Urban underground space planning collaborative management system based on live-action three-dimensional surveying and mapping

Abstract

The invention relates to the technical field of space three-dimensional data management, in particular to an urban underground space planning collaborative management system based on live-action three-dimensional surveying and mapping. The urban underground space planning collaborative management system based on live-action three-dimensional surveying and mapping comprises the following modules: a point cloud topology extraction module, a model topology anchoring module, a track topology adaptation module and a collaborative management output module. Wherein the point cloud topology extraction module is used for collecting live-action three-dimensional surveying and mapping data of an urban underground space, extracting underground entity three-dimensional topology features, associating reference coordinates of the live-action three-dimensional surveying and mapping data and generating a point cloud topology data set. According to the invention, by collecting the live-action three-dimensional surveying and mapping, identifying the entity three-dimensional topological features and establishing the live-action mapping three-dimensional network, the precise matching with the live-action surveying and mapping and the dynamic adaptation of the underground three-dimensional space are realized, so that the underground space planning fitting degree and the management reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of spatial three-dimensional data management technology, and in particular to a collaborative management system for urban underground space planning based on real-scene three-dimensional mapping. Background Technology

[0002] Real-scene 3D mapping technology is increasingly being applied to collaborative planning of urban underground space. Related technologies need to cover underground space data acquisition, topological feature processing, model adaptation, and dynamic monitoring to achieve a match between the planning scheme and the actual geometric structure. However, existing technologies have significant limitations in practical applications: Firstly, the integration of topological feature processing with real-scene data is low, lacking refined methods for dividing underground space geometric areas. The extracted 3D topological features are loosely correlated with the baseline coordinates of the real-scene mapping data, resulting in insufficient accuracy of point cloud data. Secondly, the lack of effective correlation methods in the model adaptation stage leads to low matching accuracy between the model and the real-scene 3D network, and significant deviations between planning parameters and actual structures. Thirdly, dynamic management efficiency is insufficient. During monitoring, differentiated strategies are not developed based on the complexity of the underground space geometry; a single acquisition mode is used, making it difficult to balance monitoring accuracy and efficiency. The methods for identifying and adjusting abnormal areas are crude and lack specificity, failing to adapt to changes in underground space geometry in a timely manner. Furthermore, data integration at each stage is weak and the correlations are not close, making it difficult to meet the overall needs for accurate planning and efficient management of underground space. Summary of the Invention

[0003] Therefore, it is necessary to provide a collaborative management system for urban underground space planning based on real-scene 3D mapping to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a collaborative management system for urban underground space planning based on real-scene 3D mapping includes: The point cloud topology extraction module is used to collect real-scene 3D mapping data of urban underground space, extract the 3D topological features of underground entities, associate the reference coordinates of the real-scene 3D mapping data, and generate a point cloud topology dataset. The model topology anchoring module is used to receive point cloud topology datasets, anchor the 3D topology features of entities to a preset real-world 3D network, configure visualization anchor points and establish a mapping relationship between visualization anchor points and point cloud topology datasets, and generate a real-world mapped 3D network. The trajectory topology adaptation module is used to access the real-scene mapping 3D network, identify the trajectory coordinates of point clouds, plan the underground space model based on the trajectory coordinates of point clouds and call the 3D topological features of the entity, and mark the trajectory surface of the underground space model. The collaborative management output module is used to continuously monitor the surface change rate of the trajectory surface within a preset monitoring time, integrate the underground space model, perform collaborative spatial planning management, and generate a real-world 3D planning model after collaborative planning management.

[0005] The beneficial effects of this invention are as follows: On the one hand, the point cloud topology extraction module accurately divides three types of geometric regions and extracts the three-dimensional topological features of underground entities. Combined with the model topology anchoring module, a mapping relationship between the visualization anchor points and the point cloud topology dataset is established, achieving accurate matching between topological features and the real-world three-dimensional network. The trajectory topology adaptation module ensures that the geometric parameters of underground space planning are highly consistent with the real-world three-dimensional mapping data through geometric verification, abnormal area identification, and point-by-point correction under the local geometric calibration coordinate system. This significantly improves the fit of the planning scheme to the actual underground geometric structure and reduces planning deviation.

[0006] On the other hand, the collaborative management output module sets differentiated data collection strategies based on the complexity of the underground space's geometric structure, enabling accurate and efficient monitoring of the trajectory surface change rate. By pre-setting tolerance geometric parameters, it makes targeted adjustments to abnormal areas and integrates global geometric data to optimize the connection relationship, forming a closed-loop collaborative management mechanism of "monitoring-identification-adjustment-verification". This ensures that the planning scheme can dynamically adapt to the geometric changes of the underground space, improving the reliability and operability of planning management. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the execution process of a collaborative management system for urban underground space planning based on real-scene 3D mapping. Figure 2 A schematic diagram showing the partitioning of the trajectory surface fit. Figure 3 A schematic diagram of the three-dimensional structural model of the underground space; Figure 4 This is an interface diagram of a collaborative management system for urban underground space planning based on real-scene 3D mapping. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0008] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0009] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0010] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0011] To achieve the above objectives, please refer to Figures 1 to 4 A collaborative management system for urban underground space planning based on real-scene 3D mapping is disclosed. The system comprises the following modules: a point cloud topology extraction module, a model topology anchoring module, a trajectory topology adaptation module, and a collaborative management output module. The system performs the following steps: S1: Collect real-scene 3D mapping data of urban underground space, extract 3D topological features of underground entities, associate the reference coordinates of real-scene 3D mapping data, and generate point cloud topology dataset; S2: Receive point cloud topology dataset, anchor the 3D topology features of entities to a preset real-world 3D network, configure visualization anchor points and establish a mapping relationship between visualization anchor points and point cloud topology dataset, and generate a real-world mapped 3D network. S3: Access the real-scene mapping 3D network, identify point cloud trajectory coordinates; plan the underground space model based on the point cloud trajectory coordinates and call the entity's 3D topological features, and mark the trajectory surface of the underground space model; S4: Within the preset monitoring time, continuously monitor the surface change rate of the trajectory surface, integrate the underground space model, carry out spatial planning collaborative management, and generate a real-scene 3D planning model after planning collaborative management.

