Complex space truss positioning system and method based on three-dimensional coordinate acquisition technology

Abstract

The invention discloses a complex space truss positioning system and method based on a three-dimensional coordinate acquisition technology, and relates to the technical field of truss positioning. The complex space truss positioning system and method based on the three-dimensional coordinate acquisition technology comprises the following steps: S1, three-dimensional modeling and pre-compensation; s2, constructing and dynamically calibrating a three-dimensional control network; s3, on-site preliminary measurement and data acquisition; s4, contour trajectory speculation and connection complementation; s5, performing dynamic multi-view measurement and data integration; s6, data analysis and positioning adjustment; measuring equipment is selected according to the characteristics of the truss, node coordinates are preliminarily measured and collected, shielded parts are recorded, interference of environmental factors is controlled, and the quality of basic measurement data is guaranteed. An image processing and mode recognition technology is utilized to analyze a contour track, connection and node positions of a shielding part are speculated in combination with structural characteristics, and verification is performed after mechanical analysis optimization, so that the measurement problem caused by shielding is effectively solved, and data integrity is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of truss positioning, and in particular to a complex space truss positioning system and method based on three-dimensional coordinate acquisition technology. Background Art

[0002] Space truss structures are widely used in large-scale buildings, bridges, aerospace, and other fields. Their precise positioning is key to ensuring structural safety and performance compliance. Space trusses are typically composed of multiple rods and nodes, forming a complex three-dimensional structure. Accurate positioning ensures that each rod and node is installed precisely according to design requirements, avoiding structural instability or safety hazards caused by installation deviations. Mechanical property optimization: Through precise positioning, the mechanical properties of the truss, such as stiffness and damping, can be optimized to ensure that the structure exhibits good mechanical properties when subjected to loads, improving the overall stability and reliability of the structure.

[0003] Publication No. CN119169075A discloses a steel structure positioning system and method based on computer vision and AI, including a preset module, a shooting module, an image processing module, a sensor module and an adjustment module. The preset module is used to set the preset rectangular structure of the steel structure corridor and the accurate posture of the corresponding four corner points. The shooting module is used to capture the first image of the steel structure corridor in real time. The image processing module is used to obtain the real-time offset angle of the steel structure corridor. The sensor module is used to obtain the real-time posture of the four corner points of the steel structure corridor and obtain the real-time posture difference of the four corner points. The adjustment module is used to adjust the steel structure corridor in real time until the real-time offset angle is less than the preset offset angle threshold and the real-time posture difference of the four corner points is less than the preset posture difference threshold. The invention can ensure that all nodes are of qualified quality during a single installation process to ensure the consistency and stability of the steel structure corridor positioning in a long-distance, highly complex environment.

[0004] As shown in the above technology, the current positioning of steel structure trusses is usually achieved through image vision recognition. However, for complex spatial truss structures, many parts of the trusses will be obscured, and the built trusses are not in their normal state. They will be deformed by the influence of forces, and the construction process will also change. Therefore, simply using visual shooting and recognition is difficult to accurately locate all trusses. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a complex space truss positioning system and method based on three-dimensional coordinate acquisition technology, which solves the problems of the prior art.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A complex space truss positioning method based on three-dimensional coordinate acquisition technology, comprising the following steps:

[0007] S1 3D modeling and pre-compensation: Use professional software to build accurate 3D models, covering key nodes, connectors, and structural details; analyze historical data to identify error sources and pre-compensate, and extract theoretical node coordinates as a benchmark;

[0008] S2 3D control network construction and dynamic calibration: Select benchmark points at the construction site for precise measurement, and use relevant equipment to build a control network based on the benchmark points to cover all nodes to be measured. Consider the dynamic characteristics of the equipment and use a real-time feedback mechanism to calibrate the measurement data.

[0009] S3 On-site preliminary measurement and data collection: Select measurement equipment according to the characteristics and requirements of the truss, preliminarily measure and collect node coordinates, record the position and outline of the obstructed parts; control the impact of environmental factors on the measurement;

[0010] S4 Contour trajectory inference and connection supplement: Use image processing technology to analyze the contour trajectory, combine structural characteristics and known coordinates to infer the connection relationship and node position of the occluded part, and perform mechanical analysis and optimization verification;

[0011] S5 dynamic multi-view measurement and data integration: Dynamic multi-view measurement is used to ensure that nodes are measured from different viewpoints; data fusion algorithms are used to integrate data and compensate for errors to form a complete data set;

[0012] S6 Data Analysis and Positioning Adjustment: Use machine learning algorithms to analyze data, model mechanical properties and stability, and adjust the truss node positions and angles accordingly, taking environmental factors into account. After verification, feedback is collected for improvement.

[0013] Preferably, step S1 specifically includes:

[0014] S1.1 Accurate 3D modeling: Use professional software such as CAD and SolidWorks to build a 3D model of a complex space truss according to the design drawings, ensuring that the model includes all key nodes, connectors and structural details. The key nodes are set as N = (N1, N2, ..., N n ), to ensure the accuracy of geometric dimensions and node positions, let the node coordinates be r i =(x i ,y i , z i ), rod length L ij =||r i -r j || (i≠j, and i and j=1, 2, ..., n);

[0015] S1.2 Error and deformation pre-compensation: Analyze historical measurement data or empirical models, identify the source of error, use finite element analysis methods, calculate deformation and perform pre-compensation based on material properties and load conditions;

[0016] S1.3 Coordinate pre-extraction: Extract the theoretical three-dimensional coordinates of each node from the three-dimensional model i ′, serving as a benchmark for subsequent measurement and positioning.

