groups to obtain the corresponding clustered geometric features;

[0018] Based on the geometric relationships of the clustered geometric features, the clustered geometric features are merged to obtain the corresponding road surface rectangles.

[0019] Optionally, the step of merging the trajectory geometric features according to the road surface attributes to obtain the corresponding cluster groups includes:

[0020] Detect whether the road surface attributes match the clustering attributes of the current cluster group;

[0021] If the road surface attribute matches the clustering attribute, then the trajectory geometric features are merged with the corresponding cluster group;

[0022] If the road surface attributes do not match the clustering attributes, then a clustering group other than the current clustering group is generated.

[0023] Optionally, the road surface clustering segment further includes a road surface region bounding box. After the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangle, the method further includes:

[0024] Based on the distance between different road surface rectangles, the road surface rectangles are merged to obtain the road surface area frame.

[0025] Optionally, the road surface clustering segment further includes road rectangles. After the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangles, the method further includes:

[0026] The road surface rectangle is obtained by combining the road surface rectangle with a preset second map.

[0027] Optionally, the road surface data includes horizontal axis angle data and vertical axis angle data, and the step of obtaining the road surface attributes corresponding to the road surface based on the road surface data includes:

[0028] Analyze the road surface data to determine the corresponding detection method;

[0029] Based on the detection method described above, the road surface attributes of the corresponding road surface are obtained.

[0030] Optionally, the step of analyzing the road surface data to determine the corresponding detection method includes:

[0031] Obtain the preset first velocity coefficient;

[0032] Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the first sum of variances;

[0033] If the sum of the first variances is greater than the first velocity coefficient, then the detection method is determined to be uneven road detection.

[0034] Optionally, the road surface data further includes pitch angle data, and the step of analyzing the road surface data to determine the corresponding detection method includes:

[0035] Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the second sum of variances;

[0036] If the pitch angle is greater than a preset first value and the sum of the second variances is greater than a preset second value, then the detection method is determined to be ramp detection.

[0037] Optionally, the step of analyzing the road surface data to determine the corresponding detection method includes:

[0038] Obtain the preset second velocity coefficient;

[0039] Obtain the horizontal variance of the horizontal axis angle data and the vertical variance of the vertical axis angle data;

[0040] If the vertical axis variance is greater than the second speed coefficient and the horizontal axis variance is less than a preset third value, then the detection method is determined to be speed bump detection.

[0041] Optionally, after the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface bounding boxes, the method further includes:

[0042] For the road surface clustering segments, obtain and analyze the clustering segment data collected by different vehicles on the corresponding road surfaces;

[0043] If the clustered road segment data does not conform to the detection method, the road surface clustered road segment will be deleted.

[0044] This application also proposes a road surface condition monitoring device, which includes:

[0045] The road surface recognition module is used to obtain the target road segment identifier of the target road surface based on a preset first map, wherein the target road segment identifier is obtained based on the pre-marked first map;

[0046] The road surface matching module is used to match the corresponding road surface clustering segment according to the target road segment identifier, wherein the road surface clustering segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface;

[0047] The status acquisition module is used to obtain the current status of the target road surface based on the road surface clustering segments.

[0048] This application also proposes a vehicle, which includes a memory, a processor, and a road condition monitoring program stored in the memory and executable on the processor. When the road condition monitoring program is executed by the processor, it implements the steps of the road condition monitoring method as described above.

[0049] This application also proposes a computer-readable storage medium storing a road condition monitoring program, which, when executed by a processor, implements the steps of the road condition monitoring method described above.

[0050] The road condition monitoring method, device, vehicle, and storage medium proposed in this application obtain target road segment identifiers based on a preset first map, wherein the target road segment identifiers are pre-marked based on the first map; corresponding road cluster segments are matched according to the target road segment identifiers, wherein the road cluster segments are obtained by extracting and clustering features of different vehicles on corresponding road surfaces; and the current condition of the target road surface is obtained based on the road cluster segments. By extracting and clustering features of different vehicles on corresponding road surfaces to obtain road bounding boxes, effective monitoring of road conditions can be achieved during vehicle driving, improving driving comfort and safety. Based on the solution of this application, starting from the bump patterns of roads in the real world, a road bounding box based on crowdsourced road feature extraction is designed, and the effectiveness of the road condition monitoring method proposed in this application is verified using this road bounding box, improving driving comfort and safety. Attached Figure Description

[0051] Figure 1 This is a schematic diagram of the functional modules of the road condition monitoring device belonging to this application;

[0052] Figure 2 This is a flowchart illustrating a first exemplary embodiment of the road surface condition monitoring method of this application;

[0053] Figure 3 This is a flowchart illustrating a second exemplary embodiment of the road surface condition monitoring method of this application;

[0054] Figure 4 This is a schematic diagram of the pavement condition monitoring method of this application for extracting pavement rectangles;

[0055] Figure 5 This is a flowchart illustrating a third exemplary embodiment of the road surface condition monitoring method of this application;

[0056] Figure 6 This is a flowchart illustrating a fourth exemplary embodiment of the road surface condition monitoring method of this application;

[0057] Figure 7 This is a flowchart illustrating a fifth exemplary embodiment of the road surface condition monitoring method of this application;

[0058] Figure 8 This is a flowchart illustrating a sixth exemplary embodiment of the road surface condition monitoring method of this application;

[0059] Figure 9 This is a flowchart illustrating the seventh exemplary embodiment of the road surface condition monitoring method of this application;

[0060] Figure 10 This is a flowchart illustrating the eighth exemplary embodiment of the road surface condition monitoring method of this application;

[0061] Figure 11 This is a flowchart illustrating the ninth exemplary embodiment of the road surface condition monitoring method of this application.

