Control Method, Device and Storage Medium for 3D Model of Smart City Space

Abstract

The present invention discloses a control method, device, and storage medium for a 3D model of a smart city space, including the following steps: S1, acquiring and preprocessing real-time data of the smart city space; S2, constructing a dynamic graph convolutional network to map image and point cloud data, initializing the 3D model and establishing a topological structure; S3, calculating the change rate of sensor data in combination with time characteristics to update the topological structure and state; S4, using a gated graph attention network to optimize rotation, scaling, displacement, and deformation parameters and adjust the motion trajectory and form; S5, compressing features using a variable topology aggregation mechanism to optimize the topological structure, boundary curvature, and mesh density; S6, outputting to a mobile terminal to achieve visualization and multi-terminal interaction; S7, storing and updating the 3D model to optimize feature learning and adaptive adjustment strategies. The present invention improves the dynamic adaptability and interaction accuracy of the 3D model, and realizes efficient modeling, optimization control, and multi-terminal collaborative operation of the smart city space.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent space and digital twin, and particularly to a control method, device, and storage medium for a 3D model of a smart city space. Background Art

[0002] In the construction and management of a smart city space, 3D models are widely used as a core data representation form in multiple fields such as building information modeling (BIM), intelligent manufacturing, digital twin, and smart cities. The existing control technologies for 3D models mainly rely on rule setting, manual adjustment, or predefined parameter optimization, and cannot achieve adaptive adjustment to complex dynamic environments, which limits their intelligent applications in smart city spaces. Especially when it comes to real-time interaction, precise control, and environmental adaptive optimization of large-scale 3D data, traditional methods have many limitations and are difficult to meet the requirements of high-precision, low-latency, and real-time adjustable 3D model control in smart city spaces.

[0003] Current mainstream 3D model control methods rely on traditional methods based on geometric constraints and mathematical transformations, such as using mathematical models like Euler angles and quaternions to achieve basic operations such as rotation, scaling, and displacement. These methods are usually applicable to static scenarios or manual interactions, but when 3D models need to be adaptively adjusted according to sensor data or external environmental changes, traditional mathematical transformation methods cannot effectively handle complex dynamic relationships. For example, in an intelligent factory or autonomous driving simulation, 3D models need to be updated in real time with the change of sensor data, but traditional methods cannot establish a mapping relationship from environmental data to 3D model control parameters, resulting in lag or distortion of model adjustment. In addition, traditional 3D model control methods usually rely on predefined rules or fixed thresholds and lack self-learning and optimization capabilities, making the adaptability of the models poor and unable to meet the real-time and highly dynamic change requirements of smart city spaces.

[0004] In recent years, the development of deep learning and computer vision technologies has provided new solutions for 3D model control. 3D model optimization methods based on deep learning technologies such as convolutional neural networks (CNNs) and graph neural networks (GNNs) can achieve intelligent model adjustment by extracting features from inputs such as images and point cloud data. However, existing deep learning methods still face many challenges when dealing with 3D models in smart city spaces. First, existing 3D deep learning methods are usually trained based on static datasets and are difficult to adapt to real-time changes in smart city spaces, resulting in lag in model updates. Second, when deep learning models process 3D topological structures, they usually need to regularize the models, such as using uniform sampling or fixed grid division, but this method is prone to losing key geometric details, resulting in a decrease in 3D model control accuracy.

[0005] In addition, most existing 3D deep learning methods are based on global feature learning, ignoring the local topological information of 3D models, which makes it easy to have problems such as local structure distortion or detail loss when adjusting the models.

[0006] In the smart city space, the real-time control of 3D models requires the integration of multi-source data such as Internet of Things sensors, point cloud scanning, and cameras to achieve dynamic adjustment of the models. However, the existing 3D model control methods have weak fusion capabilities for multi-source data and are difficult to achieve real-time optimization based on sensor feedback. For example, there are significant differences in the format, dimension, and temporal characteristics between the point cloud data obtained by lidar and the image data collected by cameras. When integrating these data, the existing 3D model control methods usually adopt simple interpolation or rule matching methods, which are difficult to accurately represent the dynamic changes in the smart city space. In addition, the Internet of Things sensor data has a high degree of temporal correlation, and the existing 3D model control methods lack the ability to model temporal information, resulting in the inability to accurately predict future states when dealing with the dynamic changes in the smart city space, making the adjustment of 3D models lag or unstable.

[0007] Therefore, how to provide control methods, devices, and storage media for 3D models in the smart city space is an urgent problem that needs to be solved by those skilled in the art. Summary of the Invention

[0008] An object of the present invention is to propose a control method for 3D models in the smart city space. The present invention combines a dynamic graph convolutional network, a gated graph attention network, and a variable topology aggregation mechanism, and details the dynamic adjustment, intelligent optimization, and interactive control methods of 3D models in the smart city space, with the advantages of high real-time performance, high accuracy, and strong adaptability.

