Road recognition method and device, electronic equipment, storage medium and program product

Abstract

The invention discloses a road recognition method and device, electronic equipment, a storage medium and a program product. The method relates to the field of data processing and automatic driving and comprises the steps that original boundary data of a target area are acquired, the original boundary data are used for representing the boundary shape of the target area, and the target area comprises at least one operation area and at least one road connected with different operation areas; triangulation is conducted on the original boundary data, an initial triangular mesh set is obtained, and triangular meshes in the initial triangular mesh set are used for representing at least one road; based on a preset filtering strategy corresponding to a preset type of road, filtering the initial triangular mesh set to obtain a target triangular mesh set; and clustering the target triangular mesh set, and determining a preset type of road in the at least one road. According to the invention, the technical problem of low road identification efficiency in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the fields of data processing and autonomous driving, and in particular to a road recognition method, device, electronic equipment, storage medium and program product. Background Art

[0002] In the field of autonomous driving, it is necessary to perform path planning and road recognition generation on the roads on which the vehicle is traveling in order to determine the road conditions for the vehicle's subsequent driving, thereby helping the vehicle make correct driving decisions and plan the driving path. Through road recognition generation, the autonomous driving system can more accurately understand the surrounding environment and improve driving safety and efficiency.

[0003] At present, the identification and generation of roads in related technologies usually requires professional and technical personnel to manually perform road marking and road data analysis. The types of road-related data that professional and technical personnel need to process are diverse and complex, the data volume is large, and the task volume is large, resulting in low efficiency in road identification.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a road recognition method, device, electronic device, storage medium and program product to at least solve the technical problem of low efficiency in road recognition in related technologies.

[0006] According to one aspect of an embodiment of the present invention, a road recognition method is provided, comprising: obtaining original boundary data of a target area, wherein the original boundary data is used to characterize the boundary shape of the target area, the target area includes at least one work area, and at least one road connected to different work areas; triangulating the original boundary data to obtain an initial triangular mesh set, wherein the triangular meshes in the initial triangular mesh set are used to characterize at least one road; filtering the initial triangular mesh set based on a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set; clustering the target triangular mesh set to determine a preset type of road in at least one road.

[0007] According to another aspect of an embodiment of the present invention, a road identification device is also provided, including: an acquisition module, used to acquire original boundary data of a target area, wherein the original boundary data is used to characterize the boundary shape of the target area, and the target area includes at least one working area, and at least one road connected to different working areas; a subdivision module, used to triangulate the original boundary data to obtain an initial triangular mesh set, wherein the triangular meshes in the initial triangular mesh set are used to characterize at least one road; a filtering module, used to filter the initial triangular mesh set based on a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set; and a clustering module, used to cluster the target triangular mesh set to determine a preset type of road in at least one road.

[0008] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present invention is executed when the program is running.

[0009] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0010] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0011] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0012] According to another aspect of the embodiments of the present invention, a computer program is further provided. When the computer program is executed by a processor, the methods in the embodiments of the present invention are implemented.

[0013] In an embodiment of the present invention, first, original boundary data for describing the boundary shape of a target area is obtained, which includes at least one operation area and at least one road connecting different operation areas; then, these original boundary data are triangulated to obtain an initial triangular mesh set; then, according to a preset filtering strategy corresponding to a preset type of road, the initial triangular mesh set is filtered to obtain a target triangular mesh set; finally, the target triangular mesh set is clustered to determine the position and shape of a road that meets the preset type. It is easy to notice that the present application can more accurately characterize the boundary shape of the target area by obtaining the original boundary data of the target area and performing triangulation, and can filter the initial triangular mesh set by a preset filtering strategy corresponding to a preset type of road to exclude non-road parts, thereby obtaining a target triangular mesh set, improving the accuracy of road recognition, and by performing triangulation and filtering operations on the original boundary data, the amount of data to be processed can be reduced, the complexity and time consumption of calculation can be reduced, and the target triangular mesh set can be clustered to more quickly determine the shape and type of the road, improve the efficiency of road recognition, improve the accuracy and efficiency of road recognition, and reduce the misjudgment rate, thereby solving the technical problem of low efficiency of road recognition in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0015] Figure 1 is a flow chart of a road recognition method according to an embodiment of the present invention;

[0016] Figure 2 is a flow chart of an optional road identification method according to an embodiment of the present invention;

[0017] Figure 3 is a schematic diagram of an optional road recognition result generated based on a road recognition method according to an embodiment of the present invention;

[0018] Figure 4 is a schematic diagram of a road recognition device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] According to one aspect of an embodiment of the present invention, a road recognition method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0022] Figure 1 is a flow chart of a road recognition method according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0023] Step S102, obtaining original boundary data of the target area.

[0024] The original boundary data is used to characterize the boundary shape of the target area, and the target area includes at least one working area and at least one road connecting different working areas.

[0025] The above-mentioned operation area may refer to a specific area where various types of vehicles such as engineering vehicles or mining vehicles need to operate when performing operations, and may refer to an area where excavation, transportation or other operations are required. The operation area may include excavation pits, loading areas, unloading areas, mining areas, etc. For example, in a mining scene, the operation area may refer to a specific area in the mining area where blasting, excavation, and ore loading are required. The operation area may be defined by actual terrain, obstacles, fences or marking lines. Various types of vehicles such as engineering vehicles or mining vehicles need to perform corresponding operations within these operation areas. The operation area may include narrow or wide areas, etc. The specific operation area can be determined according to actual needs and is not limited here.

