A method and system for detecting defects in a mine dump retaining wall

Abstract

The application belongs to the technical field of mine construction, and specifically discloses a mine dump retaining wall defect detection method and system, which comprises a processor, a laser radar, a GPS positioning device and a management platform, and adopts the following steps: a rectangular detection area is generated according to the length of the parking line and the width of the retaining wall; retaining wall point clouds in the detection area are obtained, and multiple frames of the retaining wall point clouds are overlapped to obtain dense point clouds; the detection area is cut along the length direction at equal intervals to obtain grids, the dense point clouds are separated according to the intervals and projected into the corresponding grids respectively; the dense point clouds in each grid are judged to identify invalid grids; and the invalid grids are subjected to retaining wall risk judgment; and the method has the following advantages: the highest point, the distribution size and the slope of the extracted gridded retaining wall point clouds are used to judge the overall missing of the retaining wall from three dimensions, so that the safety hidden danger is reduced.

Description

Technical Field

[0001] This invention relates to the field of mining construction technology, and more specifically, to a method and system for detecting defects in retaining walls of mining spoil heaps. Background Technology

[0002] The spoil heap is one of the most important operating areas in the mining area. During the unloading process of mining vehicles at the spoil heap, it is necessary to control the safe parking of mining vehicles based on the distance between the end marker - the retaining wall and the mining vehicles to prevent mining vehicles from crossing the boundary. If the spoil heap retaining wall fails, it will seriously affect the safe parking of mining vehicles, especially the safe parking of intelligent driving vehicles. How to detect the risk of failure of the retaining wall in a timely manner, or whether the retaining wall has failed, is a very important issue.

[0003] In existing technologies, the distance between the vehicle and the retaining wall is usually measured at the spoil heap to find a suitable parking point before unloading begins. However, there is a lack of a process to check whether the retaining wall is missing, which poses a great safety hazard.

[0004] To address these issues, a method and system for detecting defects in retaining walls at mine spoil heaps are proposed. Summary of the Invention

[0005] The present invention aims to provide a method and system for detecting defects in retaining walls at mining spoil heaps, in order to solve or improve the problem mentioned above of the lack of a process for detecting whether retaining walls are missing, which poses a great safety hazard and improves the safety of vehicles in mining areas.

[0006] In view of this, the first aspect of the present invention is to provide a method for detecting defects in retaining walls of mine spoil heaps.

[0007] A second aspect of the present invention is to provide a defect detection system for retaining walls in mining spoil heaps.

[0008] The first aspect of the present invention provides a method for detecting defects in retaining walls at mining spoil heaps, comprising the following steps: S1, cutting the boundary line of the retaining wall opposite to the vehicle as a stop line, and generating a rectangular detection area based on the length of the stop line and the width of the retaining wall; S2, segmenting the point cloud acquired by the vehicle-mounted lidar to obtain the retaining wall point cloud within the detection area, and overlapping multiple frames of the retaining wall point cloud to obtain a dense point cloud; S3, cutting the detection area at equal intervals along its length to obtain a grid, separating the dense point cloud according to the intervals and projecting it onto the corresponding grid; S4, judging the vertical highest point, the distribution size along the width of the retaining wall, and the distribution slope of the dense point cloud in each grid to identify invalid grids; S5, performing a retaining wall risk assessment on the invalid grids, and issuing an alarm that the current retaining wall has defects based on the risk assessment result.

[0009] This invention provides a method for detecting defects in retaining walls at mining spoil heaps. Specifically, it employs a multi-dimensional judgment method for retaining wall missing detection. By extracting information from the rasterized retaining wall point cloud in three dimensions—the highest point, distribution size, and slope—the overall missingness of the retaining wall is simultaneously determined from these three dimensions. Specifically, the three-dimensional judgment requires first extracting the highest point, distribution size, and slope information from the rasterized retaining wall point cloud in the vertical direction, and then judging the continuity of information that does not meet the set conditions in the horizontal direction. An alarm is triggered when the continuous area that does not meet the conditions exceeds a certain threshold. This multi-dimensional judgment method for retaining wall missing detection can comprehensively determine the safety of the retaining wall from multiple angles.