[0009] According to the control method of the 3D model in the smart city space according to the embodiments of the present invention, the following steps are included:

[0010] S1. Obtain the real-time data of the smart city space and preprocess the real-time data;

[0011] S2. Construct a dynamic graph convolutional network, map the preprocessed image data and point cloud data to a unified feature space, calculate the spatial structure information, initialize the 3D model, and establish the topological structure of the 3D model;

[0012] S3. Combine the time characteristics of the dynamic graph convolutional network and the preprocessed sensor data, calculate the change rate of the sensor data, update the topological structure and state change trend of the 3D model, and make the 3D model keep dynamic consistency with the smart city space;

[0013] S4. Based on the preprocessed instruction parameters, calculate the difference between the control target and the 3D model state, use the improved gated graph attention network to calculate the importance weights of the control target in the topological structure, adjust the rotation, scaling, displacement, and deformation parameters of the 3D model, optimize the motion trajectory of the 3D model, and adjust the change range of the 3D model's shape based on the topological structure;

[0014] S5. Use the improved variable topology aggregation mechanism to perform feature compression on the 3D model topological structure, extract local geometric features, and adjust the edge connection method between adjacent vertices based on the deformation distribution of the motion trajectory to correct the boundary curvature and mesh density of the 3D model;

[0015] S6. Output the 3D model to the mobile device to achieve visual display, and synchronize it to the remote computing platform to support multi-user collaborative interaction, achieve multi-person remote sharing control, and record interaction data;

[0016] S7. Store the 3D model state, input the interaction data into the dynamic graph convolutional network, update the feature learning ability of the 3D model, optimize the adaptive adjustment strategy, and improve the response accuracy to the real-time environment and the interaction experience.

[0017] Optionally, the real-time data includes Internet of Things sensor data, image data collected by a camera, point cloud data obtained by a lidar, and control instructions input by a user. The preprocessing includes performing noise filtering and normalization processing on the sensor data, performing denoising, histogram equalization, and feature extraction on the image data, performing downsampling, surface reconstruction, and normal vector calculation on the point cloud data, and parsing the control instructions input by the user to extract rotation, scaling, displacement, and deformation parameters.

[0018] Optionally, the specific steps of S2 include:

[0019] S21. Based on the preprocessed image data and point cloud data, construct the 3D model topological structure , where is the vertex set in the 3D model, is the edge set between vertices, and initialize the feature vector of each vertex and the feature vector of each edge;

[0020] S22. Use the dynamic graph convolutional network to perform topological structure learning on , calculate the connection relationship between vertices using the dynamic adjacency matrix, define as the association weight between vertex and , and adjust the adjacency matrix through normalization:

[0021] ;

[0022] Among them, is the normalized adjacency matrix, is the adjacency matrix at time is the degree matrix of , and ;

[0023] S23. Update the eigenvector of each vertex. Adopt the dynamic graph convolution calculation method, and perform feature aggregation through the message passing mechanism to calculate the vertex state:

[0024] ;

[0025] Among them, is the eigenvector of vertex at time, is the training parameter, is the non-linear activation function, is the neighborhood set of vertex , is the normalized correlation weight between vertex and , is the eigenvector of vertex at time;

[0026] S24. Update the topological structure of the 3D model by updating the eigenvector of the vertex , and calculate the spatial structure information:

[0027] ;

[0028] Among them, is the local curvature value of the 3D model vertex at time, is the weight between vertex and , is the curvature value at vertex , is the curvature value at vertex .

[0029] Optionally, the S3 specifically includes:

[0030] S31. Define the sensor data set as , where represents the data of the th sensor at time, is the number of sensors, and calculate the state vector at the current time step:

[0031] ;

[0032] where is the state vector at time is the mapping weight matrix, is the bias term;

[0033] S32. Calculate the rate of change between the state vector at the current time step and the state vector at the previous time step, and define the rate of change calculation formula:

[0034] ;

[0035] where is the state vector at time is the rate of change;

[0036] Take as the input of the dynamic graph convolutional network, adjust the adjacency matrix in the topological structure, and update the adjacency relationship:

[0037] ;

[0038] where is the normalized adjacency matrix, is the adjacency matrix at time is the topological structure adjustment matrix calculated according to , is the degree matrix;

[0039] S33. Based on the updated topological structure , use the dynamic graph convolutional network to calculate the state change trend, and define the vertex feature update formula:

[0040] ;

[0041] where is the feature vector of vertex at time , is the training parameter, is the non-linear activation function, is the neighborhood set of vertex , is the normalized correlation weight between vertex and , is the feature vector of vertex at time The feature vector at a moment, is the bias term;

[0042] S34. Calculate the change in the spatial position of the updated vertex feature in the 3D model, and define the vertex position update formula in the three-dimensional space:

[0043] ;

[0044] where, is the coordinate position of vertex at the moment, is the coordinate position of vertex at the moment, is the step size adjustment coefficient, represents the position change amount of vertex , is the local curvature value of the 3D model vertex at the moment;

[0045] S35. Adjust the 3D model state to keep the 3D model in dynamic consistency with the smart city space.