[0026] The above-mentioned original boundary data may refer to the basic data of the boundary shape of the target area, which may include but is not limited to the geometric information and spatial relationships of geographical elements such as the outer boundary line, roads, rivers, buildings, etc. of the target area. The data type of the original boundary data may include point cloud data, terrain data, satellite image data, etc. Point cloud data may be a three-dimensional spatial data set composed of a large number of discrete points, which can be used to represent the terrain and surface shape of the target area. The specific type of the original boundary data can be determined according to actual needs and is not limited here.

[0027] In an optional embodiment, obtaining the original boundary data of the target area can be achieved in a variety of ways. Specifically, remote sensing technologies such as satellite images, aerial images, or drone images can be used to obtain image data of the target area; the boundary data of the target area can be obtained through field measurements and surveys. Measurement points can be set around the target area, and field measurements can be conducted to obtain accurate boundary data of the target area; a geographic information system can be used to obtain data such as topographic data, topographic maps, land use data, etc., and then obtain the boundary data of the target area. In actual applications, the above methods can also be combined to obtain the original boundary data of the target area for subsequent road identification and work area analysis. By obtaining accurate original boundary data, the spatial structure and shape characteristics of the target area can be better understood, thereby providing more accurate data support for engineering vehicle and mine car road identification.

[0028] Step S104, triangulating the original boundary data to obtain an initial triangular mesh set.

[0029] The triangular meshes in the initial triangular mesh set are used to represent at least one road.

[0030] The above-mentioned triangulation may refer to the process of connecting a set of points on a plane into a set of non-intersecting triangles according to certain rules. In the application of engineering vehicle and mining vehicle road identification, triangulation may be used to process terrain data for terrain analysis, terrain modeling, visualization and other applications. The basic principle of triangulation may be to ensure that the generated triangles do not intersect and are as close as possible to the distribution of the original data points. The triangulation algorithm may include Delaunay triangulation, minimum angle triangulation, etc. The specific method of triangulation may be determined according to actual needs and is not limited here.

[0031] In an optional embodiment, the original boundary data can be triangulated. Specifically, the original boundary data can be preprocessed, including data cleaning, denoising and correction, to ensure the accuracy and integrity of the data. The original boundary data can be divided into smaller areas or fragments for subsequent processing. Then, a suitable triangulation algorithm can be selected to process the segmented boundary data. According to the selected algorithm, a triangulation operation can be performed on each segmented boundary data area to obtain an initial triangular mesh set. The triangular mesh used to characterize the road can be screened from the initial triangular mesh set. The obtained triangular mesh may have specific features such as shape, size, angle, etc., which can be identified and extracted by image processing, feature extraction and other methods. The identified road triangular meshes can be connected to form a complete road path. Through the above steps, the original boundary data is triangulated and an initial triangular mesh set is obtained, which can provide data support for subsequent road recognition applications.

[0032] Step S106 , filtering the initial triangular mesh set based on a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set.

[0033] The above-mentioned preset types may refer to various pre-set road types. The preset types may be determined based on information such as road width, slope, road surface type, road curvature, real-time road conditions, etc., and may be determined based on actual needs. This is not limited here. In actual applications, the preset types of roads may be determined by users, or may be automatically determined by navigation and other software systems based on user driving needs. This is not limited here.

[0034] The above-mentioned preset filtering strategy may refer to a filtering strategy determined based on a preset type of road. For example, by setting a standard width range for mine roads, a triangular mesh set that meets the range can be screened out to exclude roads that are too narrow. The slope in the triangular mesh set can be screened according to the slope limit of the mine road to exclude roads that are too steep or too gentle. The road surface features in the triangular mesh set can be identified and filtered according to the road surface conditions of the mine road, such as mud, stones, etc., to screen out qualified roads. The above-mentioned multiple conditions can also be comprehensively considered. The preset filtering strategy can be determined according to actual needs and is not limited here.

[0035] In an optional embodiment, the initial triangular mesh set can be filtered based on the preset filtering strategy corresponding to the preset type of road to obtain the target triangular mesh set. Specifically, the preset filtering strategy corresponding to the preset type of road can be determined, and the preset filtering strategy can include road width, slope, curvature and other information; the initial triangular mesh set can be traversed to determine whether it meets the preset filtering strategy one by one. For each triangular mesh, it can be matched with the preset filtering strategy according to its attribute information, and the characteristic values ​​of each triangular network, such as length, angle, etc., can be calculated. If it meets the preset conditions, it will be retained to obtain the target triangular mesh set, otherwise it will be discarded; in the filtering process, image processing technology and machine learning algorithms can be combined for auxiliary judgment and screening to improve the accuracy and efficiency of filtering. After multiple screening and optimization, the target triangular mesh set obtained is the triangular mesh corresponding to the road that meets the preset type, which can be used for subsequent road recognition and analysis tasks. Through the implementation of the above steps, the initial triangular mesh set can be effectively filtered to obtain a road triangular mesh set that meets the preset type, providing a data basis for engineering vehicle and mining vehicle road recognition.