[0010] By combining continuous GPS positioning data and using a multi-frame overlay method, a denser retaining wall point cloud is constructed, compensating for the sparsity of the LiDAR point cloud. The retaining wall point cloud and the corresponding GPS information within the ROI region are cached for N consecutive frames. Based on the coordinate transformation matrix from LiDAR point cloud to GPS data and the inverse of the matrix, the cached N consecutive frames of point cloud are transformed to the current coordinate system. The point cloud after overlay at the current time is obtained by accumulating the point cloud data. The multi-frame overlay point cloud provides a more detailed description of the retaining wall, extracts feature information more accurately, and avoids false detections caused by insufficient point cloud accuracy or partial missing data.

[0011] In addition, the technical solutions provided by embodiments of the present invention may also have the following additional technical features:

[0012] In any of the above technical solutions, when there is no dense point cloud in the grid, the height of the ground point cloud corresponding to the vertical direction of the current grid is taken as the height of the current grid, the width and slope of the current grid are set to zero, and the current grid is recorded as an invalid grid.

[0013] In this technical solution, in order to ensure the smooth implementation of the method, when the retaining wall collapses or breaks and there is exposed ground in the longitudinal area of ​​the retaining wall, the point cloud of the retaining wall cannot be detected. Therefore, the height of the ground point cloud corresponding to the longitudinal direction is used as the grid height, and the grid width and slope are set to zero so that subsequent calculations and judgments can be carried out smoothly. At the same time, when the ground is too high, the actual height can also be included in the statistics of the point cloud.

[0014] In any of the above technical solutions, the step of determining the highest vertical point specifically includes: obtaining the standard height of the retaining wall; extracting the highest point of the dense point cloud in each grid; comparing the highest point of each grid with the standard height, and if it is lower than the standard height, the current grid is an invalid grid.

[0015] In this technical solution, the highest point of the point cloud in each grid of each vertical column is extracted as the grid height \(H_{g1}, H_{g2}, \cdots, H_{gN}\). If there is no point cloud in the grid, it is replaced by the ground height value point in the same coordinate system.

[0016] The obtained grid heights are compared with the standard height \(dh\) of the retaining wall in sequence. The grids that meet the requirement of \(H_{g}>dh\) are recorded as valid grids, and the grids that meet \(H_{g}<dh\) are recorded as invalid grids.

[0017] In any of the above technical solutions, the steps for judging the distribution size along the width direction of the retaining wall are as follows: specifically, take half of the product of the reference retaining wall width and the allocation value based on the soil slope as the standard width; obtain the distribution size along the width direction of the dense point cloud greater than half of the standard height in each of the grids; compare the distribution size along the width direction with the standard width respectively. If it is smaller, the current grid is an invalid grid.

[0018] In this technical solution, calculate the horizontal distribution widths \(W_{g1}, W_{g2}, \cdots, W_{gN}\) of the grids where the height of the point cloud in each vertical column is greater than \(dh / 2\). The horizontal distribution width of the grid point cloud , is recorded as a valid grid, and the horizontal distribution width of the grid point cloud , is recorded as an invalid grid. Among them, \(ds\) is the reference retaining wall width requirement, is the allocation value based on the soil slope.

[0019] In any of the above technical solutions, the steps for judging the distribution slope are as follows: specifically, obtain the standard slope, and the specific calculation rule is: standard slope = standard height / (allocation value based on soil slope 0.5 × reference retaining wall width); obtain the distance of the dense point cloud distributed along the width direction of the retaining wall and the height of the highest point along the longitudinal direction of the grid in a single grid, and take the ratio of the height of the highest point to the distance distributed along the width direction of the retaining wall as the distribution slope; compare the distribution slope of each grid with the standard slope respectively. If it is less, the current grid is an invalid grid.

[0020] In this technical solution, use the distance \(Y\) in the length direction and the longitudinal height \(Z\) information of the point cloud to calculate and record the slope size of the point cloud distribution in each vertical column, \(Slope_{g1}, Slope_{g2}, \cdots, Slope_{gN}\). The slope of the point cloud distribution \(Slope_{g}>dh / ( ds)\) is recorded as a valid grid, and \(Slope_{g}<dh / ( ds)\) is recorded as an invalid grid. Among them, \(dh\) is the standard height of the retaining wall, \(ds\) is the reference retaining wall width requirement, [[ID=

[0021] In any of the above technical solutions, the current grid is a valid grid when the highest point of the grid is greater than or equal to the standard height, the size distributed along the width of the retaining wall is greater than or equal to the standard width, and the distribution slope is greater than or equal to the standard slope.

[0022] In this technical solution, a grid is defined as valid when all three factors are valid, thus ensuring that the determination of a valid grid matches the actual situation.

