Uav-based three-dimensional laser radar underground map construction method and device

Abstract

The application provides an unmanned aerial vehicle (UAV)-based three-dimensional laser radar underground map construction method and device. The method comprises the following steps: controlling a laser radar of the UAV to scan an underground operation area to obtain a scanned point cloud data set of the underground operation area; obtaining a measured point cloud data set of the underground operation area according to a relative position between the UAV and a positioning base station of the underground operation area; and generating a target three-dimensional map of the underground operation area according to the scanned point cloud data set and the measured point cloud data set. In the application, the target three-dimensional map is constructed based on the combination of the point cloud data sets, the completeness of the target three-dimensional map is optimized, the accuracy of the construction of the target three-dimensional map is improved, the UAV can realize real-time positioning with high precision based on the target three-dimensional map, and the positioning effect of the UAV is optimized.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular to a method and apparatus for constructing a three-dimensional lidar downhole map based on an unmanned aerial vehicle (UAV). Background Technology

[0002] As society develops, people have increasingly higher requirements for underground operations, and are relying more and more on drones for these operations.

[0003] In related technologies, the positioning of drones operating underground is inaccurate and has a low accuracy rate. Therefore, how to achieve high-precision positioning of drones operating underground is a problem that needs to be solved. Summary of the Invention

[0004] This application aims to address one of the technical problems in related technologies to a certain extent. Therefore, one objective of this application is to propose a method and apparatus for constructing a 3D lidar downhole map based on an unmanned aerial vehicle (UAV), thereby solving the problems of low positioning accuracy and weak real-time performance of UAVs during downhole operations. The technical solution of this application is as follows:

[0005] The first aspect of this application proposes a method for constructing a three-dimensional LiDAR downhole map based on an unmanned aerial vehicle (UAV), comprising: controlling the UAV's LiDAR to scan an downhole working area to obtain a scanned point cloud dataset of the downhole working area; obtaining a measured point cloud dataset of the downhole working area based on the relative position of the UAV and a positioning base station in the downhole working area; and generating a target three-dimensional map of the downhole working area based on the scanned point cloud dataset and the measured point cloud dataset.

[0006] The method for constructing a 3D lidar downhole map based on an unmanned aerial vehicle (UAV) proposed in the first aspect of this application also has the following additional features:

[0007] According to one embodiment of this application, the method of controlling the lidar of the UAV to scan the downhole working area to obtain the scan point cloud dataset of the downhole working area includes: obtaining the scan point set of the lidar and determining the curvature of each scan point in the scan point set; extracting effective feature points from the scan point set according to the curvature of each scan point, and generating the scan point cloud dataset of the downhole working area based on the point cloud data of the effective feature points.

[0008] According to one embodiment of this application, the step of extracting effective feature points from the set of scan points based on the curvature of each scan point includes: determining edge points and plane points from the set of scan points based on the curvature of the scan points, as the effective feature points.

[0009] According to one embodiment of this application, generating the scanned point cloud dataset of the downhole operating area based on the point cloud data of the effective feature points includes: acquiring scan frames collected by the lidar; determining, based on the curvature of the effective feature points, effective feature points in the scan frames that match the previous scan frame, as target feature points of the scan frames; determining the point cloud data of the target feature points, and determining the point cloud data of the effective feature points of the scan frames according to the relative positions between the target feature points and the remaining effective feature points; and generating the scanned point cloud dataset of the downhole operating area based on the point cloud data of the effective feature points of each scan frame.

[0010] According to one embodiment of this application, before determining the effective feature points in the scan frame that match the previous scan frame based on the curvature of the effective feature points as the target feature points of the scan frame, the method further includes: determining whether the scan frame is a distorted frame; and in response to the scan frame being the distorted frame, compensating the distorted frame based on the acquisition data corresponding to the inertial measurement unit to obtain a compensated scan frame.

[0011] According to one embodiment of this application, the step of compensating the distorted frame based on the acquired data corresponding to the inertial measurement unit to obtain a compensated scan frame includes: determining the distortion timestamp of the distorted frame; acquiring the acquired data of the inertial measurement unit at the distortion timestamp; performing pose calculation on the acquired data at the distortion timestamp to generate a compensated attitude trajectory corresponding to the distorted frame; and generating the compensated scan frame corresponding to the distorted frame based on the effective feature points of the previous scan frame corresponding to the distorted frame and the compensated attitude trajectory.

[0012] According to one embodiment of this application, generating the compensated scan frame corresponding to the distorted frame based on the effective feature points of the previous scan frame corresponding to the distorted frame and the compensation attitude trajectory includes: controlling the effective feature points of the previous scan frame to move along the compensation attitude trajectory to obtain a target compensation point formed after the effective feature points move; determining a target distortion point corresponding to the target compensation point on the distorted frame based on the compensation attitude trajectory; determining a remaining compensation point corresponding to the remaining distortion point based on the relative position of the remaining distortion point and the target distortion point; and generating the compensated scan frame corresponding to the distorted frame based on the target compensation point and the remaining compensation point.

[0013] According to one embodiment of this application, generating a target 3D map of the downhole work area based on the scanned coordinate set and the measured coordinate set includes: obtaining the union of point cloud data of the scanned point cloud dataset and the measured point cloud dataset; and generating the target 3D map of the downhole work area based on the union of point cloud data.

[0014] According to one embodiment of this application, the method further includes: obtaining a first set of operating coordinates for the UAV based on the scanned point cloud dataset; obtaining a second set of operating coordinates for the UAV based on the relative position of the UAV and the positioning base station; and determining the operating position of the UAV in the underground working area based on the first set of operating coordinates and the second set of operating coordinates.

[0015] A second aspect of this application proposes a UAV-based 3D LiDAR downhole map construction device, comprising: a scanning module for controlling the UAV's LiDAR to scan an downhole work area to obtain a scanned point cloud dataset of the downhole work area; a calculation module for obtaining a calculated point cloud dataset of the downhole work area based on the relative position of the UAV and a positioning base station in the downhole work area; and a generation module for generating a target 3D map of the downhole work area based on the scanned point cloud dataset and the calculated point cloud dataset.

