Mine high chute unmanned aerial vehicle scanning inspection method

Abstract

This invention belongs to the field of mining engineering technology, specifically disclosing a method for unmanned aerial vehicle (UAV) scanning and inspection of high-pass shafts in mines. The method involves placing a target ball in the roadway, measuring and recording the target ball's coordinates using a control traverse and a total station; starting the UAV and synchronizing its onboard LiDAR and IMU, then using the LiDAR to scan the deployed target ball, matching and fusing the target ball data with the IMU's coordinates; controlling the UAV to enter the high-pass shaft, using the LiDAR to perform a full-coverage scan of the shaft's inner wall, and matching and fusing the IMU's coordinates with the LiDAR scan data; importing the scanned point cloud data and target ball coordinates into TRW software, fitting the scanned target ball positions within the point cloud, assigning the target ball coordinates measured by the total station to known coordinates, and using analytical modeling to generate a solid model from the high-pass shaft point cloud data and exporting the graphic representation of the solid model. This invention features high efficiency, strong adaptability, low data processing volume, and no need for manual intervention.

Description

Technical Field

[0001] This application belongs to the field of mining engineering technology, specifically relating to a high-efficiency, highly adaptable, low-data-processing, and no-human-assisted-intervention method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines. Background Technology

[0002] High ore passes in mines are crucial channels that utilize their own weight to transport ore and waste rock from the mining area to the lower levels or transfer them to storage bins. Due to long-term impact from ore, wear and tear on the shaft walls, and geological stress, high ore passes are prone to collapses and blockages. Therefore, mine safety regulations (such as the "Safety Regulations for Metal and Non-metal Mines") explicitly require regular inspection and maintenance of high ore passes. Inspections identify structural deformations, cracks, and other safety hazards in high ore passes, allowing for targeted repair plans to prevent serious accidents such as ore runoff and roof collapses, effectively ensuring the safety of personnel and equipment and normal underground production operations.

[0003] Traditionally, ore passes are inspected manually. This involves observing the condition of the shaft walls, ore flow, and signs of arching or large blockages at the unloading point, venting point, maintenance shaft, or platform using a strong flashlight and reflector. Long poles are used to tap the shaft walls, and the sound is used to determine the presence of cavities, spalling, or cracks. While this method is the lowest-cost and simplest, requiring no complex equipment, it is highly dependent on the experience and judgment of personnel. Furthermore, because personnel must work close to the ore pass or in hazardous areas, there are risks of falls, falling rocks, dust, and harmful gases. Visual inspection can only examine a localized area near the inspection point and cannot provide a comprehensive view of the entire shaft.

[0004] In existing technologies, to address the shortcomings of manual inspection, long poles or cables are used at the unloading and venting points, maintenance risers, or platforms of high-speed ore passes to lower camera equipment or 3D laser scanning equipment (such as "Refined Scanning and Modeling Technology for Ore Passes in Pulang Copper Mine" and "Application of 3D Laser Scanning Technology in Ore Pass Measurement at Zijinshan Gold and Copper Mine") into the shaft. High-precision 3D point cloud models are generated by capturing videos, photographs, or scanning, thus providing a non-contact inspection method that visually reflects the geometry, deformation, wear, spalling, and blockage contours and volumes of the shaft wall. While this allows personnel to stay away from hazardous areas, provides a comprehensive view of the entire ore pass, and reduces or even eliminates reliance on human experience, it also presents challenges such as the difficulty and complexity of lowering long poles or cables, the high cost of professional-grade 3D laser scanners, complex post-processing of data, and the deterioration or even inability to use data due to dust and water mist underground.

