CN116500643A - A method and system for detecting road deformation defects based on single-line laser point cloud - Google Patents

Abstract

The invention discloses a method and a system for detecting highway deformation diseases based on single-line laser point clouds, which are characterized in that firstly, pose of an unmanned aerial vehicle at different moments is recorded, and laser radar acquires road point cloud data in a cross section form and then fits a road surface reference cross section; fitting a road reference surface and a road curved surface model, determining the position of a disease point cloud, and dividing a 10m pavement area; and finally, calculating the maximum depth and deformation volume parameters of the diseases from a three-dimensional angle, and drawing up an index and evaluating the highway deformation diseases. According to the invention, the calculation of the deformation disease depth is realized from a three-dimensional angle, and compared with the traditional detection of the pavement deformation disease, the method has higher precision, and the three-dimensional evaluation index of the pavement deformation disease is provided, so that the evaluation method of the deformation disease is enriched, and the position of the road section with the disease to be maintained is primarily positioned.

Description

A method and system for detecting road deformation defects based on single-line laser point cloud

technical field

The invention belongs to the field of road engineering, and in particular relates to a method and system for detecting road deformation defects based on a single-line laser point cloud.

Background technique

With the rapid economic development, asphalt roads have gradually become the main road surface type of expressways in our country. However, the continuous increase of traffic volume and the aggravation of problems such as overloading and heavy loading have caused various damages to asphalt pavement, especially highway asphalt pavement.

Deformation damage is a disease with obvious deformation of the pavement due to the influence of external force, and it is one of the common diseases of asphalt highway pavement. On the one hand, deformation-related diseases lead to a large amount of deformation of the road surface, causing a sudden change in driving elevation, which affects the comfort and stability of driving. A series of other diseases, causing drainage problems, etc., reduce the strength and service life of the pavement, and reduce the safety of the pavement. Therefore, it is necessary to conduct a systematic and comprehensive assessment of road surface deformation disasters and make targeted maintenance measures.

At present, the evaluation indicators for deformation diseases at home and abroad are often based on the depth and width indicators of cross-sectional diseases. There is no unified evaluation index for deformation diseases. At the same time, due to the lack of pavement detection technology, the accuracy and density of collected data are difficult to guarantee. , seldom consider the damage of deformation diseases from a three-dimensional perspective.

In the past, due to the backwardness of electronic technology, manual methods were the main method for detecting road surface defects, mainly measuring the depth and width of ruts with a ruler. However, manual measurement relies on experience and has large errors, so it has been eliminated now. In recent years, with the continuous development of electronic technology, deformation-related damage detection methods based on deep learning or image processing have gradually been put into use. The detection vehicle is mainly used as a mounting device, and multiple multi-line laser radars are arranged to realize road deformation damage detection. detection. However, this method still has some disadvantages, such as: being easily affected by other vehicles on the road, the cost and maintenance of detecting vehicles and multi-line radar are high, and the speed requirement is strict. Although the way the lidar measures the road surface point cloud can ensure the accuracy of the data, the mounting device limits the performance of the radar. Compared with inspection vehicles, UAVs can achieve lower cost, lighter weight and higher efficiency. In theory, they are better mounting devices for lidar. At the same time, the UAV detects the deformation of the road surface from a high altitude, and the single-line radar is sufficient to meet the detection needs. Compared with multi-line radar, the data processing of single-line lidar is simpler and simpler, and it also has more advantages in terms of calculation volume. After certain point cloud data processing, single-line radar can also generate high-density and accurate point clouds.

Contents of the invention

In order to solve the technical problems mentioned in the above-mentioned background technology, the present invention proposes a method and system for detecting road deformation defects based on single-line laser point cloud.

In order to realize above-mentioned technical purpose, technical scheme of the present invention is:

A method for detecting road deformation defects based on a single-line laser point cloud, comprising the following steps:

S1. Mount the single-line lidar based on triangular ranging on the UAV, record the pose of the UAV at different times while the UAV is flying along the specified route, and collect cross-sectional road point cloud data.

S2. There is noise interference in the data collection process, resulting in a series of isolated points and interference points. Denoise the collected road point cloud data, effectively eliminate outliers, obtain a single frame point cloud of the road, filter the point cloud of the outer lane, fit it to the standard cross-section in a straight line, and insert it according to the slope of the fitted line The single-frame point cloud of the inner lane is spliced and inserted into the point cloud and the point cloud of the outer lane to obtain a 3D point cloud of the road.

S3. According to the three-dimensional point cloud of step S2, the road is divided into specific intervals of the road surface area, and the reference driving surface of the road is fitted by the thin-plate spline interpolation method, and the area where the disease is located is located on the divided road surface area, so as to study the extent of the disease. Severity, and the actual surface equation of the road is determined by thin-plate spline interpolation.

S4. Calculate the maximum subsidence, maximum protrusion of the disease and corresponding deformation volume of the divided diseased area, determine the severity of the disease, and judge whether maintenance is required.

Further, the specific steps of step S2 are as follows:

S201. Use the radius filtering algorithm to make a circle with its center for each point cloud. If the number of points contained in the circle is greater than a fixed value, it will be retained, and if it is less than a fixed value, it will be deleted. Using a statistical filtering algorithm, for each point cloud, solve its to Calculate the average value of the distances of all points in the field K, and calculate the mean value μ and variance σ of all average values, set μ+nv as the threshold, n as the specified multiple, and use this threshold as the screening value to initially eliminate the road point cloud Data noise data to obtain a single-frame point cloud of the road; determine the range of the road surface point cloud based on the road cross slope and the elevation of the lowest point on the road surface, thereby extracting an accurate cross-sectional point cloud.

S202. Using the method of extracting the outer point cloud of single frame data based on the point cloud coordinates, screening the point cloud of 0.5m-3m outside the cross-sectional direction in the single frame point cloud, as the emergency lane point cloud, and fitting the emergency lane points in a straight line Cloud, forming a standard cross-section, and according to the slope of the line, insert the single-frame point cloud inside the emergency lane at point intervals within the road range to simulate the reference driving surface.

S203. Since the UAV equipped with radar will be offset and twisted due to wind factors during operation and driving, it is necessary to use sensors to know the specific running track and pose of the mounted device in order to locate its acquisition. The three-dimensional position of the cross section of . Based on the single-line radar pose recorded in step S1, the cross-sectional point cloud extracted in S201, and the single-frame point cloud in S202, through NED (North East Down, north east coordinate system) to LLA (longitude, latitude, altitude, longitude and latitude high coordinates) System) coordinate transformation, stitching together point cloud data of consecutive frames to form 3D road point cloud data and 3D point cloud data of reference driving surface.

