Indoor Map Construction Method and System of BIM and LiDAR

Abstract

The present invention discloses a method and system for building an indoor map of BIM and laser radar, and the method includes: an engineer interactive control interface sends an indoor map building signal to a BIM and laser radar multimodal data monitoring and acquisition platform; after the BIM and laser radar multimodal data monitoring and acquisition platform receives the indoor map building signal, multiple data acquisition sensors collect multiple radar setting parameters and BIM setting parameters of BIM and laser radar at each unit time, and data processing and fusion parameters of the working status of BIM and laser radar at each unit time, and store and send the collected data to the engineer interactive control interface; the engineer interactive control interface determines whether the multiple radar setting parameters and BIM setting parameters, data processing and fusion parameters are within a preset error allowable threshold; and feedback is given to the BIM and laser radar room scanning point calibration inspection algorithm unit. The present invention can timely send a prompt signal for BIM and laser radar to be optimized, thereby improving the efficiency of indoor map construction.

Description

Technical Field

[0001] The present invention relates to the field of indoor map construction, and particularly to a method and system for indoor map construction using BIM and lidar. Background Art

[0002] The combination of BIM (Building Information Modeling) and lidar technology provides an efficient and accurate solution for indoor map construction. Lidar can quickly obtain high-density three-dimensional point cloud data and capture the geometric features of buildings, while BIM provides a comprehensive view of the building by integrating building information and data management. This combination not only improves the modeling efficiency but also provides data support for subsequent building maintenance and management.

[0003] However, existing methods still face challenges in data processing and model fusion. The point cloud data generated by lidar often contains noise and inconsistencies, and complex registration and filtering steps are required to improve the data quality. In addition, the fusion process of the point cloud and the BIM model may be restricted by the algorithm efficiency, resulting in a long calculation time and difficulty in realizing real-time applications.

[0004] In addition, current methods often rely on traditional geometric processing techniques and lack an in-depth understanding of the complex structures and features in the point cloud. This limits their application capabilities in complex environments, unable to effectively process point cloud data with different scales and densities, and thus affecting the accuracy and practicality of indoor maps. Therefore, seeking new algorithms and technical means to improve the processing accuracy and efficiency is the key direction for future development.

[0005] To solve the above problems, a method and system for indoor map construction using BIM and lidar are proposed in this application. Summary of the Invention

[0006] To overcome the disadvantages and deficiencies of the existing technology, the present invention provides a method and system for indoor map construction using BIM and lidar.

[0007] An embodiment of the present invention provides a method for indoor map construction using BIM and lidar, which includes:

[0008] An engineer interaction control interface sends an indoor map construction signal to a BIM and lidar multi-modal data monitoring and acquisition platform;

[0009] After receiving the indoor map construction signal, the BIM and lidar multi-modal data monitoring and acquisition platform uses various data acquisition sensors to collect various radar setting parameters of BIM and lidar and BIM setting parameters per unit time, as well as data processing and fusion parameters of the working states of BIM and lidar per unit time, and stores and sends the collected data to the engineer interaction control interface;

[0010] The engineer interaction control interface determines whether the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within the preset error tolerance threshold; if the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway.

[0011] Further, the multiple radar setting parameters and BIM setting parameters include: scanning frequency, scanning range, point cloud resolution, data acquisition rate, geometric data, attribute information, hierarchical structure, time information.

[0012] Further, the data processing and fusion parameters include: registration algorithm, data filtering, feature extraction, model generation.

[0013] Further, it also includes: pre-setting a map construction model data cloud in the engineer interaction control interface, and storing the error tolerance ranges of the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters, as well as information on the map construction type in the map construction model data cloud.

[0014] Further, when the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, generating a BIM and lidar optimization prompt signal and feeding it back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway includes:

[0015] The engineer interaction control interface evaluates the received multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters against the map construction type in the map construction model data cloud;

[0016] The engineer interaction control interface sends a corresponding BIM and lidar optimization prompt signal to the edge computing gateway according to the evaluation result.