[0023] In any of the above technical solutions, the risk judgment adopts the following rules: calculate the proportion of the number of invalid grids to the total number of grids in the current retaining wall, and if the proportion is greater than 30%, the current retaining wall is defective; and / or determine the product of the consecutive number of invalid grids and the spacing and half the width of the rear wheel of a vehicle, and if it is greater than half the width of the rear wheel of a vehicle, the current retaining wall is defective.

[0024] In this technical solution, two rules are used for separate judgments, which can ensure that the final judgment result can be applied to vehicle safety in practice to the greatest extent and reduce the damage caused by judgment errors to vehicle operation.

[0025] In any of the above technical solutions, before step S1, the following steps are also included: continuously detecting whether the minimum distance between the vehicle and the parking line, and the angle between the vehicle's heading and the parking line meet the standard; if so, the vehicle stops proceeding to step S1.

[0026] In this technical solution, the minimum distance between the vehicle and the parking line, as well as the angle between the vehicle's heading and the parking line, are used to determine whether the vehicle and the retaining wall are close enough and whether their postures are appropriate, so that this method can be applied stably.

[0027] In any of the above technical solutions, the standard is that the minimum distance between the vehicle and the parking line is less than 15 meters, and the angle between the vehicle's heading and the parking line is greater than 75 degrees.

[0028] In this technical solution, the preset 15-meter distance and 75-degree angle ensure that the implementation system can carry out the method normally.

[0029] The first aspect of the present invention provides a method for detecting defects in retaining walls at mining waste dumps, comprising: a processor, a lidar, a GPS positioning device, and a management platform; the lidar is used to detect the retaining wall behind the vehicle and generate a point cloud when the vehicle is reversing; the GPS positioning device is used to acquire the vehicle's location information; the management platform is used to distribute a boundary map file output by the GPS positioning device, the boundary map file including the boundary line information of the retaining wall; the processor is used to load the point cloud and the location information, convert the point cloud to the coordinate system of the GPS positioning device, filter and extract the retaining wall points in the point cloud according to the boundary map file, and determine whether the retaining wall has defects by combining the vehicle size information and the retaining wall standard; wherein, the mining waste dump retaining wall defect detection system is used to implement the mining waste dump retaining wall defect detection method described in any of the technical solutions of the first aspect.

[0030] In this technical solution, the lidar is installed directly behind the vehicle, enabling it to detect the barrier behind the vehicle when reversing; the GPS positioning device is installed on the vehicle to obtain the vehicle's location information; the lidar data coordinate system and the GPS positioning device data coordinate system have a measurable and calculable mutual conversion relationship; the processor can load the point cloud detected by the lidar and the GPS positioning device data.

[0031] The LiDAR uses point clouds to characterize the 3D environment behind the vehicle; GPS positioning devices are used for precise positioning and trigger detection algorithms; the management platform sends map boundary files to the processor, and ROI regions are created using the map boundaries provided by the map boundary files, accurately extracting LiDAR point clouds from the retaining wall for targeted data analysis; the high detection accuracy of LiDAR and GPS positioning devices, and the ROI regions created using map boundaries sent by the management platform can eliminate external interference.

[0032] The beneficial effects of this invention compared to the prior art are as follows:

[0033] By combining continuous GPS positioning data with a multi-frame overlay method, a denser retaining wall point cloud is constructed, compensating for the sparsity of the LiDAR point cloud. The retaining wall point cloud and its corresponding GPS information within N consecutive frames of the ROI are cached. Based on the coordinate transformation matrix from LiDAR point cloud to GPS data and its inverse, the cached N consecutive frames of point cloud are transformed to the current coordinate system. The point cloud is then accumulated and overlaid at the current time. This multi-frame overlay provides a more detailed description of the retaining wall, extracts more accurate feature information, and avoids false detections caused by insufficient point cloud precision or partial missing data.

[0034] By extracting information from the three dimensions of the rasterized retaining wall point cloud—the highest point, distribution size, and slope—the overall incompleteness of the retaining wall can be determined simultaneously from these three dimensions, enabling the detection of defects in the retaining wall and reducing safety hazards.

[0035] Additional aspects and advantages of embodiments of the invention will become apparent in the following description or may be learned by practice of embodiments of the invention. Attached Figure Description

[0036] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0037] Figure 1 This is a flowchart of an embodiment of the retaining wall missing detection method of the present invention;

[0038] Figure 2 This is a schematic diagram illustrating the detection of different damage conditions of the retaining wall according to the present invention;

[0039] Figure 3 This is a schematic diagram of the structure of the defect detection system for the retaining wall of the mine spoil heap of the present invention.