[0017] Preferably, the implementation steps of step S1.2 using the finite element analysis method to calculate the deformation and perform pre-compensation according to material properties and load conditions include:

[0018] Import model: import 3D CAD model into finite element analysis software;

[0019] Meshing: Mesh the model and select the required element type;

[0020] Material property definition: Input the elastic modulus E and Poisson's ratio ν of the material;

[0021] Boundary conditions and load application: Boundary conditions and loads are applied according to actual conditions. Boundary conditions include but are not limited to fixed constraints and displacement constraints, and loads include but are not limited to external forces, pressure, and temperature.

[0022] Solve the model: Use finite element analysis software to solve the model and obtain the displacement of each node;

[0023] The equilibrium equation is: Ku = f; where K is the stiffness matrix, u is the displacement vector, and f is the external force vector. By solving the equilibrium equation, the displacement vector u of each node is obtained;

[0024] Deformation Δr i It can be expressed as: Among them, u i is the displacement vector of node i, is the initial displacement vector of node i;

[0025] Calculate the pre-compensation amount: Calculate the pre-compensation amount of each node according to the displacement result. The node coordinate after pre-compensation is r i ′=r i +Δr i ; Among them, r i is the original coordinate of node i;

[0026] Update node coordinates: Apply the pre-compensation amount to the node coordinates to obtain the pre-compensated model.

[0027] Preferably, step S2 specifically includes:

[0028] S2.1 Reference point setting: Select a stable and easily identifiable point on the construction site as the reference point, set as B = (B1, B2, ..., B m ) to ensure that the benchmarks are reasonably distributed and cover the entire measurement area; use high-precision measuring equipment to accurately measure the benchmarks and record the three-dimensional coordinates of the benchmarks and k=1, 2, ..., m;

[0029] S2.2 Construction of 3D control network: Based on the reference points, a 3D control network is constructed using a total station and laser tracker. The overall transformation matrix T of the control network is fitted using the least squares method to minimize the error between the measured coordinates and the theoretical coordinates. The transformation matrix T and the error vector e are designed to satisfy min||e|| 2 =min Among them, b′ k is the coordinate of the theoretical reference point;

[0030] S2.3 Dynamic calibration: Considering the dynamic characteristics of the measuring device's moving velocity v and acceleration a, the measurement data is dynamically calibrated. Using the Kalman filter algorithm, the state transfer matrix F and the process noise covariance Q are used to predict the state vector and covariance matrix at the next moment. The observation matrix H and the observation noise covariance R are combined with the observation vector Z to update the predicted state vector and covariance matrix, estimate the device state in real time, and correct the measurement data.

[0031] 5. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, wherein step S3 specifically comprises:

[0032] S3.1 Selection of measuring equipment: Select appropriate measuring equipment based on the structural characteristics of the truss and measurement requirements;

[0033] S3.2 Preliminary measurement: Perform preliminary measurement of the truss and collect the actual three-dimensional coordinate data of each node as much as possible. i For the parts that cannot be directly measured due to occlusion, record the position and approximate outline to provide a basis for subsequent inference and connection. Let the measurement error be δr i =r i ″-r i ′, requiring ||δr i ||≤∈, ∈ is the allowable error threshold;

[0034] S3.3 Environmental interference control: During the measurement process, pay attention to controlling the impact of environmental factors on the measurement data, including temperature T, humidity H, vibration v vib .

[0035] Preferably, step S4 specifically includes:

[0036] S4.1 Contour trajectory analysis: Use advanced image processing and pattern recognition technology to analyze the truss contour trajectory obtained in the preliminary measurement, identify the contour features of the obscured parts, and provide data support for the inferred connection supplement. Let the contour feature vector be f, which is obtained through the feature extraction algorithm;

[0037] S4.2 Infer connection supplement: Based on the structural characteristics of the truss and the coordinate information of the known nodes, combined with the results of the contour trajectory analysis, the connection relationship and node position of the obscured part are inferred, and the mechanical model is used for mechanical analysis and optimization to ensure the rationality and reliability of the supplemented part; let the inferred node coordinates be r pred , the structural mechanics equilibrium equation Ku=F must also be satisfied, where K is the stiffness matrix, u is the displacement vector, and F is the external force vector;

[0038] S4.3 Verification of Supplementation: Verify the presumed supplemented parts to ensure that they are consistent with the actual structure and meet the stability requirements. Use simulation software or actual testing to verify and compare the structural performance differences before and after the supplementation.

[0039] Preferably, step S5 specifically includes:

[0040] Multi-view measurement: Using dynamic multi-view measurement method, by moving the measuring equipment or adjusting the measuring angle, the truss is measured in all directions and at multiple angles, ensuring that each node can be measured from at least two different perspectives, thus improving the accuracy and reliability of the measurement. Let the perspective vector be v view , measurement matrix M;

[0041] Data integration: Use data fusion algorithm to integrate the three-dimensional coordinate data obtained from different perspectives, consider the errors and differences between the measurement data from different perspectives, perform error compensation and optimization, and form a complete and accurate three-dimensional coordinate data set R = {r final，i};

[0042] Design parameters: weight vector ω, satisfying Among them, n view is the number of viewing angles, r il is the measured coordinate of node i at the lth viewing angle.

[0043] Preferably, step S6 specifically includes:

[0044] Data analysis: Let the eigenvector X feat To extract features from the integrated 3D coordinate data, including coordinate values, relative distances between coordinates, angles, speeds, and accelerations, the selection of features is determined based on the specific application scenario and the requirements of the classification and recognition tasks. Let the label vector Y be the corresponding label assigned to the 3D coordinate data, including category labels and numerical labels. The selection of labels is consistent with the goals of classification, recognition, and prediction.

