Automatic mounting method and system for cambered surface wood board based on dynamic point cloud analysis
By using dynamic point cloud analysis, the adaptability and accuracy issues of static fitting in the installation of curved wooden planks were solved, enabling efficient and accurate installation process control and quality assessment, and improving the accuracy and engineering efficiency of curved wooden plank installation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-03-27
AI Technical Summary
Existing technologies for installing wooden boards on curved structures suffer from poor static fitting adaptability, low accuracy in detecting minute features, and a lack of dynamic closed-loop control, resulting in high rework rates and a lack of process visualization monitoring and quality assessment.
A method based on dynamic point cloud analysis is adopted. By establishing a joint coordinate system of the robotic arm and the point cloud acquisition device, the initial center and radius are obtained by using a one-sided point cloud fitting algorithm. Z-value mutation feature detection is performed to identify the edge position of the installed components. The board pose parameters are dynamically compensated according to the deviation, the number of installations is adaptively adjusted, and real-time visual monitoring and quality assessment are provided.
It achieves high-precision identification of minute mutations, dynamically compensates for errors, improves installation accuracy and efficiency, provides full-process visual monitoring and quality assessment, and reduces installation rework rate.
Smart Images
Figure CN121733207A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of three-dimensional point cloud processing and engineering application, and particularly relates to an arc surface wood board automatic installation method and system based on dynamic point cloud analysis. BACKGROUND
[0002] In the fields of industrial manufacturing, building installation and automatic assembly, the modeling and installation planning of complex arc surface structures are key links. With the development of three-dimensional scanning technology, point cloud data provides a basis for engineering application. However, there are still significant limitations in existing technical means when installing components such as wood boards on arc surface structures (such as a disc).
[0003] Early technologies mainly rely on manual measurement using traditional tools such as tape measures and angle gauges, which have the problems of low efficiency, poor accuracy and difficulty in handling complex arc surfaces. Subsequent traditional digitization stages introduce laser scanning and CAD assistance to generate parameters through static point cloud fitting. However, such static fitting methods have the following serious defects: first, they cannot automatically identify and exclude abnormal mutation areas, such as small height differences at wood board joints, which interfere with fitting accuracy; second, they lack real-time adjustment capabilities and cannot adapt to cumulative errors or local deformation of the base during installation.
[0004] In existing technologies, although there are feature detection based on curvature changes or vision-based positioning systems, they are usually only suitable for obvious geometric edges and are not sensitive to small mutations in the Z-axis direction, or they rely on pre-set templates and lack flexibility. In actual engineering, this leads to high installation rework rates, lack of process visualization monitoring, and inability to automatically evaluate installation quality. Therefore, there is an urgent need for an automated installation solution that can dynamically adapt to environmental changes, accurately identify small mutations, and compensate for errors in real time. SUMMARY
[0005] The purpose of the present application is to provide an arc surface wood board automatic installation method and system based on dynamic point cloud analysis to solve the problems of poor adaptability of static fitting, low precision of small feature detection and lack of dynamic closed-loop control in existing technologies.
[0006] To achieve the above purpose, a specific embodiment of the present application provides an arc surface wood board automatic installation method based on dynamic point cloud analysis, which comprises: establishing a joint coordinate system of a mechanical arm and a point cloud acquisition device, and obtaining a calibration matrix through a specific motion trajectory of the mechanical arm; performing initial scanning on an arc surface base without installed wood boards, and performing single-sided circular arc fitting using a single-sided point cloud fitting algorithm to obtain initial center coordinates and a radius; entering a dynamic installation loop, collecting local point cloud data of the current installation area, and performing Z value mutation feature detection to identify the edge position of the installed components; Calculate the actual installation deviation based on the edge position, and dynamically compensate the pose parameters of the wood board to be installed according to the deviation; Adaptively adjust the installation quantity of the remaining wood boards according to the real-time updated arc surface radius, and generate control instructions of the wood boards to be installed in batches and send the control instructions to the mechanical arm to execute the installation until all the laying is completed.