Further, the specific steps of step S3 are as follows:

S301, based on the point cloud of the emergency lane collected in step S202, extract the plane coordinates of the point cloud, use the cubic curve method to fit the overall line shape of the road, reflect the trend of the road, divide the curve into 10m subsections, and extract x at the subsections , y coordinates, and the normal plane of the curve at the section, take the point on the inside of the normal plane on the xOy projection straight line 20m away from the trend curve, and determine the road area clamped by the normal plane, so as to realize the division of a 10m section of road surface area on the reference driving surface ; The coordinates of points taken on the normal plane are marked as:

QD=[[x ₀₁ , y ₀₁ , x ₀₂ , y ₀₂ ],  …]

Among them, x ₀₁ represents the abscissa of the intersection point of the trend curve and the first normal plane, y ₀₁ represents the ordinate of the intersection point of the trend curve and the first normal plane, x ₀₂ represents the abscissa of the point inside the first normal plane, and y ₀₂ represents the vertical coordinate of the point inside the first normal plane.

S302. Since the point cloud of the emergency lane preserves the road elevation information at the design location relatively well, which is convenient for subsequent calculations, based on the three-dimensional point cloud data of the reference traffic surface in step S203, a plane fitting algorithm is used to fit the reference traffic surface of the road, Obtain the parameters A ₀ , A ₁ , A ₂ of the plane equation of the reference traffic surface, the specific formula is:

z＝A ₀ x+A ₁ y+A ₂

Among them, A ₀ , A ₁ , and A ₂ are the parameters of the plane about the x and y coordinates and the intercept of the z coordinate of the plane when the x and y coordinates are equal to 0, respectively.

S303. Based on the three-dimensional road point cloud data in step S203 and the road reference driving surface equation in step S302, calculate the elevation difference between the three-dimensional road point cloud data and the reference surface corresponding to xOy coordinates.

S304. Based on the grid method, divide the xOy coordinates of the road datum plane into 1dm×1dm grids, if the absolute value of the elevation difference of the point cloud is greater than the preset threshold, and the point is located below the datum plane, then the point is a subsidence Disease point, the grid area where the point is located is a subsidence disease area; if the absolute value of the point cloud elevation difference is greater than the preset threshold, and the point is above the datum, then the point is a convex disease point, and the point is located The grid area is the raised disease area; thus the subsidence diseased area Sa and the raised diseased area Sb are preliminarily divided.

S305. Based on the point cloud of the subsidence diseased area and the raised diseased area, the thin-plate spline interpolation method is used to fit the three-dimensional surface equation of the road diseased area. The specific formula is:

U(x)＝r ² lnr

Among them, p(x, y) is any point on the surface, U(x) is the radial basis function, ||pp _i || indicates the distance from point p to a certain control point, known control points 1, 2 , 3..., N, ω _i represent the weighting of different radial basis, m ₀ , m ₁ , m ₂ are the parameters of this plane.

S306, establish the control point matrix and the height matrix of the point cloud data, the specific formula is:

(1) Control point matrix

Among them, n is the number of control points, and the second and third columns represent the (x, y) coordinates of the control points.

(2) Height matrix

Among them, v ₁ to v _n represent the coordinates of each control point in the z direction.

S307. Calculate the radial basis function value of any two control points, the specific formula is:

Among them, r _ij represents the distance between control points i and j, and U(r _ij ) is the value of the radial basis function corresponding to the distance r _ij .

S308. Define matrix L as:

Then the above matrix has the following relationship:

Y=L*(ω ₁ , . . . ω _N , m ₀ , m ₁ , m ₂ ) ^T .

Based on the thin-plate spline interpolation principle, the conditional function about the control points is substituted, all parameters of the three-dimensional surface equation of the road are calculated, and the interpolation is completed. The specific matrix is:

Among them, ω _ij is the weight of the j-th radial basis on the i-th segment, m _i0 , m _i1 , m _i2 are m ₀ , m ₁ , m ₂ coefficients on the i-th segment.

Further, the specific steps of step S4 are:

S401. Extract the points M and N corresponding to the maximum elevation difference in step S3, and use the grid where the M and N points are located and the 8 surrounding grids as the selected area, and solve the point M based on the steepest descent/ascent method , N out of the gradient, the specific formula is:

f = z _actual - z _reference

Among them, f is the difference between the actual surface elevation of the road and the road reference driving surface, z _actual is the actual surface elevation of the road, and z _reference is the reference driving surface of the road; let the known minimum point in the selected disease area be M, after k times The point obtained by iteration is denoted as M ^k , P ^k represents the direction of the maximum surface change rate at point M ^k , and d ^k represents the gradient.

S402. According to the formula in step S305, obtain the z _reference value, solve the differential for the z _reference value formula, and obtain the following formula:

d ^k is finally expressed as:

Among them, N is the number of control points for datum surface fitting, Q is the number of control points for the actual road surface fitting, M ^k (x, y) is the starting point of the current iteration; x _i , y _i are the corresponding control points coordinates; r is the distance from M ^k (x, y) to each control point.

S403. Solve the elevation of the next point with a step size of 0.01m, and perform iterations until an extreme point appears, and the elevation difference between the extreme point and the reference plane corresponding to the x0y coordinate is the disease depth/disease height on the road surface area.

S404, regularly take 9 points in each disease grid, and calculate the average value of its elevation take /> The product of the grid area and the grid area is the deformation volume V _i of the grid, and the specific formula is:

Among them, S is the grid area of 4cm ² ; V _i is the deformation volume of grids with deformation defects from left to right and then from bottom to top in each pavement area.

The sum of all grid deformation volumes is the subsidence/convex volume V _a , V _b of the pavement area.

S405. Based on the depth/height of the road surface disease and the subsidence/bulge volume of the road surface area, complete the three-dimensional evaluation of road deformation-type diseases.

Further, the present invention also proposes a detection system for highway deformation defects based on single-line laser point cloud, including

The information acquisition module is used to mount the single-line lidar on the UAV, record the pose of the UAV at different times during the flight of the UAV along the specified route, and collect road point cloud data in the form of cross-sections.

The road 3D point cloud acquisition module is used to denoise the collected road point cloud data, obtain a single frame point cloud of the road, filter the point cloud of the outer lane, and fit it to the standard cross-section in a straight line, and according to the fitting line Insert the single-frame point cloud of the inner lane with the slope of , and splicing the inserted point cloud and the point cloud of the outer lane to obtain the 3D point cloud of the road.

The disease area positioning module is used to divide the road into specific interval road surface areas according to the three-dimensional point cloud of the road, fit the reference driving surface of the road through the thin plate spline interpolation method, locate the disease area on the divided road surface area and determine the road The actual surface equation of .

The deformation volume calculation module of the diseased area is used to calculate the maximum subsidence, the maximum protrusion of the diseased area and the corresponding deformation volume of the divided diseased area.

