Road obstacle detection method, maintenance method, system, device, and medium

Abstract

The application discloses a kind of detection method of road obstacle, maintenance method, system, equipment and medium, the detection method includes: obtaining the point cloud data of the road;From the point cloud data, the road surface point cloud of the road and the point cloud to be detected are separated, and the road boundary line of the road is determined according to the road surface point cloud;The road boundary line is fitted to obtain the simulated road boundary line;According to the simulated road boundary line and road parameter, the obstacle-free pipeline of the road is generated, and the point cloud to be detected in the obstacle-free pipeline is obtained to obtain the information of road obstacle.The detection method of the application realizes the automatic detection of the obstacle on the road that obstructs the field of view, improves the detection efficiency and accuracy of the field of view obstacle, simultaneously constructs the obstacle-free pipeline of different height and shape based on the difference of road, meets the detection needs of different roads, and improves the detection accuracy of the field of view obstacle.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, maintenance method, system, equipment and medium for detecting road obstacles. Background Technology

[0002] In recent years, my country's transportation development has been rapid, placing higher demands on infrastructure construction. Roadside facilities in cities, such as street trees, billboards, and bushes, often become obstacles obstructing traffic signs due to unreasonable distribution and improper management and maintenance, affecting drivers' ability to read information and creating certain traffic safety hazards. Improperly distributed obstacles can create blind spots for drivers, leading to misjudgments of the surrounding environment.

[0003] Traditionally, the analysis and judgment of whether visual obstructions will cause traffic safety hazards generally rely on manual on-site visual measurement. However, the existing methods for analyzing obstructions encroaching on unobstructed pipelines are not only inefficient and dependent on manual visual judgment, but also have low accuracy and may even affect traffic safety. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art, which relies on manual visual measurement of obstacles on the road that obstruct the driver's vision, resulting in low efficiency and low accuracy. The invention provides a method, maintenance method, system, equipment and medium for detecting road obstacles.

[0005] The present invention solves the above-mentioned technical problems through the following technical solution:

[0006] This invention provides a method for detecting road obstacles, the method comprising:

[0007] Obtain the point cloud data of the road;

[0008] The road surface point cloud and the point cloud to be detected are separated from the point cloud data, and the road boundary line of the road is determined based on the road surface point cloud.

[0009] Fit the road boundary line to obtain a simulated road boundary line;

[0010] Based on the simulated road boundary line and road parameters, an accessible pipeline of the road is generated, and the point cloud to be detected within the accessible pipeline is obtained to obtain information about road obstacles.

[0011] Preferably, the step of separating the road surface point cloud and the point cloud to be detected from the point cloud data includes:

[0012] The road surface point cloud is determined based on the changes in the geometric features of the point cloud data, and other point cloud data in the point cloud data are identified as point clouds to be detected.

[0013] Preferably, the step of determining the road boundary line of the road based on the road surface point cloud includes:

[0014] Project the road surface point cloud onto a two-dimensional plane;

[0015] The road boundary line of the road is determined based on the contour features of the set of all road surface point clouds on the two-dimensional plane.

[0016] Preferably, the step of fitting the road boundary line to obtain a simulated road boundary line includes:

[0017] The road boundary line is divided into several sections;

[0018] The fitted curve for each segment is obtained by curve fitting;

[0019] At preset intervals, a simulated road boundary point is obtained based on the fitted curve, and all the simulated road boundary points are connected to obtain the simulated road boundary line.

[0020] Preferably, the road parameters include road grade and traffic standards; the step of generating the accessible pipeline of the road based on the simulated road boundary line and road parameters includes:

[0021] Using the symmetrical points on the boundary line of the simulated road as the bottom points, the cross-section of the barrier-free pipeline is constructed based on the bottom points and the corresponding top points; the distance between the bottom points and the top points is determined according to the road grade and traffic standards of the road.

[0022] The barrier-free pipeline is constructed based on several adjacent cross-sections;

[0023] The barrier-free pipeline is obtained by connecting several adjacent pipeline segments.

[0024] The present invention also provides a method for maintaining road obstacles, the method comprising:

[0025] Information on road obstacles is obtained using the road obstacle detection method described above, in order to update the information on obstacles requiring maintenance;

[0026] Road obstacles are maintained based on the updated maintenance obstacle information.

[0027] The present invention also provides a road obstacle detection system, the detection system comprising:

[0028] A point cloud data acquisition module is used to acquire point cloud data of the road.

[0029] A boundary line generation module is used to separate the road surface point cloud and the point cloud to be detected from the point cloud data, and to determine the road boundary line of the road based on the road surface point cloud.

[0030] A boundary line optimization module is used to fit the road boundary line to obtain a simulated road boundary line;

[0031] The pipeline generation module is used to generate barrier-free pipelines for the road based on the simulated road boundary lines and road parameters.