[0045] Training model: Select a machine learning algorithm based on task requirements; use the feature vector X featAnd label vector Y train the model to obtain model parameters θ; use the trained model and model parameters θ to classify, identify and predict the new three-dimensional coordinate data; input the new feature vector X feat , the model outputs the corresponding prediction results; compares the model's prediction results with the theoretical model, and calculates the deviation and error between the measured data and the theoretical model; the deviation and error are expressed as the difference between the predicted value and the theoretical value; analyzes the source of the deviation and error and finds out the possible cause;

[0046] Mechanical properties and stability modeling: Model and predict the mechanical properties and stability of truss structures using machine learning algorithms or traditional mechanical methods to ensure that the adjusted truss structures meet mechanical requirements and stability standards. Mechanical properties include stiffness and damping, and stability indicators include critical load and instability mode.

[0047] Positioning adjustment: Use the trained machine learning model or traditional mechanical methods to input new mechanical model parameters and predict the mechanical response of the truss structure;

[0048] Stability assessment: Determine whether the structure is stable by comparing the predicted critical load with the actual load; fine-tune the node positions and angle parameters of the truss to ensure accurate docking and installation of each node; consider the impact of environmental factors on the truss structure and improve the stability of the structure through environmental monitoring and compensation technology;

[0049] Assume that the adjustment vector is ΔR={Δr i}, satisfying the adjusted node coordinate r adj，i =r final，i +Δr i ;

[0050] Verification and feedback: Verify the adjusted truss to ensure that it meets the design requirements and usage standards. Verification indicators include position accuracy, stability, and feedback vector; collect feedback data f back , including verification results and improvement suggestions, providing an improvement basis for subsequent measurement and positioning.

[0051] The present invention also discloses a complex space truss positioning system, comprising:

[0052] 3D modeling module, used to build 3D models of complex space trusses based on drawings;

[0053] 3D control network construction and dynamic calibration module, used to construct a 3D control network at the construction site and perform dynamic calibration;

[0054] On-site measurement and data acquisition module, used to perform preliminary measurements of the truss on site and collect the actual three-dimensional coordinate data of each node;

[0055] The contour trajectory estimation and supplementation module is used to estimate the contour trajectory of the parts that cannot be directly measured in the preliminary measurement and supplement the connection relationship;

[0056] The multi-view measurement and data integration module uses a dynamic multi-view measurement method to measure the truss in all directions and at multiple angles, and integrate the data;

[0057] The data analysis and positioning adjustment module is used to analyze the integrated three-dimensional coordinate data and perform positioning adjustments based on the analysis results.

[0058] Preferably, the data processing process of the system is processed and stored on the cloud platform, and the three-dimensional model established according to the drawing and the three-dimensional model constructed by actual measurement are displayed separately on the same interface, and the size and error data are displayed.

[0059] The present invention provides a complex space truss positioning system and method based on three-dimensional coordinate acquisition technology. Compared with the existing technology, it has the following advantages:

[0060] This complex spatial truss positioning system and method, based on 3D coordinate acquisition technology, uses a 3D modeling module to precisely construct models and pre-compensate for errors, providing an accurate benchmark for subsequent positioning and reducing positioning errors caused by model deviations. Modules for 3D control network construction and dynamic calibration, multi-view measurement, and data integration ensure all-round, multi-angle measurement, effectively compensating for measurement gaps caused by occlusions, and improving data integrity and accuracy. The data analysis and positioning adjustment module utilizes machine learning algorithms to deeply analyze data and model mechanical properties, enabling scientific positioning adjustments and ensuring truss structural stability. Data processing is performed on a cloud platform for easy storage and analysis. The intuitive and clear display of dual models and error data facilitates timely problem identification and resolution, improving positioning efficiency and quality.

[0061] 2. This complex space truss positioning system and method based on three-dimensional coordinate acquisition technology uses precise three-dimensional modeling to ensure that the model includes all key nodes, connectors and structural details, accurately sets node coordinates and rod lengths, provides a reliable basis for subsequent analysis, and reduces errors caused by model simplification. In the error and deformation pre-compensation link, the finite element analysis method is used to calculate the deformation based on material properties and load conditions, making the pre-compensation more targeted. By importing the model, meshing, defining material properties, applying boundary conditions and loads, solving the model and other steps, the node displacement is accurately obtained, and then a reasonable pre-compensation amount is calculated, effectively reducing errors caused by material deformation, manufacturing tolerances, etc. Coordinate pre-extraction extracts the theoretical three-dimensional coordinates of each node from the three-dimensional model as a benchmark for subsequent measurement and positioning, ensuring the accuracy and consistency of the benchmark, helping to improve the accuracy and reliability of the entire positioning process and ensure the accuracy of the positioning of complex space trusses.

[0062] 3. This complex spatial truss positioning system and method, based on 3D coordinate acquisition technology, rationally selects stable, easily identifiable benchmarks at the construction site and accurately measures 3D coordinates, providing a reliable foundation for subsequent measurements and ensuring comprehensive coverage. Based on the benchmarks, a 3D control network is constructed using advanced equipment. The least-squares method is used to fit the transformation matrix, minimizing the error between measured and theoretical coordinates and improving measurement accuracy. Taking into account the dynamic characteristics of the measurement equipment, a Kalman filter algorithm is used for dynamic calibration. The state transition matrix and observation matrix are used to estimate the equipment state in real time and correct the data, effectively reducing dynamic errors and further improving the accuracy and reliability of the measurement data.

[0063] 4. This complex spatial truss positioning system and method, based on 3D coordinate acquisition technology, selects measurement equipment based on truss characteristics, initially measures and collects node coordinates, records obscured areas, and controls environmental interference to ensure the quality of basic measurement data. Image processing and pattern recognition techniques are used to analyze contour trajectories, and the locations of obscured connections and nodes are inferred based on structural characteristics. This is verified after mechanical analysis and optimization, effectively resolving measurement challenges caused by obstructions and improving data integrity. Dynamic multi-view measurement ensures multi-view coverage of nodes, improving measurement accuracy and reliability. A data fusion algorithm integrates data from different viewpoints, compensating for and optimizing errors to form a complete and accurate 3D coordinate dataset, providing more comprehensive and accurate data support for truss measurement.