[0005] In addition, existing technologies also include the use of industrial-grade drones equipped with cameras or small LiDAR to conduct inspections within mine shafts. For example, the "Promotion Catalogue of Advanced and Applicable Technologies and Equipment for Mine Safety (2024)" introduces drone-based underground space 3D scanning and measurement technology. This technology utilizes drones equipped with LiDAR or photogrammetry equipment to autonomously or semi-autonomously fly through complex spaces such as underground mine roadways and goafs, rapidly acquiring high-precision 3D point cloud data to construct digital models. However, due to its high dependence on SLAM technology (such as laser SLAM and visual SLAM), errors in position and attitude gradually accumulate as the flight distance increases, affecting the overall accuracy of large-area scanning. To address this, a fixed flight path design is used to avoid accumulated errors, but this makes it difficult to adapt to complex structures such as inclined shaft walls and narrow shaft openings. For example, the patent CN105953867B, titled "A Method for Measuring and Visualizing Material Level in Mine Passes Based on Small Unmanned Aerial Vehicles," employs a quadcopter drone equipped with a lidar and a vision camera. It is remotely controlled or autonomously descends from the wellhead, achieving precise hovering in GPS-free environments through SLAM. The lidar scans the material surface in real time, and the vision camera collects texture information. The data is wirelessly relayed back to a ground terminal, and finally, a 3D model of the material surface is generated using a point cloud stitching algorithm to calculate the material level height. However, in deep wells and narrow shafts, factors such as limited space, high heat, and high humidity affect the transmission of the drone's wireless signal, resulting in poor adaptability. Furthermore, the large number of point clouds affects computer processing speed, leading to a long data processing cycle. Summary of the Invention

[0006] This invention addresses the shortcomings of existing technologies by proposing a highly efficient, adaptable, low-data-processing, and human-intervention-free method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines.

[0007] The UAV scanning and inspection method for high ore passes in mines of the present invention is implemented as follows: it includes starting point coordinate measurement, scanning target sphere coordinates, high ore pass scanning, and data processing steps, as detailed below: A. Starting point coordinate measurement: Place the target ball on the ground of the roadway connected to the high chute to be measured, and use the control traverse laid out on site and a total station to measure and record the target ball coordinates; B. Scanning target ball coordinates: Start the UAV and synchronize the onboard lidar and inertial measurement unit. Then, use the lidar to scan the target ball deployed on site in the takeoff area. Match and fuse the scanned target ball data with the relative coordinate positioning of the inertial measurement unit to form the starting point cloud data. C. High chute scanning: The drone is controlled to enter the high chute from the starting point. Then, the onboard lidar is used to gradually rise or fall to perform a full-coverage scan of the inner wall of the high chute. The relative coordinates of the inertial measurement unit are matched and fused with the lidar scan data to form high chute point cloud data. D. Data Processing: Download the UAV's starting point cloud data, the high ore well point cloud data, and the target sphere coordinates measured by the total station to the computer. Then, import the downloaded point cloud data into TRW software. Use the target analysis function in the aforementioned software to fit the scanned target sphere position in the point cloud data. Then, use the geographic coordinate transformation function to use the total station to measure the target sphere control point coordinates to transform all point cloud data into a real coordinate system. Next, use analysis and modeling to generate a solid model from the high ore well point cloud data. Finally, export the graphic of the high ore well solid model through TRW software.

[0008] Furthermore, in step A, 2 to 4 target balls are placed sequentially at intervals on the flat ground of the roadway connected to the high chute to be tested, and the target balls are not placed on the same line in the direction of roadway extension.

[0009] Furthermore, in step B, matching and fusing the scanned target ball data with the relative coordinate positioning of the inertial measurement unit (IMU) is achieved using the geographic coordinate transformation function in the TRW software. The measurement coordinates corresponding to the target ball are input, thus matching and fusing the scanned target ball data with the relative coordinate positioning of the IMU. Similarly, in step C, matching and fusing the relative coordinates of the IMU with the data scanned by the lidar is achieved using the geographic coordinate transformation function in the TRW software. The measurement coordinates corresponding to the target ball are input, thus matching and fusing the relative coordinates of the IMU with the data scanned by the lidar.

[0010] Furthermore, in steps B and C, the UAV equipped with a fixed lidar uses a self-rotating flight scan or a figure-eight flight scan to scan both the target sphere and the inner wall of the high chute. During the flight scan, the UAV's flight speed is <2m / s.

[0011] Furthermore, in step C, the UAV performs a self-rotating flight scan or a figure-eight flight scan every 10±2m of ascent or descent in the high chute, and the overlap rate of the laser radar carried by the UAV scanning the inner wall during ascent or descent in the high chute is ≥30%.