Further, in the road 3D point cloud acquisition module, the specific steps are as follows:

Step 1. Use the radius filtering algorithm to make a circle with its center for each point cloud. If the number of points contained in the circle is greater than a certain value, it will be retained, and if it is less than a certain value, it will be deleted. Using a statistical filtering algorithm, for each point cloud, its Calculate the average value of the distances to all points in the domain K, and calculate the mean value μ and variance σ of all average values, set μ+nσ as the threshold, n as the specified multiple, and use the threshold as the screening value to initially eliminate road points Cloud data noise data to obtain a single-frame point cloud of the road; determine the range of the road surface point cloud based on the road cross slope and the elevation of the lowest point of the road surface, thereby extracting an accurate cross-sectional point cloud.

Step 2. Use the method of extracting the outer point cloud of single-frame data based on the point cloud coordinates, and select the point cloud of 0.5m-3m outside the cross-sectional direction in the single-frame point cloud as the point cloud of the emergency lane, and fit the emergency lane in a straight line The point cloud forms a standard cross-section, and according to the slope of the straight line, inserts the single-frame point cloud inside the emergency lane at point intervals within the road range.

Step 3. Based on the single-line radar pose, cross-sectional point cloud and single-frame point cloud, through the coordinate transformation from NED to LLA coordinate system, the point cloud data of consecutive frames are combined to form 3D road point cloud data and 3D point cloud of reference driving surface data.

Further, in the disease area location module, the specific steps are as follows:

Step 1. Based on the point cloud of the emergency lane, extract the plane coordinates of the point cloud, use the cubic curve method to fit the overall line shape of the road, divide the curve into 10m segments, and extract the x and y coordinates of the segments and the curves of the segments The normal plane of the normal plane, take the point on the inner side of the normal plane on the xOy projection straight line 20m away from the trend curve, determine the road area clamped by the normal plane, so as to realize the division of a section of road surface area of 10m on the reference driving surface; the coordinates of the points taken by the normal plane are marked as :

QD=[[x ₀₁ , y ₀₁ , x ₀₂ , y ₀₂ ],  …]

Among them, x ₀₁ represents the abscissa of the intersection point of the trend curve and the first normal plane, y ₀₁ represents the ordinate of the intersection point of the trend curve and the first normal plane, x ₀₂ represents the abscissa of the point inside the first normal plane, and y ₀₂ represents the vertical coordinate of the point inside the first normal plane.

Step 2. Based on the 3D point cloud data of the reference traffic surface, use the plane fitting algorithm to fit the reference traffic surface of the road, and obtain the parameters A ₀ , A ₁ , and A ₂ of the plane equation of the reference traffic surface. The specific formula is:

z＝A ₀ x+A ₁ y+A ₂

Among them, A ₀ , A ₁ , and A ₂ are the parameters of the plane about the x and y coordinates and the intercept of the z coordinate of the plane when the x and y coordinates are equal to 0, respectively.

Step 3. Based on the three-dimensional road point cloud data and the road datum driving surface equation, calculate the elevation difference between the three-dimensional road point cloud data and the datum plane corresponding to the xOy coordinates.

Step 4. Based on the grid method, the xOy coordinates of the road reference plane are divided into 1dm×1dm grids. If the absolute value of the point cloud elevation difference is greater than the preset threshold and the point is located below the reference plane, then the point is The subsidence disease point, the grid area where the point is located is the subsidence disease area; if the absolute value of the elevation difference of the point cloud is greater than the preset threshold, and the point is above the reference plane, then the point is a convex disease point, and the point is located at The grid area at is the raised diseased area; thus the subsidence diseased area Sa and the raised diseased area Sb are preliminarily divided.

Step 5. Based on the point clouds of the subsidence diseased area and the raised diseased area, the thin-plate spline interpolation method is used to fit the three-dimensional surface equation of the road diseased area. The specific formula is:

U(x)=r ² lnr

Among them, p(x,y) is any point on the surface, U(x) is the radial basis function, ||pp _i || indicates the distance from point p to a certain control point, known control points 1, 2 , 3..., N, ω _i represent the weighting of different radial basis, m ₀ , m ₁ , m ₂ are the parameters of this plane.

Step 6. Establish the control point matrix and height matrix of the point cloud data. The specific formula is:

(1) Control point matrix

Among them, n is the number of control points, and the second and third columns represent the (x, y) coordinates of the control points.

(2) Height matrix

Among them, v ₁ to v _n represent the coordinates of each control point in the z direction.

Step 7, calculate the radial basis function value of any two control points, the specific formula is:

Among them, r _ij represents the distance between control points i and j, and U(r _ij ) is the value of the radial basis function corresponding to the distance r _ij .

Step 8, define the matrix L as:

Then the above matrix has the following relationship:

Y=L*(ω ₁ , . . . ω _N , m ₀ , m ₁ , m ₂ ) ^T .

Based on the thin-plate spline interpolation principle, the conditional function about the control points is substituted, all parameters of the three-dimensional surface equation of the road are calculated, and the interpolation is completed. The specific matrix is:

Among them, ω _ij is the weighting of the jth radial basis on the i-th segment, m _i0 , m _i1 , m _i2 are m ₀ , m ₁ , m ₂ coefficients on the i-th segment.

Further, in the calculation module of the deformed volume of the diseased area, the specific steps are as follows:

Step 1. Extract the points M and N corresponding to the maximum elevation difference, take the grid where the M and N points are located and the 8 surrounding grids as the selected area, and solve the points M and N based on the steepest descent/ascent method The gradient out of N, the specific formula is:

f = z _actual - z _reference

Among them, f is the difference between the actual surface elevation of the road and the road reference driving surface, z _actual is the actual surface elevation of the road, and z _reference is the reference driving surface of the road; let the known minimum point in the selected disease area be M, after k times The point obtained by iteration is denoted as M ^k , P ^k represents the direction of the maximum surface change rate at point M ^k , and d ^k represents the gradient.

S402. According to the formula in step S305, obtain the z _reference value, solve the differential for the z _reference value formula, and obtain the following formula:

d ^k is finally expressed as:

Among them, N is the number of control points for datum surface fitting, Q is the number of control points for the actual road surface fitting, M ^k (x, y) is the starting point of the current iteration; x _i , y _i are the corresponding control points coordinates; r is the distance from M ^k (x, y) to each control point.

Step 3. Solve the elevation of the next point with a step size of 0.01m, and iterate until an extreme point appears. The elevation difference between the extreme point and the reference plane corresponding to xOy coordinates is the disease depth/disease height on the road surface area .

Step 4. Take 9 points regularly in each disease grid and calculate the average value of their elevations take /> The product of the grid area and the grid area is the deformation volume V _i of the grid, and the specific formula is:

Among them, S is the grid area of 4cm ² ; V _i is the deformation volume of grids with deformation defects from left to right and then from bottom to top in each pavement area.

The sum of all grid deformation volumes is the subsidence/convex volume V _a , V _b of the pavement area.

Step 5. Based on the depth/height of the road surface disease and the subsidence/bulge volume of the pavement area, the three-dimensional evaluation of road deformation-type diseases is completed.

