Fabricated bridge virtual intelligent pre-assembly method and system based on LiDAR

Abstract

The invention relates to an assembled bridge virtual intelligent pre-assembling method and system based on LiDAR. The method comprises the steps that component point cloud data of at least two prefabricated components forming an assembled bridge are obtained based on three-dimensional laser radar scanning; obtaining a BIM model of the fabricated bridge, and determining model point cloud data based on the BIM model; performing point cloud registration on the component point cloud data and the model point cloud data, and determining at least two point cloud components corresponding to the component point cloud data based on registration result semantic segmentation; and acquiring a plurality of feature points of the at least two point cloud components, constructing an error control relationship based on the plurality of feature points of the at least two point cloud components, and controlling to realize virtual pre-assembly of the at least two point cloud components based on the error control relationship.

Description

Technical Field

[0001] The disclosed embodiments relate to the field of intelligent building detection technology, and in particular to a LiDAR-based virtual intelligent pre-assembly method and system for prefabricated bridges. Background Art

[0002] At present, bridges, as important transportation infrastructure, play a vital role in the national economy and social development. With the acceleration of urbanization and the expansion of transportation networks, the demand for bridge construction continues to grow. Prefabricated bridge technology has gradually become an important means in engineering construction due to its advantages such as fast construction speed, controllable quality and environmental protection.

[0003] The on-site construction of prefabricated bridges, such as fully prefabricated steel-concrete (CRT) composite bridges, mainly involves the assembly and connection of prefabricated components. However, during the actual on-site construction process, due to the accumulated errors in the processing of each component, the assembly of the prefabricated bridge may encounter problems such as misalignment or collision, making it difficult to assemble and connect, which will have a great impact on the construction, such as rework, extended construction period, and increased costs. Therefore, it is usually possible to conduct pre-assembly inspection of prefabricated bridges before on-site construction. However, since different components in prefabricated bridges, such as steel beams and bridge decks, are often produced in different prefabrication plants, the existing pre-assembly method usually uses physical components for manual pre-assembly inspection, which is time-consuming, labor-intensive, increases labor costs, is inefficient, and has low accuracy in pre-assembly results.

[0004] Related technologies, such as the invention patent document with publication number CN117610132A, propose a technical solution for virtual pre-assembly, which saves time and labor, and improves the efficiency of assembly detection and the accuracy of results compared to manual methods. However, it is only applicable to steel structure truss bridges and only considers the impact of component surface features on three-dimensional point cloud acquisition data. It is not applicable to prefabricated bridges such as steel-concrete composite bridges. Therefore, how to improve the accuracy of virtual pre-assembly results of prefabricated bridges has become an urgent problem to be solved. Summary of the invention

[0005] In order to solve the above technical problems or at least partially solve the above technical problems, the embodiments of the present disclosure provide a LiDAR-based virtual intelligent pre-assembly method and system for prefabricated bridges.

[0006] In a first aspect, an embodiment of the present disclosure provides a LiDAR-based virtual intelligent pre-assembly method for assembled bridges, comprising:

[0007] Acquire component point cloud data of at least two prefabricated components constituting the prefabricated bridge based on three-dimensional laser radar scanning; acquire a BIM model of the prefabricated bridge, and determine model point cloud data based on the BIM model;

[0008] Performing point cloud registration on the component point cloud data and the model point cloud data, and determining at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result;

[0009] Acquire multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control and implement virtual pre-assembly of the at least two point cloud components based on the error control relationship.

[0010] In one embodiment, the model point cloud data includes point cloud data of multiple different components and corresponding point cloud identifiers; the determining at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result includes:

[0011] Using a kd-tree search tree to construct an index structure of the model point cloud data, traversing each point cloud in the component point cloud data based on the index structure and using the kd-tree search tree to find the nearest neighbor point cloud of each point cloud within a preset radius, and updating the point cloud identifier of the corresponding point cloud in the component point cloud data based on the point cloud identifier of the nearest neighbor point cloud;

[0012] After the traversal is completed, at least two point cloud components corresponding to the component point cloud data are determined based on the point cloud identification segmentation of each point cloud in the updated component point cloud data.