[0062] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0063] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0064] The main solution of this application embodiment is as follows: Based on a preset first map, obtain the target road segment identifier of the target road surface, wherein the target road segment identifier is pre-marked based on the first map; match the corresponding road surface clustering segments according to the target road segment identifier, wherein the road surface clustering segments are obtained by extracting and clustering features of different vehicles on the corresponding road surface; obtain the current condition of the target road surface based on the road surface clustering segments. By extracting and clustering features of different vehicles on the corresponding road surface to obtain road surface rectangles, effective monitoring of road conditions can be achieved during vehicle driving, improving driving comfort and safety. Based on the solution of this application, starting from the bump patterns of roads in the real world, a road surface rectangle based on crowdsourced road surface feature extraction is proposed, and the effectiveness of the road condition monitoring method proposed in this application is verified using this road surface rectangle, improving driving comfort and safety.

[0065] Technical terms used in the embodiments of this application:

[0066] A Global Navigation Satellite System (GNSS) is a space-based radio navigation and positioning system that provides users with all-weather 3D coordinates, velocity, and time information at any location on the Earth's surface or in near-Earth space. It consists of one or more satellite constellations and the augmentation systems required to support specific tasks.

[0067] The International Committee on Global Navigation Satellite Systems (ICG) has announced the four major global satellite navigation system providers: China's BeiDou Navigation Satellite System (BDS), the United States' Global Positioning System (GPS), Russia's GLONASS, and the European Union's Galileo. GPS was the world's first globally established system for navigation and positioning; GLONASS, after a rapid resurgence, has become the world's second-largest satellite navigation system, and both are currently undergoing modernization. Galileo is the first fully civilian-use satellite navigation system and is in the experimental phase. BDS is China's independently developed and operated global satellite navigation system, providing global users with all-weather, all-time, high-precision positioning, navigation, and timing services.

[0068] An inertial measurement unit (IMU) is a device that measures the three-axis attitude angles (or angular rates) and acceleration of an object.

[0069] Gyroscopes and accelerometers are the main components of an IMU (Inertial Measurement Unit), and their accuracy directly affects the accuracy of the inertial system. In practical operation, various unavoidable interference factors cause errors in the gyroscopes and accelerometers. From the initial alignment, the navigation error increases over time, especially the position error, which is a major drawback of inertial navigation systems. Therefore, external information is needed to assist in achieving integrated navigation, effectively reducing the problem of error accumulation over time. To improve reliability, more sensors can be equipped for each axis. Generally, the IMU should be mounted at the center of gravity of the object being measured.

[0070] Typically, an IMU contains three single-axis accelerometers and three single-axis gyroscopes. The accelerometers measure pitch, roll, and yaw angles, while the gyroscopes measure the vertical axis (y-axis), horizontal axis (x-axis), and vertical axis (z-axis).

[0071] An accelerometer detects the acceleration signals of an object along three independent axes of the carrier's coordinate system, while a gyroscope detects the angular velocity signal of the carrier relative to the navigation coordinate system. By measuring the object's angular velocity and acceleration in three-dimensional space, the object's attitude can be calculated. This has significant application value in navigation.

[0072] IMUs are mostly used in devices that require motion control, such as automobiles and robots. They are also used in applications that require precise displacement calculations based on attitude, such as inertial navigation systems for submarines, aircraft, missiles, and spacecraft.

[0073] The station-centered coordinate system (station coordinate system, East-North-Sky ENU, local Cartesian coordinate system) is primarily used to understand the motion patterns of other objects centered on the observer. It's a coordinate system with the station as its origin, meaning a prepared base is used to pinpoint the location for observation and measurement, and is generally used in construction projects. It's also used to understand the motion patterns of other objects centered on the observer, such as the viewing angle, azimuth, and distance of GPS satellites visible to a receiver.

[0074] The Earth-Centered, Earth-Fixed (ECEF) coordinate system is a Cartesian coordinate system with the Earth's center as the origin. The origin O (0,0,0) is the Earth's center of mass. The z-axis is parallel to the Earth's axis and points towards the North Pole. The x-axis points towards the intersection of the Prime Meridian and the equator. The y-axis is perpendicular to the xOz plane (the intersection of 90°E and the equator), forming a right-handed coordinate system.

[0075] Principal Component Analysis (PCA) is a statistical method that uses orthogonal transformations to convert a set of potentially correlated variables into a set of linearly uncorrelated variables; the transformed set of variables is called the principal components.

[0076] In practical research, in order to comprehensively analyze the problem, many related variables (or factors) are proposed, because each variable reflects certain information about the problem to different degrees.

[0077] Principal component analysis was first introduced by Karl Pearson for non-random variables, and later H. Hotling extended this method to the case of random vectors. The amount of information is usually measured by the sum of squared deviations or variance.

[0078] This application takes into account that there are some very bumpy road surfaces in actual vehicle driving, such as potholes, steep slopes, and speed bumps, which not only affect the user's driving comfort, but may also scrape the vehicle chassis.

[0079] Therefore, the embodiments of this application, starting from the practical problem of monitoring road conditions, combine the labeling capabilities of crowdsourcing, and propose a road feature extraction road rectangle based on crowdsourcing by extracting and clustering features from different vehicles on corresponding road data. This solves the technical problem of lacking road condition monitoring during vehicle driving and improves vehicle driving comfort and safety.

[0080] Specifically, refer to Figure 1 , Figure 1 This is a schematic diagram of the functional modules of the road condition monitoring device belonging to this application. The road condition monitoring device can be a stand-alone device capable of extracting road features, and it can be integrated into the device in hardware or software form. The device can be a smart mobile terminal with data processing capabilities, such as a vehicle, mobile phone, or tablet computer, or it can be a fixed device or server with data processing capabilities.

[0081] In this embodiment, the road condition monitoring device includes at least an output module 110, a processor 120, a memory 130, and a communication module 140.

[0082] The memory 130 stores the operating system and road condition monitoring program. The road condition monitoring device can identify the target road surface, obtain the target road segment identifier, obtain the current condition of the target road surface by matching the corresponding road surface rectangle based on the target road segment identifier, and store information such as road surface rectangles obtained by feature extraction and clustering based on different vehicles on the corresponding road surface in the memory 130. The output module 110 can be a display screen, etc. The communication module 140 can include a WIFI module, a mobile communication module, and a Bluetooth module, etc., and communicates with external devices or servers through the communication module 140.

