Variable cross-section steel structure mounting and positioning method and system based on unmanned aerial vehicle laser point cloud scanning

Abstract

The invention discloses a variable cross-section steel structure mounting and positioning method and system based on unmanned aerial vehicle laser point cloud scanning. The method comprises the steps that a theoretical BIM model is established; point cloud data of the steel structure are obtained through laser scanning of the unmanned aerial vehicle; processing the point cloud to generate a live-action model; performing intelligent registration comparison on the real scene model and the BIM model, and calculating an installation deviation; outputting a visual result to guide adjustment; and rechecking until the product is qualified. The system comprises a data acquisition module, a data processing and control module and a feedback execution module. The data acquisition module comprises a GPS positioning device and an unmanned aerial vehicle carrying a laser scanning radar. And data processing and control: a mobile workstation or a server. And feeding back and executing a gantry crane or a crawler crane. According to the method, the problems of low efficiency, poor precision, blind area and information lag in the mounting process of the variable cross-section steel structure are solved, real-time, accurate and digital closed-loop quality control of the mounting process is realized, and the construction efficiency and safety are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of construction engineering construction measurement, in particular to a method and system for realizing high-precision and rapid installation positioning of variable cross-section steel structural members based on unmanned aerial vehicle (UAV) airborne laser radar (LiDAR) technology, three-dimensional point cloud processing and BIM model comparative analysis. BACKGROUND

[0002] Variable cross-section steel structures (such as shuttle-shaped columns, variable cross-section trusses, spatial curved surface nets, variable cross-section arch ribs, etc.) are increasingly widely used in modern large public buildings due to their beautiful lines, mechanical properties and appearance. However, the installation and positioning of such structures is a major technical challenge in the construction process: 1. Complex shape: the size and angle of the member continuously change with the position, and each cross-section is different, and the traditional total station and other measurement methods have very low lofting efficiency (about 2-3 hours for full-section measurement), and it is difficult to obtain the overall spatial attitude of the member, and can only be used for spatial cross-section review.

[0003] 2. Measurement blind area: total station will encounter measurement blind area in measurement, site member shielding, need to turn point, and high-altitude, special-shaped node measurement has many blind areas, manual measurement is high-risk and difficult.

[0004] 3. Cumulative error: In the process of segmented installation, small installation errors will accumulate, which may eventually lead to difficulty in closing or deviation of the internal force distribution of the structure from the design value, causing serious consequences.

[0005] 4. Information lag: the traditional "measurement-adjustment-remeasurement" process is long, which cannot provide data support for real-time adjustment and installation, and the construction efficiency is low.

[0006] The existing related patent technologies also have obvious defects: 1. Patent CN113722789B discloses a virtual assembly method for steel structure bridge based on 3D laser scanning, which uses Leica ScanStation P30 / P40 ground three-dimensional laser scanner to rely on fixed station for point cloud collection, which cannot cover the measurement blind area of high-altitude and special-shaped nodes, and this method focuses on the error analysis of virtual assembly, and does not involve real-time adjustment and dynamic closed-loop control in the installation process, the information feedback is lagging behind, and it is not suitable for the installation and positioning of variable cross-section steel structures; at the same time, its data processing is for the assembly error of conventional steel structures, and does not consider the cross-section change rule of variable cross-section members, and cannot accurately extract the key features such as axis and port center of variable cross-section members.

[0007] 2. Patent CN118587350A discloses a TLS-based assembled ring truss VTA automation method, which uses ground-based laser radar for point cloud scanning, and the scanning range is limited. It is difficult to measure and has low safety when working at high altitudes. This method is mainly used for quality inspection of virtual pre-assembly. The core is the extraction and deviation analysis of assembly points such as bolt holes and ear plate boundaries. It does not design an adaptive solution for the cross-section change characteristics of variable cross-section steel structures, cannot meet the high-precision positioning requirements of key parameters such as variable cross-section member axis and angle, and does not support real-time adjustment and re-inspection closed loop during installation, making it difficult to solve the cumulative error problem.

[0008] In summary, although there are cases of using three-dimensional laser scanning technology for completion detection, they are mainly used for post-acceptance, not for real-time installation and positioning during the process. How to integrate advanced scanning technology into the installation process itself to achieve dynamic and closed-loop precision control is a problem that needs to be solved in this field. SUMMARY

[0009] The purpose of the present application is to overcome the shortcomings of the prior art and provide a variable cross-section steel structure installation positioning method and system based on unmanned aerial vehicle laser point cloud scanning, aiming to solve the problems of low installation efficiency, poor precision, the need for repeated measurement and correction, high measurement and installation risk, lagging measurement data information, and long measurement period during the installation of variable cross-section steel structures.