The present invention adopts above technical scheme, compared with prior art, its remarkable technical effect is as follows:

The invention fits the equations of the road curved surface and the road reference driving surface, locates the road disease area simply and effectively, completes the calculation of the depth and deformation volume of the deformed disease on the three-dimensional level, and proposes a comprehensive and effective evaluation index with reference to the specification. And based on the UAV lidar acquisition technology, using lighter and lower-cost loading equipment, and using a single-line lidar with a small amount of data, a low-computing three-dimensional deformation and disease extraction algorithm is designed to reduce computing time. , which increases the inspection efficiency, ensures the accuracy of deformation damage depth calculation, and realizes the evaluation of deformation damage from a three-dimensional perspective.

At the same time, the present invention uses thin-plate spline interpolation method to fit the road surface equation, and writes the gradient formula of the equation on the basis of the equation, and combines the principle of the steepest descent/ascent method to further improve the accuracy of deformation-type disease depth detection. Refer to In the previous norms, a comprehensive evaluation index of deformation-related diseases was proposed, which provided certain guidance and assistance for the selection of subsequent maintenance road sections and the calculation of the amount of maintenance materials used.

Description of drawings

Fig. 1 is a schematic diagram of the overall flow of the embodiment of the present invention.

Fig. 2 is a schematic diagram of a frame of cross-sectional point cloud collected by a single-line lidar according to an embodiment of the present invention.

Fig. 3 is a three-dimensional point cloud image after point cloud stitching according to an embodiment of the present invention.

Fig. 4 is a schematic diagram of point cloud disease area detection depth calculation according to an embodiment of the present invention.

Fig. 5 is a schematic diagram of the point selection of the micro-element grid for calculating the disease volume according to the embodiment of the present invention.

Detailed ways

In order to make the technical means, creative features, goals and effects of the present invention easy to understand, a method for detecting road deformation defects based on single-line laser point clouds according to the present invention will be described in detail below in conjunction with the embodiments and accompanying drawings.

A method for detecting road deformation defects based on a single-line laser point cloud according to the present invention, the flow chart of which is shown in Figure 1, specifically includes the following steps:

S1. In this embodiment, the acquisition device is DJI m600pro drone, and the laser radar is sick lms511 single-line laser radar. The laser radar is mounted under the drone, and the drone is located above the road for shooting. The flight speed is set to 1m/s for inspection.

During the flight of the UAV along the specified route, the single-line radar detects at a frequency of 25 Hz, the angular resolution of each detection is 0.1667°, and the data points collected in each frame are about 1100, recording the UAV at different times. Poses and attitudes, and the road point cloud data of 10m sections collected by lidar in the form of cross-sections.

S2. Denoise the collected road point cloud data, effectively eliminate outliers, obtain a single-frame point cloud of the road, filter the point cloud of the outer lane, and fit it to the standard cross-section in a straight line, and according to the fit line The slope is inserted into the single-frame point cloud of the inner lane, and the inserted point cloud and the point cloud of the outer lane are spliced to obtain the 3D point cloud of the road. The specific steps are as follows:

S201. The road point cloud data is road cross-sectional data collected by a single-line lidar with more than 1,000 frames, as shown in FIG. 2 . Use open3d to read a single frame point cloud, first select the lowest point in the point cloud data, which is the lowest point of the road surface point cloud, and use the elevation of this point to the elevation +30cm as the boundary to initially screen the road surface data. Using the radius filter algorithm, make a circle with its center for each point cloud. If the number of points contained in the circle is greater than a fixed value, it will be retained, and if it is less than a fixed value, it will be deleted. In this embodiment, the fixed value is set to 2cm, and the statistical filter algorithm is used. For each point cloud, calculate the average value of its distance to all points in the field, and calculate the mean value μ and variance σ of all average values, set μ+nσ as the threshold, n as the specified multiple, and use the threshold as the radius filtering algorithm The point cloud is screened by the value of , and combined with the road surface elevation and the normal vector of the point cloud, the single-frame point cloud noise data is preliminarily eliminated, and the single-frame point cloud of the road is obtained.

The normal vector of the point cloud is the normal vector of a plane fitted by a point and the point cloud in its neighborhood. For the road surface, the direction of the normal vector is mainly distributed in the z coordinate. The normal vector is calculated by the estimate normals function of open3d, the modulus length of the normal vector is corrected to 1, and the direction is the positive direction of the z coordinate.

If the z-coordinate data of the normal vector is greater than 0.9, the point is a disease-free road point; extract the maximum and minimum values of the single-frame point cloud coordinates in the y direction, and all point clouds within this range are regarded as road points. Based on the cross slope of the road and the elevation of the lowest point of the road surface, the point cloud range of the road surface is determined, so as to extract the accurate cross-sectional point cloud, which is used to merge the 3D point cloud data.

S202. Since most of the emergency lanes will not cause deformation and damage during road operation, the cross section of the outer emergency lane is extended as a straight line, and the result close to the standard cross section can be obtained. In the absence of original design data, it can be viewed Make a standard cross section.

Use open3d to read the processed road surface point cloud, and use the method of extracting the point cloud outside the single frame data based on the point cloud coordinates, and select the point cloud 0.5m-3m outside the cross-section direction in the single frame point cloud as the emergency lane point cloud, Fit the point cloud of the emergency lane with a straight line to form a standard cross-section, and according to the slope of the line, insert the single-frame point cloud of the inner side of the emergency lane at a point interval of 2cm within the road range to simulate the reference traffic surface. Record the inserted point cloud and the frame number in the array DY:

DY=[[x, y, z, (number of frames)], ...].

S203. Based on the single-line radar pose recorded in step S1, the cross-sectional point cloud extracted in S201, and the single-frame point cloud in S202, through the coordinate transformation from NED to LLA coordinate system, the point cloud data of consecutive frames are combined to form a three-dimensional road point cloud. data and the 3D point cloud data of the reference driving surface.

In order to facilitate point cloud modeling, chainage setting, and area division, the single-line radar pose recorded in step S1, the cross-sectional point cloud extracted in S201, and the single-frame point cloud in S202 are converted into an array DY1 through numpy, and are included in all point cloud data A new column of data is added to mark the original frame number of the point cloud. DY1 is:

DY1=[[x, y, z, (number of frames)],  …].

S3. According to the three-dimensional point cloud in step S2, the road is divided into pavement areas with an interval of 10m, and the reference driving surface of the road is fitted by the thin plate spline interpolation method, and the area where the disease is located is located on the divided pavement area, so as to study the extent of the disease. severity, and determine the actual surface equation of the road through the thin plate spline interpolation method, the specific steps are as follows:

Fig. 3 is a three-dimensional point cloud image of this embodiment, and the black part of the circled area is the diseased point cloud area of this embodiment.

