3D map construction methods, systems, machinery, equipment and media

Abstract

This invention provides a method, system, machinery, equipment, and medium for constructing a 3D map, including: acquiring current point cloud data and inertial data; determining the motion change information of the machinery based on the inertial data; correcting the current point cloud data using the motion change information to obtain target point cloud data; extracting point cloud features from the target point cloud data; performing point cloud registration between the current frame and the previous frame based on the point cloud features; determining the relative motion information between the current frame and the previous frame based on the result of the point cloud registration; and constructing a 3D map based on the relative motion information. This invention addresses the shortcomings of existing technologies where the scanning device is fixed to the periphery of the construction area, making it unsuitable for large-scale work scenarios and creating scanning blind spots.

Description

Technical Field

[0001] This invention relates to the field of 3D construction technology for work machinery, and in particular to a 3D map construction method, system, work machinery, equipment and medium. Background Technology

[0002] Earthmoving is a crucial initial step in construction and a common operating scenario for excavators. According to standard operating procedures, during earthmoving, the excavator operator needs to excavate to a depth approximately 30 centimeters below the target depth, based on the construction drawings. Then, manual labor or a small excavator is used to refine the excavation to the target depth. During this refinement process, the excavator operates for a period, then pauses to allow a surveyor to measure and compare the result to a standard height. This process of excavation, measurement, and comparison is repeated until the target depth is reached.

[0003] Using manual inspection methods for earthwork operations results in low work efficiency and can easily delay the construction period. In addition, the accuracy of manual inspection methods is questionable.

[0004] To address the numerous problems associated with manual inspection, existing technologies employ a positioning scanning device fixed to the perimeter of the construction area. This device monitors and scans the terrain model of the construction area to instruct the excavating equipment to carry out excavation work. However, existing technologies are limited by their fixed location, making them unsuitable for monitoring large-scale work scenarios. Furthermore, because the scanning device is fixed to the perimeter of the construction area, blind spots exist during scanning.

[0005] Therefore, how to quickly and accurately construct three-dimensional construction scenarios so that excavators can complete earthwork operations quickly and accurately is an important issue that the target industry urgently needs to address. Summary of the Invention

[0006] This invention provides a method, system, machinery, equipment, and medium for constructing three-dimensional maps, which solves the defects in the prior art where the scanning device is fixed to the periphery of the construction area, making it unsuitable for large-scale operation scenarios and creating scanning blind spots, thereby achieving rapid and accurate three-dimensional construction of construction scenarios.

[0007] This invention provides a method for constructing a three-dimensional map, including:

[0008] Acquire current point cloud data and inertial data, wherein the current point cloud data is obtained through a scanning unit installed on the operating machinery, and the inertial data is obtained through an inertial measurement unit installed on the operating machinery;

[0009] Based on the inertial data, the motion change information of the operating machinery is determined;

[0010] Using the motion change information, the current point cloud data is corrected to obtain the target point cloud data;

[0011] Extract the point cloud features from the target point cloud data;

[0012] Based on the point cloud features, perform point cloud registration operation between the current frame point cloud data and the previous frame point cloud data.

[0013] Based on the result of the point cloud registration operation, the relative motion information between the current frame point cloud data and the previous frame point cloud data is determined;

[0014] Based on the relative motion information, the three-dimensional map is constructed.

[0015] According to a three-dimensional map construction method of an embodiment of the present invention, the step of extracting point cloud features from the target point cloud data includes:

[0016] Calculate the curvature corresponding to each target point in the target point cloud data;

[0017] The target point corresponding to the curvature being less than the first preset curvature is taken as the planar feature point;

[0018] The target point corresponding to the curvature being greater than the second preset curvature is used as the corner feature point;

[0019] The planar feature points and the corner feature points are used as the point cloud features;

[0020] Wherein, the first preset curvature is less than or equal to the second preset curvature.

[0021] According to an embodiment of the present invention, a method for constructing a three-dimensional map, wherein constructing the three-dimensional map based on the relative motion information includes:

[0022] Based on the relative motion information, adjust the current frame point cloud data;

[0023] The adjusted current frame point cloud data is overlaid onto the previous frame point cloud data to complete the construction of the 3D map.