[0046] Optionally, the S4 specifically includes:

[0047] S41. Based on the 3D model topology and the preprocessed instruction parameters , define the control target set:

[0048] ;

[0049] where, is the number of control targets, represents the th control target at the moment, and the control target includes rotation, scaling, displacement and deformation parameters;

[0050] S42. Use the improved gated graph attention network to calculate the importance weight of the control target in the topology, and define the influence weight of the control target on the vertex:

[0051] ;

[0052] where, is the control target, is the vertex, is the weight of the control target on the vertex , is the vertex At time step The eigenvector of is the vertex The neighborhood set of is the natural exponential function with base e is the interaction function of the control target and the vertex feature, defined as:

[0053] ;

[0054] where and are trainable parameter matrices is the bias term;

[0055] S43. Adjust the rotation, scaling, displacement, and deformation parameters of the 3D model based on the influence weight of the control target on the vertex, and define the control update matrix:

[0056] ;

[0057] where is the new position of the vertex at time step , is the original position at time step , is the number of control targets;

[0058] S44. Calculate the adjusted 3D model motion trajectory, and define the trajectory smoothing optimization objective function:

[0059] ;

[0060] where is the trajectory smoothing loss and respectively represent the new positions of adjacent vertices and at time step , is the vertex set in the 3D model is the neighborhood set of the vertex , represents the square of the Euclidean norm;

[0061] S45. Adjust the morphological change range of the 3D model through the motion trajectory, and calculate the final morphological adjustment function:

[0062] ;

[0063] where is the final morphological adjustment position of the vertex at time step ​ is the morphological change control coefficient.

[0064] Optionally, the S5 specifically includes:

[0065] S51. Based on the 3D model topology and the adjacency matrix , form the topological feature matrix with the eigenvectors of each vertex at time:

[0066] ;

[0067] Among them, is the vertex feature matrix, is the number of vertices at indicating the eigenvector of vertex at time step ;

[0068] Improve the variable topology aggregation mechanism and adaptively update the topological feature matrix:

[0069] ;

[0070] Among them, is the topological feature matrix at time step , is the trainable weight matrix, is the transpose matrix of is the topological level adjustment parameter, is the adjacency matrix at the topological adjacency matrix of the th layer,

[0071] S52. Based on the topological aggregation matrix, obtain the motion trajectory matrix and calculate the motion deformation influence factor:

[0072] ;

[0073] ;

[0074] Among them, is the motion trajectory matrix at time step , is the motion trajectory feature of vertex at time step , is the number of vertices at is the time step The motion deformation influence factor, is the vertex of the neighborhood set, is the motion deformation control parameter, is the natural exponential function with base e, represents the square of the Euclidean norm;

[0075] S53. Combine the motion trajectory matrix and the motion deformation influence factor to calculate the boundary curvature adjustment matrix and the mesh density distribution at time step and optimize the mesh density:

[0076] ;

[0077] ;

[0078] ;

[0079] Among them, is the boundary curvature adjustment matrix at time step , is the degree matrix, is the Laplacian matrix, is the curvature smoothing coefficient, is the mesh density at time step , is the mesh density control parameter, is the vertex set in the 3D model, is the mesh density loss, is the target mesh density;

[0080] S54. Optimize the topological structure based on the boundary curvature and the mesh density, and update the 3D model state:

[0081] ;

[0082] Among them, is the 3D model state calculated at time step , is the identity matrix, and are the balance coefficients, is the degree matrix at time step , is the boundary curvature adjustment matrix calculated at time step , is the optimization step size, is the mesh density gradient descent adjustment term.

[0083] According to the control device of the 3D model of the smart city space in the embodiment of the present invention, it includes:

[0084] A network construction module, which is used to construct a dynamic graph convolutional network, map the preprocessed image data and point cloud data to a unified feature space, calculate spatial structure information, initialize a 3D model, and establish the topological structure of the 3D model;

[0085] A state update module, which is used to combine the time characteristics of the dynamic graph convolutional network and the preprocessed sensor data, calculate the change rate of the sensor data, update the topological structure and state change trend of the 3D model, and keep the 3D model in dynamic consistency with the smart city space;

[0086] An adjustment calculation module, which is used to calculate the difference between the control target and the 3D model state based on the preprocessed instruction parameters, calculate the importance weight of the control target in the topological structure by using an improved gated graph attention network, adjust the rotation, scaling, displacement and deformation parameters of the 3D model, optimize the motion trajectory of the 3D model, and adjust the change range of the 3D model shape based on the topological structure;

[0087] A topology optimization module, which is used to perform feature compression on the 3D model topological structure by using an improved variable topology aggregation mechanism, extract local geometric features, and adjust the edge connection mode between adjacent vertices based on the deformation distribution of the motion trajectory, and correct the boundary curvature and mesh density of the 3D model;

[0088] A synchronization and sharing module, which is used to output the 3D model to the mobile terminal for visual display, and synchronize it to the remote computing platform to support multi-user collaborative interaction, realize multi-person remote shared control, and record interaction data;

[0089] A storage and update module, which is used to store the 3D model state, input the interaction data into the dynamic graph convolutional network, update the feature learning ability of the 3D model, optimize the adaptive adjustment strategy, and improve the response accuracy to the real-time environment and the interaction experience.

[0090] According to an embodiment of the present invention, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor can execute the control method of the 3D model of the smart city space.

[0091] The beneficial effects of the present invention are:

[0092] First of all, the present invention establishes the time-dependent relationship of the 3D model through the dynamic graph convolutional network, enabling the model to perform adaptive adjustment based on historical sensor data and current environmental data, improving the real-time performance of model control, enabling the 3D model to accurately respond to changes in the external environment, and ensuring the continuity and stability of the adjustment process.

[0093] Secondly, by using the gated graph attention network, the present invention optimizes the dynamic distribution of the control target in the 3D model topological structure, enabling precise adjustment of rotation, scaling, displacement, and deformation parameters, avoiding the problems of local structure distortion or control response lag in traditional methods, improving the interaction accuracy, and making the operation of the 3D model smoother and more natural.