[0032] The obstacle detection module is used to acquire the point cloud to be detected within the barrier-free pipeline to obtain information about road obstacles.

[0033] Preferably, the boundary line generation module is specifically used to determine the road surface point cloud based on the changes in the geometric features of the point cloud data, and to determine other point cloud data in the point cloud data as point clouds to be detected.

[0034] Preferably, the boundary line generation module is specifically used to project the road surface point cloud onto a two-dimensional plane;

[0035] The boundary line generation module is specifically used to determine the road boundary line of the road based on the contour features of the set of all road surface point clouds on the two-dimensional plane.

[0036] Preferably, the boundary line optimization module is specifically used to divide the road boundary line into several segments;

[0037] The boundary line optimization module is specifically used to obtain a fitted curve for each segment by curve fitting.

[0038] The boundary line optimization module is specifically used to obtain a simulated road boundary point at preset intervals based on the fitted curve, and connect all the simulated road boundary points to obtain the simulated road boundary line.

[0039] Preferably, the road parameters include road grade and traffic standards; the pipeline generation module is specifically used to construct the cross-section of the barrier-free pipeline based on the bottom point and the corresponding top point on the boundary line of the simulated road, using the symmetrical point on the boundary line of the simulated road as the bottom point; the distance between the bottom point and the top point is determined according to the road grade and traffic standards of the road.

[0040] The pipeline generation module is specifically used to construct pipeline segments of the barrier-free pipeline based on several adjacent cross-sections.

[0041] The pipeline generation module is specifically used to connect several adjacent pipeline segments to obtain the barrier-free pipeline.

[0042] The present invention also provides a road obstacle maintenance system, the maintenance system comprising:

[0043] The obstacle information update module is used to obtain information about road obstacles using the road obstacle detection system described above, so as to update the information about obstacles to be maintained;

[0044] The maintenance instruction generation module is used to generate instructions for maintaining road obstacles based on the updated maintenance obstacle information.

[0045] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the road obstacle detection method or the road obstacle maintenance method as described above.

[0046] The present invention also provides a computer-readable medium having computer instructions stored thereon, which, when executed by a processor, implement the road obstacle detection method or the road obstacle maintenance method as described above.

[0047] The positive and progressive effects of this invention are as follows:

[0048] The road obstacle detection method of the present invention separates the road surface point cloud and the point cloud to be detected from the road point cloud data, obtains a simulated road boundary line based on the road surface point cloud, generates an unobstructed pipeline of the road based on the simulated road boundary line and road parameters, and obtains the point cloud to be detected within the unobstructed pipeline to obtain information on road obstacles. This realizes the automatic detection of obstacles that obstruct the view on the road, improves the detection efficiency and accuracy of visual obstacle detection, and at the same time, constructs unobstructed pipelines of different heights and shapes based on the differences in roads to meet the detection needs of different roads and improve the detection accuracy of visual obstacle detection. Attached Figure Description

[0049] Figure 1 This is the first flowchart of the road obstacle detection method in Embodiment 1 of the present invention.

[0050] Figure 2 This is the second flowchart of the road obstacle detection method in Embodiment 1 of the present invention.

[0051] Figure 3 This is a schematic diagram of the method for determining road boundary lines in Embodiment 1 of the present invention.

[0052] Figure 4 This is a schematic diagram of the first configuration of the barrier-free pipeline in this embodiment of the present invention.

[0053] Figure 5 This is a schematic diagram of the second configuration of the barrier-free pipeline in Embodiment 1 of the present invention.

[0054] Figure 6 This is a schematic diagram of the third configuration of the barrier-free pipeline in Embodiment 1 of the present invention.

[0055] Figure 7 This is a schematic diagram of the fourth configuration of the barrier-free pipeline in Embodiment 1 of the present invention.

[0056] Figure 8 This is a schematic diagram of the fifth configuration of the barrier-free pipeline in Embodiment 1 of the present invention.

[0057] Figure 9 This is a schematic diagram of the obstacle detection effect in Embodiment 1 of the present invention.

[0058] Figure 10 This is a schematic diagram of the road obstacle detection system in Embodiment 3 of the present invention.

[0059] Figure 11 This is a schematic diagram of the hardware structure of the electronic device in Embodiment 5 of the present invention. Detailed Implementation

[0060] The present invention will be further illustrated by way of embodiments below, but the present invention is not limited to the scope of the embodiments described herein.

[0061] Example 1

[0062] Please refer to Figure 1 This is the first flowchart of the road obstacle detection method in this embodiment.

[0063] Specifically, such as Figure 1 As shown, the detection method includes:

[0064] S101. Obtain the point cloud data of the road.