[0040] in, Figure 1-3 The correspondence between the reference numerals and component names in the attached drawings is as follows:

[0041] 101 Processor, 102 LiDAR, 103 GPS positioning device, 104 Management platform. Detailed Implementation

[0042] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0043] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0044] Please see Figure 1-3 The present invention provides a method and system for detecting defects in retaining walls of mining spoil heaps, wherein the method and system for detecting defects in retaining walls of mining spoil heaps includes: a lidar 102, a GPS positioning device 103, a management platform 104, and a processor 101.

[0045] The lidar 102 is installed directly behind the vehicle and can detect the barrier behind the vehicle when reversing;

[0046] The management platform 104 issues map boundary files. The map boundaries are collected and created by GPS positioning devices and include the boundary line information of the retaining wall. One way is to store them in advance at a fixed location in the processor 101, and another way is to download them to the fixed location in the processor 101 in real time through network communication.

[0047] The GPS positioning device 103 needs to be installed or calibrated to the center of the rear axle to obtain the vehicle's location information;

[0048] The data coordinate system of the lidar 102 and the data coordinate system of the GPS positioning device 103 have a measurable and calculable mutual conversion relationship;

[0049] The processor 101 loads the point cloud detected by the lidar 102 and the data from the GPS positioning device 103, converts the lidar point cloud to the coordinate system of the GPS positioning device, filters and extracts the retaining wall points in the lidar point cloud according to the map boundary issued by the management platform 104, and determines whether the retaining wall is missing by combining the vehicle size information and the retaining wall standard.

[0050] The method for detecting defects in retaining walls at mining spoil heaps includes the following steps:

[0051] After the vehicle reverses into the spoil heap, the GPS positioning device 103 detects the vehicle's location and determines the retaining wall boundary line and parking position corresponding to the vehicle's current position.

[0052] like Figure 2 (a) The vehicle's position is detected by the GPS positioning system at time 103. Based on the vehicle's GNSS positioning information, the nearest point on the parking line is searched. The vehicle is within the detection range of the lidar, specifically 15 meters away from the parking line, and the vehicle's heading is nearly perpendicular to the parking line, specifically at 75 degrees. Retaining wall detection begins:

[0053] like Figure 2 (b) Read the map boundary issued by the management platform 104. Based on the vehicle position and vehicle width, extract the map boundary line directly opposite the vehicle as the parking line. The parking line is converted to the LiDAR coordinate system to obtain the parking line L1. The length of the parking line L1 is d1.

[0054] Based on the standard width ds of the retaining wall, the parking line L1 is shifted backward by a distance ds to obtain line L2. That is, the corresponding points of line L1 and line L2 are all ds apart.

[0055] Lines L1 and L2, along with the line connecting their corresponding endpoints, form the detection region (ROI).

[0056] The lidar 102 acquires lidar point cloud and transmits it to the processor 101. Each point in the lidar point cloud contains XYZ information. By determining whether the X and Y positions of the point cloud are within the ROI area, the retaining wall point cloud within the ROI area is extracted.

[0057] Since the laser point cloud becomes sparse with increasing distance, the laser radar point cloud of the ROI area is superimposed in multiple frames based on continuous GPS positioning information to make the point cloud on the retaining wall denser, so as to facilitate subsequent calculations.

[0058] In the ROI region, divide dl horizontally into grids with a horizontal precision of deta_grid. The size of each grid in the direction of the retaining wall width can maintain an accuracy of ds / n, where n >= 1;

[0059] Project the extracted point cloud into the grid:

[0060] (1) Height factor judgment

[0061] Extract the highest point of the point cloud in each grid of each vertical column as the corresponding grid height H_g1, H_g2,... H_gN. If there is no point cloud in the grid, replace it with the ground height value point in the same coordinate system;

[0062] Compare the extracted grid heights with the standard retaining wall height dh in sequence. Those that meet the requirement of the retaining wall height H_g > dh are marked as valid, and those that meet H_g < dh are marked as invalid;

[0063] (2) Width factor judgment

[0064] Calculate the distribution width W_g1, W_g2,... W_gN in the direction of the retaining wall width of the grids where the point cloud height in each vertical column is greater than dh / 2. If there is no point cloud, record it as 0. The distribution width of the grid point cloud in the direction of the retaining wall width , is marked as a valid grid, and the distribution width of the grid point cloud in the direction of the retaining wall width , is marked as an invalid grid. Among them, ds is the reference retaining wall width requirement, is the adjustment value based on the soil slope.