[0013] In one embodiment, the point cloud identification corresponding to the point cloud data of multiple different components in the model point cloud data includes at least component labels and / or color information, and different component labels and / or color information indicate that the corresponding point cloud data belong to different components.

[0014] In one embodiment, obtaining a plurality of feature points of the at least two point cloud components includes:

[0015] Determine at least one connected component between the at least two point cloud components; and obtain a plurality of feature points of the at least one connected component.

[0016] In one embodiment, the error control relationship includes a minimum stitching error control relationship between the at least two point cloud components, and the minimum stitching error control relationship is determined by a minimum stitching error control function.

[0017] In one embodiment, the method further comprises:

[0018] After the virtual pre-assembly of the at least two point cloud components reaches an optimal virtual pre-assembly state, a collision check is performed between the at least two point cloud components and a collision check result is recorded.

[0019] In one embodiment, performing point cloud registration on the component point cloud data and the model point cloud data includes:

[0020] Using Harris corner detection to extract matching corner pairs in the component point cloud data and the model point cloud data; using the random sampling consistent RANSAC algorithm to estimate the rigid body transformation matrix between the point clouds according to the matching corner pairs, and completing the coarse registration based on the rigid body transformation matrix;

[0021] The rough registration result is finely registered based on a fine registration algorithm, wherein the fine registration algorithm at least includes an iterative closest point ICP algorithm.

[0022] In a second aspect, the present disclosure provides a LiDAR-based virtual intelligent pre-assembly system for assembled bridges, including:

[0023] A data acquisition module, used to acquire component point cloud data of at least two prefabricated components constituting the prefabricated bridge based on three-dimensional laser radar scanning; acquire a BIM model of the prefabricated bridge, and determine model point cloud data based on the BIM model;

[0024] A registration and segmentation module, used for performing point cloud registration on the component point cloud data and the model point cloud data, and determining at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result;

[0025] A virtual assembly module is used to obtain multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control the virtual pre-assembly of the at least two point cloud components based on the error control relationship.

[0026] In a third aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for virtual intelligent pre-assembly of prefabricated bridges based on LiDAR as described in any of the above embodiments is implemented.

[0027] In a fourth aspect, an embodiment of the present disclosure provides an electronic device, including:

[0028] a processor; and a memory for storing a computer program;

[0029] Wherein, the processor is configured to execute the LiDAR-based virtual intelligent pre-assembly method for prefabricated bridges described in any of the above embodiments by executing the computer program.

[0030] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0031] The LiDAR-based virtual intelligent pre-assembly method and system for prefabricated bridges provided in the embodiments of the present disclosure obtain component point cloud data of at least two prefabricated components constituting the prefabricated bridge based on three-dimensional laser radar scanning; obtain a BIM model of the prefabricated bridge, and determine the model point cloud data based on the BIM model; perform point cloud registration on the component point cloud data and the model point cloud data, and determine at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result; obtain multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control the virtual pre-assembly of the at least two point cloud components based on the error control relationship. In this embodiment, BIM is combined with three-dimensional laser radar (LiDAR) technology to collect and process three-dimensional laser point cloud data of components of prefabricated bridges, such as steel-concrete composite bridges. The point cloud data obtained by the BIM model of the prefabricated bridge is aligned and used as a reference to accurately segment the three-dimensional laser point cloud data of, for example, two components of the bridge to determine the point cloud components corresponding to each component. The point cloud components are used to construct an error control relationship to complete, for example, the virtual pre-assembly of two point cloud components. It has the characteristics of high efficiency and high precision, which makes up for the low precision and slow efficiency of the prefabricated component pre-assembly of traditional prefabricated bridges, improves the accuracy of the virtual pre-assembly results of prefabricated bridges, and can accurately guide subsequent actual assembly work, such as between the bridge deck and the steel beam of the prefabricated bridge, to avoid rework and other problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0034] Figure 1 This is a flow chart of a method for virtual intelligent pre-assembly of a prefabricated bridge based on LiDAR according to an embodiment of the present disclosure;