[0017] Further, when the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, generating a BIM and lidar optimization prompt signal and feeding it back to the BIM and lidar room scan point calibration and inspection algorithm unit also includes:

[0018] The engineer interaction control interface sends a software program run / interrupt or hardware electrical component run / interrupt signal to the BIM and lidar main board according to the evaluation result to control the running steps of the BIM and lidar per unit time.

[0019] Further, the feedback to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway includes: the edge computing gateway feeds the BIM and lidar to-be-optimized prompt signal to the BIM and lidar room scan point calibration and inspection algorithm unit in a special coding manner.

[0020] The embodiment of the present invention further provides an indoor map construction system for BIM and lidar, including a BIM and lidar control server, multiple BIM and lidar room scan points, an edge computing gateway, and a BIM and lidar room scan point calibration and inspection algorithm unit. An engineer interaction control interface is set on the BIM and lidar control server, and a BIM and lidar multimodal data monitoring and acquisition platform is set on each of the multiple BIM and lidar room scan points;

[0021] The engineer interaction control interface is used to send an indoor map construction signal to the BIM and lidar multimodal data monitoring and acquisition platform;

[0022] After receiving the indoor map construction signal, the BIM and lidar multimodal data monitoring and acquisition platform uses various data acquisition sensors to collect various radar setting parameters and BIM setting parameters of BIM and lidar per unit time, and data processing and fusion parameters of the working state of BIM and lidar per unit time, and stores and sends the collected data to the engineer interaction control interface;

[0023] The engineer interaction control interface is further used to determine whether the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within a preset error tolerance threshold; if the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar to-be-optimized prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway.

[0024] Further, the edge computing gateway is used to feed the BIM and lidar to-be-optimized prompt signal to the BIM and lidar room scan point calibration and inspection algorithm unit in a special coding manner.

[0025] Further, the engineer interaction control interface is used to evaluate the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters received with the map construction types in the map construction model data cloud; and according to the evaluation result, send a corresponding BIM and lidar to-be-optimized prompt signal to the edge computing gateway.

[0026] Beneficial effects:

[0027] The present invention provides a method and system for constructing an indoor map using BIM and lidar, which can timely send a prompt signal for BIM and lidar to be optimized, improve the efficiency of indoor map construction, reduce costs, and the method for constructing an indoor map using BIM and lidar provided by the embodiments of the present invention is mainly executed on the engineer interaction control interface and the BIM and lidar multimodal data monitoring and acquisition platform, with a fast response speed. The present invention uses a graph neural network in combination with BIM and lidar to perform calibration inspection on room scan points. The graph neural network can effectively capture the complex relationships between points in the point cloud data, and through deep learning of node features, achieve precise correction of inaccurate scan points. In addition, the graph structure of the graph neural network can process non-Euclidean space data, improving the robustness and flexibility of the model on irregular data sets, thus significantly improving the accuracy and efficiency of room scanning and providing a more reliable basis for intelligent building management. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0029] Figure 1 It is a schematic flowchart of the method for constructing an indoor map using BIM and lidar provided by the embodiments of the present invention;

[0030] Figure 2 It is a structural composition diagram of the system for constructing an indoor map using BIM and lidar provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The following will further describe the present application in detail with reference to the drawings and specific embodiments.

[0032] Please refer to Figure 1 , the schematic flowchart of the method for constructing an indoor map using BIM and lidar provided by the embodiments of the present invention, and the method includes:

[0033] Step 101: The engineer interaction control interface sends an indoor map construction signal to the BIM and lidar multimodal data monitoring and acquisition platform;

[0034] Step 102. After the BIM and lidar multimodal data monitoring and acquisition platform receives the indoor map construction signal, various data acquisition sensors collect various radar setting parameters and BIM setting parameters of BIM and lidar per unit time, and data processing and fusion parameters of the working states of BIM and lidar per unit time, and store and send the collected data to the engineer interaction and control interface;

[0035] Step 103. The engineer interaction and control interface determines whether the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within the preset error tolerance threshold; if the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization required prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway.