[0036] Step S108: clustering the target triangular mesh set to determine a preset type of road in at least one road.

[0037] In an optional embodiment, a clustering operation can be performed on the target triangular mesh set, and similar triangular networks can be classified into the same category. This can be achieved through a clustering algorithm, such as K-means clustering or hierarchical clustering. Clustering can group similar parts of the road together, so as to better identify the structure and type of the road. According to the clustering results, a preset type of road in at least one road can be determined. By analyzing and judging the triangular network after clustering, the type and structure of the road can be determined, such as a straight road, a curved road, etc., which can achieve accurate recognition and classification of the road. Through the above steps, the recognition and classification of the road can be effectively achieved, providing important support for the automatic navigation and driving of the vehicle.

[0038] In an embodiment of the present invention, first, original boundary data for describing the boundary shape of a target area is obtained, which includes at least one work area and at least one road connecting different work areas; then, these original boundary data are triangulated to obtain an initial triangular mesh set; then, the initial triangular mesh set is filtered according to a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set; finally, the target triangular mesh set is clustered to determine the position and shape of a road that meets the preset type to obtain the original boundary data of the target area, wherein the original boundary data is used to characterize the boundary shape of the target area, the target area includes at least one work area, and at least one road connecting different work areas; the original boundary data is triangulated to obtain an initial triangular mesh set, wherein the triangular meshes in the initial triangular mesh set are used to characterize at least one road; based on the preset filtering strategy corresponding to the preset type of road, the initial triangular mesh set is filtered to obtain a target triangular mesh set; the target triangular mesh set is clustered to determine a preset type of road in at least one road. It is easy to notice that the present application can more accurately characterize the boundary shape of the target area by acquiring the original boundary data of the target area and performing triangulation. By using the preset filtering strategy corresponding to the preset type of road, the initial triangular mesh set can be filtered to exclude the non-road part, thereby obtaining the target triangular mesh set, which improves the accuracy of road recognition. By performing triangulation and filtering operations on the original boundary data, the amount of data to be processed can be reduced, the complexity and time consumption of calculation are reduced, and the target triangular mesh set can be clustered to more quickly determine the shape and type of the road, thereby improving the efficiency of road recognition, improving the accuracy and efficiency of road recognition, and reducing the misjudgment rate, thereby solving the technical problem of low efficiency of road recognition in related technologies.

[0039] In the above embodiment of the present application, the original boundary data is triangulated to obtain an initial triangular mesh set, including: uniformly interpolating the original boundary data to obtain target boundary data; triangulating the target boundary data to obtain an initial triangular mesh set. Preferably, the interpolation resolution of the original boundary data is determined based on the wheel size of the working vehicle.

[0040] The above-mentioned wheel size may be the vehicle wheel diameter, radius, wheel width and other information of various vehicles such as engineering vehicles or mining vehicles. The specific representation type of the wheel size may be determined according to actual needs and is not limited here.

[0041] The above-mentioned interpolation resolution may refer to the density of data points for uniformly interpolating the original boundary data. When the interpolation resolution is high, the data points for uniformly interpolating the original boundary data are denser. When the interpolation resolution is low, the data points for uniformly interpolating the original boundary data are sparser. Preferably, the interpolation resolution of the original boundary data can be determined based on the wheel size of the working vehicle. Considering that when the vehicle wheel size is small, the accuracy of the road information is higher, a higher interpolation resolution can be set at this time; when the vehicle wheel size is large, the accuracy of the road information is lower, and a lower interpolation resolution can be set at this time to save computing resources. Through the above settings, the corresponding road accuracy can be effectively determined according to the actual situation of the vehicle to balance the contradiction between the vehicle road requirements and computing resources.

[0042] In an optional embodiment, the original boundary data can be uniformly interpolated, and the resolution of the interpolation can be determined according to the wheel size of the working vehicle, so that the target boundary data more suitable for the vehicle condition can be obtained. Then, the interpolated target boundary data can be triangulated to obtain an initial triangular mesh set. Triangulation can decompose complex boundary data into a series of simple triangular networks, which is convenient for subsequent analysis and processing. The initial triangular mesh set can be optimized according to actual conditions. Some algorithms and techniques can be used to adjust the size and shape of the triangle to ensure that the characteristics of the road can be accurately captured. Finally, the generated triangular mesh can be verified and adjusted to check whether there are errors or defects and correct them. Some auxiliary tools and software can be used to help verify and adjust to ensure the accuracy and integrity of the data. Triangulation of the original boundary data can comprehensively consider factors such as the shape, characteristics and size of the road. Through the above steps, accurate identification and analysis of engineering vehicle and mining vehicle roads can be effectively achieved.

[0043] In the above embodiment of the present application, based on the preset filtering strategy corresponding to the preset type of road, the initial triangular mesh set is filtered to obtain the target triangular mesh set, including: filtering the first triangular mesh located outside the target area in the initial triangular mesh set to obtain the intermediate triangular mesh set; determining the width of the road corresponding to the triangular mesh in the intermediate triangular mesh set; filtering the second triangular mesh in the intermediate triangular mesh set whose road width is greater than the road width threshold to obtain the target triangular mesh set. Preferably, the road width threshold is determined based on at least one of the following: the width of the working vehicle, the automatic driving accuracy of the working vehicle, and the safety index of the working vehicle.