[0012] Furthermore, in step D, the geographic coordinate transformation function is used to measure the coordinates of the target sphere control points using a total station to transform all point cloud data into a real coordinate system. Transforming the lidar point cloud data into a geographic coordinate system is a key step in point cloud processing, requiring the combination of rotation matrices, translation vectors, and sensor attitude information: First, sensor attitude information is acquired by using an inertial measurement unit (IMU) to obtain the lidar's attitude angles (such as yaw, pitch, and roll), converting Euler angles into a rotation matrix, or directly generating a rotation matrix using quaternions; then, a transformation matrix is ​​constructed: based on the relative position between the lidar and the inertial measurement unit, the translation vector is determined, and the rotation matrix and translation vector are combined into a 4×4 transformation matrix; subsequently, the transformation is applied: the transformation matrix is ​​applied to each point of the point cloud data, transforming it from the lidar coordinate system to the geographic coordinate system.

[0013] Furthermore, in steps B and C, the drone is also equipped with a high-definition camera, lighting equipment, and sonar sensors for collision avoidance, and the drone is connected to the flight controller via a wireless network.

[0014] Furthermore, before the starting point coordinate measurement step, there is also an on-site preprocessing step: the on-site survey of the target ball and the drone's placement and take-off position is carried out and safety is confirmed, then the tunnel where the target ball and the drone are placed is flushed and cleaned, and then the total station is set up and leveled.

[0015] Furthermore, in step D, fitting the target sphere position scanned within the starting point cloud data involves delineating the point cloud containing the target sphere position within the scanned point cloud to generate a sphere with a diameter of 14 centimeters. Using the geographic coordinate transformation function in the TRW software, the coordinates input from the total station are compared with the position corresponding to the target sphere, thus fitting it to the true coordinate position.

[0016] Furthermore, in step D, generating a solid model from the high chute point cloud data using analytical modeling involves using analytical modeling to transform the high chute point cloud data converted to the real coordinate system into a polygonal mesh model using a triangular mesh generation algorithm, thereby generating a solid model.

[0017] The beneficial effects of this invention are: 1. This invention features a unique "target ball + total station" absolute coordinate positioning method, combined with "UAV LiDAR scanning of the target ball + inertial measurement unit data fusion". By using the target ball as a coordinate reference point, it completely avoids the position and attitude error accumulation problem caused by existing UAVs relying on SLAM technology. This method is not only particularly suitable for complex structures such as deep wells and inclined well walls, but also significantly improves the overall accuracy of large-area scanning, ensuring that hidden dangers such as deformation and cracks in high ore well structures can be accurately identified, thus guaranteeing the overall accuracy of large-area scanning. Moreover, the use of lightweight LiDAR (non-professional-grade 3D scanner) significantly reduces hardware costs.

[0018] 2. This invention targets high ore passes with complex structures such as inclined well walls, narrow well openings, and deep wells. It adopts a self-rotating flight or figure-eight flight scanning mode, combined with a preset scanning interval and scanning overlap rate design, to ensure full coverage scanning of the inner wall of the well shaft. At the same time, through the anti-collision design of the UAV's sonar sensor and on-site flushing pretreatment to reduce dust interference, the UAV laser scanning can be stably applied to various high ore pass environments, solving the problem of poor adaptability of existing technologies to complex well shafts.

[0019] 3. Compared to traditional manual inspections that require personnel to work at the wellhead or in dangerous areas, this invention uses drones to autonomously enter high wells for scanning, keeping operators away from dangerous environments such as falls, falling rocks, dust, and harmful gases, fundamentally eliminating the safety hazards of direct exposure. Moreover, compared to the shortcomings of traditional manual inspections that rely heavily on personal experience and judgment, this invention uses point cloud data acquired by lidar and software modeling analysis to generate entity models based on objective data, making the detection results more objective and traceable, and reducing the interference of human factors on judgment.

[0020] 4. The UAV of this invention performs fully autonomous flight scanning without the need for manual remote control intervention. Combined with the standardized "scan-data download-software processing" process (TRW software rapid modeling), the process of generating entity models and exporting entity graphics can be completed in about 1 hour for a single data processing session. Moreover, compared with the complex operation of existing equipment that is lowered by long poles / cables, and the fact that existing UAVs require manual assistance, this invention significantly improves detection efficiency.

[0021] 5. This invention, through the 360°×30° field of view of the airborne lidar combined with the autorotation flight or figure-eight flight scanning mode, can ensure comprehensive scanning coverage. Combined with wall-flying and wall-washing to reduce dust interference, it effectively improves the quality of the original data. Moreover, through the subsequent TRW software to realize target sphere coordinate fitting, geographic coordinate transformation and entity model generation, it eliminates the need for complex point cloud stitching algorithms, significantly simplifies the data processing process, and solves the problems of complex and long data processing cycles in existing technologies.