[0016] The UAV-based 3D lidar downhole map construction device proposed in the second aspect of this application also has the following additional features:

[0017] According to one embodiment of this application, the scanning module is further configured to: acquire the scanning point set of the lidar and determine the curvature of each scanning point in the scanning point set; extract effective feature points from the scanning point set according to the curvature of each scanning point, and generate the scanning point cloud dataset of the downhole working area based on the point cloud data of the effective feature points.

[0018] According to one embodiment of this application, the scanning module is further configured to: determine edge points and planar points from the set of scanning points based on the curvature of the scanning points, as the effective feature points.

[0019] According to one embodiment of this application, the base scanning module is further configured to: acquire scanning frames collected by the lidar; determine, based on the curvature of the effective feature points, effective feature points in the scanning frame that match the previous scanning frame, as target feature points of the scanning frame; determine the point cloud data of the target feature points, and determine the point cloud data of the effective feature points of the scanning frame according to the relative position between the target feature points and the remaining effective feature points; and generate the scanning point cloud dataset of the downhole working area based on the point cloud data of the effective feature points of each scanning frame.

[0020] According to one embodiment of this application, the scanning module is further configured to: determine whether the scanning frame is a distorted frame; and in response to the scanning frame being the distorted frame, compensate the distorted frame based on the acquisition data corresponding to the inertial measurement unit to obtain a compensated scanning frame.

[0021] According to one embodiment of this application, the scanning module is further configured to: determine the distortion timestamp of the distorted frame; acquire the data collected by the inertial measurement unit at the distortion timestamp; perform pose calculation on the acquired data at the distortion timestamp to generate a compensated attitude trajectory corresponding to the distorted frame; and generate the compensated scan frame corresponding to the distorted frame based on the effective feature points of the previous scan frame corresponding to the distorted frame and the compensated attitude trajectory. According to one embodiment of this application, the scanning module is further configured to: control the effective feature points of the previous scan frame to move along the compensated attitude trajectory to acquire a target compensation point formed after the effective feature points have moved; determine the target distortion point corresponding to the target compensation point on the distorted frame based on the compensated attitude trajectory; determine the remaining compensation point corresponding to the remaining distortion point based on the relative position of the remaining distortion point and the target distortion point; and generate the compensated scan frame corresponding to the distorted frame based on the target compensation point and the remaining compensation point.

[0022] According to one embodiment of this application, the generation module is further configured to: obtain the union of point cloud data of the scanned point cloud dataset and the measured point cloud dataset; and generate the target three-dimensional map of the downhole work area based on the union of point cloud data.

[0023] According to one embodiment of this application, the device further includes a positioning module, configured to: obtain a first set of operating coordinates for the UAV based on the scanned point cloud dataset; obtain a second set of operating coordinates for the UAV based on the relative position of the UAV and the positioning base station; and determine the operating position of the UAV in the downhole operating area based on the first set of operating coordinates and the second set of operating coordinates.

[0024] The third aspect of this application proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method for constructing a three-dimensional lidar well map based on an unmanned aerial vehicle (UAV) as proposed in any of the first aspects above.

[0025] This application proposes a method and apparatus for constructing a 3D LiDAR-based downhole map using an unmanned aerial vehicle (UAV). The method controls a UAV's LiDAR to scan the downhole work area and acquire the corresponding scanned point cloud dataset. Simultaneously, UWB technology is used to obtain the relative positions between the points to be located in the downhole work area and the base station. Based on the point cloud data from the base station and the relative positions, point cloud data for each point to be located is calculated, resulting in a calculated point cloud dataset for the downhole work area. Based on the combined scanned and calculated point cloud datasets, a target 3D map of the downhole work area is generated. In this application, the target 3D map is constructed based on the combined point cloud dataset, optimizing the completeness of the target 3D map and improving the accuracy of its construction. This allows the UAV to achieve highly accurate real-time positioning based on the target 3D map, thus optimizing the UAV's positioning performance.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application, and do not constitute an undue limitation of this application.

[0028] Figure 1 This is a flowchart illustrating a method for constructing a 3D lidar downhole map based on an unmanned aerial vehicle (UAV) according to an embodiment of this application.

[0029] Figure 2 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method.

[0030] Figure 3 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method.

[0031] Figure 4 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method.

[0032] Figure 5 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method.

[0033] Figure 6 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method.

[0034] Figure 7 This is a schematic diagram of the structure of a UAV-based 3D lidar downhole map construction device according to an embodiment of this application;

[0035] Figure 8 This is a schematic diagram of a UAV-based 3D lidar downhole map construction device according to another embodiment of this application. Detailed Implementation

[0036] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Throughout, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0037] Figure 1 This is a flowchart illustrating a method for constructing a 3D LiDAR downhole map based on an unmanned aerial vehicle (UAV) according to an embodiment of this application. Figure 1 As shown, the method includes:

[0038] S101 controls the drone's lidar to scan the downhole work area in order to obtain a scan point cloud dataset of the downhole work area.

[0039] In practice, drones can be controlled to operate in the underground work area. Therefore, it is necessary to achieve accurate positioning of drones in the underground work area.

[0040] To achieve accurate positioning of drones, a 3D map of the underground work area can be constructed based on the drone, enabling the drone to achieve accurate positioning during subsequent operations.

[0041] In this embodiment of the application, the area of ​​the underground working area can be coordinated, and a corresponding point cloud dataset can be constructed by acquiring point cloud data of each location in the underground working area, thereby realizing the construction of a three-dimensional map.

[0042] Furthermore, the lidar of the drone can be controlled to scan the underground working area.

[0043] By controlling the movement of the drone in the underground working area, the lidar device configured on the drone can scan the underground working area, thereby obtaining the coordinate set of the underground working area scanned by the lidar. This coordinate set can be a point cloud dataset constructed based on the world coordinate system.

[0044] Optionally, a 128-line lidar can be used to scan and acquire point cloud datasets of the downhole work area.

[0045] S102, based on the relative position of the UAV and the positioning base station in the underground work area, obtain the measurement point cloud dataset of the underground work area.