[0044] The above-mentioned autonomous driving accuracy can be determined according to the actual operating scenario of the vehicle. For example, in a mining scenario, the autonomous driving accuracy of vehicles such as excavators performing loading and unloading operations can be greater than the autonomous driving accuracy of mine trucks traveling on wide roads. The autonomous driving accuracy can be determined according to actual needs and is not limited here.

[0045] The above safety indicators can be determined according to the actual operation scenarios of the vehicle, for example, different safety indicators can be determined according to the type of cargo loaded on the vehicle, such as ordinary cargo, hazardous chemicals, etc.; for another example, the safety indicators of school buses, passenger buses, etc. can be higher than those of other vehicles. The safety indicators can be determined according to actual needs and are not limited here.

[0046] In an optional embodiment, the initial triangular mesh set can be filtered based on a preset filtering strategy corresponding to a preset type of road. Specifically, the first triangular mesh in the initial triangular mesh set that is outside the target area can be filtered to obtain an intermediate triangular mesh set. The vertex coordinates of each triangle can be judged. If there is a vertex coordinate outside the target area, the triangle is removed from the set. Secondly, the width of the road corresponding to the triangular mesh in the intermediate triangular mesh set can be determined. The width of the road corresponding to the triangle can be found by calculating each triangle. The normal vector of the center line of the road can be used to calculate the width of the road, or the width can be determined by a preset road type. Next, the second triangular mesh in the intermediate triangular mesh set whose road width is greater than a road width threshold is filtered to obtain a target triangular mesh set. The road width corresponding to each triangle can be compared with a preset road width threshold. If the road width is greater than the threshold, the triangle is removed from the set. Finally, the road width threshold can be determined according to specific needs. The road width threshold can be determined based on the width of the working vehicle, the autonomous driving accuracy of the working vehicle, or the safety index of the working vehicle. For example, if the width of the working vehicle is large, the road width threshold can also be set larger to ensure that the working vehicle can pass through the road smoothly. The autonomous driving accuracy or safety index of the working vehicle can also be considered, and the road width threshold can be flexibly adjusted according to these indicators.

[0047] Through the above steps, the initial triangular mesh set can be filtered based on the preset filtering strategy corresponding to the preset type of road, and the target triangular mesh set that meets the requirements can be finally obtained by judging the position of the triangle, the road width, etc. Through the preset filtering strategy corresponding to the preset type of road, the triangular meshes outside the target area can be effectively filtered out, thereby reducing interference factors and improving the recognition accuracy of the target road. By filtering the triangular meshes whose road width is greater than the threshold, the triangular meshes that do not meet the requirements can be excluded, reducing the amount of calculation for subsequent processing and improving the efficiency of road recognition. The determination of the road width threshold can be based on factors such as the width of the operating vehicle, the accuracy of automatic driving and the safety index, so as to better adapt to the characteristics of the actual operating vehicle, ensure the accuracy of road recognition and ensure the safety of vehicle driving.

[0048] In the above embodiment of the present application, the method further includes: determining the position information of the incenter of the triangular mesh in the initial triangular mesh set; and determining whether the triangular mesh in the initial triangular mesh set is located in the target area based on the position information of the incenter.

[0049] In an optional embodiment, the initial triangular mesh set can be filtered to help determine which triangular meshes are located in the target area, thereby improving the accuracy and efficiency of road recognition. Specifically, the in-center position information of the triangular meshes in the initial triangular mesh set can be determined, and the in-center position information of each triangular mesh can be calculated. For each triangle, the coordinates of its three vertices are calculated, and the average of these three coordinates is obtained to obtain the in-center position of the triangle. The calculation of the in-center position can be solved by mathematical methods, such as obtaining the coordinates of the center of gravity of the triangle or using vector operations. Based on the in-center position information, it can be determined whether the triangular meshes in the initial triangular mesh set are located in the target area. Determining whether a triangle is located in the target area can be achieved by judging whether its in-center position is in the target area. Specifically, the boundary information of the target area can be determined, such as the coordinates of the four vertices of the rectangular area. For the in-center position of each triangle, judging whether it is in the target area can be achieved by calculating the distance from the in-center position to each boundary in the target area and judging whether it is less than or equal to zero. If the in-center position is in the target area, the triangle is added to the target triangular mesh set. By implementing the above steps, it is possible to determine whether the triangle meshes in the initial triangle mesh set are located in the target area, thereby determining which triangles are located in the target area. The above process can also be implemented through programming, using computers to perform large-scale data processing and screening to improve the accuracy and efficiency of road recognition.

[0050] In the above embodiment of the present application, determining the width of the road corresponding to the triangular mesh in the intermediate triangular mesh set includes: obtaining the lengths of multiple target edges of the triangular mesh in the intermediate triangular mesh set; determining the average value of the lengths of the multiple target edges to obtain the width of the road.

[0051] The above-mentioned multiple target edges may refer to two longer edges of a triangle in the triangulated network. The multiple target edges may also be determined according to actual needs, which is not limited here.

