Construction site robot real-time monitoring method and system based on BIM

Abstract

The invention relates to a BIM-based construction site robot real-time monitoring method and system, and relates to the technical field of building engineering monitoring, and the method comprises the steps: obtaining a real-time BIM model of a target construction body, planning monitoring points, and obtaining a monitoring point sequence and an abnormal rate sequence; monitoring and controlling a robot according to the monitoring point sequence, extracting monitoring point BIM data and inspection data, and carrying out consistency verification to obtain an initial consistency rate; predicting a movement anomaly rate according to the BIM data, configuring a re-check parameter by combining the initial consistency rate and the anomaly rate, and obtaining a re-check consistency rate through re-check; and according to the anomaly rate, the movement anomaly rate and the re-check consistency rate of the monitoring points, calculating to obtain a fusion anomaly rate, marking the monitoring points and obtaining a monitoring result. The method solves the problems that traditional construction site monitoring only adopts a single monitoring mode and does not consider related errors and probability problems, so that judgment on construction progress management reasonability is not accurate enough, result credibility is not enough, and efficient management and control requirements of a construction site cannot be met.

Description

Technical Field

[0001] The present application relates to the field of construction engineering monitoring, and in particular to a method and system for real-time monitoring of construction site robots based on BIM. Background Art

[0002] With the continuous improvement of the sophistication of construction management, the accuracy and reliability of real-time monitoring of construction sites have become key technical challenges.

[0003] Currently, traditional monitoring methods struggle to accurately reflect actual construction conditions and are subject to multiple error factors, significantly compromising the accuracy of monitoring results. Furthermore, their reliance on a single monitoring method not only reduces the practical value of monitoring data but also increases the complexity and cost of subsequent construction management and problem resolution. This makes monitoring results less reliable and unable to meet the actual needs of efficient construction site management. Summary of the Invention

[0004] In order to solve the above technical problems, this application provides a BIM-based construction site robot real-time monitoring method and system, which improves the current situation in which traditional construction monitoring is difficult to meet the needs of efficient construction site management due to insufficient accuracy and poor credibility.

[0005] The embodiments of this application disclose the following technical solutions: In a first aspect, an embodiment of the present application provides a method for real-time monitoring of a construction site robot based on BIM, the method comprising: Obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a monitoring point sequence and an abnormality rate sequence; According to the monitoring point sequence, the robot is controlled to perform monitoring. When monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted and consistency is verified with the inspection data to obtain an initial consistency rate. Based on the BIM data of the monitoring points, the movement anomaly rate is predicted to obtain the movement anomaly rate. The initial consistency rate and the anomaly rate of the monitoring points are combined to configure the review parameters, and the monitoring points are reviewed and verified for consistency to obtain the review consistency rate. According to the abnormality rate, movement abnormality rate and verification consistency rate of the monitoring points, the fusion abnormality rate is calculated, the monitoring points are marked and the monitoring results are obtained.

[0006] In a second aspect, an embodiment of the present application provides a BIM-based construction site robot real-time monitoring system, the system comprising: A BIM monitoring point planning module is used to obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a monitoring point sequence and an abnormality rate sequence; The robot initial verification module is used to control the robot to monitor according to the monitoring point sequence. When monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted and consistency verification is performed with the inspection data to obtain an initial consistency rate. A movement anomaly review module is used to predict the movement anomaly rate based on the BIM data of the monitoring point, obtain the movement anomaly rate, configure the review parameters based on the initial consistency rate and the anomaly rate of the monitoring point, and perform review consistency verification on the monitoring point to obtain the review consistency rate; The fusion anomaly labeling module is used to calculate the fusion anomaly rate based on the anomaly rate, movement anomaly rate and review consistency rate of the monitoring points, label the monitoring points and obtain the monitoring results.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application proposes a real-time monitoring method and system for construction site robots based on BIM. By obtaining the real-time BIM model of the target construction body and planning the monitoring points, the robot is controlled to monitor and verify the consistency of the monitoring point sequence, and the recheck parameters are configured in combination with the prediction of the movement anomaly rate. Finally, the fusion anomaly rate is calculated and the monitoring results are marked, thereby achieving accurate real-time monitoring of the construction site. First, the real-time BIM model of the target construction body is obtained, and the monitoring point sequence and anomaly rate sequence are planned accordingly; then the robot is controlled to perform monitoring according to the monitoring point sequence, and the BIM data of the monitoring point is extracted and verified with the inspection data for consistency to obtain the initial consistency rate; then the movement anomaly rate is predicted based on the BIM data of the monitoring point, and the recheck parameters are configured in combination with the initial consistency rate and the anomaly rate of the monitoring point, and recheck verification is performed to obtain the recheck consistency rate; finally, the fusion anomaly rate is calculated based on the anomaly rate, movement anomaly rate and recheck consistency rate of the monitoring point, and the monitoring points are marked to obtain the monitoring results.