[0065] (3) Slope factor

[0066] Utilize the longitudinal distance Y and height Z information of the point cloud to calculate and record the slope magnitude of the point cloud distribution in each vertical column, Slope_g1, Slope_g2,... Slope_gN. If there is no point cloud, record it as 0. The slope of the point cloud distribution Slope_g > dh / ( ds), is marked as a valid grid, Slope_g < dh / ( ds) is marked as an invalid grid, where dh is the standard retaining wall height, ds is the reference retaining wall width requirement, is the adjustment value based on the soil slope.

[0067] When all three factors are valid, it is marked as a valid grid; otherwise, it is marked as an invalid grid. Evaluate the safety and missing degree of the retaining wall based on the continuity of the valid grids and the proportion of the invalid grids. For example, Figure 2 In (b)(b)(d), the solid dots are the points in the valid grids, and the hollow dots are the points in the invalid grids, Figure 2(b) Determined to be a safe retaining wall; if the percentage of invalid grid columns N0 within the ROI area is greater than P, for example: N0>30%, it is considered a hazard, and a retaining wall missing alarm is triggered, such as... Figure 2 (c); or the number of consecutive invalid grid columns N0 multiplied by the vertical grid precision deta_grid, representing the length of the invalid area, is greater than the width of the rear wheel by 0.5*W. If N*deta_grid>0.5*W, then a hazard is considered, and a retaining wall missing alarm is triggered. Figure 2 (d).

[0068] The method described in this embodiment enables the detection of safety deficiencies in mine spoil heaps, such as missing, damaged, or insufficient retaining walls, providing a final guarantee for the operation of mine truck spoil heaps.

[0069] Since laser point clouds become sparser with increasing distance, this method uses continuous GPS positioning information to overlay multiple frames of the LiDAR point cloud in the ROI area, making the point cloud on the retaining wall denser and facilitating calculation.

[0070] The technical implementation is as follows: cache the retaining wall point cloud and the corresponding GPS information within the ROI region of consecutive N_data frames, such as map(lidar_data_t0,GPS_t0), map(lidar_data_t1,GPS_t1), ..., map(lidar_data_tN,GPS_tN), where map(lidar_data_tN,GPS_tN) represents the retaining wall point cloud and GPS location information acquired at the same time in the previous N periods, that is, map(lidar_data_t0,GPS_t0) is the retaining wall point cloud and GPS location information acquired at the same time in the current moment; map(lidar_data_t1,GPS_t1) is the retaining wall point cloud and GPS location information acquired at the same time in the previous period.

[0071] The coordinate transformation matrix from lidar point cloud to GPS data is T_lidar_gps, and its inverse is T_lidar_gps_inv.

[0072] Based on the current GPS positioning data GPS_t0 and the previous cycle's GPS positioning data GPS_t1, calculate the transformation matrix T_t1t0 from GPS_t1 to GPS_t0, transforming the point cloud lidar_data_t1 on the retaining wall from the previous cycle to the current laser coordinate system.

[0073]

[0074] Based on the current GPS positioning data GPS_t0 and the previous cycle's GPS positioning data GPS_t2, calculate the transformation matrix T_t2t0 from GPS_t2 to GPS_t0, transforming the point cloud lidar_data_t2 on the retaining wall from the previous two cycles to the current laser coordinate system.

[0075]

[0076] And so on:

[0077] Based on the current GPS positioning data GPS_t0 and the GPS positioning data GPS_tN from the previous N periods, calculate the transformation matrix T_tNt0 from GPS_tN to GPS_t0, and transform the point cloud lidar_data_tN on the retaining wall from the previous two periods to the current laser coordinate system.

[0078]

[0079] This allows us to obtain the superimposed point cloud at the current moment:

[0080] ...+

[0081] in, .

[0082] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this invention, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0083] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for detecting defects in retaining walls of mine spoil heaps, characterized in that, Includes the following steps: S1, the boundary line of the retaining wall opposite to the vehicle is cut off as the parking line, and a rectangular detection area is generated according to the length of the parking line and the width of the retaining wall. S2, segment the point cloud acquired by the vehicle-mounted lidar to obtain the wall point cloud in the detection area, and overlap the wall point clouds of multiple frames to obtain a dense point cloud. S3, the detection area is cut at equal intervals along its length to obtain a grid, and the dense point cloud is separated according to the interval and projected onto the corresponding grid. S4, determine the vertical highest point, distribution size and distribution slope of the dense point cloud in each grid, in order to identify invalid grids; S5, perform a retaining wall risk assessment on the invalid grid, and issue an alarm that the current retaining wall has defects based on the risk assessment result.