[0065] In one alternative implementation, a mobile laser scanning system can be used to simultaneously acquire trajectory data and laser scan data. The mobile laser scanning system may include GNSS (Global Navigation Satellite System) equipment, a panoramic camera, and a laser sensor. The mobile laser scanning system is mounted on a moving vehicle (e.g., a surveying vehicle). As the vehicle travels along the road, the GNSS equipment acquires trajectory data, the panoramic camera obtains color information, and the laser sensor acquires laser scan data of the road surface. Point cloud calculations are then used to obtain urban road point cloud data containing information such as three-dimensional coordinates (X, Y, Z), color (R, G, B), intensity, and category attributes.

[0066] S102. Separate the road surface point cloud and the point cloud to be detected from the point cloud data, and determine the road boundary line based on the road surface point cloud.

[0067] Specifically, the point cloud acquired from mobile laser scanning data includes not only road point cloud data but also point cloud data of buildings, pedestrians, vehicles, etc. This latter type needs to be excluded, and only the road surface point cloud needs to be extracted for subsequent calculations to reduce the complexity of the data processing. Furthermore, the road surface point clouds located on the road boundary lines are identified and extracted from all the road surface point clouds to determine the road boundary lines.

[0068] S103. Fit the road boundary line to obtain the simulated road boundary line; specifically, due to the obstruction of the road surface by obstacles such as pedestrians and vehicles, the road boundary line needs to be optimized.

[0069] S104. Generate an accessible pipeline for the road based on the simulated road boundary line and road parameters, and obtain the point cloud to be detected within the accessible pipeline to obtain information on road obstacles.

[0070] Please refer to Figure 2 This is the second flowchart of the road obstacle detection method in this embodiment.

[0071] Specifically, such as Figure 2 As shown, in one optional implementation, step S102 includes:

[0072] S1021. Determine the road surface point cloud based on the changes in the geometric features of the point cloud data, and identify other point cloud data in the point cloud data as the point cloud to be detected.

[0073] Specifically, since mobile laser scanning data consists of multiple scan lines, and each scan line reflects changes in road elevation, and road boundaries vary, abrupt changes in elevation or angle can occur along these boundaries. Therefore, the elevation and angle changes of each scan line in the road point cloud data can be calculated separately. If no abrupt changes occur in elevation or angle, the point cloud is identified as a road surface point cloud; otherwise, it is considered a point cloud to be detected.

[0074] In another optional implementation, other point cloud data may include point cloud data of pedestrians, vehicles, buildings, roadside trees, etc. Among them, pedestrians and vehicles that are temporarily traveling together are not considered obstacle detection objects, while buildings and their appurtenances, roadside trees, and other fixed facility obstacles are considered obstacle detection objects. Therefore, point cloud data of the road can be acquired at different times. Other point cloud data that appears repeatedly and excludes the road surface point cloud can be identified as point cloud data of fixed facility obstacles, which are the point clouds to be detected.

[0075] In one optional implementation, step S102 includes:

[0076] S1022. Project the road surface point cloud onto a two-dimensional plane; specifically, project the three-dimensional road surface point cloud, after removing the Z-axis elevation information, onto a two-dimensional plane.

[0077] S1023. Determine the road boundary line based on the contour features of the set of all road surface point clouds on the two-dimensional plane. The steps are as follows: Figure 3 As shown.

[0078] Specifically, suppose a sphere of radius r rolls around a two-dimensional plane of the road surface point cloud. Since this sphere only rolls along the two-dimensional boundary line of the road point cloud, the trajectory of the sphere is the boundary of the road, and its radius r is calculated as follows:

[0079]

[0080] In Equation (1), n ​​is the number of point clouds on the road surface, and l is the distance between two two-dimensional plane points. The significance of Equation (1) is to determine the radius of the rolling sphere by the average distance between all two-dimensional plane points, and then obtain the final road boundary line through this sphere.

[0081] In another alternative implementation, boundary points extracted from a finite unstructured road point cloud can be used with a value γ. Specifically, to select an optimal γ, a random point and its x nearest neighbors are chosen. Then, the average distance between the random point and its nearest neighbors is calculated. The x random points and their average distance are calculated in the same way. Finally, the optimal γ value is defined as the reciprocal of Dis, calculated as follows:

[0082]

[0083] In one optional implementation, step S103 includes:

[0084] S1031. Divide the road boundary line into several sections.

[0085] S1032. Obtain the fitted curve for each segment using curve fitting.

[0086] S1033. At preset intervals, obtain a simulated road boundary point based on the fitted curve, and connect all simulated road boundary points to obtain the simulated road boundary line.

[0087] In this embodiment, a density-based spatial clustering method is used to divide the road boundary line into different segments. For each segment of the road boundary line, a two-dimensional planar curve of the segment is fitted using the global least squares method, and a simulated road point is obtained at fixed intervals on the curve to form a set of points P1 of the simulated road boundary line.

[0088] P1={P1 i |i=1,2,3…N} (3)

[0089] Where N is the number of simulated road points on a simulated road boundary line.