[0035] Figure 2 A schematic diagram of the component registration and segmentation process of the prefabricated bridge according to an embodiment of the present disclosure;

[0036] Figure 3 A schematic diagram of another component registration and segmentation process of the prefabricated bridge according to an embodiment of the present disclosure;

[0037] Figure 4 It is a schematic diagram of virtual pre-assembly of two components of the prefabricated bridge according to the embodiment of the present disclosure;

[0038] Figure 5 A schematic diagram of a virtual intelligent pre-assembly system for prefabricated bridges based on LiDAR according to an embodiment of the present disclosure;

[0039] Figure 6 Schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0040] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0042] It should be understood that, in the following, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0043] Figure 1 The present invention is a flowchart of a method for virtual intelligent pre-assembly of a prefabricated bridge based on LiDAR according to an embodiment of the present invention. The method for virtual intelligent pre-assembly of a prefabricated bridge based on LiDAR can be executed by a computing device such as a computer or a mobile terminal, and specifically may include the following steps:

[0044] Step S101: acquiring component point cloud data of at least two prefabricated components constituting an assembled bridge based on three-dimensional laser radar scanning; acquiring a BIM model of the assembled bridge, and determining model point cloud data based on the BIM model.

[0045] Exemplarily, the prefabricated bridge in this embodiment can be a steel-concrete composite bridge, and the prefabricated components can include steel beams and concrete panels (i.e., bridge panels). These two components are used as examples in this embodiment, but are not limited to this. Three-dimensional laser scanning technology can obtain spatial point cloud data at high speed, and has the unique advantages of high efficiency and high precision. Specifically, according to the characteristics of prefabricated components and the needs of key parts, the scanning site can be planned in advance, and corresponding preparations can be made to ensure the comprehensiveness and accuracy of data acquisition. Take the steel beam and concrete panel as two components: the steel beam needs to consider the height of the scanner site to obtain a relatively complete top shear steel plate data; the concrete panel needs to consider suspension to obtain the bottom embedded parts data. After the scan is completed, the PTX file exported by the scanner is converted to PDF format for later data viewing and management. In some other embodiments, based on the scanning site coordinates, the bounding box can also be used to segment the scanned point cloud data, effectively remove most of the irrelevant background point clouds, and retain the point cloud data of the target component and a small amount of background information. Subsequently, the filtering algorithm can also be used to process the point cloud data to effectively remove noise and improve the purity of the data. Next, the point cloud data can be simplified through downsampling technology to reduce redundant information and data volume while retaining key geometric features, thereby providing a high-quality and efficient point cloud data foundation for subsequent data processing and analysis.

[0046] The BIM model contains rich design geometry and semantic information, and can accurately digitally express the physical and functional characteristics of the target component. The BIM model of the prefabricated bridge can be a BIM model determined in the design stage before the bridge construction. In this embodiment, the existing BIM model is used to obtain model point cloud data, that is, target point cloud data or BIM point cloud data.

[0047] Step S102: performing point cloud registration on the component point cloud data and the model point cloud data, and determining at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result.

[0048] Exemplarily, after obtaining BIM point cloud data and scanned point cloud data, i.e., component point cloud data, the BIM point cloud data and scanned point cloud data can be loaded for point cloud registration, such as rough registration first and then fine registration. Based on the registration result, semantic segmentation is used to determine at least two point cloud components corresponding to the component point cloud data, that is, the point cloud data of each component is segmented from the scanned point cloud data (at least including background point cloud, noise point cloud, etc.) to determine the point cloud component, i.e., the component model. In this embodiment, the point cloud data obtained by the BIM model of the prefabricated bridge is used for registration and used as a reference to accurately segment the three-dimensional laser point cloud data of, for example, two components of the bridge to determine the point cloud components corresponding to each of the two components.

[0049] Step S103: Acquire multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control and implement virtual pre-assembly of the at least two point cloud components based on the error control relationship.