[0036] In the embodiment of the present invention, the BIM and lidar room scan point calibration and inspection algorithm unit uses a graph neural network for room scan point calibration and inspection, including:

[0037] Graph construction: For point cloud data , each node , where i = 1, 2,..., n, has a feature vector , where i = 1, 2,..., n;

[0038] Node features: , T is the matrix transpose operation,

[0039] Edge features: Edge is connected to nodes , , and the edge feature is defined as their Euclidean distance :

[0040] ,

[0041] Graph neural network layer: Each layer of the graph neural network updates the node features through graph convolution. Assuming that the node feature of the +1 layer is , the update formula is:

[0042] ,

[0043] Where: is the set of neighbor nodes of node , is the normalization factor, is the weight matrix of the layer, is layer bias term, is the activation function. After passing through multiple GNNs in the output layer, the final representation of the nodes in the -th layer is obtained. Then, a final classification or regression task is performed through a fully connected layer:

[0044] ,

[0045] where is the prediction result of node , is the weight matrix of the +1-th layer, is the node feature of the +1-th layer, is the -th layer bias term;

[0046] Loss function: The loss function is used to measure the difference between the prediction result and the true label, and the expression is:

[0047] ,

[0048] where is the mean squared error of the loss function, is the true label;

[0049] Optimization process: The backpropagation algorithm is used to update the model parameters. Assuming the use of the stochastic gradient descent optimizer, the parameter update rule is:

[0050] ,

[0051] where is the learning rate;

[0052] Calibration test: After completing the training, the model is used for calibration test. For new scan points, the model is input, and the calibration value is obtained and compared with the BIM model to evaluate the calibration effect.

[0053] In the embodiment of the present invention, the engineer interactive control interface can send an indoor map construction signal to the BIM and laser radar multimodal data monitoring platform. The BIM and laser radar multimodal data monitoring platform can control the collection of multiple radar setting parameters and BIM setting parameters, data processing and fusion parameters according to the indoor map construction signal and feedback to the engineer interactive control interface. In this way, the engineer interactive control interface can determine whether the multiple radar setting parameters and BIM setting parameters, data processing and fusion parameters are within the error allowable threshold. If not, a BIM and laser radar to be optimized prompt signal is automatically generated and fed back to the BIM and laser radar room scanning point calibration inspection algorithm unit through the edge computing gateway. The management personnel can timely understand the map construction information of BIM and laser radar per unit time, and process the map construction in time to prevent damage to BIM and laser radar per unit time. The method provided in the embodiment of the present invention can automatically perform inspections, and can timely send BIM and laser radar to be optimized prompt signals, improve the efficiency of indoor map construction, and reduce costs. The indoor map construction method of BIM and laser radar provided in the embodiment of the present invention is mainly executed on the engineer interactive control interface and the BIM and laser radar multimodal data monitoring platform, and has a fast response speed.

[0054] Specifically, in step 101, the engineer interactive control interface can send the indoor map construction signal to the BIM and lidar multimodal data monitoring platform in different ways, for example, sending the indoor map construction signal in a timed manner, that is, sending the indoor map construction signal once every same period of time, which can be the default duration of the device or a custom duration.

[0055] In addition, the indoor map building signal can be sent irregularly, that is, the indoor map building signal is sent every different time intervals. In this way, it can be manually started by the management personnel, that is, the management personnel can evaluate whether it is necessary to send the indoor map building signal based on their own experience or actual working status. It is also possible to divide each cycle into several different time periods in advance, and set a time interval for different time periods, so that within a time period, the indoor map building signal is sent regularly according to the corresponding time interval. This method is actually a scheduled inspection, but the time interval is different in different time periods. The advantage of setting different time periods is that the sampling frequency can be set in a targeted manner.

[0056] In step 102, the multiple data acquisition sensors are started, and multiple radar setting parameters and BIM setting parameters of the BIM and the lidar at each unit time, and data processing and fusion parameters of the working status of the BIM and the lidar at each unit time are collected according to the type of each sensor in the sensor group.