[0090] In another alternative implementation, for each segment of the road boundary, the number of point clouds and the slope within that block can be calculated. If the number is too low or the slope changes abruptly, that section is considered a false boundary and eliminated. Furthermore, considering the gaps in the road boundary, a smoothing spline algorithm is used to add simulated road boundary points.

[0091] In one optional implementation, the road parameters include road grade and traffic standards; step S104 includes:

[0092] S1041. Using the points symmetrical to the boundary line of the simulated road as the bottom points, construct the cross-section of the barrier-free pipeline based on the bottom points and the corresponding top points; the distance between the bottom points and the top points is determined according to the road grade and traffic standards.

[0093] S1042. Construct a barrier-free pipeline segment based on several adjacent cross sections.

[0094] S1043. Connect several adjacent pipe segments to obtain an unobstructed pipe.

[0095] like Figure 4 As shown, several adjacent cross-sections constitute a section of an accessible pipeline, and several adjacent sections are connected to form an accessible pipeline, such as... Figure 5 As shown, each pipe segment and cross-section consists of two types of feature points: a bottom point (P1) and a top point (P2). The bottom point P1 is obtained from the simulated road boundary line, while the top point is calculated from its geometric relationship with the corresponding bottom point. P1 is a point with two-dimensional coordinates. The three-dimensional coordinates of P2 are obtained by acquiring the elevation Z of P1. The elevation value Z is calculated as follows:

[0096]

[0097] Among them, h i Let n be the elevation of the nearest 3D boundary point to each 2D boundary point on the simulated road boundary line, and let n represent the sum of points on the simulated road boundary line. After assigning the Z value to the 2D boundary point P1, the point set P1 transforms from 2D planar points into 3D spatial points. For each pipe, the top point P2 corresponds one-to-one with P1, and the formula for calculating P2 is as follows:

[0098] P2=P1+T*f (5)

[0099] In formula (4), T can be determined based on the road grade and traffic standards, or it can be defined by the user. That is, the pipeline can be stretched and raised. For some roads with high vehicle speeds and strict visibility requirements, such as highways, T should be increased appropriately. For some roads with low vehicle speeds and less strict visibility requirements, such as rural roads, T should be decreased appropriately. The distance of the ground normal vector (f) is determined by the normal vectors of the three nearest boundary points of each P1 point.

[0100] like Figure 6 As shown, the pipes can be stretched according to the user's needs and the road grade; in other words, Figure 6 The left-side pipe is designed for rural roads where visibility is limited and driving speeds are slow; the barrier-free pipe can be lowered appropriately. Figure 6 The right-side pipe is designed for roads with extremely high driving speeds, such as highways and inner ring elevated roads, where visibility requirements are correspondingly higher, so the pipe should be raised accordingly.

[0101] In another alternative implementation, such as Figure 7 As shown, each pipe segment and cross-section consists of three feature points: bottom point (Bp), top point (Up), and top point (Tp). The bottom point is obtained from the road boundary line, while the top and top points are calculated from their geometric relationships with the bottom point. The road boundary points are projected onto a two-dimensional plane and then fitted into a smooth spline curve. Then, a parameter C is set. w To represent the width of the convex hull. Each C w The bottom is sampled from a smooth spline. To obtain the elevation value of the bottom point, for each bottom point, several nearest road boundary points are extracted, with elevation values ​​H. k The calculation is as follows:

[0102]

[0103]

[0104] Where d(p,p) i ) is the Euclidean distance between two points in a two-dimensional plane, w i Let each represent the weight of the point. For each pipeline, the top point and the vertex are given by the following formula:

[0105] Up=Bp+H1*Gnv (8)

[0106] Tp=Bp-E+H2*Gnv (9)

[0107] Here, H1, H2, and E are defined by traffic standards. The ground normal vector (Gnv) is taken into account for the nearest point of each pipeline.

[0108] In an optional implementation, step S104 can be performed in the following way:

[0109] like Figure 5 As shown, the cuboid that makes up the pipe consists of 8 points. Let the 3D coordinates of each point cloud be (a, b, c). Then the maximum and minimum values ​​of the x, y, and z axes of the eight points of the cuboid are respectively x, y, and c. max ,x min ,y max ,y min ,z max ,z min Point clouds that simultaneously satisfy equations (10), (11), and (12) are considered to be contained within the pipe, that is, point clouds that encroach on the unobstructed pipe.

[0110] x max >a>x min (10)

[0111] y max >b>y min (11)

[0112] z max >c>z min (12)

[0113] In this embodiment, an unobstructed pipeline is derived from mobile laser scanning data. By calculating the geometric relationship between the unobstructed pipeline and obstacles, the point cloud of obstacles encroaching on the unobstructed pipeline can be obtained, which is the maintenance information for generating row trees.