[0083] When the road condition monitoring program in memory 130 is executed by the processor, it performs the following steps:

[0084] Based on a preset first map, the target road segment identifier is obtained, wherein the target road segment identifier is obtained based on the pre-marked first map;

[0085] Match the corresponding road surface clustering road segment according to the target road segment identifier, wherein the road surface clustering road segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface;

[0086] Based on the road surface clustering, the current condition of the target road surface is obtained.

[0087] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0088] Acquire road surface data collected by different vehicles on corresponding road surfaces, and obtain the driving trajectories of different vehicles on corresponding road surfaces;

[0089] Based on the road surface data, the road surface attributes of the corresponding road surface are obtained;

[0090] Based on the road surface attributes, features are extracted and clustered from the driving trajectory to obtain the corresponding road surface bounding boxes.

[0091] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0092] The driving trajectory is broken into a sequence of straight line segments;

[0093] Feature extraction is performed on the sequence of line segments to obtain the corresponding trajectory geometric features;

[0094] Based on the road surface attributes, the trajectory geometric features are merged to obtain the corresponding cluster groups;

[0095] Feature extraction is performed on the clustered groups to obtain the corresponding clustered geometric features;

[0096] Based on the geometric relationships of the clustered geometric features, the clustered geometric features are merged to obtain the corresponding road surface rectangles.

[0097] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0098] Detect whether the road surface attributes match the clustering attributes of the current cluster group;

[0099] If the road surface attribute matches the clustering attribute, then the trajectory geometric features are merged with the corresponding cluster group;

[0100] If the road surface attributes do not match the clustering attributes, then a clustering group other than the current clustering group is generated.

[0101] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0102] Based on the distance between different road surface rectangles, the road surface rectangles are merged to obtain the road surface area frame.

[0103] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0104] The road surface rectangle is obtained by combining the road surface rectangle with a preset second map.

[0105] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0106] Analyze the road surface data to determine the corresponding detection method;

[0107] Based on the detection method described above, the road surface attributes of the corresponding road surface are obtained.

[0108] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0109] Obtain the preset first velocity coefficient;

[0110] Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the first sum of variances;

[0111] If the sum of the first variances is greater than the first velocity coefficient, then the detection method is determined to be uneven road detection.

[0112] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0113] Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the second sum of variances;

[0114] If the pitch angle is greater than a preset first value and the sum of the second variances is greater than a preset second value, then the detection method is determined to be ramp detection.

[0115] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0116] Obtain the preset second velocity coefficient;

[0117] Obtain the horizontal variance of the horizontal axis angle data and the vertical variance of the vertical axis angle data;

[0118] If the vertical axis variance is greater than the second speed coefficient and the horizontal axis variance is less than a preset third value, then the detection method is determined to be speed bump detection.

[0119] Furthermore, when the road condition monitoring program in memory 130 is executed by the processor, it also performs the following steps:

[0120] For the road surface clustering segments, obtain and analyze the clustering segment data collected by different vehicles on the corresponding road surfaces;

[0121] If the clustered road segment data does not conform to the detection method, the road surface clustered road segment will be deleted.

[0122] This embodiment, through the above-described scheme, specifically obtains the target road segment identifier based on a preset first map, wherein the target road segment identifier is pre-marked based on the first map; matches the corresponding road surface clustering segments according to the target road segment identifier, wherein the road surface clustering segments are obtained by extracting and clustering features of different vehicles on corresponding road surfaces; and obtains the current condition of the target road surface based on the road surface clustering segments. By extracting and clustering features of different vehicles on corresponding road surfaces to obtain road surface bounding boxes, effective monitoring of road conditions can be achieved during vehicle driving, improving driving comfort and safety. Based on the scheme of this application, starting from the bump patterns existing in real-world roads, a road surface bounding box based on crowdsourced road surface feature extraction is proposed, and the effectiveness of the road condition monitoring method proposed in this application is verified using this road surface bounding box, improving driving comfort and safety.

[0123] Based on, but not limited to, the above-described device architecture, this application proposes method embodiments.

[0124] Reference Figure 2 , Figure 2 This is a flowchart illustrating a first exemplary embodiment of the road surface condition monitoring method of this application. The road surface condition monitoring method includes:

[0125] Step S210: Based on a preset first map, obtain the target road segment identifier of the target road surface, wherein the target road segment identifier is obtained based on the pre-marked first map;

[0126] The subject executing the method in this embodiment can be a road condition monitoring device, a road condition monitoring equipment, or a server. This embodiment takes a road condition monitoring device as an example, which can be integrated into vehicles, smartphones, tablets, and other devices.

[0127] This embodiment addresses the practical problem of monitoring road conditions by combining crowdsourcing's data acquisition and labeling capabilities. It proposes a road feature extraction method based on crowdsourcing by extracting and clustering features from different vehicles on corresponding road surfaces. This method solves the technical problem of lacking road condition monitoring during vehicle driving, thereby improving driving comfort and safety.

[0128] Specifically, the first map is used to determine the current location of the target vehicle and, based on this location, to identify the corresponding target road surface. This can be a Global Navigation Satellite System (GNSS) or another Global Positioning System (GPS). In actual vehicle driving, including both autonomous and manual driving, roads often contain very bumpy sections, such as potholes, steep slopes, and frequent speed bumps. These not only affect the user's driving experience but, depending on the vehicle's configuration, may also cause damage to the vehicle's chassis, potentially damaging critical chassis components and rendering the car immobile. Therefore, before the vehicle traverses a bumpy road surface, the first map identifies the target road surface, obtaining the corresponding target road segment identifier. By matching the target road segment identifier with the corresponding road surface clusters, the current condition of the target road surface can be determined.