[0046] In practice, when the lidar equipped on the drone scans the underground working area, there may be blind spots. In these blind spots, the lidar cannot obtain point cloud data for each location.

[0047] Furthermore, while the lidar scans the downhole work area, point cloud datasets corresponding to the downhole work area can be acquired through other means.

[0048] Optionally, ultra-wideband (UWB) technology can be used for ranging and calculation, and the point cloud dataset corresponding to the downhole working area can be obtained based on the calculation results.

[0049] Among them, several base stations with known coordinates can be configured in the underground working area. Based on the relative position between the positioning tag configured on the UAV and the base station, relevant calculations are performed, and the point cloud dataset of the underground working area is obtained based on the calculation results.

[0050] This can be understood as follows: by using drones to acquire the corresponding underground work area to be located, and through the interaction of request and response signals between base stations, the relative positional relationship between the to-be-located point and the base station can be obtained. Furthermore, by using the known point cloud data of the base station, the relative positional relationship is transformed and calculated, thereby determining the point cloud data corresponding to each to-be-located point, and all the point cloud data of all to-be-located points are determined as the measurement point cloud dataset.

[0051] S103, generate a target 3D map of the downhole work area based on the scanned point cloud dataset and the measured point cloud dataset.

[0052] In this embodiment, the point cloud data in the calculated point cloud dataset may overlap with the point cloud data scanned by the lidar. Further, the two can be integrated by deduplicating and merging the scanned point cloud dataset and the calculated point cloud dataset to generate a corresponding point cloud data set for the downhole work area.

[0053] Optionally, the integration of the measured point cloud dataset and the scanned point cloud dataset can be achieved based on the extended Kalman filter algorithm, or it can be achieved based on other algorithms that can integrate the measured point cloud dataset and the scanned point cloud dataset; no limitation is made here.

[0054] The point cloud data collection includes both the point cloud data from the scanned point cloud dataset and the point cloud data from the measured point cloud dataset.

[0055] Furthermore, based on the integrated point cloud data set, a target 3D map is generated.

[0056] This application proposes a method for constructing a 3D LiDAR-based downhole map using an unmanned aerial vehicle (UAV). The method involves controlling a UAV's LiDAR to scan the downhole work area and acquire the corresponding scanned point cloud dataset. Simultaneously, UWB technology is used to obtain the relative positions between the points to be located in the downhole work area and the base station. Based on the point cloud data from the base station and the relative positions, point cloud data for each point to be located is calculated, resulting in a calculated point cloud dataset for the downhole work area. A target 3D map of the downhole work area is generated based on the combined scanned and calculated point cloud datasets. This method constructs the target 3D map based on the combined point cloud datasets, optimizing the completeness of the target 3D map and improving the accuracy of its construction. This allows the UAV to achieve highly accurate real-time positioning based on the target 3D map, thus optimizing the UAV's positioning performance.

[0057] In the above embodiments, scanning the downhole working area using lidar can acquire point cloud data of multiple feature points. However, some feature points may be invalid. Therefore, it is necessary to extract the feature points acquired by the lidar scan. This can be done by combining... Figure 2 To understand further, Figure 2 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method, as shown below. Figure 2 As shown, the method includes:

[0058] S201, acquire the set of scanning points of the lidar and determine the curvature of each scanning point in the set of scanning points.

[0059] In practice, the points scanned by the lidar can be defined as the corresponding set of scan points. Within the set of scan points, there is a possibility that adjacent feature points have small feature differences.

[0060] Furthermore, adjacent feature points with small feature differences can be further processed to extract effective feature points.

[0061] This can be achieved through the attribute parameters of each scan point.

[0062] In this embodiment of the application, it is necessary to construct a target three-dimensional map of the downhole working area using the point cloud data corresponding to each scanning point. The effective feature points can be judged and extracted by the appearance attributes of each scanning point.

[0063] Optionally, each scanning point can be determined by curvature, where the curvature corresponding to the scanning point can characterize the degree of curvature of the plane to which the scanning point belongs.

[0064] Furthermore, the curvature of each scan point can be obtained based on a set algorithm.

[0065] S202: Based on the curvature of each scan point, extract effective feature points from the scan point set, and generate a scan point cloud dataset of the downhole work area based on the point cloud data of the effective feature points.

[0066] In this embodiment of the application, the lidar can acquire multiple scanning points within the same scanning range.

[0067] Among them, edge points and planar points can be determined from the set of scan points based on the curvature of the scan points, and used as effective feature points.

[0068] In practice, the curvature can be used to determine the edge points and planar points in the scanned points.

[0069] Optionally, edge point curvature thresholds and planar point curvature thresholds can be obtained.

[0070] Furthermore, the curvature of the scan point is compared with the edge point curvature threshold. When the curvature of the scan point is greater than or equal to the edge point curvature threshold, the scan point can be determined to be an edge point.

[0071] Accordingly, the curvature of the scan point can be compared with the curvature threshold of the plane point. If the curvature of the scan point is less than or equal to the curvature threshold of the plane point, then the scan point can be determined to be a plane point.

[0072] Among them, edge points can characterize the edge features of the plane to which the scan point belongs, and plane points can characterize the planar features of the plane to which the scan point belongs. Therefore, edge points and plane points can be identified as effective feature points in the scan point set.

[0073] Furthermore, based on the point cloud data of each valid feature point, a point cloud dataset corresponding to the downhole working area is generated and identified as the scanned point cloud dataset.

[0074] In this embodiment of the application, the scanning results of the LiDAR on the downhole working area are output based on the scanning frame. By determining the point cloud data of the effective feature points in each scanning frame, the point cloud data of the effective feature points in the downhole working area can be obtained.

[0075] Optionally, the scan frames collected by the lidar can be acquired.

[0076] Specifically, the system can acquire scanned images from the output device of the lidar, which include all scanned points within the area covered by the scanned image. Effective feature points can be extracted from these scanned points to generate scanned frames acquired by the lidar.

[0077] Furthermore, based on the curvature of the effective feature points, the effective feature points in the scan frame that match the previous scan frame are determined as the target feature points of the scan frame.