[0022] In summary, this invention achieves high precision, high efficiency, low cost, and high safety and reliability in the detection of mine ore passes through the innovative combination of target ball absolute coordinate positioning technology and UAV comprehensive scanning strategy. It completely solves the risks of traditional manual inspection, the error accumulation problem of existing UAV technology, and the economic bottleneck of professional equipment, providing a universal solution for intelligent detection of mine ore passes. Attached Figure Description

[0023] Figure 1 This is a flowchart of the present invention. Figure 2 This is a tunnel plan view according to an embodiment of the present invention. Figure 3 for Figure 2 MM cross-section, In the diagram: 1-tunnel, 2-high chute, 3-waste rock unloading station, 4-baseline point, 5-target ball, 6-drone flight path; Figure 4 This is a scanned plan view of a waste rock chute according to an embodiment of the present invention; Figure 5 for Figure 4 One of the longitudinal cross-sectional views; Figure 6 for Figure 4 The second longitudinal cross-section; In the diagram: 10 - Original design waste rock chute, 11 - First inspection waste rock chute (pink), 12 - Second inspection waste rock chute (blue), 13 - Third inspection waste rock chute (red). Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this does not limit the present invention in any way. Any changes or improvements made based on the teachings of the present invention shall fall within the protection scope of the present invention.

[0025] like Figure 1 , 2 As shown in Figure 3, the UAV scanning and inspection method for high ore passes in mines according to the present invention includes the following steps: starting point coordinate measurement, scanning target sphere coordinates, high ore pass scanning, and data processing. The specific details are as follows: A. Starting point coordinate measurement: Place the target ball on the ground of the roadway connected to the high chute to be measured, and use the control traverse laid out on site and a total station to measure and record the target ball coordinates; B. Scanning target ball coordinates: Start the UAV and synchronize the onboard lidar and inertial measurement unit. Then, use the lidar to scan the target ball deployed on site in the takeoff area. Match and fuse the scanned target ball data with the relative coordinate positioning of the inertial measurement unit to form the starting point cloud data. C. High chute scanning: The drone is controlled to enter the high chute from the starting point. Then, the onboard lidar is used to gradually rise or fall to perform a full-coverage scan of the inner wall of the high chute. The relative coordinates of the inertial measurement unit are matched and fused with the lidar scan data to form high chute point cloud data. D. Data Processing: Download the UAV's starting point cloud data, the high ore well point cloud data, and the target sphere coordinates measured by the total station to the computer. Then, import the downloaded point cloud data into TRW software. Use the target analysis function in the aforementioned software to fit the scanned target sphere position in the point cloud data. Then, use the geographic coordinate transformation function to use the total station to measure the target sphere control point coordinates to transform all point cloud data into a real coordinate system. Next, use analysis and modeling to generate a solid model from the high ore well point cloud data. Finally, export the graphic of the high ore well solid model through TRW software.

[0026] In step A, 2 to 4 target balls are placed sequentially at intervals on the flat ground of the roadway connected to the high chute to be tested, and the target balls are not placed on the same line in the direction of the roadway extension.

[0027] In step B, matching and fusing the scanned target ball data with the relative coordinate positioning of the inertial measurement unit (IMU) is achieved using the geographic coordinate transformation function in the TRW software. The measurement coordinates corresponding to the target ball are input, thus matching and fusing the scanned target ball data with the relative coordinate positioning of the IMU. In step C, matching and fusing the relative coordinates of the IMU with the data scanned by the lidar is achieved using the geographic coordinate transformation function in the TRW software. The measurement coordinates corresponding to the target ball are input, thus matching and fusing the relative coordinates of the IMU with the data scanned by the lidar.

[0028] In steps B and C, the UAV is equipped with a VLP-16 or equivalent LiDAR with a field of view of 360°×30°, and a six-axis inertial measurement unit.