S301, based on the point cloud plane coordinates of the emergency lane collected in step S202, use the python library function to complete the cubic curve fitting of the overall road line type, mark the chainage of the curve, and divide it into 10m subsections, extract the x at the subsection , y coordinates, and the normal plane of the curve at the section, take the point on the inside of the normal plane on the xOy projected straight line 20m away from the trend curve, determine the road area clamped by the normal plane, and record the coordinates in the numpy array, so that the reference traffic surface Realize the division of 10m section of road surface area. The coordinates of the points taken on the normal plane are marked as:

QD=[[x ₀₁ , y ₀₁ , x ₀₂ , y ₀₂ ],  …]

Among them, x ₀₁ represents the abscissa of the intersection point of the trend curve and the first normal plane, y ₀₁ represents the ordinate of the intersection point of the trend curve and the first normal plane, x ₀₂ represents the abscissa of the point inside the first normal plane, and y ₀₂ represents the vertical coordinate of the point inside the first normal plane.

S302. Based on the three-dimensional point cloud data of the reference traffic surface in step S203, a plane fitting algorithm is used to fit the reference traffic surface of the road to obtain parameters A ₀ , A ₁ , and A ₂ about the reference traffic surface. The specific formula is:

z=A ₀ x+A ₁ y+A ₂

Among them, A ₀ , A ₁ , and A ₂ are the parameters of the plane about the x and y coordinates and the intercept of the z coordinate of the plane when the x and y coordinates are equal to 0, respectively.

S303. Based on the three-dimensional road point cloud data in step S203 and the road reference surface equation in step S302, calculate the elevation difference between the three-dimensional road point cloud data and the reference surface equation corresponding to xOy coordinates. If the value is negative, the point is below the datum, otherwise it is above the datum.

S304. Based on the grid method, divide the xOy coordinates of the road reference plane into 1dm×1dm grids.

In the array DY1 described in step S203, a new column of data is added, which is used to determine the non-deformed disease point cloud. If the absolute value of the point cloud elevation difference is greater than the preset threshold, and the point is below the datum, then the point is a subsidence disease point, and the grid area where the point is located is a subsidence disease area; if the absolute value of the point cloud elevation difference is greater than the preset threshold, and the point is above the reference plane, then the point is a raised diseased point, and the grid area where the point is located is a raised diseased area. The threshold value set in this embodiment is ±5 mm.

DY2 is obtained after adding a column of data to the array DY1:

DY2=[[x, y, z, (number of frames), (none/sink/raise)],  …]

In order to divide the position of the point cloud and facilitate subsequent calculation and processing, the disease point cloud is divided by grid. The information of the grid is represented by the array WG, specifically:

WG=[[x, y, (subsidence/bulge), (ordinal number of pavement area)],...]

Among them, the first two columns of data represent the coordinate position of the grid, such as (2,3) represents the area surrounded by x=2cm, x=4cm, y=4cm, and y=6cm in the coordinate system; the third column of data is used for Mark the subsidence and bulge of the grid, according to the fourth column data of the array DY2, if none of the point clouds in the grid is a subsidence type point cloud or a convex type point cloud, then there is no disease area in the grid; If the point cloud in the grid has subsidence or bulge, the grid is a subsidence-type disease area Sa or a bulge-type disease area Sb, which is subsequently used to calculate the subsidence volume V _a and the uplift volume V _b ; the fourth column of data Be the ordinal number of the pavement area where the grid where the disease is located, can determine the pavement area where the grid is located according to the positional relationship between the grid and the normal plane described in step S301 (be a straight line after being projected into the x0y plane), so as to be the first road area of the grid Four-column data assignment, for the disease-free grid, to simplify the calculation, the default is 0.

The method for judging the positional relationship between the grid and the normal plane is: take the midpoint G of the diseased grid, and calculate the vector from point G to the point recorded in the QD array; calculate the vector inner product of the adjacent normal plane from the starting plane, if If the vector inner product is negative, then G is between the adjacent normal planes corresponding to the vector, thereby determining the road area where the grid is located; if the vector inner product is positive, then G is outside the adjacent normal plane corresponding to the vector, and the next set of The vector inner product of the normal plane and G.

S305. Based on the point cloud of the subsidence diseased area and the raised diseased area, the thin-plate spline interpolation method is used to fit the three-dimensional surface equation of the road diseased area. The specific formula is:

U(x)＝r ² lnr

Among them, p(x, y) is any point on the surface, U(x) is the radial basis function, ||pp _i || indicates the distance from point p to a certain control point, known control points 1, 2 , 3..., N, ω _i represent the weighting of different radial basis, m ₀ , m ₁ , m ₂ are the parameters of this plane.

S306, the three-dimensional surface equation has point cloud number + 3 parameters, and establishes the control point matrix and height matrix of point cloud data. The specific formula is:

(1) Control point matrix

Among them, n is the number of control points, and the second and third columns represent the (x, y) coordinates of the control points.

(2) Height matrix

Among them, v ₁ to v _n represent the coordinates of each control point in the z direction.

S307. Calculating the radial basis function values of any two control points, and substituting them into the thin-plate spline interpolation condition function, thereby calculating all parameters of the equation, and completing the interpolation. The specific formula is:

Among them, r _ij represents the distance between control points i and j, and U(r _ij ) is the value of the radial basis function corresponding to the distance r _ij .

S308. Define matrix L as:

Then the above matrix has the following relationship:

Y=L*(ω ₁ , . . . ω _N , m ₀ , m ₁ , m ₂ ) ^T .

Based on the thin-plate spline interpolation principle, the conditional function about the control points is substituted, all parameters of the three-dimensional surface equation of the road are calculated, and the interpolation is completed. The specific matrix is:

Among them, ω _ij is the weight of the jth radial basis on the i-th segment, m _i0 , m _i1 , m _i2 are the m ₀ , m ₁ , m ₂ coefficients on the i-th segment.

S4. Calculate the maximum subsidence of the divided disease area, the maximum protrusion of the disease and the corresponding deformation volume, determine the severity of the disease, and judge whether maintenance is required. The specific steps are as follows:

S401, use the max and min functions of python to solve the point M and N corresponding to the maximum and minimum elevation difference, the grid where it is located, and the elevation difference, and the extreme value in the neighborhood of point M and N is the deformation on the road surface area the most value.

Taking the grid where the points M and N are located and the 8 grids around them as the selected area, the extreme value of the actual road model in this area is solved by the steepest descent/ascent method, and the extreme value is the disease depth/ Disease height. Solve the gradients from points M and N, as shown in Figure 4, select the grid containing the maximum elevation difference in the left figure, and use the steepest descent/ascent method in this area to solve the maximum elevation difference in the right figure. The specific formula is:

f = z _actual - z _reference

Among them, f is the difference between the actual surface elevation of the road and the road reference driving surface, z _actual is the actual surface elevation of the road, and z _reference is the reference driving surface of the road; let the known minimum point in the selected disease area be M, after k times The point obtained by iteration is denoted as M ^k , P ^k represents the direction of the maximum surface change rate of point M ^k , and d ^k represents the gradient;

S402. According to the formula in step S305, obtain the z _reference value, solve the differential for the z _reference value formula, and obtain the following formula:

d ^k is finally expressed as:

Among them, N is the number of control points for datum surface fitting, Q is the number of control points for the actual road surface fitting, M ^k (x, y) is the starting point of the current iteration; x _i , y _i are the corresponding control points coordinates; r is the distance from M ^k (x, y) to each control point.