[0012] Preferably, the point cloud topology extraction module is used to collect real-scene 3D mapping data of urban underground space, extract the 3D topological features of underground entities, associate the reference coordinates of the real-scene 3D mapping data, and generate a point cloud topology dataset. Optionally, the point cloud topology extraction module includes: Collect real-scene 3D mapping data of urban underground space; Analyze the underground spatial geometry in real-world 3D mapping data and divide it into three types of geometric regions: horizontal layers, vertical cavities, and intersection nodes. Three-dimensional topological features of underground entities are extracted for three types of geometric regions: continuous contour nodes are extracted for horizontal layers, axial connection points are extracted for vertical cavities, and multi-directional extension trajectories are extracted for intersection nodes. The reference coordinates of the three types of geometric regions are associated, and the coordinate mapping relationship between the horizontal layer and the vertical cavity is calibrated based on the coordinates of the intersection nodes to form a point cloud topology dataset.

[0013] All specific values ​​involved in this embodiment are exemplary parameters used to clearly illustrate the technical operation process and are not the only limitation of the present invention.

[0014] In one embodiment, a ground-based 3D laser scanner (scanning resolution 0.5mm@10m, scanning angle 360°×270°) and an underground pipeline detector (detection depth 0-50m, positioning accuracy ±2cm) are used to simultaneously collect real-scene 3D mapping data of urban underground space to form an original dataset of real-scene 3D mapping of urban underground space.

[0015] It should be noted that, using the noise filtering function built into the terrestrial 3D laser scanner, a noise point removal threshold of 0.3mm is set, retaining effective point clouds with a signal strength ≥80%. After removing environmental interference point clouds, the effective point cloud data is traversed point by point. Based on the actual structural characteristics of the underground space, the following judgment conditions are set: areas with an elevation difference ≥3m and a horizontal extension area ≥10㎡ are divided into horizontal layers; areas with an axial length ≥5m and a cross-sectional diameter ≤3m are divided into vertical cavities; and areas where at least two horizontal layers or vertical cavities intersect and have a spatial range ≥5m×5m×5m are divided into intersection nodes, forming a geometric region division result of horizontal layer-vertical cavity-intersection node. It should be noted that, for the three types of geometric regions, the three-dimensional topological features of underground entities are extracted respectively. For the horizontal layer, the distance between edge points is measured point by point. Continuous points with a distance between adjacent points ≤1.2mm are selected as continuous contour nodes. For the vertical cavity, the center points of the upper and lower ends and the intermediate turning points are measured. The fitting deviation between adjacent points is controlled to be ≤0.5mm as the axial connection points. For the intersection nodes, the trajectory points are measured along each extension direction at a step of 1° to form a multi-directional extension trajectory, thus forming a set of three-dimensional topological features for the three types of regions. In another embodiment, the reference coordinates of the three types of geometric regions are associated, with the center point coordinates (X0, Y0, Z0) of the intersection node as the reference, and the rotation angle is adjusted. , , Translation amount , , The scaling factor k=1.0002 is used to calibrate the coordinate mapping relationship between the horizontal layer and the vertical cavity. The three-dimensional topological feature sets of the three types of regions are then fused with the calibrated coordinate data to form a point cloud topological dataset.

[0016] Preferably, the model topology anchoring module is used to receive the point cloud topology dataset, anchor the three-dimensional topology features of the entity to the preset real-world three-dimensional network, configure the visualization anchor points and establish the mapping relationship between the visualization anchor points and the point cloud topology dataset, and generate the real-world mapped three-dimensional network. Optionally, the model topology anchoring module includes: Geometric nodes for extracting 3D topological features of underground entities from point cloud topology datasets, including cavity corner points, inter-layer connection points, and pipeline intersection points; With geometric nodes as the core reference, visual anchor points are set up in three directions: the horizontal extension direction of the layer, the vertical cavity axis, and the radial direction of the intersection nodes. Align the visualization anchor points with the corresponding geometric positions of the real-world 3D network to ensure that the distribution of visualization anchor points is consistent with the 3D topological feature outline of the underground entity. Based on the geometric correspondence between visualized anchor points and geometric nodes, a mapping association with the point cloud topology dataset is established to generate a real-scene mapped 3D network.

[0017] In this embodiment, a three-dimensional point cloud feature filtering device (positioning accuracy ±1mm, data processing rate 1 million points / second) is used to read the point cloud topology dataset. By setting feature filtering conditions: for cavity corner points, points with an included angle of ≥120° between adjacent points are selected; for inter-layer connection points, points with an elevation difference ≥2m and a horizontal distance ≤1m are selected; and for pipeline intersection points, clustered points with a point cloud density ≥500 points / ㎡ are selected. Geometric nodes of the three-dimensional topological features of the underground entity are extracted to form a set of geometric nodes of the three-dimensional topological features of the underground entity.

[0018] In one embodiment, using the extracted geometric nodes as the reference core points, a spatial coordinate positioning instrument (measurement range 0-100m, repeatability ±0.5mm) is used to lay out visual anchor points at 5m intervals along the horizontal extension direction, at 3m intervals along the vertical cavity axis, and at 4m intervals and uniformly at 360° along the radial direction of the intersecting nodes. Each visual anchor point is assigned a unique coordinate identifier and an association code with the corresponding geometric node.

[0019] In another embodiment, a laser positioning calibrator (calibration accuracy ±0.3mm, measurement distance 0-80m) is used to read the spatial coordinate system of the real-world 3D network. The coordinates of each visualized anchor point are matched and aligned with the corresponding geometric coordinates of the real-world 3D network. The spatial position of the anchor points is adjusted through real-time calibration so that the spatial deviation between the distribution of visualized anchor points and the three-dimensional topological feature contour of the underground entity is ≤2cm.