[0064] 5. This complex spatial truss positioning system and method, based on 3D coordinate acquisition technology, extracts multiple features from integrated 3D coordinate data and assigns labels, providing a rich and accurate data foundation for subsequent model training. Appropriate machine learning algorithms are selected, and parameters are optimized through cross-validation or grid search to improve model performance and enable accurate classification, identification, and prediction of new data. Mechanical properties and stability are modeled and predicted using machine learning or traditional mechanics methods to ensure that the truss structure meets requirements, and mechanical responses are predicted to assess stability. Node positions and angle parameters are fine-tuned, environmental factors are considered, structural stability is improved, and accurate node docking is achieved. The adjusted trusses are verified and feedback data is collected to provide a basis for improvement in subsequent measurement and positioning, creating a virtuous cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a system module block diagram of the present invention;

[0066] Figure 2 It is a schematic diagram of the steps of the present invention. DETAILED DESCRIPTION

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0068] See Figure 1-Figure 2 , the present invention provides the following five technical solutions:

[0069] A complex space truss positioning system, comprising:

[0070] 3D modeling module, used to build 3D models of complex space trusses based on drawings;

[0071] 3D control network construction and dynamic calibration module, used to construct a 3D control network at the construction site and perform dynamic calibration;

[0072] On-site measurement and data acquisition module, used to perform preliminary measurements of the truss on site and collect the actual three-dimensional coordinate data of each node;

[0073] The contour trajectory estimation and supplementation module is used to estimate the contour trajectory of the parts that cannot be directly measured in the preliminary measurement and supplement the connection relationship;

[0074] The multi-view measurement and data integration module uses a dynamic multi-view measurement method to measure the truss in all directions and at multiple angles, and integrate the data;

[0075] The data analysis and positioning adjustment module is used to analyze the integrated three-dimensional coordinate data and make positioning adjustments based on the analysis results;

[0076] The system's data processing is processed and stored on the cloud platform. The three-dimensional model built according to the drawing and the three-dimensional model constructed by actual measurement are displayed separately on the same interface, and the size and error data are displayed.

[0077] The present invention also discloses a complex space truss positioning method based on three-dimensional coordinate acquisition technology, comprising the following steps:

[0078] S1 3D modeling and pre-compensation: Use professional software to build accurate 3D models, covering key nodes, connectors, and structural details; analyze historical data to identify error sources and pre-compensate, and extract theoretical node coordinates as a benchmark;

[0079] S2 3D control network construction and dynamic calibration: Select benchmark points at the construction site for precise measurement, and use relevant equipment to build a control network based on the benchmark points to cover all nodes to be measured. Consider the dynamic characteristics of the equipment and use a real-time feedback mechanism to calibrate the measurement data.

[0080] S3 On-site preliminary measurement and data collection: Select measurement equipment according to the characteristics and requirements of the truss, preliminarily measure and collect node coordinates, record the position and outline of the obstructed parts; control the impact of environmental factors on the measurement;

[0081] S4 Contour trajectory inference and connection supplement: Use image processing technology to analyze the contour trajectory, combine structural characteristics and known coordinates to infer the connection relationship and node position of the occluded part, and perform mechanical analysis and optimization verification;

[0082] S5 dynamic multi-view measurement and data integration: Dynamic multi-view measurement is used to ensure that nodes are measured from different viewpoints; data fusion algorithms are used to integrate data and compensate for errors to form a complete data set;

[0083] S6 Data Analysis and Positioning Adjustment: Use machine learning algorithms to analyze data, model mechanical properties and stability, and adjust the truss node positions and angles accordingly, taking environmental factors into account. After verification, feedback is collected for improvement.

[0084] The 3D modeling module precisely constructs models and pre-compensates for errors, providing an accurate benchmark for subsequent positioning and reducing positioning errors caused by model deviations. The 3D control network construction and dynamic calibration, multi-view measurement, and data integration modules ensure all-round, multi-angle measurement, effectively compensating for measurement gaps caused by occlusions, and improving data integrity and accuracy. The data analysis and positioning adjustment module utilizes machine learning algorithms to deeply analyze data and model mechanical properties, enabling scientific positioning adjustments and ensuring the stability of the truss structure. Data processing is performed on a cloud platform for easy storage and analysis; the dual-model display and error data presentation are intuitive and clear, facilitating the timely identification and resolution of problems, and improving positioning efficiency and quality.

[0085] The second embodiment differs from the first embodiment mainly in that step S1 specifically includes:

[0086] S1.1 Accurate 3D modeling: Use professional software such as CAD and SolidWorks to build a 3D model of a complex space truss according to the design drawings, ensuring that the model includes all key nodes, connectors and structural details. The key nodes are set as N = (N1, N2, ..., N n ), to ensure the accuracy of geometric dimensions and node positions, let the node coordinates be r i =(x i ,y i , z i ), rod length L ij =||r i -r j || (i≠j, and i and j=1, 2, ..., n);

[0087] S1.2 Error and deformation pre-compensation: Analyze historical measurement data or empirical models to identify possible error sources (such as material deformation and manufacturing tolerances). Use methods such as finite element analysis (FEA) to calculate deformation and perform pre-compensation based on material properties (such as elastic modulus E and Poisson's ratio ν) and load conditions.

[0088] S1.3 Coordinate pre-extraction: Extract the theoretical three-dimensional coordinates of each node from the three-dimensional model i ′, serving as a benchmark for subsequent measurement and positioning.