[0114] Taking point cloud data of a certain area as an example, such as Figure 8 The image shows point cloud data of the surrounding street acquired through mobile laser scanning. After data acquisition, the original road scene point cloud data is obtained according to the method provided in the example above, and road boundary lines and accessible pipelines are generated sequentially. Experimental results show that the above method can generate accessible pipelines that match the corresponding city. After obtaining the accessible pipelines, the geometric relationship between the obstacle point cloud and the accessible pipelines is calculated, such as... Figure 9 As shown, we can see the point cloud of unobstructed pipes that needs trimming inside the pipe and the point cloud of unobstructed pipes that does not need trimming outside the pipe. The point cloud of unobstructed pipes that needs modification represents the maintenance information of the obstacles.

[0115] The road obstacle detection method in this embodiment separates the road surface point cloud and the point cloud to be detected from the road point cloud data, obtains a simulated road boundary line based on the road surface point cloud, generates an unobstructed pipeline based on the simulated road boundary line and road parameters, and obtains the point cloud to be detected within the unobstructed pipeline to obtain information about road obstacles. This achieves automatic detection of obstacles that obstruct the view on the road, improving the detection efficiency and accuracy of visual obstacles. At the same time, it constructs unobstructed pipelines of different heights and shapes based on road differences, meeting the detection needs of different roads and improving the detection accuracy of visual obstacles.

[0116] Example 2

[0117] This embodiment provides a method for maintaining road obstacles, the method comprising:

[0118] Information on road obstacles is obtained using the road obstacle detection method of Example 1 in order to update the information on obstacles to be maintained;

[0119] Road obstacles are maintained based on the updated maintenance obstacle information.

[0120] The road obstacle maintenance method of this embodiment utilizes the aforementioned road obstacle detection method to achieve automatic detection and maintenance of obstacles that obstruct the view on the road, thereby improving the detection efficiency and accuracy of visual obstacle obstacles. At the same time, based on the differences in roads, barrier-free pipelines of different heights and shapes are constructed to meet the detection needs of different roads, improving the detection accuracy of visual obstacle obstacles, and thus improving the efficiency and effectiveness of road maintenance.

[0121] Example 3

[0122] Please refer to Figure 10 This is a schematic diagram of the road obstacle detection system in this embodiment. Specifically, as shown... Figure 10 As shown, the detection system includes:

[0123] Point cloud data acquisition module 1 is used to acquire point cloud data of roads.

[0124] In one alternative implementation, a mobile laser scanning system can be used to simultaneously acquire trajectory data and laser scan data. The mobile laser scanning system may include GNSS (Global Navigation Satellite System) equipment, a panoramic camera, and a laser sensor. The mobile laser scanning system is mounted on a moving vehicle (e.g., a surveying vehicle). As the vehicle travels along the road, the GNSS equipment acquires trajectory data, the panoramic camera obtains color information, and the laser sensor acquires laser scan data of the road surface. Point cloud calculations are then used to obtain urban road point cloud data containing information such as three-dimensional coordinates (X, Y, Z), color (R, G, B), intensity, and category attributes.

[0125] Boundary line generation module 2 is used to separate the road surface point cloud and the point cloud to be detected from the point cloud data, and to determine the road boundary line based on the road surface point cloud.

[0126] Specifically, the point cloud acquired from mobile laser scanning data includes not only road point cloud data but also point cloud data of buildings, pedestrians, vehicles, etc. This latter type needs to be excluded, and only the road surface point cloud needs to be extracted for subsequent calculations to reduce the complexity of the data processing. Furthermore, the road surface point clouds located on the road boundary lines are identified and extracted from all the road surface point clouds to determine the road boundary lines.

[0127] Boundary line optimization module 3 is used to fit the road boundary line to obtain the simulated road boundary line; specifically, due to the occlusion of the road surface by obstacles such as pedestrians and vehicles, the road boundary line needs to be optimized.

[0128] Pipeline generation module 4 is used to generate barrier-free pipelines for roads based on simulated road boundary lines and road parameters;

[0129] The obstacle detection module 5 is used to acquire the point cloud of the obstacle to be detected in the unobstructed pipeline to obtain information about road obstacles.

[0130] In one optional implementation, the boundary line generation module 2 is specifically used to determine the road surface point cloud based on changes in the geometric features of the point cloud data, and to identify other point cloud data in the point cloud data as point clouds to be detected. Specifically, since the moving laser scanning data consists of multiple scan line data, and each scan line data reflects changes in road elevation, and road boundaries are diverse, abrupt changes in elevation or angle can occur on the road boundary lines. Therefore, the elevation and angle changes of each scan line on the road point cloud data can be calculated separately. If no abrupt changes in elevation or angle occur, then the point cloud is identified as a road surface point cloud; otherwise, it is considered a point cloud to be detected.