[0020] In another embodiment, a coordinate correlator (correlation response time ≤ 0.1s, correlation accuracy ± 0.8mm) is used to bind the coordinate data of each visual anchor point to the coordinate data of the geometric nodes in the point cloud topology dataset based on the spatial coordinate correspondence between the visual anchor point and the geometric nodes, establishing a one-to-one mapping association. The bound coordinate association data is then fused with the spatial structure data of the real-scene 3D network to generate a real-scene mapped 3D network.

[0021] It should be noted that all specific values ​​involved in this embodiment are exemplary parameters used to clearly illustrate the technical operation process, and are not the only limitation of the present invention.

[0022] Of particular importance is that the mapping association is specifically as follows: Extract the geometric parameters of the geometric nodes corresponding to each visualization anchor point, including node angle, extension direction, and spatial curvature; Based on the connectivity of the three-dimensional topological features of underground entities, a three-level geometric index framework of nodes-regions-whole is constructed. The geometric parameter encoding of the visualization anchor points is bound to the corresponding storage address of the point cloud topology dataset, and a three-level geometric indexing framework is embedded. The index association path is optimized based on the proximity relationship of the three-dimensional topological features of underground entities to achieve geometric orientation association between the three-dimensional topological features of underground entities and the point cloud topological dataset.

[0023] In this embodiment, a three-dimensional geometric parameter measuring instrument (measurement accuracy 0.1° / 0.01m⁻¹) is used to read the geometric node data associated with the visualized anchor points. The device's analysis function scans the point cloud within a 1m radius around each node, extracting the node's included angle (0°-360°), extension direction (XYZ axis components, accuracy 0.1°), and spatial curvature (measurement radius 0.5m-10m), forming a set of geometric parameters for the three-dimensional topological features of the underground entity. An index building device is used to construct a three-level geometric index framework based on connectivity: node-region-overall. The node-level index is associated with anchor point coordinates; the region-level index integrates the three types of geometric region node indexes and labels the boundary range; and the overall-level index is associated with the connecting nodes of each region. The index hierarchy spacing is set to 1m for node level, 5m for region level, and 10m for overall level. A data binding device (binding accuracy ±0.5mm) is used to generate unique identifiers for the anchor point geometric parameters using 8-bit binary encoding, binding them one-to-one with the corresponding storage address of the point cloud topology dataset. The corresponding nodes are embedded according to the index hierarchy, with an embedding positioning accuracy ≤0.3mm. The path optimization device is used to calculate the spatial straight-line distance between adjacent nodes. A proximity threshold of 3m is set, and priority association paths with a distance ≤3m are marked and their weights are adjusted to twice that of non-priority paths. Invalid links are eliminated to achieve geometric orientation association between the three-dimensional topological features of underground entities and the point cloud topological dataset. The association response time is ≤0.5s.

[0024] It should be noted that all specific values ​​involved in this embodiment are exemplary parameters used to clearly illustrate the technical operation process, and are not the only limitation of the present invention.

[0025] Preferably, the trajectory topology adaptation module is used to access the real-scene mapping 3D network, identify the point cloud trajectory coordinates, plan the underground space model based on the point cloud trajectory coordinates and call the entity 3D topological features, and mark the trajectory surface of the underground space model; Optionally, the trajectory topology adaptation module, which plans the underground space model based on the point cloud trajectory coordinates and calls the entity's 3D topological features, includes: Grouped by geometric region identifiers of point cloud trajectory coordinates, the underground space geometric constraint parameters of the corresponding region in the real scene mapping 3D network are called, including minimum clearance, cross-sectional shape, and extension slope. Using point cloud trajectory coordinates as architecture points, the spatial extension direction and cross-sectional dimensions of the architecture points are limited by geometric constraint parameters; The three-dimensional topological features of the corresponding region in the real-scene mapping three-dimensional network are invoked, and the geometric boundaries of the three-dimensional topological features are used as rigid geometric constraints for the underground space model. A three-dimensional framework for the underground space model is constructed along the geometric constraint boundaries, so that the three-dimensional framework of the underground space model conforms to the actual geometric structure of the underground space.

[0026] In this embodiment, a trajectory coordinate grouping device with a data processing rate of 500,000 points / second and a grouping accuracy of ±0.2mm is used to read the point cloud trajectory coordinate set. The geometric region identifier of each coordinate is extracted through the built-in identifier recognition function of the device. The point cloud trajectory coordinates are divided into three groups according to the identifiers of horizontal layer, vertical cavity, and intersection node, forming a grouped point cloud trajectory coordinate set. The terminal accesses the real scene mapping 3D network through the parameter calling terminal with a response time ≤0.1s and a transmission accuracy of ±0.1mm, and retrieves the underground space geometric constraint parameters of the corresponding area. Among them, the minimum clearance of the horizontal layer is 2.5m, the cross-sectional shape is a rectangle with a length of 5m and a width of 4m, and the extension slope is ≤3°; the minimum clearance of the vertical cavity is 2m, the cross-sectional shape is a circle with a diameter of 3m, and the extension slope is ≤2°; the minimum clearance of the intersection node is 3m, the cross-sectional shape is a polygon with an inscribed circle diameter of 6m, and the extension slope is ≤1°, forming a set of underground space geometric constraint parameters.

[0027] It should be noted that a spatial positioning calibration device with a positioning accuracy of ±0.3mm and a measurement range of 0-80m is used. The coordinates of the grouped point cloud trajectory are used as the framework points. The spatial extension direction of the framework points is limited according to the geometric constraint parameters of the corresponding area. The horizontal layer is adjusted ±5° along the horizontal direction, the vertical cavity is adjusted ±3° along the axial direction, and the intersection node is adjusted ±2° along the radial direction. At the same time, the cross-sectional dimensions are strictly matched with the cross-sectional shape and size range in the constraint parameters to ensure that the spatial distribution of the framework points meets the constraint requirements.