[0057] Among them, the various radar setting parameters and BIM setting parameters include: scanning frequency, scanning range, point cloud resolution, data acquisition rate, geometric data, attribute information, hierarchy, and time information. The data processing and fusion parameters include: registration algorithm, data filtering, feature extraction, and model generation.

[0058] Scanning frequency: Affects the point cloud density and accuracy. Scanning range: The distance and angle that the lidar can cover. Point cloud resolution: The number and accuracy of points generated by each scan. Data acquisition rate: The amount of data that can be acquired per unit time. Geometric data: The shape, size, layout, etc. of the building. Attribute information: Material, structure, equipment information, etc. Hierarchy: The hierarchical relationship of each part of the building. Time information: Data on the construction progress or building life cycle.

[0059] Registration algorithm: Aligns the point cloud data collected at different times or angles.

[0060] Data filtering: Removes noise and unnecessary data points to improve the point cloud quality.

[0061] Feature extraction: Extracts the feature information of the building from the point cloud, such as walls, doors, and windows.

[0062] Model generation: Converts the point cloud data into a BIM model, which usually involves geometric modeling and attribute association.

[0063] The above data can be collected by the corresponding sensors. The combination of each sensor can be collectively referred to as a sensor group. A sensor group can be set on BIM and the lidar per unit time, so as to collect the various radar setting parameters and BIM setting parameters of BIM and the lidar per unit time, and the data processing and fusion parameters of the working states of BIM and the lidar per unit time, thereby realizing the remote construction function of BIM and the lidar per unit time.

[0064] The functions of the above various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are different. For example, for the various radar setting parameters and BIM setting parameters, they can directly reflect the current state of BIM and the lidar per unit time, and can determine whether map construction occurs for BIM and the lidar per unit time. For the data processing and fusion parameters of the working states of BIM and the lidar per unit time, they can directly reflect the current working state of BIM and the lidar per unit time, such as whether there are significant changes in the working states of BIM and the lidar per unit time, or whether there are phenomena such as deterioration of the working state. In this way, it can be expected whether map construction will occur for BIM and the lidar per unit time, so as to process it in a timely manner.

[0065] The BIM and LiDAR multimodal data monitoring and acquisition platform can store the data collected by the sensor group and eventually send it to the engineer's interactive control interface.

[0066] In step 103, after receiving the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters, the engineer interactive control interface can determine whether the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within a preset error tolerance threshold.

[0067] If the multiple radar setting parameters and BIM setting parameters, data processing and fusion parameters are not within the preset error tolerance threshold, that is, the multiple radar setting parameters and BIM setting parameters are not within the preset error tolerance threshold, and the data processing and fusion parameters are not within the preset error tolerance threshold, then it is necessary to generate a BIM and lidar optimization prompt signal and feed it back to the BIM and lidar room scanning point calibration inspection algorithm unit through the edge computing gateway.

[0068] The BIM and laser radar room scanning point calibration and verification algorithm unit can be the BIM and laser radar room scanning point calibration and verification algorithm unit of the maintenance personnel, so that the maintenance personnel can promptly process the map construction that is occurring or is about to occur by BIM and laser radar per unit time. The BIM and laser radar room scanning point calibration and verification algorithm unit can also be the BIM and laser radar room scanning point calibration and verification algorithm unit of the management personnel, so that the management personnel can arrange personnel to process according to the area to which the BIM and laser radar belong per unit time.

[0069] Calibration verification algorithms for BIM and LiDAR room scan points are often used to ensure that the geometric characteristics of BIM and LiDAR or physical prototypes meet the design requirements during the manufacturing process. The main goal of such algorithms is to calibrate the measurement data, correct errors, and perform accurate construction accuracy assessment. A typical calibration verification algorithm process may include the following steps:

[0070] Using multiple data acquisition sensors to collect multiple radar setting parameters and BIM setting parameters of BIM and LiDAR at each unit time, data processing and fusion parameters of BIM and LiDAR working status at each unit time, the calibration inspection algorithm can ensure the consistency and accuracy of BIM and LiDAR in the manufacturing process, and help to timely detect deviations or equipment abnormalities in production. Depending on the specific production environment and sensor type, the algorithm may be different.