[0050] Exemplarily, in this embodiment, after accurately segmenting the three-dimensional laser point cloud data of, for example, two components of the bridge to determine the point cloud components corresponding to each of the two components, multiple feature points of, for example, the point cloud components of the two point cloud components, such as a steel beam and a concrete panel, are obtained. The multiple feature points can be, for example, feature points on the opposite surfaces when the two point cloud components are connected, and / or feature points on the connecting component between the two point cloud components. Based on the positions of the multiple feature points of the two point cloud components, an accurate error control relationship can be constructed, and the error control relationship is used to represent the error when the two point cloud components are assembled. Finally, based on the error control relationship, the virtual pre-assembly of, for example, the two point cloud components can be controlled and the assembly results can be recorded.

[0051] In this embodiment, BIM is combined with three-dimensional laser radar (LiDAR) technology to collect and process three-dimensional laser point cloud data of components of prefabricated bridges, such as steel-concrete composite bridges. The point cloud data obtained by the BIM model of the prefabricated bridge is aligned and used as a reference to accurately segment the three-dimensional laser point cloud data of, for example, two components of the bridge to determine the point cloud components corresponding to each component. The point cloud components are used to construct an error control relationship to complete, for example, the virtual pre-assembly of two point cloud components. It has the characteristics of high efficiency and high precision, which makes up for the low precision and slow efficiency of the prefabricated component pre-assembly of traditional prefabricated bridges, improves the accuracy of the pre-assembly results of prefabricated bridges, and can accurately guide subsequent actual assembly work, such as between the bridge deck and the steel beam of the prefabricated bridge, to avoid rework and other problems.

[0052] In order to further improve the accuracy of the pre-assembly inspection results of the prefabricated bridge, based on the above embodiments, in one embodiment, the model point cloud data may include point cloud data of multiple different components and corresponding point cloud identifiers. In one embodiment, the point cloud identifiers corresponding to the point cloud data of multiple different components in the model point cloud data at least include component labels and / or color information, and different component labels and / or color information indicate that the corresponding point cloud data belong to different components.

[0053] Exemplarily, for the above-mentioned BIM model, for example, the software Tekla Structure can be used to create the model of each component such as a steel beam, and then the OBJ format point cloud file is exported and converted into a txt format (x, y, z) point cloud, and then labels and colors are added to different components according to different file names. For example, the definition: l=1 means that the point cloud belongs to component 1 and is marked as purple; l=2 means that the point cloud belongs to component 2 and is marked as orange; l=3 means that the point cloud belongs to component 3 and is marked as red; l=4 means that the point cloud belongs to component 4 and is marked as green; l=5 means that the point cloud belongs to component 5 and is marked as yellow. The same method can be used to process the model of components in the BIM model, such as concrete panels. After converting them into txt format (x, y, z) point clouds, labels and colors can be added to different components according to different file names.

[0054] Correspondingly, in the above-mentioned step S102, determining at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the alignment result may specifically include the following steps: using a kd-tree search tree to construct an index structure of the model point cloud data, traversing each point cloud in the component point cloud data based on the index structure and using the kd-tree search tree to find the nearest neighbor point cloud of each point cloud within a preset radius, and updating the point cloud identifier of the corresponding point cloud in the component point cloud data based on the point cloud identifier of the nearest neighbor point cloud; when the traversal is completed, determining at least two point cloud components corresponding to the component point cloud data by segmentation based on the updated point cloud identifier of each point cloud in the component point cloud data.

[0055] Exemplarily, after the registration of BIM point cloud data, i.e., model point cloud data, and scanning point cloud data, i.e., component point cloud data, such as coarse registration and fine registration, automatic semantic segmentation is performed. To this end, a kd tree search tree is used to construct an index structure of the target point cloud (the point cloud converted from the BIM model, i.e., the model point cloud data), so that the nearest neighbor point cloud can be efficiently searched. On this basis, each point cloud in the source point cloud (three-dimensional laser scanning point cloud, i.e., component point cloud data) is traversed, and the k-dtree search tree is used to find the nearest neighbor point cloud within a certain radius. If the nearest neighbor point cloud is found, the color of the corresponding point cloud in the corresponding source point cloud is updated according to the label and color information of the nearest neighbor point cloud to ensure that it is consistent with the color of the nearest neighbor point cloud. If the nearest neighbor point cloud cannot be found, the color of the corresponding point cloud is set to black, indicating that the point cloud has no corresponding matching point in the model point cloud data.