[0052] In an optional embodiment, the average value of the lengths of the multiple target edges of the triangular meshes in the intermediate triangular mesh set can be determined to obtain the width of the road. For example, the lengths of the two long sides of the triangle can be calculated, and the method of fitting the local road width with the mean value can be used to obtain the road width. Specifically, each triangular mesh can be traversed and the lengths of its three sides can be calculated. When obtaining the lengths of the multiple target edges of the triangular meshes in the intermediate triangular mesh set, the two long sides of each triangular mesh can be selected as the target sides. These two long sides are usually perpendicular to the direction of the road. By calculating the lengths of these two long sides, the road width of each triangular mesh can be obtained. The road width of each triangular mesh is averaged, so that the width of the entire road can be obtained. Determining the average value of the lengths of multiple target edges can help estimate the width of the road more accurately, thereby providing a better reference for the path planning of the vehicle. Finally, the width information of the road can be applied to the autonomous navigation system of the engineering vehicle or the mining vehicle to help the vehicle avoid obstacles and plan the optimal path. In the above manner, the rapid and accurate acquisition of the road width can be achieved, and the driving efficiency and safety of the vehicle can be improved.

[0053] In the above embodiment of the present application, clustering the target triangular mesh set to determine a preset type of road in at least one road includes: clustering the triangular meshes in the target triangular mesh set based on the spatial position information of the triangular meshes in the target triangular mesh set to obtain at least one initial polygon, wherein each initial polygon is used to represent a preset type of road; filtering at least one initial polygon based on an area filtering parameter to obtain at least one target polygon; and determining that the road corresponding to the at least one target polygon is a preset type of road. Preferably, the area filtering parameter is determined based on the length and width of the work vehicle.

[0054] In an optional embodiment, the target triangular mesh set can be clustered to determine the preset type of road in at least one road. The triangular meshes in the target triangular mesh set can be clustered based on the spatial position information of the triangular meshes in the target triangular mesh set to obtain at least one initial polygon, wherein each initial polygon is used to represent a preset type of road. The triangular meshes can be clustered according to the spatial position information of the triangular meshes, and each cluster obtained by clustering can represent an initial polygon, that is, a preset type of road. Then, at least one initial polygon can be filtered based on an area filtering parameter to obtain at least one target polygon. The area filtering parameter can be determined according to the length and width of the working vehicle, and is used to filter out the initial polygons with too small or too large areas. By setting appropriate area filtering parameters, the target polygons that meet the requirements can be screened out, that is, representing the preset type of road. Finally, it can be determined that the road corresponding to at least one target polygon is a preset type of road. For each target polygon, its corresponding road type can be determined according to its spatial position information and shape characteristics. The map data or road database can be used to verify whether the road corresponding to the target polygon is a preset type of road.

[0055] Through the above steps, the target triangular mesh set can be clustered to determine the preset type of roads in at least one road, and further filter out the target polygons that meet the requirements, thereby realizing the process of road recognition. By clustering the target triangular mesh set, the preset type of roads in at least one road can be effectively determined, thereby improving the accuracy of road recognition. By performing area filtering on at least one target polygon, the recognition result of the road type can be further optimized to ensure that the recognized road meets the preset type. The area filtering parameters are determined based on the length and width of the working vehicle. The filtering parameters can be adjusted according to the specific characteristics of the working vehicle to adapt to the needs of different working vehicles and improve the applicability and practicality of road recognition. By clustering, filtering and determining the preset type of roads for the target triangular mesh set, the accuracy, efficiency and applicability of road recognition can be improved, thereby providing better technical support for the technical development in the field of autonomous driving and road recognition.

[0056] In the above embodiments of the present application, the original boundary data is generated based on an original map of the target area or measurement data of the target area.

[0057] In an optional embodiment, the original boundary data of the target area can be obtained by obtaining the original map of the target area or field measurement data, and the boundary information of the target area, including important data such as the shape, size, and position of the road, can be obtained, which is convenient for accurately identifying the road and the surrounding environment, thereby providing accurate navigation and driving paths for the autonomous driving vehicle. Collecting the original map data, the map data of the target area, including information such as roads and buildings, can be obtained through satellite maps, aerial images, etc. Field measurement data can be performed, including the use of laser radar, cameras and other equipment to obtain real-time data of the target area, and the collected data can be processed and integrated to generate the original boundary data of the target area, including information such as the geometry of the road, the location of the intersection, and the number of lanes. By obtaining the original boundary data of the target area, the surrounding environment can be better understood, and important information such as roads, lanes and traffic signs can be identified, thereby providing support for autonomous navigation and safe driving of the vehicle.

[0058] In the above embodiment of the present application, the method also includes: compiling the target polygons corresponding to the preset type of roads to generate a map layer corresponding to the preset type of roads; updating the original map of the target area based on the map layer corresponding to the preset type of roads to obtain an updated map.

[0059] In an optional embodiment, target polygons for preset types of roads can be compiled. Preset types of roads can be determined, and target polygons for these roads can be compiled using map data and image recognition technology to generate target polygons corresponding to each road type. These target polygons can include road boundaries, lane lines, intersections and other information. Then, based on the generated target polygons, they can be matched with the original map data to generate map layers corresponding to the preset types of roads. These map layers can include information such as the shape, length, width, and traffic signs of the road, providing basic data for subsequent road identification and navigation. Finally, the original map can be updated based on the map layers corresponding to the preset types of roads. The generated map layers can be compared with the original map, and the parts of the original map corresponding to the preset types of roads can be updated. Through the updated map, the road where the vehicle is located can be more accurately identified and tracked, thereby improving the positioning accuracy and path planning capabilities of the autonomous driving system. Through the above-mentioned processes such as determining the road type, compiling the target polygon, generating the map layer and updating the original map, accurate road information can be provided to the autonomous driving system, improving the performance and safety of the system. By updating the original map based on the map layer corresponding to the preset type of road, the map information can be updated faster, making the map information more timely and accurate, thereby improving the autonomous driving system's perception of the road environment, improving the accuracy, stability and reliability of the system, and reducing the system's demand for computing resources, providing better support for the development and application of autonomous driving technology.