S403. In order to ensure the accuracy of the solution, the search step size is λ=1cm, and the search is performed along the direction d ^k at M ^k , and M ^k+1 =M ^k +λd ^k . Check whether the modulus of the gradient of the original point is less than 0.01. If it is satisfied, the point is an extreme point. If not, continue to iterate until the modulus of the gradient is less than 0.01, and end the iteration. The output f and M ^k are the coordinates corresponding to the maximum value H _a or H _b of the deformation depth. In this embodiment, the maximum lesion depth calculated by this method is 82.9 mm.

S404. Take 9 points regularly in each disease grid, as shown in Figure 5, and calculate its elevation mean value Pick The product of the grid area and the grid area is the deformation volume V _i of the grid, and the specific formula is:

in, is the average value of the elevation sum of the nine points taken, S is the grid area 4cm ² ; V _i is the deformation volume of grids with deformation defects from left to right and then from bottom to top in each pavement area .

If V _i is positive, the grid calculates a raised volume; if V _i is negative, a sunken volume is obtained. According to the positive and negative of V _i and the road area to which the grid belongs, add up all the grid deformation volumes respectively, that is, the subsidence/bulge volume V _a and V _b of the road area.

The road sections with large volume indexes V _a and V _b may develop serious diseases in the future, and can play a certain role in helping and guiding the selective maintenance of subsequent roads.

The volume of the subsidence disease calculated in this example is 0.1532m ³ , and no protrusion type of disease was detected.

S405. Based on the depth/height of the road surface disease and the subsidence/bulge volume of the road surface area, complete the three-dimensional evaluation of road deformation-type diseases.

The evaluation indexes of deformed diseases such as bulging, waves, ruts, subsidence, etc., mostly use 5mm, 15mm, and 25mm as the boundary. The maximum deformation depths H _a and H _b of the present invention also use 5mm, 15mm, and 25mm as boundaries to divide the severity of deformation, as shown in Table 1. The difference is that the commonly used indicators of deformation-related diseases are the distance from the line to the lowest point by the virtual drawing method or the virtual ruler method, and the maximum deformation depth H _a and H _b in the present invention are the distance from the fitted road reference plane to the lowest point of the disease. point distance. Therefore, for the same road section, the calculated values of the maximum deformation depth H _a and H _b are relatively smaller.

Table 1 Classification table of disease severity

Combined with Table 1, in the detection of actual road deformation defects, H _a , H _b , V _a , V _b , and the severity of each road section are listed in Table 2 to evaluate the overall road deformation. Among them, H _a and H _b determine the severity of road deformation, and Va and Vb are the volumes of subsidence and bulge. As for the volume indicators V _a and V _b , they can provide certain information support for the maintenance and monitoring scheme. For road sections with medium and low severity, if there is a large deformation volume, the disease of this type of road section is more likely to develop into a high-severity disease in the future, and it needs to be paid attention to in the follow-up monitoring and maintenance. label in .

Table 2 Deformation depth and volume summary table of pavement area

The embodiment of the present invention also proposes a road deformation-type disease detection system based on a single-line laser point cloud, including an information collection module, a road three-dimensional point cloud acquisition module, a disease area positioning module, a disease area deformation volume calculation module and a processor that can A computer program that runs. It should be noted that each module in the above system corresponds to the specific steps of the method provided by the embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method. For technical details not described in detail in this embodiment, refer to the method provided in the embodiment of the present invention.

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein. The solutions in the embodiments of the present application can be realized by using various computer languages, for example, the object-oriented programming language Java and the literal translation scripting language JavaScript.

The present application is described with reference to flowcharts and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

While preferred embodiments of the present application have been described, additional changes and modifications to these embodiments can be made by those skilled in the art once the basic inventive concept is appreciated. Therefore, the appended claims are intended to be construed to cover the preferred embodiment and all changes and modifications which fall within the scope of the application.

Obviously, those skilled in the art can make various changes and modifications to the application without departing from the spirit and scope of the application. In this way, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims (8)