[0056] Through the above-mentioned process based on registration and semantic segmentation, efficient automatic alignment and color update of the source point cloud and the target point cloud (i.e., the model point cloud data) can be achieved, so that the final point cloud data is not only geometrically aligned, but also highly consistent in semantic information. This method has both high efficiency and robustness while maintaining accuracy, and is particularly suitable for point cloud processing and analysis in complex building environments. Therefore, the scheme based on the above-mentioned embodiment can more accurately segment the three-dimensional laser point cloud data of, for example, two components of the bridge to accurately determine the point cloud components corresponding to each component, and construct an error control relationship based on the point cloud components to complete the virtual pre-assembly of, for example, two point cloud components. Since the three-dimensional laser point cloud data of each component obtained by segmentation is more accurate, the point cloud components of each component corresponding to the determined three-dimensional laser point cloud data are more accurate, and the error control relationship constructed based on this is more accurate, so the final virtual pre-assembly result is more accurate, thereby further improving the accuracy of the virtual pre-assembly result of the prefabricated bridge.

[0057] In a specific example, if Figure 2 As shown in the figure, (a) shows the initial spatial relationship between the source point cloud and the target point cloud of the steel beam before registration; (b) shows the state after rough registration; (c) shows the result after ICP fine registration; (d) shows the point cloud component obtained after automatic semantic segmentation. Figure 3 As shown in the figure, (a) represents the initial spatial relationship between the source point cloud and the target point cloud of the bridge deck before registration; (b) is the state after coarse registration; (c) is the result after ICP fine registration; (d) is the point cloud component obtained after automatic semantic segmentation.

[0058] Based on any one of the above embodiments, in one embodiment, obtaining multiple feature points of the at least two point cloud components in step S103 may specifically include: determining at least one connecting component between the at least two point cloud components; and obtaining multiple feature points of the at least one connecting component.

[0059] Exemplarily, based on the results of semantic segmentation, feature points of connection components such as two point cloud components such as a bridge deck and a steel beam are extracted respectively. The steel beam and the bridge deck in the prefabricated bridge are prefabricatedly connected using PPNG (Perforated plate nesting-grout connection structure and installation process of prefabricated composite beams). According to the characteristics of the connection components, the two end points of the bottom of the bridge deck, such as the transverse connecting steelplate, are extracted, marked in orange, and numbered B1-32; the two end points of the shear-resistant steel plate in the steel beam are extracted, marked in green, and numbered L1-16; in addition, the two end points of the bulkhead in the steel beam can also be extracted, marked in blue, and numbered L17-20, as a side baffle plate spacing indicator for detecting the steel beam. The above endpoints can be used as multiple feature points of connection components such as transverse connecting steelplates, shear-resistant steelplates, etc.

[0060] It should be noted that due to the influence of high stress concentration, the connection points between prefabricated components often become the weak link in the assembly of the overall bridge structure. The accumulated errors in the processing of superimposed components may cause the assembly of the prefabricated bridge structure to encounter problems such as misalignment or collision. To this end, in this embodiment, when determining the error control relationship before virtual pre-assembly, the error control relationship is determined based on the positions of multiple feature points on at least one connecting component between the components. The error control relationship constructed in this way is more accurate and can be used to accurately control the virtual pre-assembly results, making the final virtual pre-assembly results more accurate, thereby further improving the accuracy of the virtual pre-assembly results of the prefabricated bridge.

[0061] Based on any one of the above embodiments, in one embodiment, the error control relationship includes a minimum stitching error control relationship between the at least two point cloud components, and the minimum stitching error control relationship is determined by a minimum stitching error control function.