[0029] The UAV is equipped with a VLP-16 or equivalent LiDAR, with the following parameters: 16 laser lines, independent transmit and receive channels; maximum measurement distance of 100m; error range between radar measurement results and actual values ​​of ±3cm; angular coverage range of ±15° in the vertical plane; complete angular coverage range of 360° in the horizontal plane; angular interval of 2° between adjacent laser lines in the vertical direction; minimum distinguishable angular difference of 0.1° to 0.4° in the horizontal direction; measurement frequency of 5 to 20Hz per second; laser safety rating of Class 1; laser wavelength of 905nm; power consumption of 8W during normal operation; operating voltage range of 9-32V; weight of 830g; dustproof and waterproof rating of IP67; operating temperature range of -10℃ to +60℃; output of 300,000 data points per second; and a six-axis inertial measurement unit.

[0030] In steps B and C, the UAV, equipped with a fixed lidar, performs a self-rotating flight scan or a figure-eight flight scan on both the target sphere and the inner wall of the high chute. During the flight scan, the UAV's flight speed is <2m / s.

[0031] In step C, the UAV performs a self-rotating flight scan or a figure-eight flight scan every 10±2m as it ascends or descends in the high chute, and the overlap rate of the laser radar carried by the UAV scanning the inner wall is ≥30% when it ascends or descends in the high chute.

[0032] In step D, the geographic coordinate transformation function is used to measure the coordinates of the target sphere control points using a total station to transform all point cloud data into a real coordinate system. Transforming the lidar point cloud data into a geographic coordinate system is a key step in point cloud processing, requiring the combination of rotation matrices, translation vectors, and sensor attitude information: First, the sensor attitude information is acquired by using an inertial measurement unit (IMU) to obtain the lidar's attitude angles (such as yaw, pitch, and roll), converting Euler angles into a rotation matrix, or directly generating a rotation matrix using quaternions; then, a transformation matrix is ​​constructed: based on the relative position between the lidar and the IMU, the translation vector is determined, and the rotation matrix and translation vector are combined into a 4×4 transformation matrix; subsequently, the transformation is applied: the transformation matrix is ​​applied to each point of the point cloud data, transforming it from the lidar coordinate system to the geographic coordinate system.

[0033] In steps B and C, the drone is also equipped with a high-definition camera, lighting equipment, and sonar sensors for collision avoidance. The drone is connected to the flight controller via a wireless network.

[0034] Before the starting point coordinate measurement step, there is also an on-site pre-processing step: conduct on-site surveys and confirm safety of the target ball and drone placement and take-off positions, then flush and wash the tunnel where the target ball and drone are placed, and then complete the setup and leveling of the total station.

[0035] In step D, fitting the target sphere position scanned within the starting point cloud data involves delineating the point cloud containing the target sphere position to generate a sphere with a diameter of 14 centimeters. Using the geographic coordinate transformation function in TRW software, the coordinates input from the total station are matched with the corresponding position of the target sphere, thus fitting it to the true coordinate position.

[0036] In step D, generating a solid model from the high chute point cloud data using analytical modeling involves using a triangulation generation algorithm to transform the high chute point cloud data, which has been converted into a real coordinate system, into a polygonal mesh model to generate a solid model.

[0037] Example

[0038] The waste rock chute in the -20 section of the western mining section of a copper mine of Yuxi Mining Co., Ltd. was put into use in 2016. The chute has an elevation of 179.4m at the top, -21.2m at the bottom, a diameter of 3 meters, and a total length of 200.6 meters. During ore discharge, large chunks frequently became stuck at the vibratory feeder inlet at the chute's foot. Visual inspection during chute emptying revealed potential wall collapse, resulting in large chunks. Due to the chute's height, visual observation was insufficient to effectively monitor the internal conditions, potentially leading to roof falls, spalling, and slag spills during ore discharge. Therefore, in 2018, 2022, and 2023, the mine high-passage unmanned aerial vehicle (UAV) scanning inspection method of this invention was used three times to conduct a comprehensive scan of the chute. The specific process is as follows: S100: Conduct on-site surveys and confirm safety of the placement and takeoff positions of target ball 5 and UAV. Then, flush and wash the walls of the tunnel 1 where target ball 5 and UAV are placed to ensure the safe conduct of the scanning work and reduce the high dust concentration when the UAV takes off. Afterwards, complete the installation and leveling of the total station.

[0039] S200: Place 2 to 4 target balls 5 at intervals on the ground of the roadway 1 connected to the high chute 2 to be measured, and ensure that each target ball 5 is not on the same line extending from the roadway 1. Then, use the control traverse laid out on site and a total station to measure and record the coordinates of the target balls 5.