[0078] In practice, the lidar scans the downhole work area based on the movement of the drone, with each scan frame having its corresponding scan position.

[0079] To improve the scanning coverage of the downhole working area, the distance between adjacent scanning positions of the lidar is relatively short, and there is an overlap of effective feature points between adjacent scanning frames acquired by the lidar.

[0080] It should be noted that when the lidar on the control drone scans the downhole work area, the point cloud data of the scan points obtained at the initial scan position has already been determined.

[0081] Furthermore, by matching effective feature points between adjacent scan frames, the point cloud data of effective feature points in the scan frame can be determined based on the matching results.

[0082] Optionally, effective feature points between adjacent scan frames can be matched based on curvature.

[0083] For example, you can set it to acquire point cloud data of valid feature points in the second scan frame.

[0084] Specifically, effective feature points in the second scan frame can be matched with effective feature points in the first scan frame based on curvature. Two effective feature points with the same curvature in the two scan frames are identified as two matched effective feature points, thereby obtaining the target feature points in the second scan frame that have matching effective feature points.

[0085] This can be understood as follows: when a valid feature point in the second scan frame matches a valid feature point in the first scan frame, it can be determined that the valid feature point in the second scan frame and the valid feature point in the first scan frame that it matches are the same valid feature point.

[0086] Furthermore, the point cloud data of the target feature points is determined, and the point cloud data of the effective feature points of the scan frame is determined based on the relative positions between the target feature points and the remaining effective feature points.

[0087] In this embodiment of the application, the valid feature points that match in two adjacent scan frames can be determined as the same valid feature point. Based on the point cloud data of the valid feature points in the previous scan frame, the point cloud data of the target feature point in the current scan frame can be determined.

[0088] For example, based on the example above, the point cloud data of the target feature points in the second scan frame is the same as the point cloud data of the valid feature points in the first scan frame that it matches.

[0089] Among them, the target feature points are some of the valid feature points in the scan frame to which they belong. For the remaining valid feature points that are not target feature points, the corresponding point cloud data can be determined by the relative positional relationship between them and the target feature points.

[0090] Furthermore, based on the point cloud data of the target feature points and the point cloud data of the remaining valid feature points, the point cloud data of all valid feature points of the scan frame can be determined.

[0091] It should be noted that the point cloud data of valid feature points in each scan frame can be determined frame by frame. This can be understood as follows: after acquiring the point cloud data of valid feature points in the current scan frame, the point cloud data of valid feature points in the next scan frame is determined.

[0092] Furthermore, based on the point cloud data of the effective feature points of each scan frame, a scan point cloud dataset of the downhole working area is generated.

[0093] In this embodiment of the application, after acquiring the point cloud data of the effective feature points in the scan frame frame by frame, the point cloud data of the effective feature points in each scan frame can be integrated, and the integrated point cloud data of all effective feature points can be determined as the scan point cloud dataset corresponding to the downhole working area.

[0094] This application proposes a method for constructing a 3D LiDAR downhole map based on an unmanned aerial vehicle (UAV). It extracts effective feature points from the LiDAR-acquired scan point set based on the curvature of each scan point. Then, based on the matching between scan frames and the previous scan frame, and the relative positional relationships between effective feature points within the scan frames, it determines the point cloud data of the effective feature points in each scan frame, thereby obtaining a scan point cloud dataset of the downhole working area acquired by the LiDAR. In this application, the extraction of effective feature points reduces the number of scan points, effectively reducing the computational load for matching between subsequent scan frames and improving the efficiency of acquiring the scan point cloud dataset.

[0095] In practice, the scan frames acquired by the lidar may be distorted. Therefore, before matching a scan frame with the previous scan frame, the distorted frames need to be processed. This can be done by combining... Figure 3 To understand further, Figure 3 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method, as shown below. Figure 3 As shown, the method includes:

[0096] S301, determine whether the scan frame is a distorted frame.

[0097] In practice, the lidar moves while emitting lasers for scanning, and the laser scanning data at each angle is not acquired instantaneously. Therefore, motion distortion may occur when the lidar is scanning.

[0098] This can be understood as follows: the scanning frames acquired by the LiDAR have corresponding timestamps. Between adjacent timestamps during the LiDAR's scanning, the LiDAR's position will shift and its pose will change. Therefore, there is a possibility of motion distortion when the LiDAR is scanning.

[0099] Furthermore, scan frames exhibiting motion distortion can be identified as distorted frames of the lidar.

[0100] In this embodiment of the application, it is possible to determine whether the current scan frame has experienced distortion by using relevant set standards.

[0101] For example, since the scanning position difference between adjacent scan frames is small, it is possible to determine whether the current scan frame is distorted based on the attribute parameter information in the previous scan frame. Optionally, the relevant judgment criteria can be stored in a set location and called before matching the scan frame with the previous scan frame, thereby realizing the judgment of whether the current scan frame is a distorted frame.

[0102] S302, in response to the scan frame being a distorted frame, the distorted frame is compensated based on the acquired data corresponding to the inertial measurement unit to obtain the compensated scan frame.

[0103] In this embodiment, there is a discrepancy between the point cloud data of distorted points in the distorted frame and the point cloud data of actual effective feature points. Therefore, the accuracy of the 3D map constructed based on the distorted point cloud data is poor. To achieve accurate construction of the target 3D map, a correction compensation frame corresponding to the distorted frame can be obtained to compensate for the distortion of the LiDAR.

[0104] Optionally, distortion compensation for lidar scanning can be achieved through an inertial measurement unit (IMU). The IMU can be a 9-axis IMU or any other measuring device that can acquire lidar acceleration and angular velocity information; there is no limitation here.

[0105] It should be noted that the IMU carrier device and the lidar are relatively fixed. This can be understood as the IMU and lidar not undergoing relative displacement. It can be understood as the IMU and lidar being in a tight coupling relationship, and the acceleration and angular velocity acquired by the IMU are the same as those acquired by the lidar.

[0106] Optionally, the data acquired by the IMU can be pre-integrated to compensate for the distortion of the lidar.