[0062] For example, in the semantic segmentation stage, the preliminary registration of the point cloud data of two point cloud components, such as steel beams and bridge decks, has been achieved. After completing the extraction of the control points, i.e., feature points, of the virtual pre-assembly, a minimum splicing error control function, such as the minimum splicing error control equation, is established. Using the Pugh analysis method, the shear steel plates and cross-connecting steel plates using the PPNG connection method are finely pre-assembled, the bridge deck posture is adjusted, and the steel beams are relatively static, such as Figure 4 As shown in (a) in the figure, the displacement of the bridge deck is controlled within the reserved error range in the design. When the splicing error is minimized, the optimal virtual pre-assembly state is obtained.

[0063] Based on any one of the above embodiments, in one embodiment, the method may further include: after the virtual pre-assembly of the at least two point cloud components reaches an optimal virtual pre-assembly state, performing a collision check between the at least two point cloud components and recording the collision check result.

[0064] For example, after reaching the optimal virtual pre-assembly state, a collision check may be performed. Figure 4 As shown in (b), after reaching the optimal pre-assembly state, if the shear steel plate and the cross-connection steel plate still collide, the characteristic points of the shear steel plate at the corresponding position are marked in red, and the collision check results are recorded and visualized. This can fully reflect the virtual pre-assembly and collision check results, and improve detection efficiency and detection accuracy.

[0065] Based on any one of the above embodiments, in one embodiment, point cloud registration of the component point cloud data and the model point cloud data is performed in step S102, which may specifically include: using Harris corner detection to extract matching corner point pairs in the component point cloud data and the model point cloud data; using a random sampling consistent RANSAC algorithm to estimate the rigid body transformation matrix between the point clouds according to the matching corner point pairs, and completing a coarse registration based on the rigid body transformation matrix; and performing a fine registration on the coarse registration result based on a fine registration algorithm, wherein the fine registration algorithm at least includes a closest point iterative ICP algorithm.

[0066] Exemplarily, Harris corner detection is used to extract corner features in the source point cloud and the target point cloud. Harris corner detection algorithm is a classic feature extraction method that captures corners with significant features in the geometric structure by evaluating the gradient changes in the local area. These corners provide key features in point cloud registration. Through Harris corner detection, a preliminary feature correspondence can be established for the source point cloud and the target point cloud. After the feature extraction is completed, a coarse registration strategy based on corner matching is adopted. Using the RANSAC algorithm, the rigid body transformation matrix (including rotation and translation transformation matrices) between the point clouds is estimated based on the matched corner pairs, thereby completing the coarse registration step. This step quickly achieves the preliminary alignment of the two sets of point clouds through the matching of the geometric features of the corners, and effectively reduces the computational complexity of the subsequent registration. In order to further improve the registration accuracy, the ICP algorithm is then used for fine registration. The ICP algorithm iteratively calculates the Euclidean distance between the nearest points in the two sets of point clouds, adjusts the transformation matrix of the point cloud, and gradually minimizes the alignment error between the point clouds. This process is optimized through continuous iteration to make the alignment of the source point cloud and the target point cloud more accurate, and finally obtain a highly consistent registration result.

[0067] In this embodiment, the two-step registration method combining Harris corner point detection and ICP can effectively use corner point features for rapid rough registration, and fine-tune the registration results through the ICP algorithm. This combination greatly reduces the initial error of ICP, makes the registration process more efficient and accurate, and performs more robustly when processing point cloud data containing noise, so that the accuracy of the point cloud registration results can be improved, and then the three-dimensional laser point cloud data of, for example, two components of the bridge can be more accurately segmented to accurately determine the point cloud components corresponding to each component, and the error control relationship is constructed based on the point cloud components to complete the virtual pre-assembly of, for example, two point cloud components. Since the three-dimensional laser point cloud data of each component obtained by segmentation is more accurate, the point cloud components of each component corresponding to the determined three-dimensional laser point cloud data are more accurate, and the error control relationship constructed based on this is more accurate, so the final virtual pre-assembly result is more accurate, thereby further improving the accuracy of the virtual pre-assembly result of the prefabricated bridge.