[0089] Step S1.2 uses the finite element analysis method to calculate the deformation and perform pre-compensation based on material properties and load conditions. The implementation steps include:

[0090] Import model: import 3D CAD model into finite element analysis software;

[0091] Meshing: Mesh the model and select the required element type (such as tetrahedral element, hexahedral element, etc.);

[0092] Material property definition: Input the material's elastic modulus E (which describes the proportional relationship between stress and strain within the material's elastic deformation range) and Poisson's ratio ν (which describes the ratio of transverse strain to longitudinal strain when the material is under tension or compression).

[0093] Boundary conditions and load application: Boundary conditions and loads are applied according to actual conditions. Boundary conditions include but are not limited to fixed constraints and displacement constraints, and loads include but are not limited to external forces, pressure, and temperature.

[0094] Solve the model: Use finite element analysis software to solve the model and obtain the displacement of each node;

[0095] The equilibrium equation is: Ku = f; where K is the stiffness matrix, u is the displacement vector, and f is the external force vector. By solving the equilibrium equation, the displacement vector u of each node is obtained;

[0096] Deformation Δr i It can be expressed as: Among them, u i is the displacement vector of node i, is the initial displacement vector of node i (usually 0);

[0097] Calculate the pre-compensation amount: Calculate the pre-compensation amount of each node according to the displacement result. The node coordinate after pre-compensation is r i ′=r i +Δr i ; Among them, r i is the original coordinate of node i;

[0098] Update node coordinates: Apply the pre-compensation amount to the node coordinates to obtain the pre-compensated model.

[0099] Accurate 3D modeling ensures that the model includes all key nodes, connectors and structural details, accurately sets node coordinates and rod lengths, provides a reliable basis for subsequent analysis, and reduces errors caused by model simplification. In the error and deformation pre-compensation link, the finite element analysis method is used to calculate the deformation based on material properties and load conditions, making the pre-compensation more targeted. By importing the model, meshing, defining material properties, applying boundary conditions and loads, solving the model and other steps, the node displacement is accurately obtained, and then a reasonable pre-compensation amount is calculated, effectively reducing errors caused by material deformation, manufacturing tolerances, etc. Coordinate pre-extraction extracts the theoretical 3D coordinates of each node from the 3D model as a benchmark for subsequent measurement and positioning, ensuring the accuracy and consistency of the benchmark, helping to improve the accuracy and reliability of the entire positioning process and ensure the accuracy of the positioning of complex spatial trusses.

[0100] The third embodiment differs from the first embodiment mainly in that step S2 specifically includes:

[0101] S2.1 Reference point setting: Select a stable and easily identifiable point on the construction site as the reference point, set as B = (B1, B2, ..., B m ) to ensure that the benchmarks are reasonably distributed and cover the entire measurement area; use high-precision measuring equipment to accurately measure the benchmarks and record the three-dimensional coordinates of the benchmarks and k=1, 2, ..., m;

[0102] S2.2 Construction of 3D control network: Based on the reference points, a 3D control network is constructed using total stations, laser trackers, and other equipment. The overall transformation matrix T of the control network is fitted using methods such as the least squares method to minimize the error between the measured coordinates and the theoretical coordinates. The transformation matrix T and error vector e are designed to satisfy min||e|| 2 =min Where b′ k is the coordinate of the theoretical reference point;

[0103] S2.3 Dynamic calibration: Considering the dynamic characteristics of the measuring device's moving velocity v and acceleration a, the measurement data is dynamically calibrated. Using the Kalman filter algorithm, the state transfer matrix F and the process noise covariance Q are used to predict the state vector and covariance matrix at the next moment. The observation matrix H and the observation noise covariance R are combined with the observation vector Z to update the predicted state vector and covariance matrix, estimate the device state in real time, and correct the measurement data.

[0104] The state vector X contains the dynamic state information of the measurement device, including position, velocity, and acceleration, and is defined as:

[0105]

[0106] Where w is the position, v is the velocity, and a is the acceleration;

[0107] The observation vector Z is the observation data obtained by the measuring device. The measuring device provides the observation values of position and velocity. The observation vector is defined as:

[0108]

[0109] Among them, z w is the observed value of the position, z v is the observed value of velocity;

[0110] The state transition matrix F describes the evolution of the device state from one moment to the next. It is a state vector containing position, velocity, and acceleration and is defined as:

[0111]

[0112] Where Δt is the time step;

[0113] The observation matrix G describes the relationship between the observation vector and the state vector, including the observation values of position and velocity, and is defined as:

[0114]

[0115] The process noise covariance Q describes the noise in the state transition process and is set according to the dynamic characteristics and noise characteristics of the device and is defined as:

[0116]

[0117] in, are the process noise variances of position, velocity, and acceleration, respectively;

[0118] The observation noise covariance R describes the noise in the observation data; it is set according to the precision of the measurement equipment and the noise characteristics and is defined as:

[0119]

[0120] in, are the noise variances of the position and velocity observations, respectively.

[0121] At the construction site, stable and easily identifiable benchmarks are rationally selected and their three-dimensional coordinates are accurately measured, providing a reliable foundation for subsequent measurements and ensuring comprehensive coverage. Based on these benchmarks, a three-dimensional control network is constructed using advanced equipment. The least-squares method is used to fit the transformation matrix, minimizing the error between measured and theoretical coordinates and improving measurement accuracy. Considering the dynamic characteristics of the measurement equipment, a Kalman filter algorithm is used for dynamic calibration. The state transition matrix and observation matrix are used to estimate the equipment state in real time and calibrate the data, effectively reducing dynamic errors and further improving the accuracy and reliability of the measurement data.