[0094] Finally, the present invention adopts a variable topology aggregation mechanism to finely optimize the local structure of the 3D model, enabling uniform grid density and smooth boundary curvature to be maintained during the model deformation process, enhancing the geometric stability and morphological consistency of the 3D model, thereby improving its applicability in complex smart city spaces, achieving high-precision, real-time, and intelligent dynamic adjustment of the 3D model, and enhancing the digital management and interaction experience of the smart city space. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0096] Figure 1 is a flowchart of the control method for the 3D model in the smart city space proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0097] The present invention will now be described in further detail with reference to the drawings. These drawings are all simplified schematic diagrams, only showing the basic structure of the present invention in a schematic manner, and thus only showing the components related to the present invention.

[0098] Refer to Figure 1 , the control method for the 3D model in the smart city space includes the following steps:

[0099] S1. Obtain the real-time data of the smart city space and preprocess the real-time data;

[0100] S2. Construct a dynamic graph convolutional network, map the preprocessed image data and point cloud data to a unified feature space, calculate the spatial structure information, initialize the 3D model, and establish the topological structure of the 3D model;

[0101] S3. Combine the time characteristics of the dynamic graph convolutional network and the preprocessed sensor data, calculate the change rate of the sensor data, update the topological structure and state change trend of the 3D model, and keep the 3D model in dynamic consistency with the smart city space;

[0102] S4. Calculate the difference between the control target and the 3D model state based on the preprocessed instruction parameters, calculate the importance weights of the control target in the topological structure using the improved gated graph attention network, adjust the rotation, scaling, displacement, and deformation parameters of the 3D model, optimize the motion trajectory of the 3D model, and adjust the change range of the 3D model's shape based on the topological structure;

[0103] S5. Use the improved variable topology aggregation mechanism to perform feature compression on the 3D model topological structure, extract local geometric features, and adjust the edge connection method between adjacent vertices based on the deformation distribution of the motion trajectory to correct the boundary curvature and mesh density of the 3D model;

[0104] S6. Output the 3D model to the mobile terminal to achieve visual display, and synchronize it to the remote computing platform to support multi-user collaborative interaction, achieve multi-person remote sharing control, and record interaction data;

[0105] S7. Store the 3D model state, input the interaction data into the dynamic graph convolutional network, update the feature learning ability of the 3D model, optimize the adaptive adjustment strategy, and improve the response accuracy to the real-time environment and the interaction experience.

[0106] In this embodiment, the real-time data includes Internet of Things sensor data, image data collected by a camera, point cloud data obtained by a lidar, and control instructions input by a user. The preprocessing includes performing noise filtering and normalization processing on the sensor data, performing denoising, histogram equalization, and feature extraction on the image data, performing downsampling, surface reconstruction, and normal vector calculation on the point cloud data, and parsing the control instructions input by the user to extract rotation, scaling, displacement, and deformation parameters.

[0107] In this embodiment, the specific steps of S2 are as follows:

[0108] S21. Based on the preprocessed image data and point cloud data, construct a 3D model topological structure , where is the vertex set in the 3D model, is the edge set between vertices, and initialize the feature vector of each vertex and the feature vector of each edge;

[0109] S22. Use the dynamic graph convolutional network to perform topological structure learning on , calculate the connection relationship between vertices using the dynamic adjacency matrix, define as the association weight between vertex and , and adjust the adjacency matrix through normalization:

[0110] ;

[0111] Among them, is the normalized adjacency matrix, is the adjacency matrix at time is the degree matrix of , and ;

[0112] S23. Update the eigenvector of each vertex. Adopt the dynamic graph convolution calculation method and perform feature aggregation through the message passing mechanism to calculate the vertex state:

[0113] ;

[0114] Among them, is the eigenvector of vertex at time, is the training parameter, is the non-linear activation function, is the neighborhood set of vertex , is the normalized correlation weight between vertex and , is the eigenvector of vertex at time;

[0115] S24. Update the topological structure of the 3D model by updating the eigenvector of the vertex, and calculate the spatial structure information:

[0116] ;

[0117] Among them, is the local curvature value of the 3D model vertex at time, is the weight between vertex and , is the curvature value at vertex , is the curvature value at vertex .

[0118] In this embodiment, the specific steps of S3 are as follows:

[0119] S31. Define the sensor data set as , where represents the data of the th sensor at time, Let \(n\) be the number of sensors, and calculate the state vector at the current time step:

[0120] ;

[0121] where, is the state vector at time , is the mapping weight matrix, is the bias term;

[0122] S32. Calculate the rate of change between the state vector at the current time step and the state vector at the previous time step, and define the rate of change calculation formula:

[0123] ;

[0124] where, is the state vector at time , is the rate of change;

[0125] Take as the input of the dynamic graph convolutional network, adjust the adjacency matrix in the topological structure, and update the adjacency relationship:

[0126] ;

[0127] where, is the normalized adjacency matrix, is the adjacency matrix at time , is the topological structure adjustment matrix calculated according to , is the degree matrix;

[0128] S33. Based on the updated topological structure , use the dynamic graph convolutional network to calculate the state change trend, and define the vertex feature update formula:

[0129] ;

[0130] where, is the feature vector of vertex at time , is the training parameter, is the non-linear activation function, is the neighborhood set of vertex , is the normalized correlation weight between vertex and , is the feature vector of vertex at time The feature vector at a moment, is the bias term;

[0131] S34. Calculate the change in the spatial position of the updated vertex feature in the 3D model, and define the vertex position update formula in three-dimensional space:

[0132] ;

[0133] where, is the coordinate position of vertex at moment is the coordinate position of vertex at moment is the step adjustment coefficient, represents the position change amount of vertex , is the local curvature value of the 3D model vertex at moment;

[0134] S35. Adjust the 3D model state to keep the 3D model in dynamic consistency with the smart city space.