[0107] In practice, the IMU's operating frequency can be 100-1000Hz, while the lidar's scanning frequency can be 10Hz. The IMU's data acquisition frequency is higher than the lidar's scanning frequency for the downhole working area. Therefore, pre-integration processing can be performed on the IMU's acquired data based on the time interval between each lidar scan frame.

[0108] Furthermore, the distortion timestamp of the distorted frame can be determined.

[0109] The scanning frames output by the lidar have corresponding timestamps.

[0110] Alternatively, the distortion timestamp corresponding to the distortion frame can be determined by reading the attribute list of the distortion frame.

[0111] Furthermore, the data collected by the inertial measurement unit at the distortion timestamp is obtained.

[0112] In this embodiment of the application, the IMU can acquire and store acceleration and angular velocity data at a set location through its configured gyroscope and accelerometer. The acquired data of the IMU has a timestamp, and the acquired data with the same timestamp as the distortion timestamp can be obtained from the IMU data storage location.

[0113] Furthermore, pose calculation is performed on the collected data of the distorted timestamps to generate the compensated pose trajectory corresponding to the distorted frame.

[0114] In this embodiment of the application, the compensated attitude trajectory corresponding to the distorted frame can be obtained by performing pose calculation on the data collected by the IMU.

[0115] This can be understood as follows: after the pose is calculated using an IMU, the distorted points in the distorted frame can be corrected and compensated based on the compensated pose trajectory obtained from the calculation. The compensated pose trajectory can be understood as the pose change trajectory of the effective feature points in the previous scan frame of the distorted frame when no motion distortion occurred.

[0116] Furthermore, based on the valid feature points and compensated attitude trajectory of the previous scan frame corresponding to the distorted frame, a compensated scan frame corresponding to the distorted frame is generated.

[0117] In this embodiment, the compensated scan frame can be obtained by using the effective feature points in the previous scan frame of the distorted frame and the corresponding compensated attitude trajectory.

[0118] Among them, the effective feature points of the previous scan frame can be controlled to move along the compensation attitude trajectory to obtain the target compensation point formed after the effective feature points are moved.

[0119] In this embodiment, the effective feature points in the previous scan frame of the distorted frame may have corresponding compensated attitude trajectories. It can be understood that the effective feature points with compensated attitude trajectories have matching effective feature points in the distorted frame.

[0120] Furthermore, it is possible to control the valid feature points in the previous scan frame that have a compensation attitude trajectory, move along the corresponding compensation attitude trajectory, obtain the feature points corresponding to the position after the movement ends, and determine the feature points as the target compensation points.

[0121] It should be noted that there is a correspondence between the compensated attitude trajectory and the valid feature points in the previous scan frame, as well as the distorted feature points in the distorted frame. Based on this correspondence, the valid feature points and distorted feature points corresponding to each compensated attitude trajectory can be determined.

[0122] Accordingly, the target distortion point corresponding to the target compensation point on the distortion frame is determined based on the compensation attitude trajectory, and the remaining compensation point corresponding to the remaining distortion point is determined based on the relative position of the remaining distortion point and the target distortion point.

[0123] In this embodiment of the application, the target distortion point corresponding to each target compensation point can be determined based on the correspondence between each compensated attitude trajectory and the distortion point in the distortion frame.

[0124] In implementation, the target compensation point is the compensation point corresponding to a portion of the distorted points in the distorted frame. For the remaining distorted points, the corresponding remaining compensation point can be obtained based on their relative positional relationship with the target distorted point.

[0125] Furthermore, based on the target compensation point and the remaining compensation points, a compensated scan frame corresponding to the distorted frame is generated.

[0126] In this embodiment, the target compensation point and the remaining compensation point realize the correction and compensation of all the distorted points in the distorted frame. Based on the target compensation point and the remaining compensation point, the compensated scan frame corresponding to the distorted frame can be generated.

[0127] Furthermore, the distorted frame is replaced by the compensated scan frame, and it is matched with the previous scan frame to obtain the point cloud data of the effective feature points in the compensated scan frame. Based on the point cloud data of the effective feature points in the compensated scan frame, the point cloud data of the effective feature points in the next scan frame is determined.

[0128] The proposed method for constructing 3D LiDAR downhole maps based on UAVs involves using an IMU (Integrated Measurement Unit) to compensate for motion distortion in LiDAR frames, generating compensated scan frames. The tight coupling between the IMU and the LiDAR allows the IMU to acquire acceleration and angular velocity data from the LiDAR, improving the accuracy of distortion compensation.

[0129] In the above embodiments, the construction of the target 3D map can be combined with... Figure 4 To understand further, Figure 4 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method, as shown below. Figure 4 As shown, the method includes:

[0130] S401, obtain the union of point cloud data from the scanned point cloud dataset and the measured point cloud dataset.

[0131] In this embodiment of the application, the point cloud data of each effective feature point in the downhole working area can be highly covered by the scanned point cloud dataset and the measured point cloud dataset. The scanned point cloud dataset and the measured point cloud dataset are integrated to obtain the corresponding point cloud data union.

[0132] The point cloud data includes both the point cloud data in the scanned point cloud dataset and the point cloud data in the measured point cloud dataset.

[0133] S402 generates a target 3D map of the downhole work area based on the union of point cloud data.

[0134] In this embodiment of the application, the point cloud data of each point in the point cloud data can be placed into a set coordinate system, thereby generating a target three-dimensional map of the downhole working area.

[0135] Optionally, the world coordinate system can be defined as the set coordinate system corresponding to the union of point cloud data, and the union of point cloud data can be placed into the world coordinate system. Based on the world coordinate system and the union of point cloud data, a target 3D map corresponding to the downhole working area can be generated.

[0136] The UAV-based 3D LiDAR downhole map construction method proposed in this application constructs a target 3D map by using point cloud data union, which optimizes the construction effect of the target 3D map of the downhole operation area and improves the coverage of the target 3D map of the downhole operation area.