[0010] To achieve the above-mentioned purpose, the present application adopts the following technical solutions: A variable cross-section steel structure installation positioning method based on unmanned aerial vehicle laser point cloud scanning, comprising the following steps: Step 1: creating a theoretical BIM model of the variable cross-section steel structure to be installed; Step 2: using an unmanned aerial vehicle onboard laser radar to scan the installed steel structure and obtain three-dimensional point cloud data; Step 3: preprocessing and fusing the point cloud data to generate a real scene point cloud model; Step 4: intelligently registering and comparing the real scene point cloud model with the theoretical BIM model to calculate the deviation between the current installation position and the theoretical design position; Step 5: outputting visual deviation results to guide installation adjustment; Step 6: based on the adjusted results, re-inspecting and repeating steps 2 to 6 until all deviations are controlled within the design allowable tolerance range, and completing the final fixation.

[0011] Further, in Step 1, the creation of the theoretical BIM model includes: establishing a high-precision three-dimensional BIM model of the variable cross-section steel structure to be installed according to design drawings, and clearly defining the three-dimensional design coordinates of key feature points and feature lines in the model.

[0012] Further, the key feature points include member port centers and interface flange center points, and the key feature lines include member axis.

[0013] Further, in step 2, the data acquisition module needs to be realized in the form of a UAV carrying a laser scanning radar, including hoisting the to-be-installed component to a temporary fixed state, operating the UAV carrying the laser radar (LiDAR) to fly around the current installation section and the adjacent installed structure for multi-view flight scanning, and automatically collecting high-density three-dimensional point cloud data.

[0014] Further, in step 3, the collected multi-station and multi-view point cloud data is preprocessed by denoising and filtering, the multi-station point cloud is accurately registered and fused by using the UAV POS system and the ground control points, and a complete high-precision on-site real scene point cloud model is generated.

[0015] Further, in step 4, the feature points and feature surfaces of the installed structure are automatically extracted from the real scene point cloud, the intelligent registration of the real scene features and the corresponding features of the theoretical BIM model is realized by using the iterative closest point (ICP) algorithm, the coordinate conversion parameters are calculated, the theoretical BIM model of the to-be-installed component is mapped to the real scene coordinate system through the conversion parameters, the accurate alignment of the theory and the reality is realized, and the deviation value of the actual position of the to-be-installed component and the theoretical design position is calculated.

[0016] Further, the deviation value includes an axis deviation, an angle deflection and an interface misalignment amount.

[0017] Further, in step 5, the deviation analysis result is visualized and output in the form of a chromatogram and a millimeter-level data report.

[0018] Further, in step 6, the construction personnel adjusts the component according to the visualized deviation result, the component is locally and quickly scanned by the UAV after adjustment, and new deviation reports are generated by repeating steps S3-S5.

[0019] A variable cross-section steel structure installation positioning system for realizing the above method comprises a data acquisition module composed of a UAV, a laser radar and a positioning system, the UAV carries the laser radar to realize point cloud data acquisition, the positioning system comprises a GPS+IMU combined positioning unit and a ground control point, and is used for providing high-precision positioning data; a data processing and control center which is a cloud server or a mobile workstation, is connected with the data acquisition module in a wireless communication mode, is used for point cloud data processing, BIM model management, data comparison and analysis and deviation result visualization display; and a feedback execution module comprising a jack, a hand-operated hoist and a fixing device, which is used for adjusting the position and attitude of the variable cross-section steel structure component according to the visualized deviation result, and finally fixing the component.

[0020] Compared with the prior art, the present application has the following remarkable beneficial effects: 1. Solve the problem of measurement blind area: adopt unmanned aerial vehicle to carry out air scanning with laser radar, break through the scanning limit of ground fixed station and ground-based radar, can flexibly cover traditional measurement blind area such as high altitude and special-shaped node, without manual high-altitude operation, reduce the risk of measurement, and the scanning coverage rate reaches 100%.

[0021] 2. Improve installation efficiency: the full-section scanning time of variable cross-section component is shortened to 10-20 minutes (compared with 2-3 hours of traditional total station), data processing and deviation output are not more than 30 minutes, and local retest scanning only needs 10 minutes, which greatly shortens the "measurement-adjustment" cycle and improves the construction efficiency by more than 60%.