[0024] According to an embodiment of the present invention, a three-dimensional map construction method further includes, before completing the construction of the three-dimensional map based on the relative motion information:

[0025] Based on the continuity of point cloud data in adjacent frames, a first motion constraint is established;

[0026] Based on the aforementioned motion change information, a second motion constraint is established;

[0027] Each frame of point cloud data is matched with preset point cloud data, and a third motion constraint is established based on the matching results.

[0028] After constructing the 3D map based on the relative motion information, the process further includes:

[0029] The three-dimensional map is adjusted using the first motion constraint, the second motion constraint, and the third motion constraint.

[0030] According to an embodiment of the three-dimensional map construction method of the present invention, after acquiring the current point cloud data and inertial data, the method further includes:

[0031] The current point cloud data is classified to obtain at least one point cloud set;

[0032] The extraction of point cloud features from the target point cloud data includes:

[0033] Extract the point cloud features for each of the point cloud sets respectively;

[0034] The point cloud registration operation based on the point cloud features, performing the current frame point cloud data and the previous frame point cloud data, includes:

[0035] Based on the point cloud features of each point cloud set, point cloud registration operation is performed between the current point cloud set and the previous point cloud set.

[0036] A three-dimensional map construction method according to an embodiment of the present invention,

[0037] The acquisition of current point cloud data includes:

[0038] Acquire first point cloud data and second point cloud data, wherein the first point cloud data is obtained through the first scanning component of the scanning unit, and the second point cloud data is obtained through the second scanning component of the scanning unit;

[0039] Based on preset calibration parameters, the first point cloud data and the second point cloud data are fused to obtain the current point cloud data.

[0040] This invention also provides a three-dimensional map construction system, including: a machine body, a control unit installed on the machine body, and a scanning unit and an inertial measurement unit installed on the machine body, wherein the scanning unit and the inertial measurement unit communicate with the control unit respectively;

[0041] The scanning unit is used to scan and obtain the current point cloud data, and send the current point cloud data to the control unit;

[0042] The inertial measurement unit is used to scan and obtain inertial data, and then send the inertial data to the control unit;

[0043] The control unit is configured to acquire the current point cloud data and the inertial data; determine the motion change information of the operating machinery based on the inertial data; perform correction processing on the current point cloud data using the motion change information to obtain target point cloud data; extract point cloud features from the target point cloud data; perform point cloud registration operation between the current frame point cloud data and the previous frame point cloud data based on the point cloud features; determine the relative motion information between the current frame point cloud data and the previous frame point cloud data based on the result of the point cloud registration operation; and complete the construction of the three-dimensional map based on the relative motion information.

[0044] This invention also provides a working machine, including the aforementioned three-dimensional map construction system.

[0045] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the three-dimensional map construction method.

[0046] This invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the three-dimensional map construction method.

[0047] The three-dimensional map construction method, system, operating machinery, equipment, and medium provided in this invention acquire current point cloud data and inertial data. The current point cloud data is obtained through a scanning unit installed on the machinery, and the inertial data is obtained through an inertial measurement unit installed on the operating machinery. As can be seen, by installing the scanning unit and the inertial measurement unit on the operating machinery, this invention can acquire point cloud data and inertial data corresponding to the operating machinery in real time as the operating machinery moves. This makes the invention applicable to any range of operating scenarios and effectively solves the problem of blind spots in the prior art due to fixing the scanning device on the outer perimeter of the construction area.

[0048] Furthermore, based on inertial data, the motion change information of the operating machinery is determined; using this motion change information, the current point cloud data is corrected to obtain the target point cloud data. It is evident that this invention corrects the point cloud data using inertial data, resulting in higher accuracy when constructing a 3D map using the point cloud data later. Further, this invention extracts the point cloud features of the target point cloud data; based on these features, point cloud registration is performed between the current frame and the previous frame; based on the result of the point cloud registration, the relative motion information between the current and previous frame is determined; based on this relative motion information, the 3D map is constructed. It is evident that this invention uses a cyclical approach to match the current and previous frame point cloud data to complete the 3D map construction. Since the amount of data in a single frame is smaller than that in multiple frames, the data processing speed is faster. Therefore, it can effectively improve the construction speed of the 3D map, ultimately achieving rapid and accurate 3D construction of the construction scene. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is one of the flowcharts illustrating a three-dimensional map construction method provided in an embodiment of the present invention;