[0129] Step S220: Match the corresponding road surface clustering road segment according to the target road segment identifier, wherein the road surface clustering road segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface;

[0130] Specifically, the target road segment identifier is used to match the road cluster segments corresponding to the target road surface. When each participating vehicle performs feature extraction and clustering on the corresponding bumpy road surface to obtain the corresponding road cluster segments, a road segment identifier corresponding to each road cluster segment is generated accordingly. The vehicles are the participating vehicles, used to obtain the different road surfaces marked by each vehicle. The road cluster segments are obtained by extracting and clustering features from the driving trajectories of different vehicles on the corresponding road surfaces, used to obtain the current condition of the target road surface. By obtaining the marking of bumpy road surfaces by different participating vehicles, the corresponding road cluster segments can be obtained. Thus, the road cluster segments can be obtained by the participating vehicles providing corresponding inertial measurement unit data and Global Navigation Satellite System or other Global Positioning System data.

[0131] Step S230: Based on the road surface clustering, obtain the current condition of the target road surface.

[0132] Specifically, before a user drives their vehicle over a bumpy road, the system can notify them in advance of the current road conditions by matching corresponding road clusters. For example, if there are potholes or frequent speed bumps ahead, the system can remind the user or the vehicle to automatically adjust the height and stiffness of the chassis, thereby improving the user's driving comfort, preventing chassis scrapes, and enhancing driving safety.

[0133] This embodiment, through the above-described scheme, specifically obtains target road segment identifiers based on a preset first map, wherein the target road segment identifiers are pre-marked based on the first map; matches corresponding road surface cluster segments according to the target road segment identifiers, wherein the road surface cluster segments are obtained by extracting and clustering features of different vehicles on corresponding road surfaces; and obtains the current condition of the target road surface based on the road surface cluster segments. Obtaining road surface bounding boxes by extracting and clustering features of different vehicles on corresponding road surfaces can improve vehicle driving comfort and safety.

[0134] Reference Figure 3 , Figure 3 This is a flowchart illustrating a second exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 2 In the illustrated embodiment, the road surface clustering segment includes a road surface rectangle. Step S220 involves matching the corresponding road surface clustering segment based on the target road surface identifier. Before the road surface clustering segment is obtained by extracting and clustering features from different vehicles on the corresponding road surface, the method further includes:

[0135] Step S310: Obtain road surface data collected by different vehicles on the corresponding road surface, and obtain the driving trajectory of different vehicles on the corresponding road surface.

[0136] In this embodiment, step S310 is implemented before step S210. In other embodiments, step S310 may also be implemented between step S210 and step S220.

[0137] Specifically, road surface data refers to the sensor states of the participating crowdsourced vehicles as they traverse bumpy road sections, used to obtain road surface attributes. Driving trajectories are used for geometric feature extraction of the road surface. By marking the bumpy road sections traversed by each participating vehicle, corresponding road surface data is obtained. The driving trajectory and vehicle position are then obtained via a Global Navigation Satellite System (GNSS) or other Global Positioning Systems (GPS). For example, when there are abrupt changes in the vehicle's inertial measurement unit (INS) data, a low-pass filter is used to filter out high-frequency noise such as its own vibration. Based on the different states of the vehicle's sensors and its position, bumpy road sections are marked and distinguished on the map.

[0138] Step S320: Based on the road surface data, obtain the road surface attributes of the corresponding road surface;

[0139] Specifically, road surface attributes refer to the properties of bumpy road surfaces, used to cluster different bumpy road surfaces. By analyzing road surface data based on the natural patterns of road bumps, the road surface attributes corresponding to bumpy road sections can be detected.

[0140] Step S330: Based on the road surface attributes, extract features from the driving trajectory and cluster them to obtain the corresponding road surface rectangles.

[0141] Specifically, based on road surface attributes, features are extracted and clustered from the vehicle trajectory to obtain the corresponding road surface bounding boxes. (Reference) Figure 4 , Figure 4 This is a schematic diagram of the road surface condition monitoring method of this application for extracting road surface rectangles. The diagram specifically shows road surface rectangle 1, road surface rectangle 2, road surface rectangle 3, road surface rectangle 4, and road surface rectangle 5. Each road surface rectangle contains several driving trajectories.

[0142] This embodiment, through the above-described scheme, specifically obtains road surface data collected by participating crowdsourced vehicles on corresponding road surfaces, and acquires the driving trajectories of different vehicles on corresponding road surfaces through a global navigation satellite system; based on the road surface data, it obtains the road surface attributes of the corresponding road surface; based on the road surface attributes, it extracts features from the driving trajectories and clusters them to obtain corresponding road surface bounding boxes, which can reduce the cost of road condition monitoring and improve the accuracy of road condition monitoring.

[0143] Reference Figure 5 , Figure 5 This is a flowchart illustrating a third exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 3 In the illustrated embodiment, the road surface data includes horizontal axis angle data and vertical axis angle data. Step S320 involves obtaining the road surface attributes corresponding to the road surface based on the road surface data, including:

[0144] Step S510: Analyze the road surface data to determine the corresponding detection method;

[0145] This application uses a six-axis inertial measurement unit as an example. In other embodiments, it can also be an inertial measurement unit with other axes.

[0146] Specifically, road surface data includes, but is not limited to, pitch, yaw, and roll angle data from the inertial measurement unit, and horizontal, vertical, and longitudinal axis data from the gyroscope. When a vehicle travels over a bumpy road surface, high-frequency noise such as its own vibration is filtered out by the vehicle's low-pass filter to obtain the corresponding road surface data. Based on the natural laws governing road surface bumps, the road surface data is analyzed to determine the detection method for bumpy roads.

[0147] Step S520: Obtain the road surface attributes of the corresponding road surface according to the detection method.

[0148] Specifically, the detection method is used to detect bumpy road surfaces and determine the corresponding road surface attributes, including but not limited to uneven road detection, slope detection, and frequent speed bump detection. By detecting the road surface data of bumpy road surfaces through the detection method, the road surface attributes can be obtained.

[0149] Further, in step S510, the road surface data is analyzed to determine the corresponding detection method, including:

[0150] Step A11: Obtain the preset first velocity coefficient;

[0151] Specifically, the first speed coefficient is used to determine whether the sum of the variances of the vehicle's lateral axis angle data and longitudinal axis angle data is greater than a preset first speed coefficient. This embodiment uses a first speed coefficient of 30*delta as an example, where delta is:

[0152]

[0153] In the formula, v represents the vehicle speed, and max represents the maximum value.