[0007] In one or more embodiments of the present application, the calibration matrix is obtained through a specific motion trajectory of the mechanical arm, specifically comprising: Control the mechanical arm clamp to execute a plurality of groups of motions in a posture of keeping the Y-axis coordinate fixed and only changing the X-axis and Z-axis positions and the Rx elevation angle; Synchronously collect position data of the mechanical arm end in the world coordinate system and midpoint coordinates and normal vectors of the corresponding feature planes in the point cloud coordinate system; Use the separated least square method to construct an affine transformation equation to solve the conversion matrix and the translation vector of the point cloud coordinate system to the mechanical arm world coordinate system.
[0008] In one or more embodiments of the present application, the single-sided point cloud fitting algorithm is specifically used, comprising: According to a preset width threshold, intercept the point cloud data of the edge area on the left or right side of the arc surface base; Perform Gaussian filtering and noise reduction processing on the intercepted point cloud data; Preliminarily fit the data after noise reduction using the least square method, and on this basis, use the least median square algorithm to suppress abnormal value interference by minimizing the median of the square of the distance of the data points to the circle to output the final center coordinates and the radius.
[0009] In one or more embodiments of the present application, the Z value mutation feature detection is specifically executed, comprising: Calculate the difference value of adjacent points in the local point cloud data in the Z-axis direction , wherein and are the Z-axis coordinate values of the two adjacent sampling points, respectively; Mark the points with greater than a preset mutation threshold as candidate mutation points, and the mutation threshold is set to ±0.5mm.
[0010] In one or more embodiments of the present application, the edge position of the installed component is identified, further comprising density clustering analysis on the candidate mutation points: Group the candidate mutation points using the DBSCAN clustering algorithm, and set the neighborhood radius and the minimum number of neighborhood points; Remove the discrete points marked as noise, and extract the effective mutation point cluster; calculating the geometric center of the effective mutation point cluster, and taking the coordinate of the geometric center as the actual physical edge position of the currently installed board.
[0011] In one or more embodiments of the present application, the pose parameters of the to-be-installed board are dynamically compensated according to the deviation, specifically, a deviation value between the actual installation position and the theoretical installation position is calculated, and the installation angle is compensated according to the deviation value, and the calculation formula of the deviation value comprises: ; wherein ， is the actual coordinate / theoretical coordinate of x, x is the actual coordinate / theoretical coordinate of z; The compensation formula is:
[0012] wherein, is the corrected installation angle, is the original theoretical angle, is the current fitting radius.
[0013] In one or more embodiments of the present application, the control instructions of the to-be-installed board are batch generated, specifically: taking the currently confirmed actual physical edge position as a reference, the central angle, normal vector and spatial coordinates of the subsequent N boards are calculated at one time, wherein N≥3; the parameters of the N boards are converted into mechanical arm control instructions and stored in an execution queue, and the next round of point cloud data is processed in parallel during the execution of the mechanical arm.
[0014] In one or more embodiments of the present application, the installation number of the remaining boards is adaptively adjusted, comprising: according to the latest fitted real-time radius R' and the standard width of the board w , the total number of boards is updated according to the following formula wherein, , the distribution gap of the remaining boards is re-planned according to the updated total number N' .
[0015] In one or more embodiments of the present application, the method further comprises a visual interactive step: a point cloud model of the disc and the board is drawn in real time by using a three-dimensional rendering engine; a heat map is generated on the visual interface according to the mutation feature detection result, and the confidence of the mutation point is displayed by using different color scales; a quality evaluation report containing fitting residual, maximum gap value and installation deviation is automatically generated.