[0022] 3. Ensure positioning accuracy: through centimeter-level unmanned aerial vehicle positioning, millimeter-level laser radar ranging, ICP precise registration (error ≤1mm) and millimeter-level deviation analysis, high-precision positioning of variable cross-section steel structure is realized, the axis deviation is controlled within 2mm, the angle deflection error is controlled within 0.1°, and the cumulative error is effectively avoided.

[0023] 4. Realize real-time closed-loop control: build a closed-loop process of "scanning-analysis-adjustment-retest", data information is fed back in real time, construction personnel can accurately adjust, without repeated measurement and correction, solve the pain point of information lag in traditional technology, and ensure the stability of installation quality.

[0024] 5. Adapt to the characteristics of variable cross-section structure: set cross-section control points through BIM model, extract variable cross-section characteristics (axis, port center, etc.), specially adapt to the shape characteristics of variable cross-section steel structure, overcome the limitation of existing technology which is only for conventional steel structure or specific assembly point, and have wider application range.

[0025] In summary, the problems of low efficiency, poor accuracy, blind area and information lag in the installation of variable cross-section steel structure are solved, real-time, accurate and digital closed-loop quality control of the installation process is realized, and the construction efficiency and safety are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0026] Fig. 1 It is a schematic diagram of a variable cross-section steel structure theoretical BIM model; Fig. 2 It is a schematic diagram of actual unmanned aerial vehicle field scanning operation; Fig. 3 It is a schematic diagram of digital closed-loop control process; Fig. 4 It is a schematic diagram of point cloud model and BIM model comparison. DETAILED DESCRIPTION

[0027] In order to make the technical scheme of the present application clearer and easier to understand, the following will be described in detail in combination with specific embodiments: As Figs. 1 to 4 shown, a variable cross-section steel structure installation positioning method based on unmanned aerial vehicle laser point cloud scanning of the application comprises the following steps: S1: Create a theoretical BIM model: According to the design drawing, a high-precision theoretical three-dimensional BIM model of the variable cross-section steel structure to be installed is established, and the three-dimensional design coordinates of the key feature points and feature lines (such as component axis, port center, interface flange face, etc.) in the model are determined.

[0028] S2: Unmanned aerial vehicle laser scanning data acquisition: After the installation of the component to be installed (which can be in a temporary fixed state), operate the unmanned aerial vehicle equipped with a laser radar (LiDAR) to fly around the current installation section and the adjacent installed structure for scanning, and automatically collect high-density three-dimensional point cloud data.

[0029] S3: Point cloud data preprocessing and fusion: The collected multi-station, multi-angle point cloud data is preprocessed, including denoising and filtering, and the multi-station point cloud data is accurately registered and fused using the POS system (positioning and orientation system) of the unmanned aerial vehicle and the ground control points, to generate a complete, high-precision on-site real scene point cloud model.

[0030] S4: Intelligent registration and comparison of point cloud and BIM model: From the fused real scene point cloud, the feature points and feature surfaces of the installed structure are automatically extracted. The extracted real scene features are intelligently registered with the corresponding features in the theoretical BIM model using the Iterative Closest Point (ICP) algorithm, and the coordinate transformation parameters between the two are calculated. The theoretical BIM model of the component to be installed is mapped to the real scene coordinate system through the above transformation parameters, realizing the accurate alignment of theory and reality.

[0031] S5: Installation deviation analysis and visual output: The deviation value between the current actual position of the component to be installed (based on the point cloud) and the theoretical design position (the BIM model after transformation) is calculated, including but not limited to: axis deviation (ΔX, ΔY, ΔZ), angle deflection (Δα, Δβ, Δγ), interface misalignment, etc. The deviation analysis results are visualized in the form of a chromatogram and a data report. The chromatogram can visually display the deviation size and trend of each part of the component, and the data report provides numerical guidance accurate to the millimeter level.

[0032] S6: Real-time guidance adjustment and reinspection: Construction personnel quickly and accurately fine-tune the components (such as adjusting the jack, hand-operated hoist, etc.) according to the visual deviation results. After adjustment, the UAV can be triggered again for rapid scanning (local scanning), and the steps S3-S5 are repeated to generate a new deviation report until all installation deviations are controlled within the design allowable tolerance range. After confirmation, final fixation (such as welding, bolt final tightening) is performed.

[0033] Preferably, the data acquisition module is in the form of a UAV carrying a laser scanning radar.

[0034] Preferably, the intelligent registration and comparative analysis uses the Iterative Closest Point (ICP) algorithm to match the features extracted from the real scene point cloud with the corresponding features in the theoretical BIM model, calculate the coordinate conversion parameters and calculate the deviation.