[0135] Optionally, the S4 specifically includes:

[0136] S41. Based on the 3D model topology and the preprocessed instruction parameters , define the control target set:

[0137] ;

[0138] where, is the number of control targets, represents the th control target at moment

[0139] S42. Use the improved gated graph attention network to calculate the importance weight of the control target in the topology, and define the influence weight of the control target on the vertex:

[0140] ;

[0141] where, is the control target, is the vertex, is the weight of the control target on the vertex , is the vertex At time step The eigenvector of is the vertex The neighborhood set of is the natural exponential function with base e is the interaction function that controls the interaction between the target and vertex features, defined as:

[0142] ;

[0143] Among them, and are trainable parameter matrices, is the bias term;

[0144] S43. Based on the influence weight of the control target on the vertex, adjust the rotation, scaling, displacement, and deformation parameters of the 3D model, and define the control update matrix:

[0145] ;

[0146] Among them, is the vertex At time step The new position of is the time step The original position of is the number of control targets;

[0147] S44. Calculate the adjusted 3D model motion trajectory, and define the trajectory smoothing optimization objective function:

[0148] ;

[0149] Among them, is the trajectory smoothing loss, and respectively represent the new positions of adjacent vertices and At time step The new position of is the vertex set in the 3D model, is the vertex The neighborhood set of represents the square of the Euclidean norm;

[0150] S45. Adjust the morphological change range of the 3D model through the motion trajectory, and calculate the final morphological adjustment function:

[0151] ;

[0152] Among them, is the vertex At time step The final morphological adjustment position of is the morphological change control coefficient.

[0153] In this embodiment, S5 specifically includes:

[0154] S51. Based on the 3D model topological structure and the adjacency matrix , form a topological feature matrix with the feature vectors of each vertex at time:

[0155] ;

[0156] Among them, is the vertex feature matrix, is the number of vertices at represents the vertex at the time step feature vector;

[0157] Improve the variable topology aggregation mechanism and adaptively update the topological feature matrix:

[0158] ;

[0159] Among them, is the topological feature matrix at the time step , is the trainable weight matrix, is the transpose matrix of is the topological hierarchy adjustment parameter, is the adjacency matrix at is the topological adjacency matrix of the layer,

[0160] S52. Based on the topological aggregation matrix, obtain the motion trajectory matrix and calculate the motion deformation influence factor:

[0161] ;

[0162] ;

[0163] Among them, is the motion trajectory matrix at the time step , is the motion trajectory feature of the vertex at the time step , is the number of vertices at is the time step The motion deformation influence factor is the vertex of the neighborhood set is the motion deformation control parameter is the natural exponential function with base e represents the square of the Euclidean norm

[0164] S53. Combine the motion trajectory matrix and the motion deformation influence factor to calculate the boundary curvature adjustment matrix and the grid density distribution at time step and optimize the grid density:

[0165] ;

[0166] ;

[0167] ;

[0168] where is the boundary curvature adjustment matrix at time step , is the degree matrix is the Laplacian matrix is the curvature smoothing coefficient is the grid density at time step , is the grid density control parameter is the vertex set in the 3D model is the grid density loss is the target grid density

[0169] S54. Optimize the topological structure based on the boundary curvature and the grid density, and update the 3D model state:

[0170] ;

[0171] where is the 3D model state calculated at time step , is the identity matrix and are the balance coefficients is the degree matrix at time step , is the boundary curvature adjustment matrix calculated at time step , is the optimization step size is the grid density gradient descent adjustment term

[0172] The control device of the 3D model in the smart city space includes:

[0173] A data processing module, configured to obtain real-time data of the smart city space and preprocess the real-time data;

[0174] A network construction module, configured to construct a dynamic graph convolutional network, map the preprocessed image data and point cloud data to a unified feature space, calculate spatial structure information, initialize a 3D model, and establish a topological structure of the 3D model;

[0175] A state update module, configured to combine the time characteristics of the dynamic graph convolutional network and the preprocessed sensor data, calculate the change rate of the sensor data, update the topological structure and state change trend of the 3D model, and keep the 3D model in dynamic consistency with the smart city space;

[0176] An adjustment calculation module, configured to calculate the difference between the control target and the 3D model state based on the preprocessed instruction parameters, calculate the importance weight of the control target in the topological structure by using an improved gated graph attention network, adjust the rotation, scaling, displacement and deformation parameters of the 3D model, optimize the motion trajectory of the 3D model, and adjust the change range of the 3D model shape based on the topological structure;

[0177] A topology optimization module, configured to perform feature compression on the 3D model topological structure by using an improved variable topology aggregation mechanism, extract local geometric features, and adjust the edge connection mode between adjacent vertices based on the deformation distribution of the motion trajectory, and correct the boundary curvature and mesh density of the 3D model;

[0178] A synchronization and sharing module, configured to output the 3D model to a mobile terminal for visual display, and synchronize it to a remote computing platform to support multi-user collaborative interaction, realize multi-person remote shared control, and record interaction data;

[0179] A storage and update module, configured to store the 3D model state, input the interaction data into the dynamic graph convolutional network, update the feature learning ability of the 3D model, optimize the adaptive adjustment strategy, and improve the response accuracy to the real-time environment and the interaction experience.