[0016] In another aspect of the present invention, an automated installation system for curved wooden boards based on dynamic point cloud analysis is provided, for implementing an automated installation method for curved wooden boards based on dynamic point cloud analysis, the system comprising: The data acquisition module includes a laser scanning device for acquiring high-frequency point cloud data of curved surfaces and wooden boards; The execution module, which includes a six-degree-of-freedom robotic arm and an end effector, is used to grip the wooden board and perform installation actions; The central processing module is communicatively connected to the data acquisition module and the execution module. The central processing module is configured to execute computer instructions including point cloud calibration, abrupt change feature detection, dynamic error compensation, and adaptive path planning.
[0017] Compared with the prior art, the automated installation method and system for curved wooden boards based on dynamic point cloud analysis of the present invention has the following significant advancements: This invention uses Z-value mutation detection and DBSCAN clustering to accurately identify minute features such as wood panel seams, and eliminates accumulated errors through dynamic compensation.
[0018] This invention utilizes the LMedS algorithm and a one-sided fitting strategy to effectively reduce the impact of noise and substrate deformation on parameter calculation.
[0019] This invention enables batch generation of control commands and adaptive quantity adjustment, significantly reducing the waiting time for calculation and execution.
[0020] This invention provides real-time visual monitoring and quality quantification assessment of the entire process, which facilitates project management and fault diagnosis. Attached Figure Description
[0021] 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 only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0022] Figure 1 This is a flowchart of an automated installation method for curved wooden boards based on dynamic point cloud analysis according to one embodiment of the present invention; Figure 2 This is an algorithm flowchart of an automated installation method for curved wooden boards based on dynamic point cloud analysis according to one embodiment of the present invention; Figure 3 This is a schematic diagram of the calibration motion path of the robotic arm in one embodiment of the present invention; Figure 4This is a schematic diagram of a single-sided fitting sampling point of a disk in one embodiment of the present invention; Figure 5 This is a visual schematic diagram illustrating the ideal and complete laying of a circular dot cloud and a wooden board in one embodiment of the present invention. Figure 6 This is a schematic diagram of the process of laying a circular dot cloud and a wooden board in one embodiment of the present invention; Figure 7 This is a graph showing the detection of Z-value abrupt change points on one side of a disc with a wooden board in one embodiment of the present invention. Figure 8 This is a schematic diagram (a) of the circular cyclic laying result in one embodiment of the present invention; Figure 9 This is a schematic diagram (II) of the circular cyclic laying result in one embodiment of the present invention; Figure 10 This is a schematic diagram of a calibration matrix according to one embodiment of the present invention; Figure 11 This is a schematic diagram of a coordinate report according to one embodiment of the present invention. Detailed Implementation
[0023] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.
[0024] As described in the background section, existing technologies suffer from poor static fitting adaptability, low accuracy in detecting minute features, and a lack of dynamic closed-loop control.
[0025] In response to the above technical problems, such as Figures 1-2 As shown, an automated installation method for curved wooden boards based on dynamic point cloud analysis is described, the method comprising the following steps: Step 1: Establish a joint coordinate system between the robotic arm and the point cloud acquisition device, and obtain the calibration matrix through the specific motion trajectory of the robotic arm; Step 2: Perform an initial scan on the curved base without the wooden board installed, and use a one-sided point cloud fitting algorithm to fit a one-sided circular arc to obtain the initial center coordinates and radius; Step 3: Enter the dynamic installation loop, collect local point cloud data of the current installation area, and perform Z-value abrupt change feature detection to identify the edge position of the installed components; Step 4: Calculate the actual installation deviation based on the edge position, and dynamically compensate the pose parameters of the wooden board to be installed according to the deviation; Step 5: Adaptively adjust the number of remaining wooden planks to be installed based on the real-time updated radius of the arc surface, and generate control commands for the wooden planks to be installed in batches, sending them to the robotic arm to execute the installation until all the planks are laid. In a further embodiment, such as Figure 3 As shown, to achieve precise alignment between the point cloud coordinate system and the robot arm's world coordinate system, this embodiment employs a simplified calibration method. Considering that during the board mounting process, the robot arm mainly operates in... X shaft and Z Axis movement, Y The value is fixed, and only in the end pose Rx (Elevation angle) changes. Therefore, the robotic arm is controlled to perform more than 9 sets of standard posture movements while gripping the wooden board in the middle region of the curved surface. In each posture, the world coordinates of the robotic arm's end effector and the midpoint coordinates and normal vector of the wooden board plane in the point cloud data acquired by the 3D camera are recorded simultaneously. An affine transformation equation is constructed using the separated least squares method to solve for the transformation matrix and translation vector from the point cloud coordinate system to the robotic arm's base coordinate system. This method simplifies the calibration process, achieving a calibration accuracy of ±0.05mm.