[0051] Figure 2 This is a schematic diagram of the excavator structure provided in an embodiment of the present invention;

[0052] Figure 3 This is a second schematic flowchart of a three-dimensional map construction method provided in an embodiment of the present invention;

[0053] Figure 4 This is the third flowchart illustrating a three-dimensional map construction method provided in this embodiment of the invention;

[0054] Figure 5 This is a schematic diagram of the structure of a three-dimensional map construction system provided in an embodiment of the present invention;

[0055] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] The following is combined with Figures 1 to 4 This invention describes a method for constructing a three-dimensional map according to an embodiment of the present invention.

[0058] This invention provides a method for constructing a three-dimensional map. This method can be applied to machinery such as excavators and loaders, and also to servers. The following description uses the application of this method in an excavator as an example; however, it should be noted that this is merely illustrative and not intended to limit the scope of protection of this invention. Other descriptions in this invention are also illustrative and not intended to limit the scope of protection of this invention, and will not be described in detail thereafter. The specific implementation of this method is as follows: Figure 1 As shown:

[0059] Step 101: Obtain the current point cloud data and inertial data.

[0060] The current point cloud data is obtained through a scanning unit installed on the operating machinery, and the inertial data is obtained through an inertial measurement unit installed on the operating machinery.

[0061] Specifically, the current point cloud data includes multiple frames of point cloud data.

[0062] Specifically, let's take an excavator as an example of the operating machinery. The scanning unit can be a LiDAR or a depth camera, where LiDAR includes mechanical LiDAR and solid-state LiDAR.

[0063] The scanning unit of this invention includes a first scanning component and a second scanning component. The first scanning component is used to scan the motion information of the operating machinery, and the second scanning component is used to scan the operating area of ​​the operating machinery. The following description uses a mechanical lidar 201 as the first scanning component and a solid-state lidar 202 as the second scanning component as an example. See details below. Figure 2 .

[0064] Among them, the mechanical lidar 201 is installed on the top of the excavator for 360-degree panoramic scanning, and the solid-state lidar 202 is installed at the front of the top of the cab for scanning the work area.

[0065] Additionally, the inertial measurement unit (IMU sensor) 203 is mounted on the excavator's operator's seat to measure the excavator's position and attitude; see details below. Figure 2 .

[0066] In one specific embodiment, after the mechanical lidar 201 and the solid-state lidar 202 are installed on the excavator, a joint calibration operation is required to obtain calibration parameters. After calibration, when the excavator is actually working, the mechanical lidar 201 scans the first point cloud data, and the solid-state lidar 202 scans the second point cloud data, respectively sending their corresponding first and second point cloud data to the control unit. The control unit receives the first and second point cloud data, and using the pre-calibrated calibration parameters, fuses the first and second point cloud data to obtain the current point cloud data.

[0067] In one specific embodiment, after acquiring point cloud data, the point cloud data is classified to obtain at least one point cloud set. For example, the point cloud data can be classified according to the shape of a preset target object to obtain at least one point cloud set. The preset target object can be a person, a tree, the ground, a wall, etc. Taking a person, a tree, and the ground as examples, three point cloud sets can be obtained, namely the first point cloud set, the second point cloud set, and the third point cloud set.

[0068] Step 102: Based on inertial data, determine the motion change information of the operating machinery.

[0069] The inertial data includes angular acceleration and linear acceleration, while the motion change information includes the direction of motion and velocity.

[0070] Specifically, the IMU sensor 203 includes three sets of gyroscopes and accelerometers, which measure the angular acceleration and linear acceleration of the three degrees of freedom, respectively. By pre-integrating the angular acceleration and linear acceleration, the motion direction and motion speed are obtained.

[0071] Step 103: Use motion change information to correct the current point cloud data to obtain the target point cloud data.

[0072] Specifically, based on motion change information and the positions of the mechanical lidar 201 and the solid-state lidar 202, the current point cloud data is corrected to obtain the target point cloud data. For example, based on motion change information, the point cloud data of the first point cloud set, the second point cloud set, and the third point cloud set are corrected respectively to obtain the target point cloud data.

[0073] Step 104: Extract point cloud features from the target point cloud data.