[0131] In another optional implementation, other point cloud data may include point cloud data of pedestrians, vehicles, buildings, roadside trees, etc. Among them, pedestrians and vehicles that are temporarily traveling together are not considered obstacle detection objects, while buildings and their appurtenances, roadside trees, and other fixed facility obstacles are considered obstacle detection objects. Therefore, point cloud data of the road can be acquired at different times. Other point cloud data that appears repeatedly and excludes the road surface point cloud can be identified as point cloud data of fixed facility obstacles, which are the point clouds to be detected.

[0132] In one alternative implementation, the boundary line generation module 2 is specifically used to project the road surface point cloud onto a two-dimensional plane; specifically, the three-dimensional road surface point cloud is projected onto a two-dimensional plane after deleting the Z-axis elevation information.

[0133] Boundary line generation module 2 is specifically used to determine the road boundary line based on the contour features of the set of all road surface point clouds on the two-dimensional plane, and the method is as follows: Figure 3 As shown.

[0134] Specifically, suppose a ball with radius r rolls around the two-dimensional plane of the road surface point cloud. Since the ball only rolls on the two-dimensional boundary line of the road point cloud, the trajectory of the ball is the boundary of the road, and its radius r is calculated by equation (1).

[0135] In another alternative implementation, boundary points extracted from a finite unstructured road point cloud can be used with a value γ. Specifically, to select an optimal γ, a random point and its x nearest neighbors are chosen. Then, the average distance between the random point and its nearest neighbors is calculated. The x random points and their average distances are calculated in the same way. Finally, the optimal γ value is defined as the reciprocal of Dis, and γ is calculated using equation (2).

[0136] In one optional implementation, the boundary line optimization module 3 is specifically used to divide the road boundary line into several segments; the boundary line optimization module 3 is specifically used to obtain a fitting curve for each segment by curve fitting; the boundary line optimization module 3 is specifically used to obtain a simulated road boundary point at every preset distance according to the fitting curve, and connect all simulated road boundary points to obtain a simulated road boundary line.

[0137] In this embodiment, a density-based spatial clustering method is used to divide the road boundary line into different segments. For each segment of the road boundary line, the two-dimensional plane curve of the segment is fitted by the global least squares method, and a simulated road point is obtained at fixed intervals on the curve, as shown in Equation (3), to form a set of points P1 of the simulated road boundary line.

[0138] In another alternative implementation, for each segment of the road boundary, the number of point clouds and the slope within that block can be calculated. If the number is too low or the slope changes abruptly, that section is considered a false boundary and eliminated. Furthermore, considering the gaps in the road boundary, a smoothing spline algorithm is used to add simulated road boundary points.

[0139] In one optional implementation, the road parameters include road grade and traffic standards; the pipeline generation module 4 is specifically used to construct the cross-section of the barrier-free pipeline based on the bottom point and the corresponding top point on the simulated road boundary line as the bottom point; the distance between the bottom point and the top point is determined according to the road grade and traffic standards; the pipeline generation module 4 is specifically used to construct the pipeline segment of the barrier-free pipeline based on several adjacent cross-sections; the pipeline generation module 4 is specifically used to connect several adjacent pipeline segments to obtain the barrier-free pipeline.

[0140] like Figure 4 As shown, several adjacent cross-sections constitute a section of an accessible pipeline, and several adjacent sections are connected to form an accessible pipeline, such as... Figure 5 As shown, each pipe segment and cross-section consists of two types of feature points: bottom point (P1) and top point (P2). The bottom point P1 is obtained from the simulated road boundary line, while the top point is calculated from its geometric relationship with the corresponding bottom point. P1 is a point with two-dimensional coordinates. The three-dimensional coordinates of P2 are obtained by obtaining the elevation Z of P1. The elevation value Z is calculated as shown in Equation (4).

[0141] After assigning the Z value to the two-dimensional boundary point P1, the point set P1 changes from two-dimensional plane points to three-dimensional spatial points. For each pipe, the top point P2 corresponds one-to-one with P1, and the calculation method of P2 is shown in Equation (5).

[0142] In formula (4), T can be determined based on the road grade and traffic standards, or it can be defined by the user. That is, the pipeline can be stretched and raised. For some roads with high vehicle speeds and strict visibility requirements, such as highways, T should be increased appropriately. For some roads with low vehicle speeds and less strict visibility requirements, such as rural roads, T should be decreased appropriately. The distance of the ground normal vector (f) is determined by the normal vectors of the three nearest boundary points of each P1 point.

[0143] like Figure 6 As shown, the pipes can be stretched according to the user's needs and the road grade; in other words, Figure 6 The left-side pipe is designed for rural roads where visibility is limited and driving speeds are slow; the barrier-free pipe can be lowered appropriately. Figure 6 The right-side pipe is designed for roads with extremely high driving speeds, such as highways and inner ring elevated roads, where visibility requirements are correspondingly higher, so the pipe should be raised accordingly.