[0026] The data acquisition process includes using a robotic arm to grip a wooden board and perform standard posture movements, recording the data synchronously for each posture. The calibration algorithm is simplified based on the robotic arm's motion characteristics, because during the board loading process, only the position parameters are recorded. X and Z The axis is moving. Y The value is fixed, and the Euler angles of the robotic arm are... Rz and Ry Fixed, only Rx The angle of the control fixture changes, so it is only necessary to collect more than 9 sets of parameters as the robotic arm moves in the middle of the curved surface.
[0027] The calibration algorithm includes solving for the calibration matrix based on the coordinates of the midpoint of the input point cloud plank plane and the normal vector, along with the robot arm pose data, using a separate least squares fitting method. In this embodiment, the calibration matrix is: ; in, The coordinates of the midpoint of the wooden board in the point cloud (sensor coordinate system). This is the point cloud normal vector (describing the orientation of the wooden board). Let be the position of the robotic arm's end effector in the world coordinate system, and let be the Euler angles of the robotic arm (only). Rx (elevation angle change) Point cloud coordinates / robotic arm position, It represents the normal vector / Euler angle.
[0028] Because of which Y With the values fixed, the calibration matrix can be simplified to: ,in, It is a 2×2 affine transformation matrix. For example, a translation vector, such as Figure 10 As shown, the output calibration matrix is the input point cloud coordinates and world coordinates.
[0029] In a further embodiment, such as Figures 4-6 As shown, before installation begins, a laser radar is used to scan the bare disk without the wooden board. To avoid data redundancy and two-sided fitting errors caused by full-circle scanning, this embodiment adopts a one-sided point cloud fitting strategy.
[0030] Specifically, point clouds of the left or right edge region of the disk are cropped according to a preset width threshold, and point clouds of Y∈[y_min, y_min+w] or Y∈[y_max-w, y_max] are cropped according to the IsLeft parameter, where the width... w The default value is 20mm. Then, Gaussian filtering is applied to the data for noise reduction. Next, a first-level coarse fit is performed using least squares, followed by a second-level optimization fit using the Weighted Least Median Squares (LMedS) algorithm. The LMedS algorithm, by minimizing the median of the squared distances from data points to the circle, can tolerate up to 50% outliers, thus outputting high-precision initial circle center coordinates. and radius R .
[0031] In a further embodiment, during installation, the system enters a dynamic loop mode, including Z-value mutation detection, cluster analysis, deviation calculation and compensation, and adaptive quantity adjustment.
[0032] like Figure 7 As shown, the Z-value mutation detection includes acquiring the local point cloud of the currently installed area and calculating the difference between adjacent points in the Z-axis direction. : ,in, and These are the Z-axis coordinates of two adjacent sampling points; Points exceeding a preset mutation threshold are marked as candidate mutation points, and the mutation threshold is set to ±0.5 mm.
[0033] In a further embodiment, density clustering analysis is performed on the candidate mutation points: the candidate mutation points are format-converted into an N×2 array (each row has [X, Z] coordinates), the DBSCAN clustering algorithm is applied (with neighborhood radius eps=2.0mm and minimum number of points min_samples=3), noise points are removed, effective mutation point clusters are extracted, the geometric center of the effective mutation point cluster is calculated, and the coordinates of the geometric center are used as the actual physical edge position of the currently installed wooden board.