[0154] Step A12: Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the first variance sum.

[0155] Specifically, the first sum of variances is the sum of the variances of the horizontal and vertical axes of the 10Hz gyroscope calculated within a preset time period. The preset time period is set according to actual conditions; this embodiment uses a time period of 3 seconds as an example. In other embodiments, it can be less than 3 seconds or greater than 3 seconds.

[0156] Step A13: If the sum of the first variances is greater than the first velocity coefficient, then the detection method is determined to be uneven road detection.

[0157] Specifically, if the sum of the first variances is greater than 30*delta, and the length of the vehicle's bumpy trajectory is greater than a preset length, then it is considered an uneven road section. The preset length is set according to the actual situation; this embodiment uses a preset length of 3 meters as an example. In other embodiments, it can be less than or greater than 3 meters.

[0158] And / or the road surface data also includes pitch angle data. Step S510: Analyze the road surface data to determine the corresponding detection method, including:

[0159] Step A21: Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the second variance sum;

[0160] Specifically, the second variance sum is the sum of the variances of the horizontal and vertical axes obtained by calculating the angle data of the horizontal axis and the angle data of the vertical axis using a 10Hz gyroscope within a preset time period. The preset time period is set according to the actual situation; this embodiment uses a time period of 3 seconds as an example. In other embodiments, it can be less than 3 seconds or greater than 3 seconds.

[0161] Step A22: If the pitch angle is greater than a preset first value and the sum of the second variances is greater than a preset second value, then the detection method is determined to be ramp detection.

[0162] Specifically, in the station-centric coordinate system, if the pitch angle of the inertial measurement unit changes by more than a first value within 3 seconds, and the sum of the second variances is greater than a second value, then the vehicle is considered to be on a slope. In this embodiment, it is preferred that the first value is 10 degrees and the second value is 10. In other embodiments, the first value can be less than or greater than 10 degrees, and the second value can be less than or greater than 10.

[0163] And / or step S510, analyzing the road surface data to determine the corresponding detection method, including:

[0164] Step A31: Obtain the preset second velocity coefficient;

[0165] Specifically, the second speed coefficient is used to determine the variance of the vehicle's longitudinal axis angle data. Taking 25*delta as an example, the second speed coefficient is:

[0166]

[0167] In the formula, v represents the vehicle speed, and max represents the maximum value.

[0168] Step A32: Obtain the horizontal variance of the horizontal axis angle data and the vertical variance of the vertical axis angle data;

[0169] Specifically, the horizontal axis variance is the variance of the horizontal axis angle data of the 10Hz gyroscope within a preset time period, and the vertical axis variance is the variance of the vertical axis angle data of the 10Hz gyroscope within a preset time period. In this embodiment, a time period of 3 seconds is used as an example. In other embodiments, it can be less than 3 seconds or greater than 3 seconds.

[0170] Step A33: If the vertical axis variance is greater than the second speed coefficient and the horizontal axis variance is less than a preset third value, then the detection method is determined to be speed bump detection.

[0171] Specifically, if the variance of the vertical axis angle data is greater than 25*delta within 3 seconds, and the variance of the horizontal axis is less than the third value, then it is considered a speed bump section. This embodiment uses a third value of 1 as an example; in other embodiments, it can be less than 1 or greater than 1.

[0172] This embodiment, through the above-described scheme, specifically analyzes the road surface data to determine the corresponding detection method; based on the detection method, it obtains the road surface attributes. Determining the corresponding detection method by analyzing road surface data to obtain the corresponding road surface attributes can improve the efficiency of road surface condition monitoring.

[0173] Reference Figure 6 , Figure 6 This is a flowchart illustrating a fourth exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 5 In the embodiment shown, after step S330, which involves extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface bounding boxes, the method further includes:

[0174] Step S610: For the road surface clustering segment, obtain and analyze the clustering segment data collected by different vehicles on the corresponding road surface;

[0175] Specifically, the clustered road segment data consists of the sensor status of the vehicles as they pass through the road segments, used to determine whether the road surface corresponding to the generated road clusters has become a smooth road surface. The road clusters include road surface rectangles, road surface region boxes, and road rectangles; the clustered road segment data includes, but is not limited to, pitch angle, yaw angle, and roll angle data from the inertial measurement unit, and horizontal, vertical, and longitudinal axis data from the gyroscope. After generating the road clusters, when vehicles pass through the corresponding road surfaces, high-frequency noise such as their own vibrations is filtered out by the vehicle's low-pass filter. For the generated road clusters, the corresponding clustered road segment data is collected and analyzed to determine the appropriate detection method.

[0176] Step S620: If the clustered road segment data does not conform to the detection method, then the road surface clustered road segment is deleted.

[0177] Specifically, if the clustered road segment data does not meet the detection method, that is, the current road surface clustered road segment belongs to a smooth road surface, then the corresponding clustered road segment data is deleted; if the clustered road segment data meets the detection method, that is, the current road surface clustered road segment belongs to a bumpy road surface, then the corresponding road surface clustered road segment is retained or updated based on the road surface attributes determined by the detection method.

[0178] This embodiment, through the above-described scheme, specifically acquires and analyzes clustered road segment data collected by different vehicles on corresponding road surfaces for the road surface clustering segments; if the clustered road segment data does not conform to the detection method, the road surface clustered segment is deleted, which can update the current road conditions in real time and improve the accuracy of road condition monitoring.

[0179] Reference Figure 7 , Figure 7 This is a flowchart illustrating a fifth exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 3 In the embodiment shown, step S330 involves extracting and clustering features from the driving trajectory based on the road surface attributes to obtain the corresponding road surface bounding boxes, including:

[0180] Step S710: Break the driving trajectory into a sequence of straight line segments;

[0181] Specifically, since the driving trajectories of different vehicles are chaotic and non-overlapping data, a sequence of straight line segments can be obtained based on the angular offset of the driving trajectory, that is, based on the direction and angle of the driving trajectory. The driving trajectory includes one or more sequences of straight line segments. Specifically, based on the driving trajectory, any two points are selected and connected by a line; the portion between these two points that coincides with the driving trajectory is taken as the sequence of straight line segments, and the angle between the non-coincident portion and the driving trajectory is taken as the angular offset. This process is repeated to obtain at least one sequence of straight line segments.