[0068] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc. In addition, it is also easy to understand that these steps may be, for example, executed synchronously or asynchronously in multiple modules / processes / threads.

[0069] like Figure 5 As shown, the embodiment of the present disclosure provides a LiDAR-based virtual intelligent pre-assembly system for assembled bridges, comprising:

[0070] The data acquisition module 401 is used to acquire the component point cloud data of at least two prefabricated components constituting the prefabricated bridge based on the three-dimensional laser radar scanning; acquire the BIM model of the prefabricated bridge, and determine the model point cloud data based on the BIM model;

[0071] A registration and segmentation module 402 is used to perform point cloud registration on the component point cloud data and the model point cloud data, and to determine at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result;

[0072] The virtual assembly module 403 is used to obtain multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control the virtual pre-assembly of the at least two point cloud components based on the error control relationship.

[0073] In one embodiment, the model point cloud data includes point cloud data of multiple different components and corresponding point cloud identifiers; the registration and segmentation module determines at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result, including:

[0074] Using a kd-tree search tree to construct an index structure of the model point cloud data, traversing each point cloud in the component point cloud data based on the index structure and using the kd-tree search tree to find the nearest neighbor point cloud of each point cloud within a preset radius, and updating the point cloud identifier of the corresponding point cloud in the component point cloud data based on the point cloud identifier of the nearest neighbor point cloud;

[0075] After the traversal is completed, at least two point cloud components corresponding to the component point cloud data are determined based on the point cloud identification segmentation of each point cloud in the updated component point cloud data.

[0076] In one embodiment, the point cloud identification corresponding to the point cloud data of multiple different components in the model point cloud data includes at least component labels and / or color information, and different component labels and / or color information indicate that the corresponding point cloud data belong to different components.

[0077] In one embodiment, the virtual assembly module obtains multiple feature points of the at least two point cloud components, including: determining at least one connecting component between the at least two point cloud components; and obtaining multiple feature points of the at least one connecting component.

[0078] In one embodiment, the error control relationship includes a minimum stitching error control relationship between the at least two point cloud components, and the minimum stitching error control relationship is determined by a minimum stitching error control function.

[0079] In one embodiment, the system may further include a collision check module for performing a collision check between the at least two point cloud components and recording the collision check result after the virtual pre-assembly of the at least two point cloud components reaches an optimal virtual pre-assembly state.

[0080] In one embodiment, the registration and segmentation module performs point cloud registration on the component point cloud data and the model point cloud data, including: using Harris corner detection to extract matching corner point pairs in the component point cloud data and the model point cloud data; using the random sampling consistent RANSAC algorithm to estimate the rigid body transformation matrix between the point clouds according to the matching corner point pairs, and completing coarse registration based on the rigid body transformation matrix; and performing fine registration on the coarse registration result based on a fine registration algorithm, and the fine registration algorithm at least includes a closest point iterative ICP algorithm.

[0081] Regarding the system in the above embodiment, the specific manner in which each module performs operations and the corresponding technical effects brought about have been described in detail in the embodiment of the method, and will not be elaborated here.

[0082] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the present disclosure, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of a module or unit described above can be further divided into multiple modules or units for concretization. The components displayed as modules or units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. Those of ordinary skill in the art can understand and implement it without paying creative work.

[0083] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for virtual intelligent pre-assembly of prefabricated bridges based on LiDAR according to any of the above embodiments is implemented.

[0084] Exemplarily, the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0085] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0086] The present disclosure also provides an electronic device, including a processor and a memory, wherein the memory is used to store a computer program, wherein the processor is configured to execute the LiDAR-based virtual intelligent pre-assembly method for prefabricated bridges in any of the above embodiments by executing the computer program.

[0087] Refer to the following Figure 6 The electronic device 600 according to this embodiment of the present invention is described. Figure 6 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0088] like Figure 6 As shown, the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0089] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above method embodiment section of this specification. For example, the processing unit 610 can perform the following steps: Figure 1 The steps of the method shown in .