[0137] To better understand the above embodiments, it can be combined with Figure 5 , Figure 5 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method, as shown below. Figure 5 As shown, where:

[0138] The downhole work area is scanned using a lidar-equipped drone, and effective feature points are extracted from the output scan frames. Adjacent scan frames are matched to determine the point cloud data of effective feature points in each frame, generating a corresponding scan point cloud dataset. Before each matching of adjacent scan frames, distortion compensation is performed on distorted frames exhibiting motion distortion. The IMU, tightly coupled to the lidar, calculates pose based on acquired acceleration and angular velocity, obtains the compensated attitude trajectory corresponding to the distorted frame, and uses this trajectory to compensate for distortion, generating a compensated scan frame. The compensated scan frame replaces the distorted frame, and the point cloud data of the compensated points is obtained. Simultaneously, UWB technology is used to obtain the relative positional relationship between the target point and the base station. Based on the base station's point cloud data, the relative position between the target point and the base station is calculated, thereby obtaining the point cloud data for each target point and ultimately acquiring the measured point cloud dataset for the downhole work area.

[0139] Furthermore, a target 3D map of the downhole work area is generated based on the scanned point cloud dataset and the calculated point cloud dataset.

[0140] This application proposes a method for constructing a 3D LiDAR-based downhole map using an unmanned aerial vehicle (UAV). The method involves controlling a UAV's LiDAR to scan the downhole work area and acquire the corresponding scanned point cloud dataset. Simultaneously, UWB technology is used to obtain the relative positions between the points to be located in the downhole work area and the base station. Based on the point cloud data from the base station and the relative positions, point cloud data for each point to be located is calculated, resulting in a calculated point cloud dataset for the downhole work area. A target 3D map of the downhole work area is generated based on the combined scanned and calculated point cloud datasets. This method constructs the target 3D map based on the combined point cloud datasets, optimizing the completeness of the target 3D map and improving the accuracy of its construction. This allows the UAV to achieve highly accurate real-time positioning based on the target 3D map, thus optimizing the UAV's positioning performance.

[0141] Furthermore, while constructing a 3D map of the target, precise real-time positioning of the drone can be achieved, which can be combined with... Figure 6 understand, Figure 6 This is a flowchart illustrating another embodiment of the UAV-based 3D lidar downhole map construction method, as shown below. Figure 6 As shown, the method includes:

[0142] S601, obtain the first operational coordinate set of the UAV based on the scanned point cloud dataset.

[0143] In this embodiment of the application, by controlling the movement of the drone, the lidar configured on the drone can collect point cloud data of the scanning points in the downhole working area. Therefore, it is necessary to accurately locate and accurately control the movement of the drone.

[0144] Optionally, the location of the LiDAR scan can be calculated based on the point cloud data of each valid feature point in the scanned point cloud dataset.

[0145] Furthermore, the lidar outputs effective feature points in the downhole working area based on scan frames. Therefore, the scanning position of the UAV can be obtained based on the point cloud data of the effective feature points in each scan frame.

[0146] This can be understood as follows: based on the point cloud data of each effective feature point in any scan frame, calculations can be performed, and the three-dimensional coordinates of the effective feature points in the scan frame when they are scanned and acquired by the lidar can be obtained based on the calculation results.

[0147] In practice, the lidar is deployed on the drone, so the three-dimensional coordinates of the lidar's location can be determined as the drone's first operating coordinates.

[0148] Furthermore, the three-dimensional coordinates of the scanning position of the LiDAR corresponding to each scanning frame are integrated to obtain the first operational coordinate set of the UAV.

[0149] S602: Obtain the second set of operational coordinates for the UAV based on its relative position to the positioning base station.

[0150] In practice, calculating the 3D coordinates of the LiDAR scanning position based on the point cloud data may introduce errors. Therefore, alternative methods can be used to simultaneously determine the UAV's operational coordinates.

[0151] In this embodiment, a positioning base station can be deployed in the underground working area, and the three-dimensional coordinates of the positioning base station can be determined. Since there is a relative positional relationship between the drone and the base station, the coordinates of the drone's location can be calculated by obtaining the relative positional relationship between the two and combining it with the three-dimensional coordinates of the positioning base station. The calculated three-dimensional coordinates are then determined as the drone's second working coordinates.

[0152] In practice, the UAV moves continuously while scanning the underground work area, allowing the UAV to obtain the relative position between itself and the positioning base station after each movement, and to calculate the corresponding three-dimensional coordinates of the UAV.

[0153] Furthermore, based on the three-dimensional coordinates obtained after all the UAVs have moved, a second set of operational coordinates for the UAVs is generated.

[0154] Alternatively, the relative position between the positioning base station and the drone can be obtained through UWB technology.

[0155] S603, determine the operating position of the UAV in the downhole operating area based on the first operating coordinate set and the second operating coordinate set.

[0156] In this embodiment of the application, the first operation coordinate set and the second operation coordinate set have a corresponding relationship. The first operation coordinate set and the second operation coordinate set with the corresponding relationship can be determined as the same three-dimensional coordinates of the UAV.

[0157] Furthermore, the first and second working coordinates that have a corresponding relationship can be filtered and calculated, and the result of the filtering calculation can be determined as the target working coordinates of the UAV in the downhole working area.

[0158] This can be understood as follows: when there is an error in the first operation coordinate, the first operation coordinate can be compensated and corrected by filtering the second operation coordinate and the first operation coordinate, thereby determining the accurate target operation coordinate of the UAV.

[0159] Furthermore, the operating position of the UAV in the underground working area is determined based on the target operating coordinates.

[0160] Optionally, the first and second operating coordinates of the UAV can be world coordinates. After obtaining the target operating coordinates of the UAV, the operating position of the UAV in the downhole operating area can be determined based on the world coordinate system.

[0161] This application proposes a method for constructing a 3D LiDAR-based downhole map using unmanned aerial vehicles (UAVs). This method obtains a first set of operational coordinates for the UAV using point cloud data acquired through LiDAR scanning. Simultaneously, based on the relative positional relationship between the UAV and a positioning base station, a second set of operational coordinates is obtained. Finally, the target operational coordinates of the UAV are derived from both the first and second sets, thereby determining the UAV's operational position. This application employs multiple methods to determine the UAV's operational position, improving its positioning accuracy, control precision, safety, and efficiency in downhole operations.