[0182] Step S720: Extract features from the sequence of line segments to obtain the corresponding trajectory geometric features;

[0183] Specifically, the trajectory geometric features of the vehicle trajectory are used for merging to obtain the corresponding cluster groups. The sequence of straight line segments includes the corresponding trajectory coordinates. By converting the trajectory coordinates of the sequence of straight line segments into geocentric coordinates, and then calculating the geocentric coordinates through principal component analysis, the trajectory geometric features of the vehicle trajectory can be obtained.

[0184] Furthermore, when the trajectory coordinates of the straight line segment sequence are converted to geocentric coordinates, trajectory identifiers are generated accordingly. By clustering the trajectory geometric features, the trajectory identifiers are processed simultaneously to obtain the target road segment identifiers corresponding to the road surface rectangles.

[0185] Step S730: Based on the road surface attributes, merge the trajectory geometric features to obtain the corresponding cluster groups;

[0186] Specifically, clustering is used to classify different bumpy road surfaces. Based on road surface attributes, the geometric features of vehicle trajectories under the same road surface attribute are merged to obtain corresponding cluster groups. By merging road surfaces with the same attributes, the accuracy of road condition monitoring can be improved.

[0187] Step S740: Extract features from the clustered groups to obtain the corresponding clustered geometric features;

[0188] Specifically, clustering geometric features are used for merging to obtain the corresponding road surface rectangles. After grouping the trajectory geometric features of the driving trajectory to obtain the corresponding cluster groups, feature extraction is then performed on the cluster groups to obtain the corresponding clustering geometric features.

[0189] Step S750: Based on the geometric relationship of the clustered geometric features, merge the clustered geometric features to obtain the corresponding road surface rectangle.

[0190] Specifically, geometric relationships include, but are not limited to: driving trajectories in the same direction and driving trajectories with small angle differences, used to merge the geometric features of clustered groups. Based on the geometric relationships of the geometric features of the clustered groups, the geometric features of the clustered groups are merged so that when a user drives over a bumpy road, if two potholes are 20 meters apart, the two potholes on this section of road are merged into a single road rectangle through the geometric relationships of the clustered groups. This allows the user to receive a notification that there are two potholes within 20 meters and their locations, and to automatically or manually adjust the vehicle chassis based on this notification.

[0191] This embodiment, through the above-described scheme, specifically involves breaking the vehicle trajectory into a sequence of straight line segments; extracting features from the sequence of straight line segments to obtain corresponding trajectory geometric features; merging the trajectory geometric features according to the road surface attributes to obtain corresponding cluster groups; extracting features from the cluster groups to obtain corresponding cluster geometric features; and merging the cluster geometric features according to the geometric relationships of the cluster geometric features to obtain the corresponding road surface rectangles. By breaking the vehicle trajectory into a sequence of straight line segments, obtaining the geometric features of the vehicle trajectory, and merging the trajectories, the accuracy of road surface feature extraction can be improved.

[0192] Reference Figure 8 , Figure 8 This is a flowchart illustrating a sixth exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 7 In the embodiment shown, step S730 involves merging the trajectory geometric features based on the road surface attributes to obtain corresponding clustering groups, including:

[0193] Step S810: Detect whether the road surface attributes match the clustering attributes of the current clustering group;

[0194] Specifically, before checking whether the road surface attributes match the clustering attributes of the current cluster group, a clustering structure can be created in advance, and corresponding clustering attributes can be set to obtain a clustering container (ClusterGroup). The clustering container is used to group according to the clustering attributes and store the clusters of bumpy roads. The attributes of the cluster groups are those of bumpy roads, including but not limited to uneven roads, slopes, and frequent speed bumps. In this way, by checking whether the road surface attributes match the current cluster group in the clustering container, similar driving trajectories can be merged, thereby clustering the road surface rectangles.

[0195] Step S820: If the road surface attribute matches the clustering attribute, then the trajectory geometric features are merged with the corresponding cluster groups;

[0196] Specifically, if the road surface attributes match the attributes of the cluster groups in the clustering container, the geometric features of the driving trajectory corresponding to the road surface attributes and the geometric features in the cluster groups are merged.

[0197] Step S830: If the road surface attribute does not match the clustering attribute, then generate a clustering group other than the current clustering group.

[0198] Specifically, if the road surface attributes do not match the attributes of the current cluster group, a new cluster group is generated in addition to the current cluster group, and the newly generated cluster group is inserted into the cluster container.

[0199] This embodiment, through the above-described scheme, specifically detects whether the road surface attributes match the clustering attributes of the current cluster group; if the road surface attributes match the clustering attributes, the trajectory geometric features are merged with the corresponding cluster group; if the road surface attributes do not match the clustering attributes, a cluster group other than the current cluster group is generated. By detecting road surface attributes to merge corresponding driving trajectories, the efficiency of road surface feature extraction can be improved.

[0200] Reference Figure 9 , Figure 9 This is a flowchart illustrating a seventh exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 3 In the embodiment shown, the road surface clustering segment further includes a road surface region bounding box. Step S330, after extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface bounding boxes, further includes:

[0201] Step S910: Merge the road surface rectangles according to the distance between different road surface rectangles to obtain the road surface area frame.

[0202] Specifically, road surface bounding boxes are used for road condition monitoring. Based on actual needs, the geometric features of road surface rectangles that are close together are merged. That is, road surface bounding boxes are obtained by extracting features from different vehicles on corresponding bumpy road surfaces, clustering them, and merging several close road surface rectangles.