[0180] A computer-readable storage medium storing a computer program, which when executed by a processor enables the processor to execute a control method for a 3D model of a smart city space.

[0181] Embodiment 1:

[0182] To verify the feasibility of the present invention in implementation, the present invention is applied to the intelligent manufacturing system of a certain smart industrial park, which includes multiple automated production devices, industrial robots, Internet of Things sensors, and a 3D visualization monitoring platform. In this park, 3D models are used to display the operating status of equipment, production process flows, and spatial layouts in real time, helping management personnel with remote monitoring, production optimization, and anomaly warning. However, the traditional 3D model control method has serious lag, and cannot adjust the model status in real time according to the operating conditions of the equipment, resulting in information lag and affecting the precise decision-making of intelligent manufacturing. The present invention realizes the intelligent dynamic adjustment of the 3D model through a dynamic graph convolutional network, a gated graph attention network, and a variable topology aggregation mechanism, enabling the 3D model to respond to equipment status changes in real time, and improving the operating efficiency and intelligent level of the system.

[0183] In this smart industrial park, each device is equipped with high-precision Internet of Things sensors, including temperature sensors, pressure sensors, vibration sensors, and current monitoring devices. These sensors collect data at time intervals of 0.1 second and transmit the data to the central control system through an industrial network. The control method of the traditional 3D model mainly relies on manual adjustment or rule matching. Usually, management personnel need to manually modify model parameters based on sensor data, or use preset rules to automatically execute adjustments. However, the update frequency of this method is relatively low, usually between 30 seconds and 1 minute, resulting in poor real-time performance of the model and unable to meet the precise control requirements of intelligent manufacturing.

[0184] The present invention realizes precise adjustment of the model by constructing a dynamic graph convolutional network to perform time series modeling on equipment operation data and combining a gated graph attention network to calculate the importance weights of equipment status changes on the 3D model topology. For example, when a certain industrial robot has a joint angle deviation due to a mechanical failure, the dynamic graph convolutional network of the present invention can detect the joint deviation situation based on the abnormal signal of the vibration sensor and complete the adaptive adjustment of the model within 0.5 second, enabling the 3D model to accurately reflect the actual state of the robot and avoiding misjudgment and scheduling errors caused by model information lag. In addition, by adopting a variable topology aggregation mechanism, the present invention dynamically optimizes the grid density and boundary curvature during the model adjustment process, enabling the 3D model to maintain good geometric accuracy and structural stability under high-frequency updates.

[0185] During the actual test process, the application scenarios of the present invention in this smart industrial park include multiple aspects such as equipment operation status visualization, production process dynamic adjustment, and anomaly monitoring. To more intuitively demonstrate the advantages of the present invention, the following data table compares the performance of the present invention and traditional methods in different scenarios of the smart industrial park:

[0186] Table 1 Comparison data table of 3D model control in the smart industrial park

[0187] ;

[0188] By comparing the traditional method and the method of the present invention, the model adjustment speed, accuracy, and computational overhead under different operating states were evaluated. As can be seen from Table 1, when 200 devices are running simultaneously, the update period of the traditional 3D model control method is 45 seconds, and the error range is between 0.8° and 3.5°. Moreover, when the device load exceeds 90%, the model synchronization update failure rate is as high as 17%. In contrast, the method of the present invention shortens the model update period to 2.3 seconds, reduces the error range to between 0.1° and 0.9°, and even when the devices are running at full load, the model synchronization failure rate is less than 1.2%.

[0189] Meanwhile, in the production anomaly monitoring scenario, based on the dynamic graph convolutional network and the gated graph attention network, when the present invention detects abnormal vibrations in production equipment, it can complete the adjustment of the 3D model within 1.2 seconds and calculate possible fault causes in combination with historical data, enabling management personnel to locate the anomaly source within 3 minutes. In contrast, the traditional method usually requires at least 15 minutes of manual analysis to identify the fault point. In addition, the dynamic optimization mechanism of the model significantly improves the computational efficiency, reducing the computational load by approximately 46.5% compared to the traditional method, enabling the 3D model control system to operate more efficiently in the industrial field.

[0190] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, should be covered by the protection scope of the present invention.