[0060] The technical solution proposed in this application is explained below in combination with an optional application scenario. The technical solution proposed in this application can be applied to the automatic update process of the narrow road section layer of the high-precision map of the open-pit mine. Specifically, an automatic identification algorithm for narrow road sections of high-precision maps based on the Deloney triangulation network is proposed, and the narrow road section layer is updated. The automatic and accurate identification of narrow road sections is realized in the map update link, and the results can be automatically sent to the vehicle side through the high-precision map. The triangulation network of the mining area roads can be constructed based on the Deloney triangulation algorithm to form a coherent and covered road triangulation structure. Then, the filtering and clustering algorithms are used to group and cluster the triangles that meet the conditions to form polygons of narrow road sections. Finally, based on the automatic update pipeline of the high-precision map, the narrow road sections can be updated and sent as map layers in real time to assist the vehicle-side planning and decision-making, which greatly reduces the vehicle-side perception decision-making computing power requirements, and significantly improves the path planning and decision-making control capabilities of the autonomous driving system in the complex environment of the open-pit mine, and improves the safety and efficiency of the open-pit mine autonomous driving operation.

[0061] The specific implementation process of the method for automatically updating the narrow road section map layer proposed in the present application may include the following, where the narrow road section here may refer to a road with an identified width below a threshold.

[0062] First, the road boundaries of the driving area can be input. The boundary data of the road, loading area, and unloading area can be input based on a high-precision map. The boundary geometry data of the target area can also be input based on a Computer-Aided Design (CAD) drawing of the driving area or other surveying and mapping data to clearly describe the boundary shape of the road as the initial data for identifying narrow sections in the area.

[0063] Next, the Delaunay triangulation can be constructed. The Delaunay triangulation is a triangulation method in computational geometry. It has the characteristics of maximum and minimum angles, that is, the generated triangles are as close to equilateral triangles as possible, reducing the appearance of sharp angles, and can describe the geometric shape well. When the input points remain unchanged, the construction result of the Delaunay triangulation has the characteristic of uniqueness. First, the original boundary data is uniformly encrypted. Since the boundary original data in the above steps describes the geometric shape based on curvature interpolation, the points are dense in places with large curvature and sparse in places with small curvature, especially on long straight lines, there may be only two points at the beginning and end. In order to ensure the uniform construction of the triangulation network, the original boundary data can be uniformly encrypted. The encryption resolution can be directly related to the wheel, generally set to 1~2 times the wheel. For example, the wheel diameter of this solution is 1.5 meters, and the encryption resolution is set to 2.0 meters. Too large will affect the accuracy of the result, and too small will affect the efficiency of the algorithm. The uniformly encrypted point set is subjected to Delaunay triangulation to obtain the Delaunay triangulation network. The above-mentioned point density may refer to that more points are inserted, that is, the distance between the points is small; the point sparseness may refer to that more points are inserted, that is, the distance between the points is large.

[0064] Next, the triangulated network can be filtered, and the triangles outside the boundary of the driving area can be filtered. The operating area of ​​an open-pit mine is generally an irregular complex polygon. The Delony triangulated network obtained above contains many triangles outside the area. By calculating the incenter of the triangle, that is, the intersection of the angle bisectors, it is determined whether the point is within the boundary polygon, and all triangles outside the area are filtered out. Secondly, the lane area triangles that meet the width requirements are filtered and retained. The judgment of narrow sections is related to the width of the vehicle body. It is necessary to ensure that there are two lanes in opposite directions and a certain buffer area. The size depends on the accuracy and safety requirements of autonomous driving. This solution is based on a vehicle width of 5.8 meters. The narrow section width threshold is set to 15 meters. The length of the two long sides of the triangle is calculated, and its mean is used to fit the local road width. The triangles exceeding the threshold of 15 meters are filtered and discarded.

[0065] Next, polygons can be generated based on triangulated network clustering, and the remaining triangles can be spatially clustered, and the co-edge triangles can be merged to eventually generate several polygons. Each polygon represents a potential narrow road section. Finally, polygons with too small an area due to irregular road boundaries or boundary corners are filtered out. The area filtering parameter is based on the coverage area of ​​the vehicle model, that is, the vehicle length * vehicle width setting, which can be set to about twice the area. The area threshold affects the filtering effect, and ultimately accurate and clear narrow road section polygons are obtained.

[0066] Finally, the narrow road section map layer can be updated, and the narrow road section polygons obtained in the above steps can be compiled into a high-precision map narrow road section layer through map compilation, which can be sent to the vehicle through the high-precision map automatic update pipeline to assist the vehicle-side path planning and decision-making control in real time, reduce the vehicle-side perception and decision-making computing pressure, and improve the safety and efficiency of autonomous driving.