[0203] In this way, after generating the road rectangle, the user can match the corresponding road surface area box according to the target road segment identifier to obtain the current condition of the target road. For example, when the user is driving on a bumpy road, if two ramps are 100 meters apart, the two rectangles can be merged into a larger road surface area box based on the distance and / or direction between them. This allows the user to receive a single notification that there are two ramps within 100 meters and their locations, and to automatically or manually adjust the vehicle chassis based on this notification.

[0204] This embodiment, through the above-described scheme, specifically acquires road surface data collected by different vehicles on corresponding road surfaces and obtains the driving trajectories of different vehicles on corresponding road surfaces; based on the road surface data, it obtains the road surface attributes of the corresponding road surfaces; based on the road surface attributes, it extracts features from the driving trajectories and clusters them to obtain corresponding road surface bounding boxes; based on the distance between different road surface bounding boxes, it merges the road surface bounding boxes to obtain the road surface region box. Obtaining the road surface region box based on the distance between the road surface bounding boxes can improve driving comfort.

[0205] Reference Figure 10 , Figure 10 This is a flowchart illustrating an eighth exemplary embodiment of the road surface condition monitoring method of this application. Based on the above... Figure 3 In the embodiment shown, the road surface clustering segment further includes a road rectangle. Step S330, after extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangle, further includes:

[0206] Step S1010: Combine the road surface rectangle with a preset second map to obtain the road rectangle.

[0207] Specifically, road bounding boxes are used for road condition monitoring. These boxes include the road surface, the road environment, and traffic congestion. Depending on the actual situation, road bounding boxes can be combined with a map to obtain the road bounding boxes. In other words, road bounding boxes are obtained by extracting and clustering features of different vehicles on corresponding bumpy road surfaces and combining them with a map.

[0208] In this way, after generating the road rectangle, the user can match the corresponding road rectangle according to the target road segment identifier to obtain the current status of the target road. For example, when the user is driving on a bumpy road, if there is a traffic jam and multiple speed bumps within 30 meters, by combining the road rectangle with the map, the user can receive a notification that there is a traffic jam and multiple speed bumps within 30 meters. The user can then automatically or manually adjust the vehicle chassis based on this notification.

[0209] This embodiment, through the above-described scheme, specifically acquires road surface data collected by different vehicles on corresponding road surfaces and obtains the driving trajectories of different vehicles on corresponding road surfaces; based on the road surface data, it obtains the road surface attributes of the corresponding road surfaces; based on the road surface attributes, it extracts features from the driving trajectories and clusters them to obtain corresponding road surface rectangles; it combines the road surface rectangles with a preset second map to obtain the road rectangle. By combining the road surface rectangles with the map to obtain the road rectangle, driving comfort can be improved.

[0210] Reference Figure 11 , Figure 11This is a flowchart illustrating a ninth exemplary embodiment of the road condition monitoring method of this application. As shown, a clustering structure is pre-created, clustering attributes are set, and a clustering container is obtained; the detection result of a single vehicle is input, where the detection result includes the driving trajectory of the single vehicle over the current bumpy road surface; the offset angle of the trajectory is broken into at least one sequence of straight line segments, with each sequence of straight line segments corresponding to the coordinates of each trajectory; the trajectory coordinates are converted to geocentric coordinates, and trajectory identifiers are generated; the geometric features of the trajectory are obtained through principal component analysis of the geocentric coordinates; it is detected whether the current bumpy road surface belongs to an existing cluster group; if the current bumpy road surface belongs to an existing cluster group, it is merged with the existing cluster group in the clustering container; if the current bumpy road surface does not belong to an existing cluster group, a new cluster group is generated and inserted into the clustering container; the geometric features of the cluster group are obtained through principal component analysis of the cluster group; based on the intersection relationship of the geometric features of the cluster group, identical cluster groups are merged; cluster groups with close proximity are merged according to requirements to obtain a larger cluster group area. This allows the vehicle to alert the user or automatically adjust the chassis height and stiffness before traversing bumpy roads, improving the user experience and preventing chassis scrapes.

[0211] Furthermore, this application also proposes a road surface condition monitoring device, which includes:

[0212] The road surface recognition module is used to obtain the target road segment identifier of the target road surface based on a preset first map, wherein the target road segment identifier is obtained based on the pre-marked first map;

[0213] The road surface matching module is used to match the corresponding road surface clustering segment according to the target road segment identifier, wherein the road surface clustering segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface;

[0214] The status acquisition module is used to obtain the current status of the target road surface based on the road surface clustering segments.

[0215] The principle and implementation process of road condition monitoring in this embodiment are explained in the above embodiments and will not be repeated here.

[0216] Furthermore, this application also proposes a vehicle, which includes a memory, a processor, and a road condition monitoring program stored in the memory and executable on the processor. When the road condition monitoring program is executed by the processor, it implements the steps of the road condition monitoring method described above.

[0217] Since this road condition monitoring program employs all the technical solutions of all the aforementioned embodiments when executed by the processor, it possesses at least all the beneficial effects brought about by all the technical solutions of all the aforementioned embodiments, which will not be elaborated upon here.

[0218] Furthermore, embodiments of this application also propose a computer-readable storage medium storing a road condition monitoring program, which, when executed by a processor, implements the steps of the road condition monitoring method described above.

[0219] Since this road condition monitoring program employs all the technical solutions of all the aforementioned embodiments when executed by the processor, it possesses at least all the beneficial effects brought about by all the technical solutions of all the aforementioned embodiments, which will not be elaborated upon here.

[0220] Compared to existing technologies, the road condition monitoring method, device, vehicle, and storage medium proposed in this application obtain target road segment identifiers based on a preset first map, wherein the target road segment identifiers are pre-marked based on the first map; corresponding road cluster segments are matched according to the target road segment identifiers, wherein the road cluster segments are obtained by extracting and clustering features of different vehicles on corresponding road surfaces; and the current condition of the target road surface is obtained based on the road cluster segments. By extracting and clustering features of different vehicles on corresponding road surfaces to obtain road bounding boxes, effective monitoring of road conditions can be achieved during vehicle driving, improving driving comfort and safety. Based on the solution of this application, starting from the bump patterns of roads in the real world, a road bounding box based on crowdsourced road feature extraction is designed, and the effectiveness of the road condition monitoring method proposed in this application is verified using this road bounding box, improving driving comfort and safety.