[0071] In order to better analyze the map construction of BIM and lidar per unit time, embodiments of the present invention preferably collect various radar setting parameters, BIM setting parameters, and data processing and fusion parameters of BIM and lidar per unit time, as well as the working state data of BIM and lidar per unit time, and analyze the map construction based on these two types of information. When the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are all within the preset error tolerance threshold, a BIM and lidar optimization prompt signal is not generated. When any one of the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters is not within the preset error tolerance threshold, a BIM and lidar optimization prompt signal is generated. This can ensure the timely discovery of map construction and the processing of map construction.

[0072] In one embodiment, the indoor map construction method of BIM and lidar further includes:

[0073] Pre-set a map construction model data cloud in the engineer interaction control interface, and store the error tolerance ranges of the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters, as well as information on the map construction type in the map construction model data cloud.

[0074] In this embodiment, the error tolerance ranges of the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters can be pre-stored in the map construction model data cloud. In this way, the engineer interaction control interface can retrieve the corresponding error tolerance ranges at any time during map construction analysis and perform evaluations. At the same time, information on the map construction type can also be stored in the map construction model data cloud, so that the type of map construction can be determined, facilitating maintenance personnel to perform map construction processing.

[0075] In one embodiment, if the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through an edge computing gateway, including:

[0076] The engineer interaction control interface evaluates the received various radar setting parameters, BIM setting parameters, and data processing and fusion parameters with the map construction type in the map construction model data cloud;

[0077] The engineer interaction control interface sends a corresponding BIM and lidar optimization prompt signal to the edge computing gateway according to the evaluation result.

[0078] After determining that the various radar setting parameters, BIM setting parameters, data processing and fusion parameters are not within the preset error tolerance range, the various radar setting parameters, BIM setting parameters, data processing and fusion parameters and the map construction type in the map construction model data cloud can be evaluated. Then, the BIM and LiDAR optimization prompt signal is sent to the edge computing gateway, and the edge computing gateway is used to feedback to the BIM and LiDAR room scanning point calibration inspection algorithm unit. In this way, not only can the information of BIM and LiDAR map construction or map construction in unit time be obtained, but also the information of what map construction has occurred in BIM and LiDAR per unit time can be obtained, so as to facilitate targeted processing.

[0079] In one embodiment, the feedback to the BIM and laser radar room scanning point calibration verification algorithm unit through the edge computing gateway includes:

[0080] The edge computing gateway feeds back the BIM and LiDAR optimization prompt signals to the BIM and LiDAR room scanning point calibration inspection algorithm unit in a special coding manner.

[0081] The edge computing gateway can feed back the BIM and LiDAR optimization prompt signal to the BIM and LiDAR room scanning point calibration verification algorithm unit via email or a signal pop-up window. Of course, it can also be constructed in other ways, such as WeChat, mini-programs, dedicated apps, etc. for reminders.

[0082] In one embodiment, if the multiple radar setting parameters and BIM setting parameters, data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and laser radar optimization prompt signal is generated and fed back to the BIM and laser radar room scanning point calibration inspection algorithm unit through the edge computing gateway, further comprising:

[0083] The engineer interactive control interface sends software program run / interrupt or hardware electrical parts run / interrupt signals to the BIM and lidar mainboard according to the evaluation results to control the operation steps of the BIM and lidar per unit time.

[0084] In this embodiment, when it is determined that the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, and a BIM and lidar optimization prompt signal is generated and sent to the BIM and lidar room scan point calibration and inspection algorithm unit, maintenance personnel can process it in a timely manner. In the embodiment of the present invention, a software program running / interrupt signal or a hardware electrical component running / interrupt signal can also be sent from the engineer interaction control interface to the BIM and lidar main board to control the running steps of the BIM and lidar per unit time. For example, when it is detected that the BIM setting parameter of the working state of the BIM and lidar per unit time is too high, a software program running / interrupt signal can be sent to the relay in the BIM and lidar main board.