[0074] In one specific embodiment, the specific implementation of extracting point cloud features is as follows: Figure 3 As shown:

[0075] Step 301: Calculate the curvature corresponding to each target point in the target point cloud data.

[0076] Step 302: The target point corresponding to the curvature being less than the first preset curvature is taken as the planar feature point.

[0077] Step 303: The target point corresponding to the curvature being greater than the second preset curvature is taken as the corner feature point.

[0078] Step 304: Use planar feature points and corner feature points as point cloud features.

[0079] The first preset curvature is less than or equal to the second preset curvature.

[0080] In one specific embodiment, the point cloud features of the target point cloud data can be extracted separately for each point cloud set. For example, the first point cloud features of the first point cloud set can be extracted, the second point cloud features of the second point cloud set can be extracted, and the third point cloud features of the third point cloud set can be extracted.

[0081] Of course, after acquiring the point cloud data, the current point cloud data can be corrected first, and then the corrected point cloud data can be classified to obtain at least one point cloud set. This invention does not strictly limit whether a correction or classification operation is performed after acquiring the point cloud data.

[0082] Step 105: Based on the point cloud features, perform point cloud registration operation between the current frame point cloud data and the previous frame point cloud data.

[0083] In one specific embodiment, point cloud registration is performed between the current point cloud set and the previous point cloud set based on the point cloud features of each point cloud set. For example, point cloud registration is performed between the first point cloud feature and the point cloud features of the previous frame of point cloud data, point cloud registration is performed between the second point cloud feature and the point cloud features of the previous frame of point cloud data, and point cloud registration is performed between the third point cloud feature and the point cloud features of the previous frame of point cloud data.

[0084] This invention improves the speed of point cloud registration by classifying point cloud data and performing point cloud registration operations on each point cloud data set separately.

[0085] Step 106: Based on the result of the point cloud registration operation, determine the relative motion information between the current frame point cloud data and the previous frame point cloud data.

[0086] Specifically, based on the result of the point cloud registration operation, the relative motion vector of the current frame point cloud data relative to the previous frame point cloud data is obtained. This relative motion vector can be represented by (R, T), where R is the three-dimensional rotation matrix and T is the three-dimensional spatial translation vector.

[0087] In one specific embodiment, motion constraints are established in the backend to improve the accuracy of point cloud data registration and the precision of the 3D map. These motion constraints include a first motion constraint, a second motion constraint, and a third motion constraint. Specifically, the first motion constraint is established based on the continuity of point cloud data in adjacent frames; the second motion constraint is established based on motion change information; and the point cloud data of each frame is matched with preset point cloud data, and the third motion constraint is established based on the matching result. The third motion constraint is established only if the matching result is successful; otherwise, it is not established.

[0088] The preset point cloud data is a frame of point cloud data selected from the target point cloud data. The selection principle for this preset point cloud data is based on the point cloud features of each frame of the target point cloud data. Specifically, any frame of point cloud data corresponding to an index value indicating the richness of point cloud features greater than a preset index value can be used as the preset point cloud data.

[0089] Step 107: Based on the relative motion information, complete the construction of the 3D map.

[0090] In one specific embodiment, the current frame point cloud data is adjusted based on relative motion information; the adjusted current frame point cloud data is then superimposed on the previous frame point cloud data to complete the construction of the 3D map. Specifically, the current frame point cloud data is adjusted based on the relative motion vector (R, T); the adjusted current frame point cloud data is then superimposed on the previous frame point cloud data, and this process is repeated cyclically to obtain a complete 3D map.

[0091] Specifically, after the 3D map is constructed, it is adjusted using the first, second, and third motion constraints. This invention utilizes a tightly coupled SLAM algorithm based on LiDAR and IMU devices to automatically measure the progress of excavator earthmoving operations. Compared to conventional scanning methods, this improves detection accuracy and frequency, increases the efficiency of the machinery, and reduces the construction cycle of earthmoving operations.

[0092] Below, through Figure 4 The present invention will be described in detail, wherein, Figure 4 The steps described are not strictly in the correct order; this is just an example of one implementation method for clarity.

[0093] Step 401: Use lidar to acquire point cloud data.

[0094] Step 402: Acquire inertial data using an IMU sensor.

[0095] Step 403: Classify and process the point cloud data.