[0028] It should be noted that by calling the topological features with a latency of ≤0.2s and a data accuracy of ±0.5mm, the device is reconnected to the real-scene mapping 3D network, the 3D topological features of the corresponding area are retrieved, and the geometric boundary coordinates of the features are extracted, including the upper and lower surface boundaries of the horizontal layer, the side wall boundaries of the vertical cavity, and the radial boundaries of the intersection nodes. These boundary coordinates are used as rigid geometric constraints to form a set of rigid geometric constraint boundaries.

[0029] In another embodiment, a frame building device with a construction accuracy of ±0.8mm and an operating efficiency of 1000 points / minute is used to connect each frame point sequentially along the coordinate trajectory in the set of rigid geometric constraints to build a continuous three-dimensional frame structure. During the construction process, the frame coordinates are compared with the actual geometric structure coordinates of the underground space in the real-world mapping three-dimensional network in real time, and the frame node positions are adjusted to ensure that the spatial deviation between the three-dimensional frame and the actual geometric structure is ≤2cm, thus ensuring that the frame fits the actual geometric structure of the underground space.

[0030] Optionally, the trajectory surface marking the underground space model in the trajectory topology adaptation module is specifically: Scan the 3D frame surface of the underground space model, obtain the geometric coordinates of each point on the surface of the 3D frame surface of the underground space model, and calculate the fit between the surface and the geometric structure of the underground space in the 3D network of the real scene mapping. The continuous curved surface areas are divided into three different bonding areas: the first bonding area, the second bonding area, and the third bonding area, based on the degree of surface fit. Differentiated geometric labels are configured for three continuous surface regions with different fitting degrees. The differentiated geometric labels include the surface geometric curvature and the geometric deviation from the topological features. An association mapping table of geometric labels-fit degree-topological features is established.

[0031] All specific values ​​involved in this embodiment are exemplary parameters used to clearly illustrate the technical operation process and are not the only limitation of the present invention.

[0032] In one embodiment, a 3D laser scanner with a scanning accuracy of ±0.1mm and a scanning rate of 500 points / second is used to perform a full-coverage scan of the surface of the 3D frame of the underground space. The scanning interval is set to 0.5m to obtain the 3D geometric coordinates of each point on the frame surface, forming a set of geometric coordinates of the 3D frame surface of the underground space. The coordinate data of the corresponding underground space geometric structure in the real-scene mapping 3D network is retrieved by a data comparison device, and the spatial deviation value between the coordinates of each scan point and the real-scene coordinates is calculated. The overall fit of the curved surface is calculated according to the rule of fit degree = (1 - average deviation value / preset maximum allowable deviation) × 100%, and the preset maximum allowable deviation is 10cm, forming the surface fit degree calculation result.

[0033] In another embodiment, three continuous surface regions are divided according to the surface fit degree calculation results, and fit degree thresholds are set: the first fit region has a fit degree ≥95%, the second fit region has a fit degree 85%-95%, and the third fit region has a fit degree <85%. The connectivity of surface points in each region is verified by a surface continuity detection device to ensure that each region is a continuous and unbroken surface, forming a set of three continuous surface regions with different fit degrees.

[0034] It should be noted that differentiated geometric identifiers are configured for the three continuous curved surface regions. The first fitting region's geometric identifier includes a surface geometric curvature ≤0.05m⁻¹ and a geometric deviation from the topological features ≤2cm. The second fitting region includes a surface geometric curvature of 0.05m⁻¹-0.1m⁻¹ and a geometric deviation of 2cm-5cm. The third fitting region includes a surface geometric curvature >0.1m⁻¹ and a geometric deviation >5cm. Through data association equipment, the geometric identifiers of each region, the corresponding fitting range, and the associated three-dimensional topological feature information of the underground entity are bound together to establish a mapping table of geometric identifier-fitting degree-topological feature association. The mapping table includes five core contents: region number, fitting degree interval, surface geometric curvature range, geometric deviation range, and corresponding topological feature type.

[0035] Please see Figure 2This diagram illustrates the partitioning of the trajectory surface fit. The three colored surfaces within the transparent cube frame represent different fit zones: green represents the first fit zone (fit ≥ 95%, curvature ≤ 0.05m⁻¹, deviation ≤ 2cm), yellow represents the second fit zone (85%-95%, curvature 0.05-0.1m⁻¹, deviation 2-5cm), and red represents the third fit zone (< 85%, curvature > 0.1m⁻¹, deviation > 5cm). The grid lines and nodes form a topological network, presenting the correlation between surface geometry and fit, intuitively reflecting the accuracy distribution of the trajectory surface in the underground space model.

[0036] Optionally, the trajectory topology adaptation module, after marking the trajectory surface of the underground space model, further includes: By retrieving the real-scene 3D mapping data corresponding to each surface segment through the association mapping table, the geometric distribution characteristics of the real-scene 3D mapping data are extracted. By comparing the geometric parameters and geometric distribution characteristics of the surface segments, mismatched geometric anomaly regions in the surface segments can be identified. Using geometric distribution characteristics as a reference, determine the geometric correction direction and range of the abnormal region; Based on the geometric structure of the underground space in the 3D network mapped from the real scene, the surface geometric parameters of the abnormal area are adjusted to match the geometric features of the point cloud.

[0037] In this embodiment, a data retrieval device with a data retrieval rate of 100Mbps and a positioning accuracy of ±0.3mm is used to access the association mapping table of geometric identifier-fit-topology features. According to the region number and topology feature type of each surface segment, the original dataset of the real-scene 3D mapping of the urban underground space of the corresponding region is retrieved. The retrieved real-scene 3D mapping data is processed by a feature extraction device to extract geometric distribution features, including point cloud density distribution (point cloud number per unit area ≥300 points / ㎡), contour undulation amplitude (maximum undulation difference ≤5cm), and cross-sectional size change law (size fluctuation range ≤10%), forming a set of real-scene geometric distribution features corresponding to each surface segment.