[0085] In other embodiments, a neural network model can also be used to predict the collected data, so as to understand the upcoming map construction information in advance, which is beneficial to avoiding the risk of unbalanced data collection of BIM and lidar per unit time.

[0086] Generally speaking, if the BIM and lidar per unit time continuously change in a certain state or some parameters in the working state continuously deviate from the normal value during use, then the risk of damage to the BIM and lidar per unit time is relatively high. Therefore, the embodiment of the present invention can collect historical map construction data, which includes the type of map construction and multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters before the map construction occurs.

[0087] Then, the above historical map construction data is preprocessed. The content of the preprocessing can include normalization, etc., to make the data of each type belong to the same dimension, and then the historical map construction data is divided into a sample set and a test set. Then, a neural network model is used to train the sample set of the historical map construction data, and the test set is used to test the trained neural network model, and finally meet the preset accuracy requirement.

[0088] Please refer to Figure 2 , the embodiment of the present invention also provides an indoor map construction system for BIM and lidar, which includes a BIM and lidar control server, multiple BIM and lidar room scan points, an edge computing gateway, and a BIM and lidar room scan point calibration and inspection algorithm unit. An engineer interaction control interface is set on the BIM and lidar control server, and a BIM and lidar multimodal data monitoring and acquisition platform is set on each of the multiple BIM and lidar room scan points;

[0089] The engineer interaction control interface is used to send an indoor map construction signal to the BIM and lidar multimodal data monitoring and acquisition platform;

[0090] The BIM and lidar multimodal data monitoring and acquisition platform is used to collect various radar setting parameters and BIM setting parameters of BIM and lidar, as well as data processing and fusion parameters of the working states of BIM and lidar per unit time by various data acquisition sensors after receiving the indoor map construction signal, and store and send the collected data to the engineer interaction and control interface;

[0091] The engineer interaction and control interface is further used to determine whether the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within the preset error tolerance threshold; if the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization required prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway.

[0092] In one embodiment, the edge computing gateway is used to feed back the BIM and lidar optimization required prompt signal to the BIM and lidar room scan point calibration and inspection algorithm unit in a special coding manner.

[0093] In one embodiment, the engineer interaction and control interface is used to evaluate the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters received with the map construction types in the map construction model data cloud; and send corresponding BIM and lidar optimization required prompt signals to the edge computing gateway according to the evaluation results.

[0094] The various embodiments in the specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method part. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection threshold of the claims of the present invention.

Claims

1. An indoor map construction method of BIM and lidar, characterized in that The method includes: Step 101, the engineer interaction control interface sends an indoor map construction signal to the BIM and lidar multimodal data monitoring and acquisition platform; Step 102, after the BIM and lidar multimodal data monitoring and acquisition platform receives the indoor map construction signal, multiple data acquisition sensors collect various radar setting parameters of BIM and lidar, BIM setting parameters, and data processing and fusion parameters of the working states of BIM and lidar per unit time, and store and send the collected data to the engineer interaction control interface; Step 103, the engineer interaction control interface determines whether the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within the preset error tolerance threshold; if the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization required prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway; The BIM and lidar room scan point calibration and inspection algorithm unit uses a graph neural network for room scan point calibration and inspection, including: Graph construction: For point cloud data , each node , where i = 1, 2, …, n, has a feature vector , where i = 1, 2, …, n; Node features: , where T represents the matrix transpose operation; Edge feature: Edge is connected to the node , , and the edge feature is defined as their Euclidean distance :[[]] ， Graph neural network layer: Each layer of the graph neural network updates node features through graph convolution. Suppose the node features of the +1-th layer are , then the update formula is: ， Wherein: is the set of neighbor nodes of the node is the normalization factor is the weight matrix of the th layer is the bias term of the th layer. is the activation function. After passing through multiple GNNs in the output layer, the final representation of the nodes in the th layer is obtained, and a final classification or regression task is performed through a fully connected layer: ， Among them, is the prediction result of the node , is the weight matrix of the +1-th layer, is the node feature of the +1-th layer, is the layer bias term; Loss function: Use a loss function to measure the difference between the prediction result and the true label, and the expression is: ， Among them, is the mean square error of the loss function, is the true label; Optimization process: Use the backpropagation algorithm to update the model parameters. Assuming the use of a stochastic gradient descent optimizer, the parameter update rule is: ， Among them, is the learning rate; Calibration and inspection: After training is completed, use the model for calibration and inspection. For new scan points, input them into the model to obtain calibration values and compare them with the BIM model to evaluate the calibration effect.