Claims

1. Control method for the 3D model of the smart city space, characterized in that, It includes the following steps: S1. Obtain the real-time data of the smart city space and preprocess the real-time data; S2. Construct a dynamic graph convolutional network, map the preprocessed image data and point cloud data to a unified feature space, calculate the spatial structure information, initialize the 3D model, and establish the topological structure of the 3D model; S3. Combine the time characteristics of the dynamic graph convolutional network and the preprocessed sensor data, calculate the change rate of the sensor data, update the topological structure and the state change trend of the 3D model, and make the 3D model maintain dynamic consistency with the smart city space; S4. Based on the preprocessed instruction parameters, calculate the difference between the control target and the state of the 3D model, use the improved gated graph attention network to calculate the importance weight of the control target in the topological structure, adjust the rotation, scaling, displacement and deformation parameters of the 3D model, optimize the motion trajectory of the 3D model, and adjust the change range of the 3D model shape based on the topological structure, where the control target includes rotation, scaling, displacement and deformation parameters; S5. Use the improved variable topology aggregation mechanism to perform feature compression on the 3D model topological structure, extract local geometric features, and based on the deformation distribution of the motion trajectory, adjust the edge connection method between adjacent vertices, and correct the boundary curvature and mesh density of the 3D model; S6. Output the 3D model to the mobile terminal to achieve visual display, and synchronize it to the remote computing platform, support multi-user collaborative interaction, achieve multi-person remote shared control, and record the interaction data; S7. Store the 3D model state, input the interaction data into the dynamic graph convolutional network, update the feature learning ability of the 3D model, optimize the adaptive adjustment strategy, and improve the response accuracy and interaction experience for the real-time environment; The specific content of S5 includes: S51. Based on the 3D model topology and the adjacency matrix , form a topological feature matrix with the eigenvectors of each vertex at moment: ； Among them, is the vertex feature matrix, is the number of vertices at time indicating the vertex at time step feature vector; Improve the variable topology aggregation mechanism and adaptively update the topological feature matrix: ； Among them, is the topological feature matrix at time step , is the trainable weight matrix, is 's transpose matrix, is the topological level adjustment parameter, is 's adjacency matrix at time is the th layer's topological adjacency matrix, is the number of topological levels; S52. Based on the topological aggregation matrix, obtain the motion trajectory matrix and calculate the motion deformation influence factor: ； ； Among them, is the motion trajectory matrix at time step . is the motion trajectory feature of vertex at time step . is the number of vertices at time is the motion deformation influence factor at time step . is the neighborhood set of vertex . is the motion deformation control parameter is the natural exponential function with base e represents the square of the Euclidean norm; S53. Calculate the boundary curvature adjustment matrix and grid density distribution of the time step by combining the motion trajectory matrix and the motion deformation influence factor, and optimize the grid density: ； ； ； Among them, is the boundary curvature adjustment matrix at time step , is the degree matrix, is the Laplacian matrix, is the curvature smoothing coefficient, is the time step of the mesh density below, is the mesh density control parameter, is the set of vertices in the 3D model, is the mesh density loss, is the target mesh density; S54. Optimize the topological structure based on the boundary curvature and mesh density, and update the 3D model state: ； Among them, is the time step The calculated 3D model state, is the identity matrix, and is the balance coefficient, is the time step The degree matrix of, is the time step The calculated boundary curvature adjustment matrix, is the optimization step size, is the grid density gradient descent adjustment term.

2. The control method of the 3D model of the smart city space according to claim 1, wherein The real-time data includes Internet of Things sensor data, image data collected by cameras, point cloud data obtained by lidar, and control instructions input by users. The preprocessing includes performing noise filtering and normalization processing on the sensor data, performing denoising, histogram equalization and feature extraction on the image data, performing downsampling, surface reconstruction and normal vector calculation on the point cloud data, and parsing the control instructions input by users to extract rotation, scaling, displacement and deformation parameters.

3. The control method of the 3D model of the smart city space according to claim 1, characterized in that, The specific content of S2 includes: S21. Construct a 3D model topological structure based on the preprocessed image data and point cloud data , where is the set of vertices in the 3D model, is the set of edges between vertices, and initialize the feature vector of each vertex and the feature vector ; S22. Use a dynamic graph convolutional network to perform topological structure learning, calculate the connection relationship between vertices using a dynamic adjacency matrix, and define as the association weight between vertex and , and adjust the adjacency matrix through a normalization operation: ； Among them, is the normalized adjacency matrix, is the adjacency matrix at time is the degree matrix of , and ; S23. Update the feature vector of each vertex using the dynamic graph convolution calculation method, and perform feature aggregation through the message passing mechanism to calculate the vertex state: ； Among them, is the vertex at the feature vector at the moment, is the training parameter, is the non - linear activation function, is the vertex the neighborhood set of, is the vertex and the normalized correlation weight between, is the vertex at the feature vector at the moment; S24. Update the topological structure of the 3D model by updating the feature vectors of the vertices , and calculate the spatial structure information: ； Among them, is the local curvature value of the 3D model vertex at time, is the weight between vertex and , is the curvature value at vertex , is the curvature value at vertex .

4. The control method of the 3D model of the smart city space according to claim 1, characterized in that The specific content of S3 includes: S31. Define the sensor data set as , where represents the data of the -th sensor at the moment, is the number of sensors, and calculate the state vector of the current time step: ； Among them, is the state vector at the moment, is the mapping weight matrix, is the bias term; S32. Calculate the change rate between the state vector of the current time step and the state vector of the previous time step, and define the change rate calculation formula: ； Among them, is the state vector at a moment, is the rate of change; Take as the input of the dynamic graph convolutional network, adjust the adjacency matrix in the topological structure, and update the adjacency relationship: ； Among them, is the normalized adjacency matrix, is the adjacency matrix at time is the topological structure adjustment matrix calculated according to ; is the degree matrix; S33. Based on the updated topological structure , use the dynamic graph convolutional network to calculate the state change trend, and define the vertex feature update formula: ； Among them, is the vertex at the feature vector at a moment, is the training parameter, is the non-linear activation function, is the vertex the neighborhood set of, is the vertex and the normalized correlation weight between, is the vertex at the feature vector at a moment, is the bias term; S34. Calculate the change in the spatial position of the updated vertex features in the 3D model, and define the position update formula of the vertex in the three-dimensional space: ； Among them, is the coordinate position of the time vertex, is the coordinate position of the time vertex, is the step adjustment coefficient, represents the position change amount of the vertex, is the local curvature value of the 3D model vertex at the time; S35. Adjust the 3D model state to make the 3D model maintain dynamic consistency with the smart city space.