[0067] The technical solution proposed in this application can achieve high-precision identification of narrow sections, and utilize the spatial subdivision advantages and intelligent clustering of the Delaunay triangulation network to achieve high-precision identification of narrow sections, ensuring the path planning and decision-making control of the autonomous driving system in the complex environment of open-pit mines. The algorithm has strong robustness and can cope with the complex environment of different terrains and road conditions in open-pit mines to ensure the stability of the output results. It has good real-time performance, relies on the high efficiency of the recognition algorithm and the automated update process of high-precision maps, and is suitable for real-time updates of narrow section layers of large-scale road network maps in open-pit mines, and supports real-time auxiliary vehicle-side path planning and decision-making control.

[0068] Figure 2 is a flow chart of an optional road recognition method according to an embodiment of the present invention. Figure 2 As shown, it includes: S202, obtaining original boundary data of the target area; S204, triangulating the original boundary data to obtain an initial triangular mesh set; S206, filtering the initial triangular mesh set based on a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set; S208, clustering the target triangular mesh set to determine a preset type of road in at least one road; S210, compiling target polygons corresponding to the preset type of road to generate a map layer corresponding to the preset type of road; updating the original map of the target area based on the map layer corresponding to the preset type of road to obtain an updated map.

[0069] Figure 3 is a schematic diagram of an optional road recognition result generated based on a road recognition method according to an embodiment of the present invention, such as Figure 3As shown, the road recognition results generated based on the road recognition method are shown. The figure includes polygons obtained by clustering triangular meshes. The quadrilateral containing multiple triangular meshes in the figure can represent a certain preset type of road.

[0070] According to another aspect of an embodiment of the present invention, a road recognition device is further provided, which can execute the road recognition method of the above embodiment. The specific implementation method and preferred application scenario are the same as those of the above embodiment and will not be described in detail here.

[0071] Figure 4 is a schematic diagram of a road recognition device according to an embodiment of the present application, such as Figure 4 As shown, the device includes the following: an acquisition module 402 , a segmentation module 404 , a filtering module 406 and a clustering module 408 .

[0072] Among them, the acquisition module is used to obtain the original boundary data of the target area, wherein the original boundary data is used to characterize the boundary shape of the target area, and the target area includes at least one working area and at least one road connected to different working areas; the subdivision module is used to triangulate the original boundary data to obtain an initial triangular mesh set, wherein the triangular meshes in the initial triangular mesh set are used to characterize at least one road; the filtering module is used to filter the initial triangular mesh set based on a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set; the clustering module is used to cluster the target triangular mesh set to determine a preset type of road in at least one road.

[0073] The segmentation module is further used to uniformly interpolate the original boundary data to obtain target boundary data; triangulate the target boundary data to obtain an initial triangular mesh set; preferably, the interpolation resolution of the original boundary data is determined based on the wheel size of the working vehicle.

[0074] Among them, the filtering module is also used to filter the first triangular mesh in the initial triangular mesh set that is located outside the target area to obtain an intermediate triangular mesh set; determine the width of the road corresponding to the triangular mesh in the intermediate triangular mesh set; filter the second triangular mesh in the intermediate triangular mesh set whose road width is greater than the road width threshold to obtain a target triangular mesh set; preferably, the road width threshold is determined based on at least one of the following: the width of the operating vehicle, the automatic driving accuracy of the operating vehicle, and the safety index of the operating vehicle.

[0075] The acquisition module is further used to determine the position information of the incenter of the triangular mesh in the initial triangular mesh set; based on the position information of the incenter, determine whether the triangular mesh in the initial triangular mesh set is located in the target area.

[0076] The filtering module is further used to obtain the lengths of multiple target edges of the triangular meshes in the intermediate triangular mesh set; determine the average value of the lengths of the multiple target edges to obtain the width of the road.

[0077] Among them, the clustering module is also used to cluster the triangular meshes in the target triangular mesh set based on the spatial position information of the triangular meshes in the target triangular mesh set to obtain at least one initial polygon, wherein each initial polygon is used to represent a road of a preset type; filter at least one initial polygon based on an area filtering parameter to obtain at least one target polygon; determine that the road corresponding to the at least one target polygon is a road of the preset type; preferably, the area filtering parameter is determined based on the length and width of the work vehicle.

[0078] The original boundary data is generated based on an original map of the target area or measurement data of the target area.

[0079] Among them, the acquisition module is also used to compile the target polygon corresponding to the preset type of road to generate a map layer corresponding to the preset type of road; based on the map layer corresponding to the preset type of road, the original map of the target area is updated to obtain an updated map.

[0080] An embodiment of the present application further provides an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of the present invention when running.

[0081] The above-mentioned memory may refer to a device inside a computer for storing data and programs, and may include memory, hard disk, etc., wherein the memory may be used to temporarily store running programs and data, the hard disk may be used to store programs and data for a long time, and the memory may be used to enable the computer to read and write data, and execute programs; the above-mentioned processor may be responsible for executing instructions in computer programs and performing data processing, and may be responsible for controlling and executing various operations, including arithmetic operations, logical operations, data transmission, etc.