[0040] S300: Start the ELIOS3 drone (an indoor drone designed for inspection in confined spaces and complex environments), and pair the ELIOS3 with the remote controller (Bluetooth connection): Open the tablet - connect the tablet to the remote controller - start the Cockpit - start the remote controller - start the drone - select the radio tab in the settings interface of the Cockpit - press and hold the start button on the drone - click "START PAIRING" on the tablet - complete the pairing; then synchronize the LiDAR and six-axis inertial measurement unit (IMU) on the ELIOS3 drone, and then take off the ELIOS3 drone in the takeoff area at a flight speed of <2m / s, using autorotation flight scanning or figure-eight flight scanning of the target sphere 5, and use the geographic coordinate transformation function in the TRW software to input the measurement coordinates corresponding to the target sphere, so that the scanned target sphere data is matched and fused with the relative coordinate positioning of the inertial measurement unit to form the starting point point cloud data.

[0041] S400: Starting from the origin, the ELIOS3 drone enters the high chutes 2. Utilizing its onboard lidar, at a flight speed of <2m / s and a scan overlap rate of ≥30%, it performs a self-rotating or figure-eight flight scan every 10±2m of ascent or descent within the high chutes 2 (the lidar itself does not rotate, so the drone needs to actively rotate). This provides a full-coverage scan of the inner wall of the high chutes 2. Using the geographic coordinate transformation function in the TRW software, the corresponding measurement coordinates of the target sphere are input, allowing the relative coordinates of the inertial measurement unit to be matched and fused with the lidar scan data to form the high chutes point cloud data. Simultaneously, the 4K high-definition camera and 10K lumen illumination lamp on the ELIOS3 drone simultaneously capture images of the inner wall of the high chutes 2 and transmit these images in real time to the remote controller and tablet via wireless network.

[0042] S500: After the scanning of the high chute 2 was completed, the ELIOS3 UAV landed within 1m of the takeoff point. The ELIOS3 UAV automatically optimized the scanning accuracy. After the scan, the UAV was shut down, and the target sphere 5 and total station were retrieved. Then, the starting point cloud data of the UAV, the high chute point cloud data, and the target sphere coordinates measured by the total station were downloaded using the ELIOS3 software installed on the computer. The downloaded point cloud data was imported into TRW (Trimble Realworks) software. Using the registration mode - target analyzer - the scanned target sphere positions within the point cloud were fitted to generate circles. Then, the geographic coordinate transformation function was used to convert all point cloud data into a real coordinate system using the coordinates of the target sphere control points measured by the total station. Next, the analysis and modeling were used to convert the high chute point cloud data into a polygon mesh model using a triangular mesh generation algorithm, thus generating a solid model. Finally, the solid model was exported using TRW software (exporting a DWG format CAD drawing). Figure 4 , 5 Scan image of the wall of well 2 in Gaolujing 6.

[0043] The total volume of the 60-0 horizontal scan, conducted in November 2018, was 1267.0 m³. 3 The total volume at horizontal level 60-0, scanned in December 2023, was 10445.0 m³. 3 , the 60-0 horizontal wellbore collapsed 9178m from 2018 to 2023. 3 The widest point of the collapse is at level 36, and the collapse length is 21.6 meters. Figure 4 , 5 The values ​​in 6 and 7 accurately show the degree of collapse at each level.

[0044] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for unmanned aerial vehicle (UAV) scanning and inspection of high-pass shafts in mines, characterized in that: The process includes measuring the starting point coordinates, scanning the target sphere coordinates, scanning the high chute, and data processing steps, detailed as follows: A. Starting point coordinate measurement: Place the target ball on the ground of the roadway connected to the high chute to be measured, and use the control traverse laid out on site and a total station to measure and record the target ball coordinates; B. Scanning target ball coordinates: Start the UAV and synchronize the onboard lidar and inertial measurement unit. Then, use the lidar to scan the target ball deployed on site in the takeoff area. Match and fuse the scanned target ball data with the relative coordinate positioning of the inertial measurement unit to form the starting point cloud data. C. High chute scanning: The drone is controlled to enter the high chute from the starting point. Then, the onboard lidar is used to gradually rise or fall to perform a full-coverage scan of the inner wall of the high chute. The relative coordinates of the inertial measurement unit are matched and fused with the lidar scan data to form high chute point cloud data. D. Data Processing: Download the UAV's starting point cloud data, the high ore well point cloud data, and the target sphere coordinates measured by the total station to the computer. Then, import the downloaded point cloud data into TRW software. Use the target analysis function in the aforementioned software to fit the scanned target sphere position in the point cloud data. Then, use the geographic coordinate transformation function to use the total station to measure the target sphere control point coordinates to transform all point cloud data into a real coordinate system. Next, use analysis and modeling to generate a solid model from the high ore well point cloud data. Finally, export the graphic of the high ore well solid model through TRW software.

2. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to claim 1, characterized in that: In step A, 2 to 4 target balls are placed sequentially at intervals on the flat ground of the roadway connected to the high chute to be tested, and the target balls are not placed on the same line in the direction of the roadway extension.

3. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to claim 1, characterized in that: In step B, matching and fusing the scanned target ball data with the relative coordinate positioning of the inertial measurement unit (IMU) is achieved using the geographic coordinate transformation function in the TRW software. The measurement coordinates corresponding to the target ball are input, thus matching and fusing the scanned target ball data with the relative coordinate positioning of the IMU. In step C, matching and fusing the relative coordinates of the IMU with the data scanned by the lidar is achieved using the geographic coordinate transformation function in the TRW software. The measurement coordinates corresponding to the target ball are input, thus matching and fusing the relative coordinates of the IMU with the data scanned by the lidar.

4. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to claim 1, characterized in that: In steps B and C, the UAV, equipped with a fixed lidar, performs a self-rotating flight scan or a figure-eight flight scan on both the target sphere and the inner wall of the high chute. During the flight scan, the UAV's flight speed is <2m / s.

5. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to claim 4, characterized in that: In step C, the UAV performs a self-rotating flight scan or a figure-eight flight scan every 10±2m as it ascends or descends in the high chute, and the overlap rate of the laser radar carried by the UAV scanning the inner wall is ≥30% when it ascends or descends in the high chute.

6. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to claim 4, characterized in that: In step D, the geographic coordinate transformation function is used to measure the coordinates of the target sphere control points using a total station to transform all point cloud data into a real coordinate system. Transforming the lidar point cloud data to the geographic coordinate system is a crucial step in point cloud processing, requiring the combination of rotation matrices, translation vectors, and sensor attitude information: First, sensor attitude information is acquired by using an inertial measurement unit (IMU) to obtain the lidar's attitude angles, converting Euler angles into a rotation matrix, or directly generating the rotation matrix using quaternions; then, a transformation matrix is ​​constructed: based on the relative position between the lidar and the IMU, the translation vector is determined, and the rotation matrix and translation vector are combined into a 4×4 transformation matrix; subsequently, the transformation is applied: the transformation matrix is ​​applied point-by-point to the point cloud data, transforming it from the lidar coordinate system to the geographic coordinate system.

7. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to claim 4, characterized in that: In steps B and C, the drone is also equipped with a high-definition camera, lighting equipment, and sonar sensors for collision avoidance. The drone is connected to the flight controller via a wireless network.

8. The method for unmanned aerial vehicle (UAV) scanning and inspection of high ore passes in mines according to any one of claims 1 to 7, characterized in that: Before the starting point coordinate measurement step, there is also an on-site pre-processing step: conduct on-site surveys and confirm safety of the target ball and drone placement and take-off positions, then flush and wash the tunnel where the target ball and drone are placed, and then complete the setup and leveling of the total station.

9. The method for unmanned aerial vehicle (UAV) scanning and inspection of high-pass shafts in mines according to any one of claims 1 to 7, characterized in that: In step D, fitting the target sphere position scanned within the starting point cloud data involves delineating the point cloud containing the target sphere position to generate a sphere with a diameter of 14 centimeters. Using the geographic coordinate transformation function in TRW software, the coordinates input from the total station are matched with the corresponding position of the target sphere, thus fitting it to the true coordinate position.

10. The method for unmanned aerial vehicle (UAV) scanning and inspection of high-pass shafts in mines according to claim 9, characterized in that: In step D, generating a solid model from the high chute point cloud data using analytical modeling involves using a triangulation generation algorithm to transform the high chute point cloud data, which has been converted into a real coordinate system, into a polygonal mesh model to generate a solid model.

Images