[0096] Step 404: Perform pre-integration processing on the inertial data to obtain motion change information.

[0097] Step 405: Correct the point cloud data using motion change information to obtain the target point cloud data.

[0098] Step 406: Extract the point cloud features of the target point cloud data and perform point cloud registration.

[0099] Step 407: Based on the result of the point cloud registration operation, perform map registration to obtain a local map.

[0100] Step 408: Based on the tightly coupled algorithm and motion constraints, a complete 3D map is obtained.

[0101] The three-dimensional map construction method, system, operating machinery, equipment, and medium provided in this invention acquire current point cloud data and inertial data. The current point cloud data is obtained through a scanning unit mounted on the machinery, and the inertial data is obtained through an inertial measurement unit mounted on the operating machinery. Therefore, by mounting the scanning unit and inertial measurement unit on the operating machinery, this invention can acquire point cloud data and inertial data corresponding to the operating machinery in real time as the machinery moves. This makes the invention applicable to any range of operating scenarios and effectively solves the problem of blind spots in existing technologies where the scanning device is fixed to the outer perimeter of the construction area.

[0102] Furthermore, based on inertial data, the motion change information of the operating machinery is determined; using this motion change information, the current point cloud data is corrected to obtain the target point cloud data. It is evident that this invention corrects the point cloud data using inertial data, resulting in higher accuracy when constructing a 3D map using the point cloud data later. Further, this invention extracts the point cloud features of the target point cloud data; based on these features, point cloud registration is performed between the current frame and the previous frame; based on the result of the point cloud registration, the relative motion information between the current and previous frame is determined; based on this relative motion information, the 3D map is constructed. It is evident that this invention uses a cyclical approach to match the current and previous frame point cloud data to complete the 3D map construction. Since the amount of data in a single frame is smaller than that in multiple frames, the data processing speed is faster. Therefore, it can effectively improve the construction speed of the 3D map, ultimately achieving rapid and accurate 3D construction of the construction scene.

[0103] The following describes the three-dimensional map construction system provided in the embodiments of the present invention. The three-dimensional map construction system described below can be referred to in correspondence with the three-dimensional map construction method described above. Specifically, as follows... Figure 5 As shown:

[0104] The three-dimensional map building system of the present invention includes a working machine body 501, a control unit 502 installed on the working machine body 501, and a scanning unit 503 and an inertial measurement unit 504 installed on the working machine body 501. The scanning unit 503 and the inertial measurement unit 504 communicate with the control unit 502 respectively.

[0105] The scanning unit 503 is used to scan and obtain the current point cloud data, and send the current point cloud data to the control unit 502;

[0106] The inertial measurement unit 504 is used to scan and obtain inertial data and send the inertial data to the control unit 502;

[0107] The control unit 502 is used to acquire current point cloud data and inertial data; determine the motion change information of the operating machinery based on the inertial data; use the motion change information to correct the current point cloud data to obtain target point cloud data; extract point cloud features from the target point cloud data; perform point cloud registration operation between the current frame point cloud data and the previous frame point cloud data based on the point cloud features; determine the relative motion information between the current frame point cloud data and the previous frame point cloud data based on the result of the point cloud registration operation; and complete the construction of a three-dimensional map based on the relative motion information.

[0108] Specifically, the control unit 502 includes: an acquisition subunit, a first determination subunit, a correction subunit, an extraction subunit, a registration subunit, a second determination subunit, and a construction subunit.

[0109] In one specific embodiment, an extraction sub-unit is used to calculate the curvature corresponding to each target point in the target point cloud data; the target point corresponding to the curvature being less than a first preset curvature is taken as a planar feature point; the target point corresponding to the curvature being greater than a second preset curvature is taken as a corner feature point; the planar feature point and the corner feature point are taken as the point cloud feature; wherein, the first preset curvature is less than or equal to the second preset curvature.

[0110] In one specific embodiment, a sub-unit is constructed to adjust the current frame point cloud data based on the relative motion information; the adjusted current frame point cloud data is then superimposed on the previous frame point cloud data to complete the construction of the three-dimensional map.

[0111] In one specific embodiment, the construction subunit is also used to establish a first motion constraint based on the continuity of point cloud data in adjacent frames; establish a second motion constraint based on motion change information; match each frame of point cloud data with preset point cloud data, and establish a third motion constraint based on the matching result; and adjust the three-dimensional map using the first motion constraint, the second motion constraint, and the third motion constraint.