[0038] In one embodiment, the geometric parameters (including surface curvature, cross-sectional dimensions, and spatial coordinate range) of each surface segment are read and compared with the corresponding set of real-world geometric distribution features. A comparison deviation threshold is set: curvature deviation ≤ 0.02m⁻¹, size deviation ≤ 3cm, and coordinate deviation ≤ 5cm. The parameter differences between the two are compared point by point, and continuous areas exceeding the deviation threshold are marked as geometrically abnormal areas. The boundary coordinates and deviation values ​​of the abnormal areas are recorded to form a geometrically abnormal area identification result.

[0039] It should be noted that, using the extracted real-scene geometric distribution features as a reference, the deviation direction (three-dimensional direction of X-axis, Y-axis, and Z-axis) between the abnormal area and the real-scene features is analyzed by spatial positioning equipment. The correction range is determined based on the deviation value, and the correction boundary is set to exceed the boundary of the abnormal area by 1m to ensure that the corrected surface is smoothly connected with the surrounding normal area, thus forming a geometric correction scheme for the abnormal area, and clarifying the correction direction, correction range, and target parameters.

[0040] In another embodiment, a surface calibration device is used to access the real-scene mapping 3D network, read the coordinate data and constraint parameters of the corresponding underground space geometry, and adjust the surface geometry parameters of the abnormal area according to the correction scheme. During the adjustment process, the surface coordinates are corrected point by point in a step of 0.5cm. The corrected surface parameters are compared with the real-scene geometry distribution features in real time until the deviations of the surface curvature, cross-sectional size, spatial position and point cloud geometry features are all ≤ the set threshold, so as to achieve accurate matching between the surface and the point cloud geometry features.

[0041] Optionally, the adjustment of the surface geometry parameters of the abnormal region specifically includes: A local geometric calibration coordinate system is established based on the topological feature geometric nodes around the anomaly region; The geometric deviation of the abnormal region is decomposed into linear deviation and angular deviation in the coordinate system; Based on the preset gradient changes in the underground space geometry, the surface geometric parameters of the abnormal area are corrected point by point.

[0042] In this embodiment, a spatial coordinate measuring instrument with a positioning accuracy of ±0.1mm and a measurement range of 0-50m is used to identify topological feature geometric nodes within a 5m radius around the abnormal area. Continuous nodes with a node density of ≥5 nodes / m² are selected as reference nodes to form a set of topological feature geometric nodes around the abnormal area. In one embodiment, the node closest to the abnormal area in the set is taken as the origin, the X-axis is set along the horizontal layer extension direction, the Y-axis is set perpendicular to the horizontal layer direction, and the Z-axis is set in the radial direction of the intersecting nodes. A local geometric calibration coordinate system is established, and the coordinate origin, coordinate axis direction and range (X-axis 0-20m, Y-axis 0-10m, Z-axis 0-15m) of the coordinate system are defined.

[0043] It should be noted that the deviation analysis device is used to read the difference between the actual coordinates of each surface point in the abnormal area in the local geometric calibration coordinate system and the corresponding coordinates in the real scene mapping 3D network. The geometric deviation is decomposed into linear deviations (unit: cm) in the X-axis, Y-axis and Z-axis directions and angular deviations (unit: °) between the coordinate axes. A linear deviation threshold of 5 cm and an angular deviation threshold of 2 ° are set. Deviation components exceeding the thresholds are recorded to form the geometric deviation decomposition results of the abnormal area.

[0044] In another embodiment, preset gradient change parameters of the underground space geometry in the real-scene mapping 3D network are retrieved. The gradient change rate of the horizontal layer region is set to 0.2 cm / m, the vertical cavity region to 0.3 cm / m, and the intersection node region to 0.1 cm / m. A surface calibration device is used to determine the correction amount of each point in the abnormal area according to the gradient change parameters. The surface geometry parameters are adjusted point by point from the boundary of the abnormal area to the center in a step of 0.2 cm. During the adjustment process, the coordinates of the reference node in the local geometric calibration coordinate system are referenced in real time to ensure that the connection deviation between the corrected surface points and the surface points of the surrounding normal area is ≤1 cm. Finally, the surface geometry parameters of the abnormal area are consistent with the actual geometry of the underground space.

[0045] Preferably, the collaborative management output module is used to continuously monitor the surface change rate of the trajectory surface within a preset monitoring time, integrate the underground space model, perform spatial planning collaborative management, and generate a real-scene 3D planning model after planning collaborative management.

[0046] Optionally, the rate of change of the trajectory surface continuously monitored in the collaborative management output module includes: Determine the complexity of the underground space geometry in the real-scene mapping 3D network, and divide the underground space into single geometric regions and complex geometric regions according to the complexity. The complex geometric regions include intersection nodes and cavity turning points. Different trajectory surface data acquisition strategies are set for single geometric regions and complex geometric regions. In the single geometric region, a fixed number of acquisition points are used, while in the complex geometric region, the number of acquisition points is increased step by step. Within the preset monitoring time, real-time geometric data of trajectory surfaces in single and complex geometric regions are acquired according to the trajectory surface data acquisition strategy. Calculate the changes in surface geometric parameters within adjacent acquisition cycles and convert them into the surface change rate per unit time.

[0047] In this embodiment, a real-scene mapping 3D network is accessed to extract parameters such as node density, turning angle, and spatial connection method of underground space geometry. A complexity judgment standard is set: areas with node density ≤ 2 / ㎡, turning angle ≤ 15° and no multiple areas intersecting are divided into single geometric areas, while areas with node density > 2 / ㎡, turning angle > 15° or with intersecting nodes or cavity turning points are divided into complex geometric areas, thus forming the underground space geometric area division result.