[0144] In another alternative implementation, such as Figure 7 As shown, each pipe segment and cross-section consists of three feature points: bottom point (Bp), top point (Up), and top point (Tp). The bottom point is obtained from the road boundary line, while the top and top points are calculated from their geometric relationships with the bottom point. The road boundary points are projected onto a two-dimensional plane and then fitted into a smooth spline curve. Then, a parameter C is set. w To represent the width of the convex hull. Each C w The bottom is sampled from a smooth spline. To obtain the elevation value of the bottom point, for each bottom point, several nearest road boundary points are extracted, with elevation values ​​H. k The calculations are shown in equations (6) and (7).

[0145] In one alternative implementation, the obstacle detection module 5 can be implemented in the following way:

[0146] like Figure 5 As shown, the cuboid that makes up the pipe consists of 8 points. Let the 3D coordinates of each point cloud be (a, b, c). Then the maximum and minimum values ​​of the x, y, and z axes of the eight points of the cuboid are respectively x, y, and c. max ,x min ,y max ,y min ,z max ,z min Point clouds that simultaneously satisfy equations (10), (11), and (12) are considered to be contained within the pipe, that is, point clouds that encroach on the unobstructed pipe.

[0147] In this embodiment, an unobstructed pipeline is derived from mobile laser scanning data. By calculating the geometric relationship between the unobstructed pipeline and obstacles, the point cloud of obstacles encroaching on the unobstructed pipeline can be obtained, which is the maintenance information for generating row trees.

[0148] Taking point cloud data of a certain area as an example, such as Figure 8 The image shows point cloud data of the surrounding street acquired through mobile laser scanning. After data acquisition, the original road scene point cloud data is obtained according to the method provided in the example above, and road boundary lines and accessible pipelines are generated sequentially. Experimental results show that the above method can generate accessible pipelines that match the corresponding city. After obtaining the accessible pipelines, the geometric relationship between the obstacle point cloud and the accessible pipelines is calculated, such as... Figure 9 As shown, we can see the point cloud of unobstructed pipes that needs trimming inside the pipe and the point cloud of unobstructed pipes that does not need trimming outside the pipe. The point cloud of unobstructed pipes that needs modification represents the maintenance information of the obstacles.

[0149] The road obstacle detection system in this embodiment separates the road surface point cloud and the point cloud to be detected from the road point cloud data, obtains a simulated road boundary line based on the road surface point cloud, generates an unobstructed pipeline based on the simulated road boundary line and road parameters, and obtains the point cloud to be detected within the unobstructed pipeline to obtain information about road obstacles. This achieves automatic detection of obstacles that obstruct the view on the road, improving the detection efficiency and accuracy of visual obstacles. At the same time, it constructs unobstructed pipelines of different heights and shapes based on road differences, meeting the detection needs of different roads and improving the detection accuracy of visual obstacles.

[0150] Example 4

[0151] This embodiment provides a road obstacle maintenance system, the maintenance system comprising:

[0152] The obstacle information update module is used to obtain information about road obstacles using the road obstacle detection system of Embodiment 3, so as to update the information of obstacles to be maintained;

[0153] The maintenance instruction generation module is used to generate instructions for maintaining road obstacles based on the updated maintenance obstacle information.

[0154] The road obstacle maintenance system of this embodiment utilizes the aforementioned road obstacle detection system to achieve automatic detection and maintenance of obstacles that obstruct the view on the road, thereby improving the detection efficiency and accuracy of visual obstacles. At the same time, it constructs barrier-free pipelines of different heights and shapes based on the differences in roads, meeting the detection needs of different roads, improving the detection accuracy of visual obstacles, and thus improving the efficiency and effectiveness of road maintenance.

[0155] Example 5

[0156] Figure 11 This is a schematic diagram of an electronic device provided in Embodiment 5 of the present invention. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the road obstacle detection method of Embodiment 1 or the road obstacle maintenance method of Embodiment 2. Figure 11 The electronic device 30 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0157] like Figure 11 As shown, the electronic device 30 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including memory 32 and processor 31).

[0158] Bus 33 includes a data bus, an address bus, and a control bus.

[0159] The memory 32 may include volatile memory, such as random access memory (RAM) 321 and / or cache memory 322, and may further include read-only memory (ROM) 323.

[0160] The memory 32 may also include a program / utility 325 having a set (at least one) of program modules 324, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0161] The processor 31 executes various functional applications and data processing by running computer programs stored in the memory 32, such as the road obstacle detection method of Embodiment 1 or the road obstacle maintenance method of Embodiment 2.