[0112] In one specific embodiment, the acquisition subunit is further configured to classify the point cloud data to obtain at least one point cloud set; extract the point cloud features of the target point cloud data, including: extracting the point cloud features of each point cloud set respectively; and the registration subunit is configured to perform point cloud registration operation between the current point cloud set and the previous point cloud set based on the point cloud features of each point cloud set.

[0113] In one specific embodiment, the scanning unit 503 includes: a first scanning component and a second scanning component. The first scanning component is used to scan and obtain first point cloud data and send the first point cloud data to the control unit 502. The second scanning component is used to scan and obtain second point cloud data and send the second point cloud data to the control unit 502. The control unit 502 is also used to acquire the first point cloud data and the second point cloud data; and to fuse the first point cloud data and the second point cloud data based on preset calibration parameters to obtain the current point cloud data.

[0114] This invention also provides a working machine, including the three-dimensional map building system provided in the above embodiments.

[0115] In one specific embodiment, the operating machinery includes excavators and loaders, as detailed in [reference needed]. Figure 2 .

[0116] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include a processor 601, a communication interface 602, a memory 603, and a communication bus 604. The processor 601, communication interface 602, and memory 603 communicate with each other via the communication bus 604. The processor 601 can call logical instructions in the memory 603 to execute a 3D map construction method. This method includes: acquiring current point cloud data and inertial data, wherein the current point cloud data is obtained through a scanning unit installed on the machinery, and the inertial data is obtained through an inertial measurement unit installed on the working machinery; determining the motion change information of the working machinery based on the inertial data; using the motion change information to correct the current point cloud data to obtain target point cloud data; extracting point cloud features from the target point cloud data; performing point cloud registration operations between the current frame point cloud data and the previous frame point cloud data based on the point cloud features; determining the relative motion information between the current frame point cloud data and the previous frame point cloud data based on the result of the point cloud registration operation; and completing the construction of a 3D map based on the relative motion information.

[0117] Furthermore, the logical instructions in the aforementioned memory 603 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer can execute the three-dimensional map construction method provided by the above methods. The method includes: acquiring current point cloud data and inertial data, wherein the current point cloud data is obtained by a scanning unit installed on a machine, and the inertial data is obtained by an inertial measurement unit installed on the machine; determining the motion change information of the machine based on the inertial data; correcting the current point cloud data using the motion change information to obtain target point cloud data; extracting point cloud features from the target point cloud data; performing point cloud registration operation between the current frame point cloud data and the previous frame point cloud data based on the point cloud features; determining the relative motion information between the current frame point cloud data and the previous frame point cloud data based on the result of the point cloud registration operation; and completing the construction of a three-dimensional map based on the relative motion information.

[0119] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program is implemented to perform the aforementioned three-dimensional map construction methods. The method includes: acquiring current point cloud data and inertial data, wherein the current point cloud data is obtained through a scanning unit mounted on a machine, and the inertial data is obtained through an inertial measurement unit mounted on the machine; determining motion change information of the machine based on the inertial data; correcting the current point cloud data using the motion change information to obtain target point cloud data; extracting point cloud features from the target point cloud data; performing point cloud registration operation between the current frame point cloud data and the previous frame point cloud data based on the point cloud features; determining the relative motion information between the current frame point cloud data and the previous frame point cloud data based on the result of the point cloud registration operation; and completing the construction of a three-dimensional map based on the relative motion information.