[0162] Corresponding to the UAV-based 3D LiDAR downhole map construction methods proposed in the above embodiments, an embodiment of this application also proposes a UAV-based 3D LiDAR downhole map construction device. Since the UAV-based 3D LiDAR downhole map construction device proposed in this embodiment corresponds to the UAV-based 3D LiDAR downhole map construction methods proposed in the above embodiments, the implementation methods of the above UAV-based 3D LiDAR downhole map construction methods are also applicable to the UAV-based 3D LiDAR downhole map construction device proposed in this embodiment, and will not be described in detail in the following embodiments.

[0163] Figure 7 This is a schematic diagram of the structure of a UAV-based 3D lidar downhole map construction device according to an embodiment of this application, as shown below. Figure 7 As shown, the UAV-based 3D LiDAR downhole map construction device 700 includes a scanning module 71, a measurement module 72, and a generation module 73, wherein:

[0164] The scanning module 71 is used to control the UAV's lidar to scan the downhole working area in order to obtain a scan point cloud dataset of the downhole working area;

[0165] The measurement module 72 is used to obtain a measurement point cloud dataset of the underground working area based on the relative position of the UAV and the positioning base station of the underground working area;

[0166] The generation module 73 is used to generate a target 3D map of the downhole work area based on the scanned point cloud dataset and the measured point cloud dataset.

[0167] Figure 8 This is a schematic diagram of the structure of a UAV-based 3D lidar downhole map construction device according to another embodiment of this application, as shown below. Figure 8 As shown, the UAV-based 3D LiDAR downhole map construction device 800 includes a scanning module 81, a measurement module 82, a generation module 83, and a positioning module 84, wherein:

[0168] It should be noted that the scanning module 71, the measurement module 72, and the generation module 73 have the same structure and function as the scanning module 81, the measurement module 82, and the generation module 83.

[0169] In this embodiment of the application, the scanning module 81 is further configured to: acquire the scanning point set of the lidar and determine the curvature of each scanning point in the scanning point set; extract effective feature points from the scanning point set according to the curvature of each scanning point, and generate a scanning point cloud dataset of the downhole working area based on the point cloud data of the effective feature points.

[0170] In this embodiment of the application, the scanning module 81 is further configured to: determine edge points and planar points from the set of scanning points based on the curvature of the scanning points, as effective feature points.

[0171] In this embodiment of the application, the scanning module 81 is further configured to: acquire scanning frames collected by the lidar; determine effective feature points in the scanning frame that match the previous scanning frame based on the curvature of the effective feature points, and use them as target feature points of the scanning frame; determine the point cloud data of the target feature points, and determine the point cloud data of the effective feature points of the scanning frame according to the relative position between the target feature points and the remaining effective feature points; and generate a scanning point cloud dataset of the downhole working area based on the point cloud data of the effective feature points of each scanning frame.

[0172] In this embodiment of the application, the scanning module 81 is further configured to: determine whether the scanning frame is a distorted frame; and in response to the scanning frame being a distorted frame, compensate the distorted frame based on the acquisition data corresponding to the inertial measurement unit to obtain a compensated scanning frame.

[0173] In this embodiment of the application, the scanning module 81 is further configured to: determine the distortion timestamp of the distorted frame; acquire the data collected by the inertial measurement unit at the distortion timestamp; perform pose calculation on the data collected at the distortion timestamp to generate the compensated attitude trajectory corresponding to the distorted frame; and generate the compensated scanning frame corresponding to the distorted frame based on the effective feature points and the compensated attitude trajectory of the previous scanning frame corresponding to the distorted frame.

[0174] In this embodiment, the scanning module 81 is further configured to: control the effective feature points of the previous scanning frame to move along the compensation attitude trajectory, and obtain the target compensation point formed after the effective feature points are moved; determine the target distortion point corresponding to the target compensation point on the distortion frame according to the compensation attitude trajectory, and determine the remaining compensation point corresponding to the remaining distortion point according to the relative position of the remaining distortion point and the target distortion point; and generate the compensated scanning frame corresponding to the distortion frame based on the target compensation point and the remaining compensation point.

[0175] In this embodiment of the application, the generation module 83 is further configured to: obtain the union of point cloud data of the scanned point cloud dataset and the measured point cloud dataset; and generate a target three-dimensional map of the downhole work area based on the union of point cloud data.

[0176] In this embodiment of the application, the device further includes a positioning module 84, which is used to: obtain a first set of working coordinates of the UAV based on the scanned point cloud dataset; obtain a second set of working coordinates of the UAV based on the relative position of the UAV and the positioning base station; and determine the working position of the UAV in the underground working area based on the first set of working coordinates and the second set of working coordinates.

[0177] This application proposes a UAV-based 3D LiDAR downhole map construction device. It controls a UAV LiDAR to scan the downhole work area and acquire the corresponding scanned point cloud dataset. Simultaneously, it uses UWB technology to obtain the relative positions between the points to be located in the downhole work area and the base station. Based on the point cloud data from the base station and the relative positions, it calculates the point cloud data for each point to be located, thereby acquiring the corresponding measured point cloud dataset for the downhole work area. Based on the combined scanned and measured point cloud datasets, a target 3D map of the downhole work area is generated. In this application, the target 3D map is constructed based on the combined point cloud dataset, optimizing the completeness of the target 3D map and improving the accuracy of its construction. This allows the UAV to achieve highly accurate real-time positioning based on the target 3D map, thus optimizing the UAV's positioning performance.

[0178] To achieve the above embodiments, this application also proposes a computer-readable storage medium.

[0179] This application provides a computer-readable storage medium storing a computer program. When the program is executed by a processor, it implements the UAV-based three-dimensional lidar downhole map construction method proposed in the above embodiments.