[0048] It should be noted that differentiated trajectory surface data acquisition strategies are set for the two types of areas. For single geometric areas, a fixed number of acquisition points are evenly distributed, with one acquisition point laid out at a interval of 5m, and the total number of acquisition points in each area is no less than 20. For complex geometric areas, an incremental acquisition point system is adopted, initially laid out at a interval of 3m, and the interval is reduced by 0.5m after each acquisition cycle until the minimum interval is 1m. The acquisition points are evenly distributed along the surface contour and focus on covering turning points and intersections.

[0049] It should be noted that the preset monitoring time is set to 24 hours, the acquisition cycle is 30 minutes, and a real-time data acquisition instrument with an acquisition accuracy of ±0.1mm and a data transmission rate of 50Mbps is used. According to the differentiated acquisition strategy, the trajectory surface of the single geometric area and the complex geometric area are continuously acquired to obtain real-time geometric data such as the surface curvature and spatial coordinates of each acquisition point, and stored in the real-time database of underground space trajectory surface.

[0050] In another embodiment, a data processing accuracy of ±0.01 mm / h is used to calculate the change rate of the trajectory surface. The surface geometric parameters of adjacent acquisition cycles are read from the real-time database of the underground space trajectory surface. The change values ​​of parameters such as curvature change and coordinate offset of the same acquisition point in adjacent cycles are calculated. The parameter change values ​​are converted into the surface change rate per unit time (hour) by the calculation method of "surface change rate = parameter change value / acquisition cycle duration". The trajectory surface change rate monitoring results of single geometric area and complex geometric area are formed.

[0051] Optionally, the spatial planning collaborative management in the collaborative management output module specifically includes: Retrieve the preset tolerance geometric parameters of the underground space geometry in the real-scene mapping 3D network to obtain the maximum allowable deformation amplitude and angle of the curved surface; By comparing the surface change rate with the change threshold corresponding to the tolerance geometric parameters, the geometric anomaly region where the surface change rate exceeds the threshold is located, and the three-dimensional topological features and model geometric parameters of the geometric anomaly region are retrieved. Adjust the geometric shape of the model surface in the geometric anomaly area according to the tolerance geometric parameters so that the surface conforms to the geometric constraints of the underground space.

[0052] In this embodiment, a real-scene mapping 3D network is accessed to read the preset tolerance geometric parameters of the underground space geometry. The maximum allowable deformation amplitude of a single geometric area surface is ≤5cm and the maximum deformation angle is ≤3°. The maximum allowable deformation amplitude of a complex geometric area (intersection node, cavity turning point) surface is ≤3cm and the maximum deformation angle is ≤2°, thus forming a set of preset tolerance geometric parameters for the underground space geometry.

[0053] In one embodiment, a change rate comparison device with a data comparison rate of 100,000 sets / second and a comparison accuracy of ±0.01mm / h is used to read the trajectory surface change rate monitoring results and the change threshold corresponding to the preset tolerance geometric parameters (change rate threshold for a single geometric area ≤0.5cm / h, change rate threshold for a complex geometric area ≤0.3cm / h). The surface change rate is compared with the corresponding threshold region by region. Continuous areas exceeding the threshold are marked as geometrically abnormal areas. The boundary coordinates, area type, and value exceeding the threshold of the abnormal area are recorded. The feature retrieval device is used to access the real-scene mapping 3D network to retrieve the 3D topological features of the underground entity in the geometrically abnormal area (including node angles, extension direction, and geometric boundaries) and surface geometric parameters (curvature, cross-sectional dimensions, and spatial coordinates) to form a geometrically abnormal area associated data set.

[0054] In another embodiment, a geometric shape adjustment device with a surface adjustment accuracy of ±0.2mm and an adjustment rate of 500 points / minute is used. The preset tolerance geometric parameters are used as the constraint standard, and the adjustment benchmark is determined according to the three-dimensional topological characteristics of the abnormal area. The surface geometry is adjusted point by point from the boundary of the abnormal area to the center in a step of 0.3cm. During the adjustment process, the adjusted surface parameters are compared with the underground space geometric constraint requirements in real time to ensure that the deformation amplitude and angle of the adjusted surface are within the preset tolerance range, and the connection deviation with the surrounding normal area surface is ≤1cm, so that the surface fully conforms to the underground space geometric constraint.

[0055] Most importantly, the following operations are performed after the local area is adjusted: The adjusted underground space model data of all geometric anomaly areas are integrated and unified into the global geometric coordinate system of underground space. Analyze the geometric connection relationship of the underground space model corresponding to each geometric anomaly region in the global coordinate system, and identify the geometric conflicts at the connection points; Based on the pre-defined overall geometric correlation of underground space, the model geometric parameters of conflict areas are adjusted to optimize the geometric connection method; Verify the geometric compatibility of the underground space model corresponding to each geometric anomaly area after adjustment, and optimize the underground space model to conform to the overall geometric structure of the underground space.

[0056] In this embodiment, the adjusted underground space geometric data of all geometrically abnormal areas are read and connected to the global geometric coordinate system of underground space (based on the city's independent coordinate system, with the X-axis pointing east, the Y-axis pointing north, and the Z-axis pointing upward, with a coordinate accuracy of ±0.1m). Through a seven-parameter coordinate transformation (rotation angle 0.02°, translation amount 0.05m, scaling ratio 1.0002), the adjusted data of each area are unified under this coordinate system, forming a set of underground space geometric data under a unified coordinate system.

[0057] In one embodiment, the geometric data set of underground space under a unified coordinate system is read, and the coordinates, angles and cross-sectional dimensions of the connection surfaces between each geometrically abnormal area and the surrounding normal and abnormal areas are extracted. The connection conflict judgment criteria are set: the connection surface spacing <2cm, the angle deviation >3° or the cross-sectional dimension mismatch exceeds 5% is considered a geometric conflict. The boundary range, conflict type and deviation value of the conflict area are analyzed and marked segment by segment to form the underground space geometric connection conflict identification result.