1. A road deformation class disease detection method based on single-line laser point cloud, is characterized in that, comprises the following steps: S1. Mount the single-line lidar on the UAV, record the pose of the UAV at different times while the UAV is flying along the specified route, and collect road point cloud data in the form of cross-sections; S2. Denoise the collected road point cloud data to obtain a single-frame point cloud of the road, filter the point cloud of the outer lane, fit it to the standard cross-section in a straight line, and insert the single frame of the inner lane according to the slope of the fitted line Point cloud, splicing and inserting the point cloud and the point cloud of the outer lane to obtain the 3D point cloud of the road; S3. According to the three-dimensional point cloud in step S2, divide the road into specific interval road surface areas, fit the reference driving surface of the road by the thin plate spline interpolation method, locate the area where the disease is located on the divided road surface area and determine the actual curved surface of the road equation; S4. Calculate the maximum subsidence of the divided disease area, the maximum protrusion of the disease, and the corresponding deformation volume. 2. the road deformation class disease detection method based on single-line laser point cloud according to claim 1, is characterized in that, the specific steps of step S2 are as follows: S201. Use the radius filtering algorithm to make a circle with its center for each point cloud. If the number of points contained in the circle is greater than a fixed value, it will be retained, and if it is less than a fixed value, it will be deleted. Using a statistical filtering algorithm, for each point cloud, solve its to The average value of all point distances in the field, and then calculate the mean value μ and variance σ of all average values, set μ+n as the threshold, n as the specified multiple, and use the threshold as the screening value, thereby preliminarily eliminating road point cloud data Noise data to obtain a single-frame point cloud of the road; determine the range of the road surface point cloud based on the road cross slope and the elevation of the lowest point of the road surface, so as to extract an accurate cross-sectional point cloud; S202. Using the method of extracting the outer point cloud of single frame data based on the point cloud coordinates, screening the point cloud of 0.5m-3m outside the cross-sectional direction in the single frame point cloud, as the emergency lane point cloud, and fitting the emergency lane points in a straight line Cloud, forming a standard cross-section, and according to the slope of the line, insert the single-frame point cloud on the inside of the emergency lane at point intervals within the road range; S203. Based on the single-line radar pose recorded in step S1, the cross-sectional point cloud extracted in S201, and the single-frame point cloud in S202, through the coordinate transformation from NED to LLA coordinate system, the point cloud data of consecutive frames are combined to form a three-dimensional road point cloud. data and the 3D point cloud data of the reference driving surface. 3. the road deformation class disease detection method based on single-line laser point cloud according to claim 2, is characterized in that, the specific steps of step S3 are as follows: S301. Based on the point cloud of the emergency lane collected in step S202, extract the plane coordinates of the point cloud, use the cubic curve method to fit the overall line shape of the road, and divide the curve into small sections of 10m, and extract the x, y coordinates and points of the subsections. For the normal plane of the curve at the section, the point on the inside of the normal plane on the xOy projected straight line 20m away from the trend curve is taken to determine the road area clamped by the normal plane, so as to realize the division of a 10m section of road surface area on the reference driving surface; the normal plane takes points The coordinates are marked as: QD＝[[X ₀₁ ,y ₀₁ ,x ₀₂ ,y ₀₂ ],...] Among them, x ₀₁ represents the abscissa of the intersection point of the trend curve and the first normal plane, y ₀₁ represents the ordinate of the intersection point of the trend curve and the first normal plane, x ₀₂ represents the abscissa of the point inside the first normal plane, and y ₀₂ represents the ordinate of the point taken inside the first normal plane; S302. Based on the three-dimensional point cloud data of the reference traffic surface in step S203, a plane fitting algorithm is used to fit the reference traffic surface of the road to obtain parameters A ₀ , A ₁ , and A ₂ of the plane equation of the reference traffic surface. The specific formula is: z＝A _o x+A ₁ y+A ₂ Among them, A ₀ , A ₁ , and A ₂ are the parameters of the plane about the x and y coordinates and the intercept of the z coordinate of the plane when the x and y coordinates are equal to 0; S303, based on the step S203 three-dimensional road point cloud data and S302 road datum driving surface equation, calculate the elevation difference between the three-dimensional road point cloud data and the datum plane corresponding to the xOy coordinates; S304. Based on the grid method, divide the xOy coordinates of the road datum plane into 1dm×1dm grids, if the absolute value of the elevation difference of the point cloud is greater than the preset threshold, and the point is located below the datum plane, then the point is a subsidence Disease point, the grid area where the point is located is a subsidence disease area; if the absolute value of the point cloud elevation difference is greater than the preset threshold, and the point is above the datum, then the point is a convex disease point, and the point is located The grid area is the raised disease area; thus the subsidence diseased area Sa and the raised diseased area Sb are preliminarily divided; S305. Based on the point cloud of the subsidence diseased area and the raised diseased area, the thin-plate spline interpolation method is used to fit the three-dimensional surface equation of the road diseased area. The specific formula is: U(x)＝r ² ln r Among them, p(x, y) is any point on the surface, U(x) is the radial basis function, ||pp _i || indicates the distance from point p to a certain control point, known control points 1, 2 , 3..., N, ω _i represent the weighting of different radial basis, m ₀ , m ₁ , m ₂ are the parameters of this plane; S306, establish the control point matrix and the height matrix of the point cloud data, the specific formula is: (1) Control point matrix Among them, n is the number of control points, and the second and third columns represent the (x, y) coordinates of the control points; (2) Height matrix Among them, v ₁ to v _n represent the coordinates in the z direction of each control point; S307. Calculate the radial basis function value of any two control points, the specific formula is: Among them, r _ij represents the distance between control points i and j, U(r _ij ) is the value of the radial basis function corresponding to the distance r _ij ; S308. Define matrix L as: Then the above matrix has the following relationship: Y=L*(ω ₁ ,...ω _N , m ₀ , m ₁ , m ₂ ) ^T ; Based on the thin-plate spline interpolation principle, the conditional function about the control points is substituted, all parameters of the three-dimensional surface equation of the road are calculated, and the interpolation is completed. The specific matrix is: Among them, ω _ij is the weight of the jth radial basis on the i-th segment, m _i0 , m _i1 , m _i2 are the m ₀ , m ₁ , m ₂ coefficients on the i-th segment. 4. The road deformation class disease detection method based on single-line laser point cloud according to claim 3, characterized in that, the specific steps of step S4 are: S401. Extract the points M and N corresponding to the maximum elevation difference, and use the grid where the M and N points are located and the 8 surrounding grids as the selected area, and solve the points M and N based on the steepest descent/ascent method The gradient of , the specific formula is: f = z _actual - z _reference Among them, f is the difference between the actual surface elevation of the road and the road reference driving surface, z _actual is the actual surface elevation of the road, and z _reference is the reference driving surface of the road; let the known minimum point in the selected disease area be M, after k times The point obtained by iteration is denoted as M ^k , P ^k represents the direction of the maximum surface change rate of point M ^k , and d ^k represents the gradient; S402. According to the formula in step S305, obtain the z _reference value, solve the differential for the z _reference value formula, and obtain the following formula: d ^k is finally expressed as: Among them, N is the number of control points for datum surface fitting, Q is the number of control points for the actual road surface fitting, M ^k (x, y) is the starting point of the current iteration; x _i , y _i are the corresponding control points coordinates; r is the distance from M ^k (x, y) to each control point; S403, take 0.01m as the step size to solve the elevation of the next point, and iterate until an extreme point occurs, and the elevation difference between the extreme point and the datum level corresponding to the xOy coordinates is the depth of disease/disease height on the road surface area; S404, regularly take 9 points in each disease grid, and calculate the average value of its elevation take /> The product of the grid area and the grid area is the deformation volume V _i of the grid, and the specific formula is: Among them, S is the grid area of 4cm ² ; V _i is the deformation volume of grids with deformation defects from left to right and bottom to top in each pavement area; Adding up all grid deformation volumes is the subsidence/bulge volume V _a , V _b of the pavement area; S405. Based on the depth/height of the road surface disease and the subsidence/bulge volume of the road surface area, complete the three-dimensional evaluation of road deformation-type diseases. 