[0082] An embodiment of the present application further provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0083] The above-mentioned computer storage medium may refer to a medium in a computer memory used to store certain discontinuous physical quantities. Computer storage media mainly include semiconductors, magnetic cores, magnetic drums, magnetic tapes, laser disks, etc.; the stored program included in the computer-readable storage medium may be a set of instructions that can be recognized and executed by a computer, running on an electronic computer, and is an information tool that meets certain needs of people.

[0084] An embodiment of the present application further provides a computer program product, including a computer program, which implements the methods in various embodiments of the present invention when executed by a processor.

[0085] The above-mentioned computer program product may refer to a software program that has been written, tested and released and can be run on a computer or other device. The computer program product may include an application, an operating system, tool software, etc., which is used to implement specific functions or solve specific problems.

[0086] An embodiment of the present application further provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0087] The above-mentioned non-volatile computer-readable storage medium may refer to a medium for storing data. The non-volatile computer-readable storage medium can keep the data from being lost when the power is off, and can be used to store long-term data, such as operating systems, applications, and user files. The non-volatile storage medium may include hard disk drives, solid-state drives, optical disks, and flash memory storage devices, etc.

[0088] The embodiments of the present application further provide a computer program, which implements the methods in the above-mentioned embodiments of the present invention when executed by a processor.

[0089] The above-mentioned computer program may refer to a collection of instructions used to tell a computer to perform a specific task or operation. A computer program may be written by a programmer using a specific programming language and may include algorithms, data structures, logic, and control flows. A computer program may be used for a variety of purposes, including application software, operating systems, and the like.

[0090] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0091] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0093] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0094] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc. Various media that can store program codes.

[0095] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A road recognition method, characterized in that: include: Acquire original boundary data of a target area, wherein the original boundary data is used to characterize a boundary shape of the target area, and the target area includes at least one working area and at least one road connecting different working areas; Triangulating the original boundary data to obtain an initial triangular mesh set, wherein the triangular meshes in the initial triangular mesh set are used to represent the at least one road; Based on a preset filtering strategy corresponding to a preset type of road, the initial triangular mesh set is filtered to obtain a target triangular mesh set; The target triangular mesh set is clustered to determine the preset type of road in the at least one road.

2. The method according to claim 1, characterized in that The triangulating the original boundary data to obtain an initial triangular mesh set includes: Performing uniform interpolation on the original boundary data to obtain target boundary data; Triangulating the target boundary data to obtain the initial triangular mesh set; Preferably, the interpolation resolution of the original boundary data is determined based on the wheel size of the working vehicle.

3. The method according to claim 1 or 2, characterized in that: The preset filtering strategy corresponding to the preset type of road is used to filter the initial triangular mesh set to obtain a target triangular mesh set, including: Filtering the first triangular mesh located outside the target area in the initial triangular mesh set to obtain an intermediate triangular mesh set; Determine the width of the road corresponding to the triangular mesh in the intermediate triangular mesh set; Filtering the second triangular meshes in the intermediate triangular mesh set whose width of the road is greater than a road width threshold to obtain the target triangular mesh set; Preferably, the road width threshold is determined based on at least one of the following: the width of the working vehicle, the automatic driving accuracy of the working vehicle, and the safety index of the working vehicle.

4. The method according to claim 3, characterized in that: The method further comprises: Determine the position information of the center points of the triangle meshes in the initial triangle mesh set; Based on the position information of the incenter, it is determined whether the triangle meshes in the initial triangle mesh set are located in the target area.

5. The method according to claim 3, characterized in that: The determining the width of the road corresponding to the triangular mesh in the intermediate triangular mesh set includes: Obtaining the lengths of multiple target edges of the triangular meshes in the intermediate triangular mesh set; An average value of the lengths of the plurality of target edges is determined to obtain the width of the road.

6. The method according to any one of claims 1 to 5, characterized in that The clustering of the target triangular mesh set to determine the preset type of road in the at least one road includes: Clustering the triangular meshes in the target triangular mesh set based on spatial position information of the triangular meshes in the target triangular mesh set to obtain at least one initial polygon, wherein each initial polygon is used to represent a road of the preset type; Filtering the at least one initial polygon based on an area filtering parameter to obtain at least one target polygon; Determining that a road corresponding to the at least one target polygon is a road of the preset type; Preferably, the area filtering parameter is determined based on the length and width of the work vehicle.

7. The method according to any one of claims 1 to 6, characterized in that: The original boundary data is generated based on an original map of the target area or measurement data of the target area.

8. The method according to any one of claims 1 to 7, characterized in that The method further comprises: Compiling the target polygon corresponding to the preset type of road to generate a map layer corresponding to the preset type of road; The original map of the target area is updated based on the map layer corresponding to the road of the preset type to obtain an updated map.

9. A road recognition device, characterized in that: include: An acquisition module, used for acquiring original boundary data of a target area, wherein the original boundary data is used for characterizing the boundary shape of the target area, and the target area includes at least one operation area and at least one road connected with different operation areas; A triangulation module, configured to triangulate the original boundary data to obtain an initial triangular mesh set, wherein the triangular meshes in the initial triangular mesh set are used to represent the at least one road; A filtering module, configured to filter the initial triangular mesh set based on a preset filtering strategy corresponding to a preset type of road to obtain a target triangular mesh set; A clustering module is used to cluster the target triangular mesh set to determine the preset type of road in the at least one road.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 8 when running.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 8.

12. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images