[0090] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0091] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0092] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0093] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 650. In addition, the electronic device 600 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 through a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0094] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software, or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the method steps of the above-mentioned embodiments according to the implementation methods of the present disclosure.

[0095] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0096] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be 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 present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A LiDAR-based virtual intelligent pre-assembly method for assembled bridges, characterized in that: include: Acquire component point cloud data of at least two prefabricated components constituting the assembled bridge based on three-dimensional laser radar scanning; Acquire a BIM model of the prefabricated bridge, and determine model point cloud data based on the BIM model; Performing point cloud registration on the component point cloud data and the model point cloud data, and determining at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result; Acquire multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control and implement virtual pre-assembly of the at least two point cloud components based on the error control relationship.

2. The method according to claim 1, characterized in that The model point cloud data includes point cloud data of multiple different components and corresponding point cloud identifiers; the semantic segmentation based on the registration result determines at least two point cloud components corresponding to the component point cloud data, including: Using a kd-tree search tree to construct an index structure of the model point cloud data, traversing each point cloud in the component point cloud data based on the index structure and using the kd-tree search tree to find the nearest neighbor point cloud of each point cloud within a preset radius, and updating the point cloud identifier of the corresponding point cloud in the component point cloud data based on the point cloud identifier of the nearest neighbor point cloud; After the traversal is completed, at least two point cloud components corresponding to the component point cloud data are determined based on the point cloud identification segmentation of each point cloud in the updated component point cloud data.

3. The method according to claim 2, characterized in that The point cloud identifications corresponding to the point cloud data of multiple different components in the model point cloud data at least include component labels and / or color information, and different component labels and / or color information indicate that the corresponding point cloud data belong to different components.

4. The method according to any one of claims 1 to 3, characterized in that The obtaining of the plurality of feature points of the at least two point cloud components comprises: Determine at least one connected component between the at least two point cloud components; and obtain a plurality of feature points of the at least one connected component.

5. The method according to any one of claims 1 to 3, characterized in that: The error control relationship includes a minimum stitching error control relationship between the at least two point cloud components, and the minimum stitching error control relationship is determined by a minimum stitching error control function.

6. The method according to any one of claims 1 to 3, characterized in that: The method further includes: After the virtual pre-assembly of the at least two point cloud components reaches an optimal virtual pre-assembly state, a collision check is performed between the at least two point cloud components and a collision check result is recorded.

7. The method according to any one of claims 1 to 3, characterized in that The step of performing point cloud registration on the component point cloud data and the model point cloud data comprises: Using Harris corner detection to extract matching corner pairs in the component point cloud data and the model point cloud data; using the random sampling consistent RANSAC algorithm to estimate the rigid body transformation matrix between the point clouds according to the matching corner pairs, and completing the coarse registration based on the rigid body transformation matrix; The rough registration result is finely registered based on a fine registration algorithm, wherein the fine registration algorithm at least includes an iterative closest point ICP algorithm.

8. A LiDAR-based virtual intelligent pre-assembly system for assembled bridges, characterized in that: include: A data acquisition module, used for acquiring component point cloud data of at least two prefabricated components constituting the assembled bridge based on three-dimensional laser radar scanning; Acquire a BIM model of the prefabricated bridge, and determine model point cloud data based on the BIM model; A registration and segmentation module, used to perform point cloud registration on the component point cloud data and the model point cloud data, and to determine at least two point cloud components corresponding to the component point cloud data based on semantic segmentation of the registration result; A virtual assembly module is used to obtain multiple feature points of the at least two point cloud components, construct an error control relationship based on the multiple feature points of the at least two point cloud components, and control the virtual pre-assembly of the at least two point cloud components based on the error control relationship.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the LiDAR-based virtual intelligent pre-assembly method for prefabricated bridges as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: processor; as well as Memory for storing computer programs; Wherein, the processor is configured to execute the LiDAR-based virtual intelligent pre-assembly method for prefabricated bridges according to any one of claims 1 to 7 by executing the computer program.

Images