[0058] In another embodiment, based on preset overall geometric correlation parameters of the underground space (including regional connection slope ≤ 5°, cross-sectional transition curvature ≤ 0.05m⁻¹, and spatial net distance ≥ 1m), a parameter optimization device (optimization accuracy ± 0.2mm, adjustment rate 500 points / minute) is used to locally adjust the geometric parameters of the conflict area. The coordinates of the connection surface are corrected in steps of 0.2cm, the included angle is adjusted to meet the correlation requirements, and the cross-sectional transition method is optimized to be a smooth curve connection, ensuring that the conflict area and the surrounding area form a continuous and compatible geometric structure.

[0059] It should be noted that a global verification is performed on the underground space geometric data of all geometrically abnormal areas after adjustment. The geometric parameters of each area are compared with the constraints of the overall geometric structure of the underground space to confirm that the deviation at the connection point is ≤1cm and the overall spatial layout conforms to the preset geometric correlation. The verified underground space geometric data are integrated and optimized to form the final data set that conforms to the overall geometric structure of the underground space.

[0060] Most importantly, the process of generating the real-world 3D planning model specifically involves the following operations: By integrating the geometric feature information of the underground space model after global collaboration with the real-world 3D mapping, the geometric details of the underground space model are supplemented. Record the geometric adjustment trajectory, correction parameters, and geometric association links with topological features of each region of the underground space model; Based on the hierarchical relationship of the underground space geometric structure, determine the geometric topology of the underground space model and associate the geometric features of each part; By integrating the geometric topology system and adjusting the records and accuracy parameters, a complete real-world 3D planning model reflecting the geometric structure of underground space is formed.

[0061] In this embodiment, the geometric data set of underground space after global collaboration and the geometric feature information set of 3D mapping of urban underground space are read. Curved textures, small protrusions and depressions with a diameter of ≥0.5cm in the real scene are extracted. These features are accurately mapped to the same position in the geometric data of underground space through coordinate matching. The surface texture of horizontal layer, details of vertical cavity sidewalls and structure of intersection nodes are supplemented to ensure that the spatial deviation between the supplemented data and the real scene features is ≤1cm, thus forming a geometric data set of underground space with supplemented details.

[0062] In one embodiment, the coordinate adjustment trajectory of each region is recorded in 0.1cm increments, simultaneously recording parameters such as linear correction and angle correction, as well as the associated node numbers and link lengths with topological features. A record archive is established in the format of "region number - adjustment trajectory - correction parameters - associated link," and bound to the corresponding region's geometric data. A topology construction device with a topology construction accuracy of ±0.3mm is used to build a geometric topology system in three levels: overall, region, and node. The overall level covers the entire underground space, the region level corresponds to three types of geometric regions, and the node level corresponds to various types of geometric nodes. The level connection deviation is set to ≤2cm. The boundary coordinates, extension directions, and curvature parameters of each region are associated to establish geometric connections between levels and regions, forming an underground space geometric topology system.

[0063] In another embodiment, the geometric topology system, adjustment record files, and accuracy parameters (measurement accuracy ±0.1mm, correction deviation ≤1cm) are read and integrated according to the logical association of "topology system - regional data - adjustment record - accuracy parameters". Redundant data is eliminated to ensure that the data of each part corresponds consistently, and finally a real-scene 3D planning data set containing all geometric features, adjustment records, and accuracy information is formed.

[0064] Please see Figure 3 This is a schematic diagram of a three-dimensional structural model of underground space. It presents the geological structure and spatial distribution relationship through layered colors and node networks. The green area represents the surface layer, the blue nodes and lines form the topological connection network of the underground space, the red cavity is the key structural unit, and the transparent color blocks mark the interfaces of different strata, presenting a three-dimensional underground space architecture as a whole.

[0065] Please see Figure 4 This is the interface for the trajectory topology adaptation module of the urban underground space planning collaborative management system; the real-scene 3D visualization area renders the surface shape after trajectory coordinate recognition in real time based on the CGCS2000 coordinate system and marks the spatial coordinate information; the trajectory surface marking area on the right displays three areas (TR001, TR002, TR003), which correspond to the first, second, and third fitting areas in this embodiment, respectively. The surface curvature (such as 0.024, 0.187, 0.312) reflects the geometric characteristics of different fitting degree intervals, intuitively presenting the relationship between trajectory surface fitting degree and geometric marking.

[0066] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0067] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A collaborative management system for urban underground space planning based on real-scene 3D mapping, characterized in that, include: The point cloud topology extraction module is used to collect real-scene 3D mapping data of urban underground space, extract the 3D topological features of underground entities, associate the reference coordinates of the real-scene 3D mapping data, and generate a point cloud topology dataset. The model topology anchoring module is used to receive point cloud topology datasets, anchor the 3D topology features of entities to a preset real-world 3D network, configure visualization anchor points and establish a mapping relationship between visualization anchor points and point cloud topology datasets, and generate a real-world mapped 3D network. The trajectory topology adaptation module is used to access the real-scene mapping 3D network, identify the trajectory coordinates of point clouds, plan the underground space model based on the trajectory coordinates of point clouds and call the 3D topological features of the entity, and mark the trajectory surface of the underground space model. The collaborative management output module is used to continuously monitor the surface change rate of the trajectory surface within a preset monitoring time, integrate the underground space model, carry out spatial planning collaborative management, and generate a real-scene 3D planning model after planning collaborative management.

2. The urban underground space planning collaborative management system based on real-scene 3D mapping as described in claim 1, characterized in that, The point cloud topology extraction module includes: Collect real-scene 3D mapping data of urban underground space; Analyze the underground spatial geometry in real-world 3D mapping data and divide it into three types of geometric regions: horizontal layers, vertical cavities, and intersection nodes. Three-dimensional topological features of underground entities are extracted for three types of geometric regions: continuous contour nodes are extracted for horizontal layers, axial connection points are extracted for vertical cavities, and multi-directional extension trajectories are extracted for intersection nodes. The reference coordinates of the three types of geometric regions are associated, and the coordinate mapping relationship between the horizontal layer and the vertical cavity is calibrated based on the coordinates of the intersection nodes to form a point cloud topology dataset.

3. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 1, characterized in that, The model topology anchoring module includes: Geometric nodes for extracting 3D topological features of underground entities from point cloud topology datasets, including cavity corner points, inter-layer connection points, and pipeline intersection points; With geometric nodes as the core reference, visual anchor points are set up in three directions: the horizontal extension direction of the layer, the vertical cavity axis, and the radial direction of the intersection nodes. Align the visualization anchor points with the corresponding geometric positions of the real-world 3D network to ensure that the distribution of visualization anchor points is consistent with the 3D topological feature outline of the underground entity. Based on the geometric correspondence between visualized anchor points and geometric nodes, a mapping association with the point cloud topology dataset is established to generate a real-scene mapped 3D network.

4. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 1, characterized in that, The specific steps for identifying point cloud trajectory coordinates in the trajectory topology adaptation module are as follows: Retrieve underground space geometric boundary data from the point cloud topology dataset to obtain horizontal layer thickness, vertical cavity cross-sectional dimensions, and spatial range of intersection nodes; Perform geometric verification on the original trajectory data input by the preset plan, and retain the valid trajectory segments within the geometric boundary range; Identify the underground space geometric coordinate system in the real-scene mapping 3D network; decompose the effective trajectory segment into continuous point cloud trajectory coordinates based on the underground space geometric coordinate system, where each point cloud trajectory coordinate is associated with the corresponding geometric region identifier.

5. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 4, characterized in that, The trajectory topology adaptation module, which plans the underground space model based on the point cloud trajectory coordinates and calls the entity's 3D topological features, includes: Grouped by geometric region identifiers of point cloud trajectory coordinates, the underground space geometric constraint parameters of the corresponding region in the real scene mapping 3D network are called, including minimum clearance, cross-sectional shape, and extension slope. Using point cloud trajectory coordinates as architecture points, the spatial extension direction and cross-sectional dimensions of the architecture points are limited by geometric constraint parameters; The three-dimensional topological features of the corresponding region in the real-scene mapping three-dimensional network are invoked, and the geometric boundaries of the three-dimensional topological features are used as rigid geometric constraints for the underground space model. A three-dimensional framework for the underground space model is constructed along the geometric constraint boundaries, so that the three-dimensional framework of the underground space model conforms to the actual geometric structure of the underground space.

6. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 5, characterized in that, The trajectory surface marking the underground space model in the trajectory topology adaptation module is specifically as follows: Scan the 3D frame surface of the underground space model, obtain the geometric coordinates of each point on the surface of the 3D frame surface of the underground space model, and calculate the fit between the surface and the geometric structure of the underground space in the 3D network of the real scene mapping. The continuous curved surface areas are divided into three different bonding areas: the first bonding area, the second bonding area, and the third bonding area, based on the degree of surface fit. Differentiated geometric labels are configured for three continuous surface regions with different fitting degrees. The differentiated geometric labels include the surface geometric curvature and the geometric deviation from the topological features. An association mapping table of geometric labels-fit degree-topological features is established.

7. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 6, characterized in that, The trajectory topology adaptation module, after marking the trajectory surface of the underground space model, also includes: By retrieving the real-scene 3D mapping data corresponding to each surface segment through the association mapping table, the geometric distribution characteristics of the real-scene 3D mapping data are extracted. By comparing the geometric parameters and geometric distribution characteristics of the surface segments, mismatched geometric anomaly regions in the surface segments can be identified. Using geometric distribution characteristics as a reference, determine the geometric correction direction and range of the abnormal region; Based on the geometric structure of the underground space in the 3D network mapped from the real scene, the surface geometric parameters of the abnormal area are adjusted to match the geometric features of the point cloud.

8. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 7, characterized in that, The specific adjustments to the surface geometry parameters of the abnormal region are as follows: A local geometric calibration coordinate system is established based on the topological feature geometric nodes around the anomaly region; The geometric deviation of the abnormal region is decomposed into linear deviation and angular deviation in the coordinate system; Based on the preset gradient changes in the underground space geometry, the surface geometric parameters of the abnormal area are corrected point by point.

9. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 1, characterized in that, The rate of change of the trajectory surface continuously monitored in the collaborative management output module includes: Determine the complexity of the underground space geometry in the real-scene mapping 3D network, and divide the underground space into single geometric regions and complex geometric regions according to the complexity. The complex geometric regions include intersection nodes and cavity turning points. Different trajectory surface data acquisition strategies are set for single geometric regions and complex geometric regions. In the single geometric region, a fixed number of acquisition points are used, while in the complex geometric region, the number of acquisition points is increased step by step. Within the preset monitoring time, real-time geometric data of trajectory surfaces in single and complex geometric regions are acquired according to the trajectory surface data acquisition strategy. Calculate the changes in surface geometric parameters within adjacent acquisition cycles and convert them into the surface change rate per unit time.

10. The urban underground space planning collaborative management system based on real-scene 3D mapping according to claim 9, characterized in that, The spatial planning collaborative management in the collaborative management output module specifically includes: Retrieve the preset tolerance geometric parameters of the underground space geometry in the real-scene mapping 3D network to obtain the maximum allowable deformation amplitude and angle of the curved surface; By comparing the surface change rate with the change threshold corresponding to the tolerance geometric parameters, the geometric anomaly region where the surface change rate exceeds the threshold is located, and the three-dimensional topological features and model geometric parameters of the geometric anomaly region are retrieved. Adjust the geometric shape of the model surface in the geometric anomaly area according to the tolerance geometric parameters so that the surface conforms to the geometric constraints of the underground space.

Images