5. The control method of the 3D model of the smart city space according to claim 1, wherein The specific content of S4 includes: S41. Based on the 3D model topology and the preprocessed instruction parameters , define the control target set: ； Among them, is the number of control targets, denotes the th control target at time, and the control targets include rotation, scaling, displacement and deformation parameters; S42. Use the improved gated graph attention network to calculate the importance weight of the control target in the topological structure, and define the influence weight of the control target on the vertex: ； Among them, is the control target, is the vertex, is the control target for the vertex weight, is the vertex at time step eigenvector, is the vertex neighborhood set, is the natural exponential function with base e, is the interaction function between the control target and the vertex feature, defined as: ； Among them, and are trainable parameter matrices, is the bias term; S43. Based on the influence weight of the control target on the vertex, adjust the rotation, scaling, displacement and deformation parameters of the 3D model, and define the control update matrix: ； Among them, is the vertex at the new position at time step , is the original position at time step , is the number of control targets; S44. Calculate the adjusted 3D model motion trajectory and define the trajectory smoothing optimization objective function: ； Among them, is the trajectory smoothing loss, and respectively represent the new positions of adjacent vertices and at time step . is the set of vertices in the 3D model, is the neighborhood set of vertex , represents the square of the Euclidean norm; S45. Adjust the morphological change range of the 3D model through the motion trajectory and calculate the final morphological adjustment function: ； Among them, is the vertex at the final form adjustment position in the time step , and is the form change control coefficient.

6. The control device of the 3D model of the smart city space, which executes the control method of the 3D model of the smart city space according to any one of claims 1 to 5, is characterized in that Including: A data processing module for obtaining real-time data of the smart city space and preprocessing the real-time data; A network construction module for constructing a dynamic graph convolutional network, mapping the preprocessed image data and point cloud data to a unified feature space, calculating spatial structure information, initializing the 3D model, and establishing the topological structure of the 3D model; A state update module for combining the time characteristics of the dynamic graph convolutional network and the preprocessed sensor data, calculating the change rate of the sensor data, updating the topological structure and state change trend of the 3D model, and keeping the 3D model in dynamic consistency with the smart city space; An adjustment calculation module for calculating the difference between the control target and the 3D model state based on the preprocessed instruction parameters, calculating the importance weight of the control target in the topological structure using an improved gated graph attention network, adjusting the rotation, scaling, displacement, and deformation parameters of the 3D model, optimizing the motion trajectory of the 3D model, and adjusting the morphological change range of the 3D model based on the topological structure, where the control target includes rotation, scaling, displacement, and deformation parameters; A topology optimization module for performing feature compression on the 3D model topological structure using an improved variable topology aggregation mechanism, extracting local geometric features, and adjusting the edge connection mode between adjacent vertices based on the deformation distribution of the motion trajectory, and correcting the boundary curvature and mesh density of the 3D model, including: Based on the topological structure of the 3D model and the adjacency matrix , the eigenvectors of each vertex at the moment are composed into a topological feature matrix: ； Among them, is the vertex feature matrix, is the number of vertices at time indicating the vertex at time step feature vector; Improve the variable topology aggregation mechanism and adaptively update the topology feature matrix: ； Among them, is the topological feature matrix at the time step , is the trainable weight matrix, is the transpose matrix of is the topological level adjustment parameter, is the adjacency matrix at the moment is the th layer of topological adjacency matrix, is the number of topological levels; Based on the topology aggregation matrix, obtain the motion trajectory matrix and calculate the motion deformation influence factor: ； ； Among them, is the motion trajectory matrix at time step . is the vertex 's motion trajectory feature at time step . is the number of vertices at time . is the motion deformation influence factor at time step . is the neighborhood set of vertex . is the natural exponential function with base e represents the square of the Euclidean norm; Calculate the boundary curvature adjustment matrix and grid density distribution of the time step by combining the motion trajectory matrix and the motion deformation influence factor, and optimize the grid density: ； ； ； Among them, is the boundary curvature adjustment matrix at time step , is the degree matrix, is the Laplacian matrix, is the curvature smoothing coefficient, is the time step of the mesh density, is the mesh density control parameter, is the set of vertices in the 3D model, is the mesh density loss, is the target mesh density; Optimize the topological structure based on the boundary curvature and mesh density and update the 3D model state: ； Among them, is the time step The calculated 3D model state, is the identity matrix, and is the balance coefficient, is the time step The degree matrix of, is the time step The calculated boundary curvature adjustment matrix, is the optimization step size, is the grid density gradient descent adjustment term; A synchronization and sharing module for outputting the 3D model to the mobile terminal to achieve visual display, synchronizing it to the remote computing platform, supporting multi-user collaborative interaction, achieving multi-person remote shared control, and recording interaction data; A storage and update module for storing the 3D model state, inputting the interaction data into the dynamic graph convolutional network, updating the feature learning ability of the 3D model, optimizing the adaptive adjustment strategy, and improving the response accuracy to the real-time environment and the interaction experience.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor can execute the control method of the 3D model of the smart city space according to any one of claims 1 to 5.

Images