[0122] The fourth embodiment differs from the first embodiment mainly in that step S3 specifically includes:

[0123] S3.1 Selection of measuring equipment: Select appropriate measuring equipment (such as laser scanner, high-precision camera, etc.) based on the structural characteristics of the truss and measurement requirements;

[0124] S3.2 Preliminary measurement: Perform preliminary measurement of the truss and collect the actual three-dimensional coordinate data of each node as much as possible. i For the parts that cannot be directly measured due to occlusion, record the position and approximate outline to provide a basis for subsequent inference and connection. Let the measurement error be δr i =r i ″-r i ′, requiring ||δr i ||≤∈, ∈ is the allowable error threshold;

[0125] S3.3 Environmental interference control: During the measurement process, pay attention to controlling the impact of environmental factors on the measurement data, including temperature T, humidity H, vibration v vib ;

[0126] Step S4 specifically includes:

[0127] S4.1 Contour trajectory analysis: Use advanced image processing and pattern recognition technology to analyze the truss contour trajectory obtained in the preliminary measurement, identify the contour features of the obscured part, and provide data support for the inferred connection supplement. Let the contour feature vector be f, and obtain it through feature extraction algorithm (such as SIFT, SURF, etc.);

[0128] S4.2 Infer connection supplement: Based on the structural characteristics of the truss and the coordinate information of the known nodes, combined with the results of the contour trajectory analysis, the connection relationship and node position of the obscured part are inferred, and mechanical models (such as finite element models) are used for mechanical analysis and optimization to ensure the rationality and reliability of the supplemented part; let the inferred node coordinates be r pred , the structural mechanics equilibrium equation Ku=F must also be satisfied, where K is the stiffness matrix, u is the displacement vector, and F is the external force vector;

[0129] S4.3 Verification of Supplementation: Verify the presumed supplemented parts to ensure that they are consistent with the actual structure and meet the stability requirements. Use simulation software or actual testing to verify and compare the structural performance differences before and after the supplementation.

[0130] Step S5 specifically includes:

[0131] Multi-view measurement: Using dynamic multi-view measurement method, by moving the measuring equipment or adjusting the measuring angle, the truss is measured in all directions and at multiple angles, ensuring that each node can be measured from at least two different perspectives, thus improving the accuracy and reliability of the measurement. Let the perspective vector be v view , measurement matrix M (containing measurement data of each view);

[0132] Data integration: Use data fusion algorithms (such as weighted average method, Kalman filter, etc.) to integrate the three-dimensional coordinate data obtained from different perspectives, consider the errors and differences between the measurement data from different perspectives, perform error compensation and optimization, and form a complete and accurate three-dimensional coordinate data set R = {r final，i};

[0133] Design parameters: weight vector ω, satisfying Among them, n view is the number of viewing angles, r il is the measured coordinate of node i at the lth viewing angle.

[0134] Measuring equipment was selected based on the truss' characteristics. Preliminary measurements were taken to collect node coordinates, record obstructions, and control for environmental interference to ensure the quality of basic measurement data. Image processing and pattern recognition techniques were used to analyze contour trajectories. The locations of obstructed connections and nodes were inferred based on structural characteristics. This was verified through mechanical analysis and optimization, effectively resolving measurement challenges caused by obstructions and improving data integrity. Dynamic multi-view measurement ensured multi-angle coverage of nodes, enhancing measurement accuracy and reliability. A data fusion algorithm was used to integrate data from different viewpoints, compensating for and optimizing errors to form a complete and accurate 3D coordinate dataset, providing more comprehensive and accurate data support for truss measurement.

[0135] The fifth embodiment differs from the first embodiment mainly in that step S6 specifically includes:

[0136] Data analysis: Let the eigenvector X feat To extract features from the integrated 3D coordinate data, including coordinate values, relative distances between coordinates, angles, speeds, accelerations, etc., the selection of features is determined based on the specific application scenario and the requirements of the classification and recognition tasks. Let the label vector Y be the corresponding label assigned to the 3D coordinate data, including category labels (such as normal, abnormal) and numerical labels (such as deviation values). The selection of labels is consistent with the goals of classification, recognition, and prediction.

[0137] Training model: Select a machine learning algorithm based on task requirements, such as neural network, support vector machine (SVM), random forest, etc.; use the feature vector X feat And label vector Y train the model to obtain model parameters θ; during the training process, use cross-validation or grid search method to optimize model parameters and improve model performance; use the trained model and model parameters θ to classify, identify and predict new three-dimensional coordinate data; input the new feature vector X feat , the model outputs the corresponding prediction results; compares the model's prediction results with the theoretical model, and calculates the deviation and error between the measured data and the theoretical model; the deviation and error are expressed as the difference between the predicted value and the theoretical value; analyzes the source of the deviation and error and finds out the possible cause;

[0138] Mechanical properties and stability modeling: Model and predict the mechanical properties and stability of truss structures using machine learning algorithms or traditional mechanical methods to ensure that the adjusted truss structures meet mechanical requirements and stability standards. Mechanical properties include stiffness and damping, while stability indicators include critical loads and instability modes.

[0139] Positioning adjustment: Use the trained machine learning model or traditional mechanical methods to input new mechanical model parameters and predict the mechanical response of the truss structure (such as displacement, stress, strain, etc.);

[0140] Stability assessment: Determine whether the structure is stable by comparing the predicted critical load with the actual load; fine-tune the node positions and angle parameters of the truss to ensure accurate docking and installation of each node; consider the impact of environmental factors on the truss structure and improve the stability of the structure through environmental monitoring and compensation technology;

[0141] Assume that the adjustment vector is ΔR={Δr i}, satisfying the adjusted node coordinate r adj，i =r final，i +Δr i ;

[0142] Verification and feedback: Verify the adjusted truss to ensure that it meets the design requirements and usage standards. Verification indicators include position accuracy, stability, and feedback vector; collect feedback data f back , including verification results and improvement suggestions, providing an improvement basis for subsequent measurement and positioning.