5. A road deformation-type disease detection system based on a single-line laser point cloud, characterized in that it includes The information acquisition module is used to mount the single-line lidar on the UAV, record the pose of the UAV at different times during the flight of the UAV along the specified route, and collect road point cloud data in the form of cross-section; The road 3D point cloud acquisition module is used to denoise the collected road point cloud data, obtain a single frame point cloud of the road, filter the point cloud of the outer lane, and fit it to the standard cross-section in a straight line, and according to the fitting line Insert the single-frame point cloud of the inner lane with the slope of , splicing and inserting the point cloud and the point cloud of the outer lane to obtain a three-dimensional point cloud of the road; The disease area positioning module is used to divide the road into specific interval road surface areas according to the three-dimensional point cloud of the road, fit the reference driving surface of the road through the thin plate spline interpolation method, locate the disease area on the divided road surface area and determine the road The actual surface equation of ; The deformation volume calculation module of the diseased area is used to calculate the maximum subsidence, the maximum protrusion of the diseased area and the corresponding deformation volume of the divided diseased area. 6. The road deformation class disease detection system based on single-line laser point cloud according to claim 5, characterized in that, in the road three-dimensional point cloud acquisition module, the specific steps are as follows: Step 1. Use the radius filtering algorithm to make a circle with its center for each point cloud. If the number of points contained in the circle is greater than a certain value, it will be retained, and if it is less than a certain value, it will be deleted. Using a statistical filtering algorithm, for each point cloud, its Calculate the average value of the distances to all points in the domain K, and calculate the mean value μ and variance σ of all average values, set μ+nσ as the threshold, n as the specified multiple, and use the threshold as the screening value to initially eliminate road points Cloud data noise data to obtain a single-frame point cloud of the road; determine the range of the road surface point cloud based on the road cross slope and the elevation of the lowest point of the road surface, so as to extract accurate cross-sectional point clouds; Step 2. Use the method of extracting the outer point cloud of single-frame data based on the point cloud coordinates, and select the point cloud of 0.5m-3m outside the cross-sectional direction in the single-frame point cloud as the point cloud of the emergency lane, and fit the emergency lane in a straight line Point cloud, forming a standard cross-section, and according to the slope of the straight line, insert the single-frame point cloud inside the emergency lane at point intervals within the road range; Step 3. Based on the single-line radar pose, cross-sectional point cloud and single-frame point cloud, through the coordinate transformation from NED to LLA coordinate system, the point cloud data of consecutive frames are combined to form 3D road point cloud data and 3D point cloud of reference driving surface data. 7. The road deformation class disease detection system based on single-line laser point cloud according to claim 5, wherein, in the disease area positioning module, the specific steps are as follows: Step 1. Based on the point cloud of the emergency lane, extract the plane coordinates of the point cloud, use the cubic curve method to fit the overall line shape of the road, divide the curve into 10m segments, and extract the x and y coordinates of the segments and the curves of the segments The normal plane of the normal plane, take the point on the inner side of the normal plane on the xOy projection straight line 20m away from the trend curve, determine the road area clamped by the normal plane, so as to realize the division of a section of road surface area of 10m on the reference driving surface; the coordinates of the points taken by the normal plane are marked as : QD＝[[x ₀₁ ,y ₀₁ ,x ₀₂ ,y ₀₂ ],...] Among them, x ₀₁ represents the abscissa of the intersection point of the trend curve and the first normal plane, y ₀₁ represents the ordinate of the intersection point of the trend curve and the first normal plane, x ₀₂ represents the abscissa of the point inside the first normal plane, and y ₀₂ represents the ordinate of the point taken inside the first normal plane; Step 2. Based on the 3D point cloud data of the reference traffic surface, use the plane fitting algorithm to fit the reference traffic surface of the road, and obtain the parameters A ₀ , A ₁ , and A ₂ of the plane equation of the reference traffic surface. The specific formula is: z＝A ₀ x+A ₁ y+A ₂ Among them, A ₀ , A ₁ , and A ₂ are the parameters of the plane about the x and y coordinates and the intercept of the z coordinate of the plane when the x and y coordinates are equal to 0; Step 3, based on the three-dimensional road point cloud data and the road datum driving surface equation, calculate the elevation difference between the three-dimensional road point cloud data and the datum plane corresponding to the xOy coordinates; Step 4. Based on the grid method, the xOy coordinates of the road reference plane are divided into 1dm×1dm grids. If the absolute value of the point cloud elevation difference is greater than the preset threshold and the point is located below the reference plane, then the point is The subsidence disease point, the grid area where the point is located is the subsidence disease area; if the absolute value of the elevation difference of the point cloud is greater than the preset threshold, and the point is above the reference plane, then the point is a convex disease point, and the point is located at The grid area at is the convex disease area; thus the subsidence disease area Sa and the convex disease area Sb are preliminarily divided; Step 5. Based on the point clouds of the subsidence diseased area and the raised diseased area, the thin-plate spline interpolation method is used to fit the three-dimensional surface equation of the road diseased area. The specific formula is: U(x)＝r ² lnr Among them, p(x, y) is any point on the surface, U(x) is the radial basis function, ||pp _i || indicates the distance from point p to a certain control point, known control points 1, 2 , 3..., N, ω _i represent the weighting of different radial basis, m ₀ , m ₁ , m ₂ are the parameters of this plane; Step 6. Establish the control point matrix and height matrix of the point cloud data. The specific formula is: (1) Control point matrix Among them, n is the number of control points, and the second and third columns represent the (x, y) coordinates of the control points; (2) Height matrix Among them, v ₁ to v _n represent the coordinates in the z direction of each control point; Step 7, calculate the radial basis function value of any two control points, the specific formula is: Among them, r _ij represents the distance between control points i and j, U(r _ij ) is the value of the radial basis function corresponding to the distance r _ij ; Step 8, define the matrix L as: Then the above matrix has the following relationship: Y=L*(ω ₁ ,...ω _N , m ₀ , m ₁ , m ₂ ) ^T ; Based on the thin-plate spline interpolation principle, the conditional function about the control points is substituted, all parameters of the three-dimensional surface equation of the road are calculated, and the interpolation is completed. The specific matrix is: Among them, ω _ij is the weight of the jth radial basis on the i-th segment, m _i0 , m _i1 , m _i2 are the m ₀ , m ₁ , m ₂ coefficients on the i-th segment. 8. The road deformation class disease detection system based on single-line laser point cloud according to claim 5, wherein, in the disease area deformation volume calculation module, the specific steps are as follows: Step 1. Extract the points M and N corresponding to the maximum elevation difference, take the grid where the M and N points are located and the 8 surrounding grids as the selected area, and solve the points M and N based on the steepest descent/ascent method The gradient out of N, the specific formula is: f = z _actual - z _reference Among them, f is the difference between the actual surface elevation of the road and the road reference driving surface, z _actual is the actual surface elevation of the road, and z _reference is the reference driving surface of the road; let the known minimum point in the selected disease area be M, after k times The point obtained by iteration is denoted as M ^k , P ^k represents the direction of the maximum surface change rate of point M ^k , and d ^k represents the gradient; S402. According to the formula in step S305, obtain the z _reference value, solve the differential for the z _reference value formula, and obtain the following formula: d ^k is finally expressed as: Among them, N is the number of control points for datum surface fitting, Q is the number of control points for the actual road surface fitting, M ^k (x, y) is the starting point of the current iteration; x _i , y _i are the corresponding control points coordinates; r is the distance from M ^k (x, y) to each control point; Step 3. Solve the elevation of the next point with a step size of 0.01m, and iterate until an extreme point appears. The elevation difference between the extreme point and the reference plane corresponding to xOy coordinates is the disease depth/disease height on the road surface area ; Step 4. Take 9 points regularly in each disease grid and calculate the average value of their elevations take /> The product of the grid area and the grid area is the deformation volume V _i of the grid, and the specific formula is: in, is the average value of the elevation sum of the nine points taken, S is the grid area 4cm ² ; V _i is the deformation volume of grids with deformation defects from left to right and then from bottom to top in each pavement area ; Adding up all grid deformation volumes is the subsidence/bulge volume V _a , V _b of the pavement area; Step 5. Based on the depth/height of the road surface disease and the subsidence/bulge volume of the pavement area, the three-dimensional evaluation of road deformation-type diseases is completed.