[0180] The computer-readable storage medium of this application controls a UAV's lidar to scan an underground working area and acquire the corresponding scanned point cloud dataset. Simultaneously, UWB technology is used to obtain the relative position between the target point in the underground working area and the base station. Based on the point cloud data from the base station and the relative position, point cloud data for each target point is calculated, thereby acquiring a calculated point cloud dataset corresponding to the underground working area. Based on the combined scanned and calculated point cloud datasets, a target 3D map of the underground working area is generated. In this application, the target 3D map is constructed based on the combined point cloud dataset, optimizing the completeness of the target 3D map and improving the accuracy of its construction. This allows the UAV to achieve highly accurate real-time positioning based on the target 3D map, thus optimizing the UAV's positioning performance.

[0181] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not 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 application.

[0182] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0183] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0184] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0185] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0186] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for constructing a 3D lidar downhole map based on an unmanned aerial vehicle (UAV), characterized in that, include: Controlling a drone's lidar to scan an underground working area to obtain a scanned point cloud dataset of the underground working area includes: acquiring the lidar's scanned point set and determining the curvature of each scanned point in the scanned point set; based on the curvature of the scanned points, determining edge points and planar points from the scanned point set as the effective feature points; acquiring scan frames collected by the lidar; based on the curvature of the effective feature points, determining the effective feature points in the scan frame that match the previous scan frame as the target feature points of the scan frame; determining the point cloud data of the target feature points; and determining the point cloud data of the effective feature points of the scan frame based on the relative positions between the target feature points and the remaining effective feature points. The point cloud data of effective feature points is used to generate the scanned point cloud dataset of the downhole working area. Before determining the effective feature points in the scan frame that match the previous scan frame based on the curvature of the effective feature points, and using these as target feature points for the scan frame, the method further includes: determining whether the scan frame is a distorted frame; in response to the scan frame being a distorted frame, determining the distortion timestamp of the distorted frame; acquiring the data collected by the inertial measurement unit at the distortion timestamp; performing pose calculation on the acquired data at the distortion timestamp to generate a compensated attitude trajectory corresponding to the distorted frame; and generating the compensated scan frame corresponding to the distorted frame based on the effective feature points of the previous scan frame and the compensated attitude trajectory. Based on the relative position of the UAV and the positioning base station of the underground working area, the measured point cloud dataset of the underground working area is obtained. Specifically, the corresponding point to be located in the underground working area is obtained by the UAV, and the relative positional relationship between the point to be located and the base station is obtained through the interaction of request and response signals between the base stations. The relative positional relationship is transformed and solved using the known point cloud data of the base station, thereby determining the point cloud data corresponding to each point to be located, and the total point cloud data of all points to be located is determined as the measured point cloud dataset. Based on the scanned point cloud dataset and the calculated point cloud dataset, a target 3D map of the downhole work area is generated.

2. The method according to claim 1, characterized in that, The step of generating the compensated scan frame corresponding to the distorted frame based on the valid feature points of the previous scan frame corresponding to the distorted frame and the compensated pose trajectory includes: Control the effective feature points of the previous scan frame to move along the compensation attitude trajectory, and obtain the target compensation point formed after the effective feature points move; Based on the compensation attitude trajectory, the target distortion point corresponding to the target compensation point on the distorted frame is determined, and the remaining compensation point corresponding to the remaining distortion point is determined based on the relative position of the remaining distortion point and the target distortion point. Based on the target compensation point and the remaining compensation points, the compensated scan frame corresponding to the distorted frame is generated.

3. The method according to claim 1, characterized in that, The step of generating a target 3D map of the downhole work area based on the scanned coordinate set and the calculated coordinate set includes: Obtain the union of point cloud data from the scanned point cloud dataset and the measured point cloud dataset; Based on the union of the point cloud data, the target 3D map of the downhole work area is generated.

4. The method according to any one of claims 1-3, characterized in that, The method further includes: The first operational coordinate set of the UAV is obtained based on the scanned point cloud dataset; Based on the relative position of the UAV and the positioning base station, obtain the second set of operational coordinates for the UAV; The operating position of the UAV in the downhole operating area is determined based on the first operating coordinate set and the second operating coordinate set.

5. A UAV-based three-dimensional lidar downhole map construction device, characterized in that, include: A scanning module is used to control the lidar of a UAV to scan the downhole working area to obtain a scanned point cloud dataset of the downhole working area. This includes: acquiring the scanned point set of the lidar and determining the curvature of each scanned point in the scanned point set; based on the curvature of the scanned points, determining edge points and planar points from the scanned point set as the effective feature points; acquiring scan frames collected by the lidar; based on the curvature of the effective feature points, determining the effective feature points in the scan frame that match the previous scan frame as the target feature points of the scan frame; determining the point cloud data of the target feature points; and determining the point cloud data of the effective feature points of the scan frame based on the relative positions between the target feature points and the remaining effective feature points. The point cloud data of effective feature points in the scanning frame is used to generate the scanning point cloud dataset of the downhole working area. Before determining the effective feature points in the scanning frame that match the previous scanning frame based on the curvature of the effective feature points, and using these as the target feature points of the scanning frame, the method further includes: determining whether the scanning frame is a distorted frame; in response to the scanning frame being a distorted frame, determining the distortion timestamp of the distorted frame; acquiring the data collected by the inertial measurement unit at the distortion timestamp; performing pose calculation on the acquired data at the distortion timestamp to generate the compensated attitude trajectory corresponding to the distorted frame; and generating the compensated scanning frame corresponding to the distorted frame based on the effective feature points of the previous scanning frame corresponding to the distorted frame and the compensated attitude trajectory. The calculation module is used to obtain a calculation point cloud dataset of the underground working area based on the relative position of the UAV and the positioning base station of the underground working area. Specifically, the UAV acquires the corresponding points to be located in the underground working area, the relative positional relationship between the points to be located and the base station is obtained through the interaction of request and response signals between the base stations, the relative positional relationship is converted and calculated using the known point cloud data of the base station, thereby determining the point cloud data corresponding to each point to be located, and the total point cloud data of all points to be located is determined as the calculation point cloud dataset. The generation module is used to generate a target 3D map of the downhole work area based on the scanned point cloud dataset and the calculated point cloud dataset.

Images