[0143] Extract multiple features from the integrated 3D coordinate data and assign labels, providing a rich and accurate data foundation for subsequent model training. Select an appropriate machine learning algorithm and optimize parameters through cross-validation or grid search to improve model performance and accurately classify, identify, and predict new data. Model and predict mechanical properties and stability through machine learning or traditional mechanics methods to ensure the truss structure meets requirements, and predict mechanical responses to assess stability. Fine-tune point positions and angle parameters, taking environmental factors into account, to improve structural stability and achieve accurate node docking. Verify the adjusted truss and collect feedback data to provide a basis for improvement in subsequent measurement and positioning, creating a virtuous cycle.

[0144] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0145] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0146] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A complex space truss positioning method based on three-dimensional coordinate acquisition technology, characterized in that: The following steps are involved: S1 3D modeling and pre-compensation: Use professional software to build accurate 3D models, covering key nodes, connectors, and structural details; analyze historical data to identify error sources and pre-compensate, and extract theoretical node coordinates as a benchmark; S2 3D control network construction and dynamic calibration: Select benchmark points at the construction site for precise measurement, and use relevant equipment to build a control network based on the benchmark points to cover all nodes to be measured. Consider the dynamic characteristics of the equipment and use a real-time feedback mechanism to calibrate the measurement data. S3 On-site preliminary measurement and data collection: Select measurement equipment according to the characteristics and requirements of the truss, preliminarily measure and collect node coordinates, record the position and outline of the obstructed parts; control the impact of environmental factors on the measurement; S4 Contour trajectory inference and connection supplement: Use image processing technology to analyze the contour trajectory, combine structural characteristics and known coordinates to infer the connection relationship and node position of the occluded part, and perform mechanical analysis and optimization verification; S5 dynamic multi-view measurement and data integration: Dynamic multi-view measurement is used to ensure that nodes are measured from different viewpoints; data fusion algorithms are used to integrate data and compensate for errors to form a complete data set; S6 Data Analysis and Positioning Adjustment: Use machine learning algorithms to analyze data, model mechanical properties and stability, and adjust the truss node positions and angles accordingly, taking environmental factors into account. After verification, feedback is collected for improvement.

2. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, characterized in that: Step S1 specifically includes: S1.1 Accurate 3D modeling: Use professional software to build a 3D model of a complex space truss according to the design drawings, ensuring that the model includes all key nodes, connectors and structural details. The key nodes are set as N = (N1, N2, ..., N n ), to ensure the accuracy of geometric dimensions and node positions, let the node coordinates be r i =(x i ,y i , z i ), rod length L ij =||r i -r j ||, i≠j, and i and j=1, 2, ..., n; S1.2 Error and deformation pre-compensation: Analyze historical measurement data or empirical models, identify the source of error, use finite element analysis methods, calculate deformation and perform pre-compensation based on material properties and load conditions; S1.3 Coordinate pre-extraction: Extract the theoretical three-dimensional coordinates of each node from the three-dimensional model i ′, serving as a benchmark for subsequent measurement and positioning.

3. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 2, characterized in that: Step S1.2 uses the finite element analysis method to calculate the deformation and perform pre-compensation based on material properties and load conditions. The implementation steps include: Import model: import 3D CAD model into finite element analysis software; Meshing: Mesh the model and select the required element type; Material property definition: Input the elastic modulus E and Poisson's ratio ν of the material; Boundary conditions and load application: Boundary conditions and loads are applied according to actual conditions. Boundary conditions include fixed constraints and displacement constraints, and loads include external forces, pressure, and temperature. Solve the model: Use finite element analysis software to solve the model and obtain the displacement of each node; The equilibrium equation is: Ku = f; where K is the stiffness matrix, u is the displacement vector, and f is the external force vector. By solving the equilibrium equation, the displacement vector u of each node is obtained; Deformation Δr i It can be expressed as: Among them, u i is the displacement vector of node i, is the initial displacement vector of node i; Calculate the pre-compensation amount: Calculate the pre-compensation amount of each node according to the displacement result. The node coordinate after pre-compensation is r i ′=r i +Δr i ; Among them, r i is the original coordinate of node i; Update node coordinates: Apply the pre-compensation amount to the node coordinates to obtain the pre-compensated model.

4. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, characterized in that: Step S2 specifically includes: S2.1 Reference point setting: Select a stable and easily identifiable point on the construction site as the reference point, set as B = (B1, B2, ..., B m ) to ensure that the benchmarks are reasonably distributed and cover the entire measurement area; use high-precision measuring equipment to accurately measure the benchmarks and record the three-dimensional coordinates of the benchmarks and k=1, 2, ..., m; S2.2 Construction of three-dimensional control network: Based on the reference points, a three-dimensional control network is constructed using total stations and laser trackers. The overall transformation matrix T of the control network is fitted using the least squares method to minimize the error between the measured coordinates and the theoretical coordinates. The transformation matrix T and the error vector e are designed to satisfy Among them, b' k is the coordinate of the theoretical reference point; S2.3 Dynamic calibration: Considering the dynamic characteristics of the measuring device's moving velocity v and acceleration a, the measurement data is dynamically calibrated. Using the Kalman filter algorithm, the state transfer matrix F and the process noise covariance Q are used to predict the state vector and covariance matrix at the next moment. The observation matrix H and the observation noise covariance R are combined with the observation vector Z to update the predicted state vector and covariance matrix, estimate the device state in real time, and correct the measurement data.

5. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, characterized in that: Step S3 specifically includes: S3.1 Selection of measuring equipment: Select appropriate measuring equipment based on the structural characteristics of the truss and measurement requirements; S3.2 Preliminary measurement: Perform preliminary measurement of the truss and collect the actual three-dimensional coordinate data of each node as much as possible. i For the parts that cannot be directly measured due to occlusion, the position and outline are recorded to provide a basis for subsequent inference and connection. The measurement error is δr i =r i ″-r i ′, requiring ||δr i ||≤∈, ∈ is the allowable error threshold; S3.3 Environmental interference control: During the measurement process, pay attention to controlling the impact of environmental factors on the measurement data, including temperature T, humidity H, vibration v vib .

6. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, characterized in that: Step S4 specifically includes: S4.1 Contour trajectory analysis: Use advanced image processing and pattern recognition technology to analyze the truss contour trajectory obtained in the preliminary measurement, identify the contour features of the obscured parts, and provide data support for the inferred connection supplement. Let the contour feature vector be f, which is obtained through the feature extraction algorithm; S4.2 Infer connection supplement: Based on the structural characteristics of the truss and the coordinate information of the known nodes, combined with the results of the contour trajectory analysis, the connection relationship and node position of the obscured part are inferred, and the mechanical model is used for mechanical analysis and optimization to ensure the rationality and reliability of the supplemented part; let the inferred node coordinates be r pred , the structural mechanics equilibrium equation Ku=F must also be satisfied, where K is the stiffness matrix, u is the displacement vector, and F is the external force vector; S4.3 Verification of Supplementation: Verify the presumed supplemented parts to ensure that they are consistent with the actual structure and meet the stability requirements. Use simulation software or actual testing to verify and compare the structural performance differences before and after the supplementation.

7. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, characterized in that: Step S5 specifically includes: Multi-view measurement: Using dynamic multi-view measurement method, by moving the measuring equipment or adjusting the measuring angle, the truss is measured in all directions and at multiple angles, ensuring that each node can be measured from at least two different perspectives, thus improving the accuracy and reliability of the measurement. Let the perspective vector be v view , measurement matrix M; Data integration: Use data fusion algorithm to integrate the three-dimensional coordinate data obtained from different perspectives, consider the errors and differences between the measurement data from different perspectives, perform error compensation and optimization, and form a complete and accurate three-dimensional coordinate data set R = {r final，i }; Design parameters: weight vector ω, satisfying Among them, n view is the number of viewing angles, r il is the measured coordinate of node i at the lth viewing angle.

8. The complex space truss positioning method based on three-dimensional coordinate acquisition technology according to claim 1, characterized in that: Step S6 specifically includes: Data analysis: Let the eigenvector X feat To extract features from the integrated 3D coordinate data, including coordinate values, relative distances between coordinates, angles, speeds, and accelerations, the selection of features is determined based on the specific application scenario and the requirements of the classification and recognition tasks. Let the label vector Y be the corresponding label assigned to the 3D coordinate data, including category labels and numerical labels. The selection of labels is consistent with the goals of classification, recognition, and prediction. Training model: Select a machine learning algorithm based on task requirements; use the feature vector X feat And label vector Y train the model to obtain model parameters θ; use the trained model and model parameters θ to classify, identify and predict the new three-dimensional coordinate data; input the new feature vector X feat , the model outputs the corresponding prediction results; compares the model's prediction results with the theoretical model, and calculates the deviation and error between the measured data and the theoretical model; the deviation and error are expressed as the difference between the predicted value and the theoretical value; analyzes the source of the deviation and error and finds out the possible cause; Mechanical properties and stability modeling: Model and predict the mechanical properties and stability of truss structures using machine learning algorithms or traditional mechanical methods to ensure that the adjusted truss structures meet mechanical requirements and stability standards. Mechanical properties include stiffness and damping, and stability indicators include critical load and instability mode. Positioning adjustment: Use the trained machine learning model or traditional mechanical methods to input new mechanical model parameters and predict the mechanical response of the truss structure; Stability assessment: Determine whether the structure is stable by comparing the predicted critical load with the actual load; fine-tune the node positions and angle parameters of the truss to ensure accurate docking and installation of each node; consider the impact of environmental factors on the truss structure and improve the stability of the structure through environmental monitoring and compensation technology; Assume that the adjustment vector is ΔR={Δr i }, satisfying the adjusted node coordinate r adj，i =r final，i +Δr i ; Verification and feedback: Verify the adjusted truss to ensure that it meets the design requirements and usage standards. Verification indicators include position accuracy, stability, and feedback vector; collect feedback data f back , including verification results and improvement suggestions, providing an improvement basis for subsequent measurement and positioning.

9. A complex space truss positioning system for implementing the complex space truss positioning method based on three-dimensional coordinate acquisition technology according to any one of claims 1 to 8, characterized in that: include: 3D modeling module, used to build 3D models of complex space trusses based on drawings; 3D control network construction and dynamic calibration module, used to construct a 3D control network at the construction site and perform dynamic calibration; On-site measurement and data acquisition module, used to perform preliminary measurements of the truss on site and collect the actual three-dimensional coordinate data of each node; The contour trajectory estimation and supplementation module is used to estimate the contour trajectory of the parts that cannot be directly measured in the preliminary measurement and supplement the connection relationship; The multi-view measurement and data integration module uses a dynamic multi-view measurement method to measure the truss in all directions and at multiple angles and integrate the data; The data analysis and positioning adjustment module is used to analyze the integrated three-dimensional coordinate data and perform positioning adjustments based on the analysis results.

10. The complex space truss positioning system according to claim 9, characterized in that: The system's data processing is processed and stored on the cloud platform. The 3D model built according to the drawing and the 3D model constructed by actual measurement are displayed separately on the same interface, and the dimensions and error data are displayed.

Images