[0221] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0222] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0223] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a device (which may be a vehicle, mobile phone, computer, server, controlled terminal, or network device, etc.) to execute the methods of each embodiment of this application.

[0224] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for monitoring road surface conditions, characterized in that, The road surface condition monitoring method includes the following steps: Based on a preset first map, the target road segment identifier is obtained, wherein the target road segment identifier is obtained based on the pre-marked first map; Match the corresponding road surface clustering road segment according to the target road segment identifier, wherein the road surface clustering road segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface; Based on the road surface clustering, the current condition of the target road surface is obtained; The road surface clustering segment includes a road surface rectangle. Before the step of matching the corresponding road surface clustering segment according to the target road segment identifier, the method further includes: Acquire road surface data collected by different vehicles on corresponding road surfaces, and obtain the driving trajectories of different vehicles on corresponding road surfaces; Based on the road surface data, the road surface attributes of the corresponding road surface are obtained, wherein the road surface attributes are the attributes of bumpy roads, which are used to cluster different bumpy roads. Based on the road surface attributes, features are extracted and clustered from the driving trajectory to obtain the corresponding road surface bounding boxes.

2. The road surface condition monitoring method as described in claim 1, characterized in that, The step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangle includes: The driving trajectory is broken into a sequence of straight line segments; Feature extraction is performed on the sequence of line segments to obtain the corresponding trajectory geometric features; Based on the road surface attributes, the trajectory geometric features are merged to obtain the corresponding cluster groups; Feature extraction is performed on the clustered groups to obtain the corresponding clustered geometric features; Based on the geometric relationships of the clustered geometric features, the clustered geometric features are merged to obtain the corresponding road surface rectangles.

3. The road surface condition monitoring method as described in claim 2, characterized in that, The step of merging the trajectory geometric features based on the road surface attributes to obtain the corresponding cluster groups includes: Detect whether the road surface attributes match the clustering attributes of the current cluster group; If the road surface attribute matches the clustering attribute, then the trajectory geometric features are merged with the corresponding cluster group; If the road surface attributes do not match the clustering attributes, then a clustering group other than the current clustering group is generated.

4. The road surface condition monitoring method as described in claim 1, characterized in that, The road surface clustering segment also includes a road surface region bounding box. After the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangle, the method further includes: Based on the distance between different road surface rectangles, the road surface rectangles are merged to obtain the road surface area frame.

5. The road surface condition monitoring method as described in claim 1, characterized in that, The road surface clustering segment also includes road rectangles. After the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangles, the method further includes: The road surface rectangle is obtained by combining the road surface rectangle with a preset second map.

6. The road surface condition monitoring method as described in claim 1, characterized in that, The road surface data includes horizontal axis angle data and vertical axis angle data. The step of obtaining the road surface attributes based on the road surface data includes: Analyze the road surface data to determine the corresponding detection method; Based on the detection method described above, the road surface attributes of the corresponding road surface are obtained.

7. The road surface condition monitoring method as described in claim 6, characterized in that, The steps for analyzing the road surface data and determining the corresponding detection method include: Obtain the preset first velocity coefficient; Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the first sum of variances; If the sum of the first variances is greater than the first velocity coefficient, then the detection method is determined to be uneven road detection.

8. The road surface condition monitoring method as described in claim 6, characterized in that, The road surface data also includes pitch angle data. The step of analyzing the road surface data and determining the corresponding detection method includes: Calculate the sum of the variances of the horizontal axis angle data and the vertical axis angle data to obtain the second sum of variances; If the pitch angle is greater than a preset first value and the sum of the second variances is greater than a preset second value, then the detection method is determined to be ramp detection.

9. The road surface condition monitoring method as described in claim 6, characterized in that, The steps of analyzing the road surface data and determining the corresponding detection method include: Obtain the preset second velocity coefficient; Obtain the horizontal variance of the horizontal axis angle data and the vertical variance of the vertical axis angle data; If the vertical axis variance is greater than the second speed coefficient and the horizontal axis variance is less than a preset third value, then the detection method is determined to be speed bump detection.

10. The road surface condition monitoring method as described in claim 6, characterized in that, After the step of extracting features and clustering the driving trajectory based on the road surface attributes to obtain the corresponding road surface rectangles, the method further includes: For the road surface clustering segments, obtain and analyze the clustering segment data collected by different vehicles on the corresponding road surfaces; If the clustered road segment data does not conform to the detection method, the road surface clustered road segment will be deleted.

11. A road surface condition monitoring device, characterized in that, The road surface condition monitoring device includes: The road surface recognition module is used to obtain the target road segment identifier of the target road surface based on a preset first map, wherein the target road segment identifier is obtained based on the pre-marked first map; The road surface matching module is used to match the corresponding road surface clustering segment according to the target road segment identifier, wherein the road surface clustering segment is obtained by extracting and clustering features of different vehicles on the corresponding road surface; The status acquisition module is used to obtain the current status of the target road surface based on the road surface clustering segments; The road surface clustering segment includes a road surface rectangle, and the road surface condition monitoring device is further used for: Acquire road surface data collected by different vehicles on corresponding road surfaces, and obtain the driving trajectories of different vehicles on corresponding road surfaces; Based on the road surface data, the road surface attributes of the corresponding road surface are obtained, wherein the road surface attributes are the attributes of bumpy roads, which are used to cluster different bumpy roads. Based on the road surface attributes, features are extracted and clustered from the driving trajectory to obtain the corresponding road surface bounding boxes.

12. A vehicle, characterized in that, The vehicle includes a memory, a processor, and a road condition monitoring program stored in the memory and executable on the processor, wherein the road condition monitoring program, when executed by the processor, implements the steps of the road condition monitoring method as described in any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a road condition monitoring program, which, when executed by a processor, implements the steps of the road condition monitoring method as described in any one of claims 1-10.

Images