[0034] In a further embodiment, deviation calculation and compensation includes calculating the deviation between the actual installation position and the theoretical installation position, and compensating for the installation angle based on the deviation. The formula for calculating the deviation includes: ; in ， for x Actual coordinates / theoretical coordinates The actual coordinates / theoretical coordinates of z; The compensation formula is as follows: ：
[0035] in, The corrected mounting angle, For the original theoretical angle, This is the current fitting radius; used to correct the installation angle of subsequent wooden boards.
[0036] In a further embodiment, the value is determined based on the latest fitted real-time radius R' and the standard width of the wooden board. w Update the total number of planks using the following formula. ,in, Recalculate the total number of remaining planks required. This helps to avoid gaps that are too large or too small at the end of the installation process.
[0037] To improve efficiency, the system calculates and outputs the pose parameters (central angle, normal vector, and spatial coordinates) of the next three wooden boards in each loop. These parameters are converted into control commands for the robotic arm and sent to the execution queue. While the robotic arm performs the installation action, the system processes the point cloud data for the next round in parallel, realizing a streamlined operation.
[0038] Simultaneously, a 3D rendering engine is used to draw point cloud models of the disk and wooden planks in real time; based on the mutation feature detection results, a heatmap is generated in the visualization interface, and different color levels are used to display the confidence level of mutation points; a quality assessment report containing fitting residuals, maximum gap values, and installation deviations is automatically generated, such as... Figures 8-9 The installation process utilizes the Open3D engine to render the installation effect in real time, and displays the confidence level of mutation points in the form of a heatmap on a visualization interface, automatically generating data such as... Figure 11 The quality assessment report shown includes key indicators such as fitting residuals (typical value 0.28 mm) and maximum gap (1.5 mm).
[0039] In a further embodiment, the present invention also includes a system for implementing an automated installation method for curved wooden boards based on dynamic point cloud analysis, the system comprising: The data acquisition module includes a laser scanning device for acquiring high-frequency point cloud data of curved surfaces and wooden boards; The execution module, which includes a six-degree-of-freedom robotic arm and an end effector, is used to grip the wooden board and perform installation actions; The central processing module is communicatively connected to the data acquisition module and the execution module. The central processing module is configured to execute computer instructions including point cloud calibration, abrupt change feature detection, dynamic error compensation, and adaptive path planning.
[0040] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0041] This invention solves the problems of traditional static fitting being unable to identify seam abrupt changes, being sensitive to local deformation, and lacking real-time correction capabilities by using an adaptive fitting and dynamic compensation mechanism driven by mutation features. It significantly improves the installation accuracy and engineering efficiency of curved surface packaging and provides a visualized quality assessment method.
[0042] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.
[0043] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0044] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0045] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.
[0046] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.
Claims
1. An automated installation method for curved wooden boards based on dynamic point cloud analysis, characterized in that, The method includes: Establish a joint coordinate system between the robotic arm and the point cloud acquisition device, and obtain the calibration matrix through the specific motion trajectory of the robotic arm; An initial scan is performed on the curved base without the wooden board installed, and a one-sided point cloud fitting algorithm is used to fit the one-sided circular arc to obtain the initial center coordinates and radius; Enter the dynamic installation loop, collect local point cloud data of the current installation area, and perform Z-value abrupt feature detection to identify the edge position of the installed components; The actual installation deviation is calculated based on the edge position, and the pose parameters of the wooden board to be installed are dynamically compensated according to the deviation. The remaining number of wooden planks to be installed is adaptively adjusted based on the real-time updated radius of the arc surface, and control instructions for the wooden planks to be installed are generated in batches and sent to the robotic arm to execute the installation until all the planks are laid.
2. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The process of obtaining the calibration matrix through a specific motion trajectory of the robotic arm specifically includes: The robotic arm gripper is controlled to perform multiple sets of movements while keeping the Y-axis coordinate fixed and only changing the X-axis and Z-axis positions and the Rx elevation angle. Simultaneously acquire the position data of the robotic arm end effector in the world coordinate system, as well as the midpoint coordinates and normal vector of the corresponding feature plane in the point cloud coordinate system; The affine transformation equations are constructed using the separable least squares method, and the transformation matrix and translation vector from the point cloud coordinate system to the robot arm's world coordinate system are solved.
3. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The single-sided point cloud fitting algorithm specifically includes: Based on a preset width threshold, extract point cloud data from the left or right edge region of the curved base. Gaussian filtering is applied to the captured point cloud data for noise reduction. The least squares method is used to initially fit the denoised data. Based on this, the least median square algorithm is used to suppress outlier interference by minimizing the median of the squared distances from the data points to the circle and output the final center coordinates and radius of the circle.
4. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The Z-value mutation feature detection process specifically includes: Calculate the difference between adjacent points in a local point cloud data along the Z-axis. : ,in, and These are two adjacent sampling points. Z Axis coordinate values; Will Points exceeding a preset mutation threshold are marked as candidate mutation points, and the mutation threshold is set to ±0.5 mm.
5. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 4, characterized in that, The process of identifying the edge locations of installed components also includes density clustering analysis of candidate mutation points: The DBSCAN clustering algorithm is used to group candidate mutation points, and the neighborhood radius and minimum number of neighborhood points are set. Remove discrete points marked as noise and extract clusters of effective mutation points; Calculate the geometric center of the effective mutation point cluster, and use the coordinates of this geometric center as the actual physical edge position of the currently installed wooden board.
6. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The dynamic compensation of the pose parameters of the wooden board to be installed based on the deviation specifically involves calculating the deviation between the actual installation position and the theoretical installation position, and compensating for the installation angle based on the deviation. The formula for calculating the deviation is as follows: include: ； in ， for x Actual coordinates / theoretical coordinates The actual coordinates / theoretical coordinates of z; The compensation formula is as follows: ： in, The corrected mounting angle, For the original theoretical angle, This is the current fitted radius.
7. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The control instructions for batch generating the wooden boards to be installed are as follows: Based on the currently confirmed actual physical edge position, calculate the central angle, normal vector and spatial coordinates of the subsequent N wooden boards in one go, where N≥3; The parameters of the N wooden boards are converted into robotic arm control commands and stored in the execution queue. The next round of point cloud data is processed in parallel during the execution of the robotic arm.
8. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The adaptive adjustment of the number of remaining wooden boards to be installed includes: Based on the latest fitted real-time radius R' and the standard width of the wooden board. w Update the total number of planks using the following formula. ,in, Based on the updated total number N' The spacing between the remaining planks needs to be redesigned.
9. The automated installation method for curved wooden boards based on dynamic point cloud analysis according to claim 1, characterized in that, The method also includes a visual interaction step: The point cloud models of the disk and the wooden board are drawn in real time using a 3D rendering engine; Based on the mutation feature detection results, a heat map is generated in the visualization interface, and the confidence level of the mutation points is displayed using different color levels; Automatically generate a quality assessment report that includes fitting residuals, maximum gap values, and installation deviations.
10. An automated installation system for curved wooden boards based on dynamic point cloud analysis, used to implement the automated installation method for curved wooden boards based on dynamic point cloud analysis as described in any one of claims 1 to 9, characterized in that, The system includes: The data acquisition module includes a laser scanning device for acquiring high-frequency point cloud data of curved surfaces and wooden boards; The execution module, which includes a six-degree-of-freedom robotic arm and an end effector, is used to grip the wooden board and perform installation actions; The central processing module is communicatively connected to the data acquisition module and the execution module. The central processing module is configured to execute computer instructions including point cloud calibration, abrupt change feature detection, dynamic error compensation, and adaptive path planning.