[0120] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0121] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a three-dimensional map, characterized in that, include: The system acquires current point cloud data and inertial data. The current point cloud data is obtained through a scanning unit installed on the operating machinery, and the inertial data is obtained through an inertial measurement unit installed on the operating machinery. The scanning unit includes a first scanning component and a second scanning component. The first scanning component is used to scan the motion information of the operating machinery and is a mechanical lidar. The second scanning component is used to scan the operating area of ​​the operating machinery and is a solid-state lidar. The inertial data includes angular acceleration and linear acceleration. By pre-integrating the angular acceleration and the linear acceleration, the motion change information of the working machinery is obtained, including the motion direction and the motion speed. Using the motion change information, the current point cloud data is corrected to obtain the target point cloud data; Extract the point cloud features from the target point cloud data; Based on the point cloud features, perform point cloud registration operation between the current frame point cloud data and the previous frame point cloud data. Based on the result of the point cloud registration operation, the relative motion information between the current frame point cloud data and the previous frame point cloud data is determined; Based on the relative motion information, the three-dimensional map is constructed. The acquisition of current point cloud data includes: Acquire first point cloud data and second point cloud data, wherein the first point cloud data is obtained through the first scanning component of the scanning unit and the second point cloud data is obtained through the second scanning component of the scanning unit; Based on preset calibration parameters, the first point cloud data and the second point cloud data are fused to obtain the current point cloud data.

2. The three-dimensional map construction method according to claim 1, characterized in that, The extraction of point cloud features from the target point cloud data includes: Calculate the curvature corresponding to each target point in the target point cloud data; The target point corresponding to the curvature being less than the first preset curvature is taken as the planar feature point; The target point corresponding to the curvature being greater than the second preset curvature is used as the corner feature point; The planar feature points and the corner feature points are used as the point cloud features; Wherein, the first preset curvature is less than or equal to the second preset curvature.

3. The three-dimensional map construction method according to claim 1, characterized in that, The construction of the 3D map based on the relative motion information includes: Based on the relative motion information, adjust the current frame point cloud data; The adjusted current frame point cloud data is overlaid onto the previous frame point cloud data to complete the construction of the 3D map.

4. The three-dimensional map construction method according to claim 3, characterized in that, Before constructing the 3D map based on the relative motion information, the process further includes: Based on the continuity of point cloud data in adjacent frames, a first motion constraint is established; Based on the aforementioned motion change information, a second motion constraint is established; Each frame of point cloud data is matched with preset point cloud data, and a third motion constraint is established based on the matching results. After constructing the 3D map based on the relative motion information, the process further includes: The three-dimensional map is adjusted using the first motion constraint, the second motion constraint, and the third motion constraint.

5. The three-dimensional map construction method according to any one of claims 1-4, characterized in that, After acquiring the current point cloud data and inertial data, the process further includes: The current point cloud data is classified to obtain at least one point cloud set; The extraction of point cloud features from the target point cloud data includes: Extract the point cloud features for each of the point cloud sets respectively; The point cloud registration operation based on the point cloud features, performing the current frame point cloud data and the previous frame point cloud data, includes: Based on the point cloud features of each point cloud set, point cloud registration operation is performed between the current point cloud set and the previous point cloud set.

6. A three-dimensional map construction system, characterized in that, include: The machine body, the control unit mounted on the machine body, and the scanning unit and the inertial measurement unit mounted on the machine body, wherein the scanning unit and the inertial measurement unit communicate with the control unit respectively, and the scanning unit includes a first scanning component and a second scanning component. The first scanning component is used to scan the motion information of the machine and is a mechanical lidar. The second scanning component is used to scan the working area of ​​the machine and is a solid-state lidar. The scanning unit is used to scan and obtain the current point cloud data, and send the current point cloud data to the control unit; The inertial measurement unit is used to scan and obtain inertial data, and send the inertial data to the control unit. The inertial data includes angular acceleration and linear acceleration. The control unit is used to acquire the current point cloud data and the inertial data; to obtain the motion change information of the working machinery by pre-integrating the angular acceleration and the linear acceleration, the motion change information including the motion direction and the motion speed; to correct the current point cloud data using the motion change information to obtain target point cloud data; and to extract the point cloud features of the target point cloud data. Based on the point cloud features, a point cloud registration operation is performed between the current frame point cloud data and the previous frame point cloud data; based on the result of the point cloud registration operation, the relative motion information between the current frame point cloud data and the previous frame point cloud data is determined; based on the relative motion information, the construction of the three-dimensional map is completed. The control unit is also used to acquire first point cloud data and second point cloud data; and to fuse the first point cloud data and the second point cloud data based on preset calibration parameters to obtain the current point cloud data.

7. A type of operating machinery, characterized in that, Includes the three-dimensional map construction system as described in claim 6.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the three-dimensional map construction method as described in any one of claims 1 to 5.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the three-dimensional map construction method as described in any one of claims 1 to 5.

Images