2. The indoor map construction method of BIM and lidar according to claim 1, characterized in that The various radar setting parameters and BIM setting parameters include: scan frequency, scan range, point cloud resolution, data acquisition rate, geometric data, attribute information, hierarchical structure, time information.

3. The indoor map construction method of BIM and lidar according to claim 1, characterized in that The data processing and fusion parameters include: registration algorithm, data filtering, feature extraction, model generation; The registration algorithm aligns the point cloud data collected at different times or angles; The data filtering is used to remove noise and unnecessary data points to improve the quality of the point cloud; The feature extraction extracts the feature information of the building from the point cloud; The model generation converts the point cloud data into a BIM model, including geometric modeling and attribute association.

4. The indoor map construction method of BIM and lidar according to claim 1, characterized in that, It also includes: Pre-set a map construction model data cloud in the engineer interaction control interface, and store the error tolerance ranges of the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters, and the information of the map construction type in the map construction model data cloud.

5. The indoor map construction method of BIM and lidar according to claim 4, characterized in that, The part that if the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization required prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway includes: The engineer interaction control interface evaluates the various radar setting parameters, BIM setting parameters, and data processing and fusion parameters received with the map construction type in the map construction model data cloud; According to the evaluation results, the engineer interaction control interface sends corresponding BIM and lidar optimization prompt signals to the edge computing gateway.

6. The indoor map construction method of BIM and lidar according to claim 5, characterized in that, If the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway. It also includes: According to the evaluation results, the engineer interaction control interface sends software program running / interrupt or hardware electrical component running / interrupt signals to the BIM and lidar main board to control the running steps of BIM and lidar per unit time.

7. The indoor map construction method of BIM and lidar according to claim 1, characterized in that, The feedback to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway includes: The edge computing gateway feeds back the BIM and lidar optimization prompt signal to the BIM and lidar room scan point calibration and inspection algorithm unit in a special coding manner.

8. An indoor map construction system of BIM and lidar, characterized in that, It includes a BIM and lidar control server, multiple BIM and lidar room scan points, an edge computing gateway, and a BIM and lidar room scan point calibration and inspection algorithm unit; An engineer interaction control interface is set on the BIM and lidar control server, and a BIM and lidar multimodal data monitoring and acquisition platform is set on each of the multiple BIM and lidar room scan points; The engineer interaction control interface is used to send indoor map construction signals to the BIM and lidar multimodal data monitoring and acquisition platform; After receiving the indoor map construction signal, the BIM and lidar multimodal data monitoring and acquisition platform uses multiple data acquisition sensors to collect multiple radar setting parameters and BIM setting parameters of BIM and lidar per unit time, as well as data processing and fusion parameters of the working state of BIM and lidar per unit time, and stores and sends the collected data to the engineer interaction control interface; The engineer interaction control interface is also used to determine whether the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are within the preset error tolerance threshold; if the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters are not within the preset error tolerance threshold, a BIM and lidar optimization prompt signal is generated and fed back to the BIM and lidar room scan point calibration and inspection algorithm unit through the edge computing gateway.

9. The indoor map construction system of BIM and lidar according to claim 8, characterized in that, The edge computing gateway is used to feed back the BIM and lidar optimization prompt signal to the BIM and lidar room scan point calibration and inspection algorithm unit in a special coding manner; The engineer interaction control interface is used to evaluate the multiple radar setting parameters, BIM setting parameters, and data processing and fusion parameters received with the map construction types in the map construction model data cloud; And according to the evaluation results, send corresponding BIM and lidar optimization prompt signals to the edge computing gateway.

Images