[0162] Electronic device 30 can also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). This communication can be performed via input / output (I / O) interface 35. Furthermore, the model-generated device 30 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 36. As shown, network adapter 36 communicates with other modules of the model-generated device 30 via bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the model-generated device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0163] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0164] Example 6

[0165] This embodiment provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the road obstacle detection method of Embodiment 1 or the road obstacle maintenance method of Embodiment 2.

[0166] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0167] In a possible implementation, the present invention can also be implemented as a program product comprising program code, which, when the program product is run on a terminal device, is used to cause the terminal device to execute the road obstacle detection method of Embodiment 1 or the road obstacle maintenance method of Embodiment 2.

[0168] The program code for executing the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0169] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. A method for detecting road obstacles, characterized in that, The detection method includes: Obtain the point cloud data of the road; Separating the road surface point cloud and the point cloud to be detected from the point cloud data includes: The road surface point cloud is determined based on the changes in the geometric features of the point cloud data, and other point cloud data in the point cloud data are determined as point clouds to be detected. Determining the road boundary line of the road based on the road surface point cloud includes: Project the road surface point cloud onto a two-dimensional plane; The road boundary line of the road is determined based on the contour features of the set of all road surface point clouds on the two-dimensional plane; Fitting the road boundary line to obtain a simulated road boundary line includes: The road boundary line is divided into several sections; The fitted curve for each segment is obtained by curve fitting; At preset intervals, a simulated road boundary point is obtained based on the fitted curve, and all the simulated road boundary points are connected to obtain the simulated road boundary line; Based on the simulated road boundary line and road parameters, an accessible pipeline of the road is generated, and the point cloud to be detected within the accessible pipeline is obtained to obtain information about road obstacles. The road parameters include road grade and traffic standards; the step of generating the accessible pipeline of the road based on the simulated road boundary line and road parameters includes: Using the symmetrical points on the boundary line of the simulated road as the bottom points, the cross-section of the barrier-free pipeline is constructed based on the bottom points and the corresponding top points; the distance between the bottom points and the top points is determined according to the road grade and traffic standards of the road. The bottom point is obtained from the simulated road boundary line, and the top point is calculated from its geometric relationship with the corresponding bottom point. The barrier-free pipeline is constructed based on several adjacent cross-sections; The barrier-free pipeline is obtained by connecting several adjacent pipeline segments.

2. A method for maintaining road obstacles, characterized in that, The maintenance methods include: The road obstacle detection method as described in claim 1 is used to obtain information about road obstacles in order to update the information about obstacles to be maintained; Road obstacles are maintained based on the updated maintenance obstacle information.

3. A road obstacle detection system, characterized in that, The detection system includes: A point cloud data acquisition module is used to acquire point cloud data of the road. A boundary line generation module, used to separate the road surface point cloud and the point cloud to be detected from the point cloud data, includes: The road surface point cloud is determined based on the changes in the geometric features of the point cloud data, and other point cloud data in the point cloud data are determined as point clouds to be detected. The boundary line generation module is also used to determine the road boundary line of the road based on the road surface point cloud. include: Project the road surface point cloud onto a two-dimensional plane; The road boundary line of the road is determined based on the contour features of the set of all road surface point clouds on the two-dimensional plane; Boundary line optimization module, used to fit the road boundary line to obtain a simulated road boundary line, includes: The road boundary line is divided into several sections; The fitted curve for each segment is obtained by curve fitting; At preset intervals, a simulated road boundary point is obtained based on the fitted curve, and all the simulated road boundary points are connected to obtain the simulated road boundary line; The pipeline generation module is used to generate accessible pipelines for the road based on the simulated road boundary lines and road parameters; the road parameters include road grade and traffic standards. The pipeline generation module is specifically used to construct the cross-section of an accessible pipeline based on symmetrical points on the simulated road boundary line as bottom points and corresponding top points. The distance between the bottom and top points is determined according to the road grade and traffic standards. The bottom points are obtained from the simulated road boundary line, and the top points are calculated from their geometric relationship with the corresponding bottom points. The pipeline generation module is also used to construct pipeline segments of the accessible pipeline based on several adjacent cross-sections. Finally, the pipeline generation module connects several adjacent pipeline segments to obtain an accessible pipeline. The obstacle detection module is used to acquire the point cloud to be detected within the barrier-free pipeline to obtain information about road obstacles.

4. A road obstacle maintenance system, characterized in that, The maintenance system includes: An obstacle information update module is used to obtain information about road obstacles using the road obstacle detection system as described in claim 3, so as to update the information about obstacles to be maintained; The maintenance instruction generation module is used to generate instructions for maintaining road obstacles based on the updated maintenance obstacle information.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes a computer program, it implements the road obstacle detection method as described in claim 1 or the road obstacle maintenance method as described in claim 2.

6. A computer-readable medium storing computer instructions thereon, characterized in that, The computer instructions, when executed by the processor, implement the road obstacle detection method as described in claim 1 or the road obstacle maintenance method as described in claim 2.

Images