2. The method for detecting defects in retaining walls of mine spoil heaps according to claim 1, characterized in that, When there is no dense point cloud in the grid, the height of the ground point cloud corresponding to the vertical direction of the current grid is taken as the height of the current grid, the width and slope of the current grid are set to zero, and the current grid is recorded as an invalid grid.

3. The method for detecting defects in retaining walls of mine spoil heaps according to claim 2, characterized in that, The steps for determining the highest vertical point are as follows: Obtain the standard height of the retaining wall; Extract the highest point of the dense point cloud in each of the grids; The highest point of each grid cell is compared with the standard height. If the highest point is lower than the standard height, the current grid cell is considered invalid.

4. The method for detecting defects in retaining walls of mine spoil heaps according to claim 3, characterized in that, The steps for determining the size distribution along the width of the retaining wall are as follows: The standard width is half the product of the reference retaining wall width and the adjustment value based on the soil slope. Obtain the distribution size of dense point clouds with a height greater than half the standard height along the width of the retaining wall in each of the grid cells; The size of the grid distributed along the width of the retaining wall is compared with the standard width. If the size is smaller, the current grid is an invalid grid.

5. A method for detecting defects in retaining walls of mine spoil heaps according to claim 4, characterized in that, The steps for determining the slope of the distribution are as follows: The standard slope is obtained by following these calculation rules: Standard slope = Standard height / (Adjustment value based on soil slope × 0.5 × Reference retaining wall width); Obtain the distance of the dense point cloud distributed along the width of the retaining wall and the height of the highest point along the longitudinal direction of the grid within a single grid, and use the ratio of the height of the highest point to the distance distributed along the width of the retaining wall as the distribution slope; The distribution slope of each grid cell is compared with the standard slope. If the slope is less than the standard slope, the current grid cell is considered invalid.

6. A method for detecting defects in retaining walls of mine spoil heaps according to claim 5, characterized in that, The grid is considered a valid grid when its highest point is greater than or equal to the standard height, its size along the width of the retaining wall is greater than or equal to the standard width, and its slope is greater than or equal to the standard slope.

7. A method for detecting defects in retaining walls of mine spoil heaps according to claim 1, characterized in that, The risk assessment adopts the following rules: Calculate the ratio of the number of invalid grid cells to the total number of grid cells in the current retaining wall. If the ratio is greater than 30%, the current retaining wall is defective; and / or The product of the consecutive number of invalid grids and the spacing is determined by the size of half the width of the vehicle's rear wheel. If it is greater than half the width of the vehicle's rear wheel, then the current retaining wall is defective.

8. A method for detecting defects in retaining walls of mine spoil heaps according to claim 1, characterized in that, Before step S1, the following steps are also included: The system continuously checks whether the minimum distance between the vehicle and the parking line, as well as the angle between the vehicle's heading and the parking line, meet the standards. If so, the vehicle stops and proceeds to step S1.

9. A method for detecting defects in retaining walls of mine spoil heaps according to claim 8, characterized in that, The standard is that the minimum distance between the vehicle and the stop line is less than 15 meters, and the angle between the vehicle's heading and the stop line is greater than 75 degrees.

10. A defect detection system for retaining walls in mining spoil heaps, characterized in that, include: Processor (101), LiDAR (102), GPS positioning device (103), and management platform (104); A lidar (102) is used to detect the barrier behind the vehicle and generate a point cloud when reversing; The GPS positioning device (103) is used to obtain the vehicle's location information; The management platform (104) is used to distribute the boundary map file output by the GPS positioning device (103), the boundary map file including the boundary line information of the retaining wall; The processor (101) is used to load the point cloud and the location information, convert the point cloud to the coordinate system of the GPS positioning device (103), filter and extract the retaining wall points in the point cloud according to the boundary map file, and judge whether the retaining wall has defects by combining the vehicle size information and the retaining wall standard. The aforementioned mining area spoil heap retaining wall defect detection system is used to implement the mining area spoil heap retaining wall defect detection method as described in any one of claims 1-9.

Images