[0035] Preferably, the deviation includes linear displacement deviation and angular deflection deviation, and the visual deviation result is presented in the form of deviation color spectrum and data report.

[0036] Preferably, the rechecking step includes UAV laser scanning and deviation analysis again after adjustment to form a closed-loop control process.

[0037] A variable cross-section steel structure installation positioning system for implementing the above method, comprising: Data acquisition module: including a UAV, a laser radar and a positioning system, for collecting point cloud data; Data processing and control center: for point cloud processing, model management, data comparison and visual display; Feedback execution module: for adjusting the steel structure according to the visual results.

[0038] Preferably, the data processing and control center is a cloud server or a mobile workstation, which is connected with the data acquisition module through wireless communication.

[0039] The protection scope of the present application is not limited to the above embodiments, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A variable cross-section steel structure installation positioning method based on unmanned aerial vehicle laser point cloud scanning, characterized in that, The method comprises the following steps: step 1, creating a theoretical BIM model of the variable cross-section steel structure to be installed; step 2, using an unmanned aerial vehicle (UAV) equipped with a laser radar to scan the installed steel structure and obtaining on-site three-dimensional point cloud data; step 3, preprocessing and fusing the point cloud data to generate a real scene point cloud model; step 4, intelligently registering and comparing the real scene point cloud model with the theoretical BIM model to calculate the deviation between the current installation position and the theoretical design position; step 5, outputting the visualized deviation result to guide the installation adjustment; and step 6, based on the adjusted result, repeating steps 2 to 6 until all deviations are controlled within the design allowable tolerance range, and finally completing the fixation.

2. The method of claim 1, wherein, In step 1, the creation of the theoretical BIM model comprises: establishing a high-precision three-dimensional BIM model of the variable cross-section steel structure to be installed according to design drawings, and clearly defining the three-dimensional design coordinates of key feature points and feature lines in the model.

3. The method of claim 2, wherein, The key feature points include the center of the component port and the center point of the interface flange face, and the key feature lines include the component axis.

4. The method of claim 1, wherein, In step 2, the data acquisition module needs to be realized in the form of an unmanned aerial vehicle (UAV) equipped with a laser scanning radar, including after the component to be installed is hoisted to a temporary fixed state, operating the UAV equipped with a laser radar (LiDAR) to perform multi-view flight scanning around the current installation section and the adjacent installed structure, and automatically collecting high-density three-dimensional point cloud data.

5. The method of claim 1, wherein, In step 3, the multi-station, multi-view point cloud data collected is preprocessed by denoising and filtering, and the multi-station point cloud is accurately registered and fused by using the UAV POS system and ground control points to generate a complete high-precision on-site real scene point cloud model.

6. The method of claim 1, wherein, In step 4, the feature points and feature surfaces of the installed structure are automatically extracted from the real scene point cloud, the iterative closest point (ICP) algorithm is used to intelligently register the real scene features with the corresponding features of the theoretical BIM model, the coordinate conversion parameters are calculated, the theoretical BIM model of the component to be installed is mapped to the real scene coordinate system through the conversion parameters, the theoretical and real alignment is realized, and the deviation value between the actual position of the component to be installed and the theoretical design position is calculated.

7. The method of claim 6, wherein, The deviation value includes axis deviation, angle deflection and interface misalignment amount.

8. The method of claim 1, wherein, In step 5, the deviation analysis result is visualized and output in the form of a chromatogram and a millimeter-level data report.

9. The method of claim 1, wherein, In step 6, the construction personnel make fine adjustments to the component according to the visualized deviation result, perform local rapid scanning through the UAV after the adjustment, and repeat steps S3-S5 to generate a new deviation report.

10. A variable cross-section steel structure installation positioning system implementing the method of any one of claims 1-9, characterized by, The method comprises the following steps: The data acquisition module is composed of a UAV, a laser radar and a positioning system, the UAV is loaded with the laser radar to realize point cloud data acquisition, the positioning system includes a GPS+IMU combined positioning unit and a ground control point, and is used for providing high-precision positioning data; the data processing and control center is a cloud server or a mobile workstation, is connected with the data acquisition module in a wireless communication mode, is used for point cloud data processing, BIM model management, data comparison and analysis and deviation result visual display; the feedback execution module includes a jack, a hand-operated hoist and a fixing device, is used for adjusting the position and attitude of the variable cross-section steel structure member according to the visual deviation result, and finally completes the member fixing.

Images