[0008] The technical solution of this application solves the problem of insufficient monitoring accuracy caused by BIM errors, robot recognition errors, movement error probabilities, etc. in traditional construction site monitoring by integrating the monitoring point planning of the BIM model, robot monitoring and consistency verification, dynamic configuration of movement anomaly rate prediction and review verification, and multi-dimensional anomaly rate fusion analysis, thereby realizing high-precision real-time monitoring of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1A flowchart of a BIM-based construction site robot real-time monitoring method provided in an embodiment of the present application; Figure 2 A schematic structural diagram of a BIM-based construction site robot real-time monitoring system provided in an embodiment of the present application.

[0011] In the accompanying drawings, the components represented by the reference numerals are described as follows: BIM monitoring point planning module 01, robot initial verification module 02, movement anomaly review module 03, fusion anomaly marking module 04. DETAILED DESCRIPTION

[0012] This application provides a real-time monitoring method and system for construction site robots based on BIM, which is used to solve the technical problems existing in the existing technology of insufficient accuracy and poor credibility of construction site monitoring, difficulty in accurately judging the rationality of construction progress management, and inability to meet the needs of efficient construction site management and control.

[0013] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0014] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the described features. In the description of this application, "plurality" means two or more, unless otherwise specifically specified.

[0015] In the description of this application, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0016] Example 1, as shown in the attached Figure 1As shown, the present application provides a real-time monitoring method for a construction site robot based on BIM, the method comprising the following steps: S110: Obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a monitoring point sequence and an abnormality rate sequence; In the embodiment of the present application, in the scenario of real-time monitoring of construction site robots based on BIM, in order to accurately plan monitoring points and obtain the basis for initial abnormality assessment, it is necessary to divide units through real-time BIM models and conduct abnormality rate analysis in combination with historical construction data to establish a scientific monitoring sequence basis.

[0017] Specifically, the real-time BIM model of the target construction object is first retrieved through the BIM management platform at the construction site to ensure real-time synchronization of the model data with the on-site construction progress, providing an accurate digital foundation for subsequent analysis.

[0018] Next, the real-time BIM model is structured and divided into multiple independent model units. At the same time, the key model features of each model unit, such as structural type, construction stage, material properties, etc., are extracted as the core basis for abnormality rate analysis.

[0019] Furthermore, based on these model characteristics, the most similar model unit sets are matched in the historical construction database, and the abnormal proportion of inconsistency between actual construction and BIM model in each set is counted to determine the unit abnormality rate of each model unit.

[0020] Finally, the model units are sorted in descending order according to the unit abnormality rate, and a preset number of high abnormality risk units are screened out as key monitoring objects to form a monitoring point sequence and the corresponding abnormality rate sequence.

[0021] This step achieves targeted planning of monitoring points through the refined disassembly of the real-time BIM model and the correlation analysis of historical construction data, providing a priority basis for the efficient implementation of subsequent robot inspections and accurate identification of anomalies, and ensuring the accuracy and reliability of construction site monitoring.

[0022] Step S110 of the method provided in the embodiment of the present application includes: Obtain the real-time BIM model of the target construction body; Dividing the real-time BIM model to obtain a plurality of model units, and acquiring model features of the plurality of model units; Analyze the unit abnormality rates of multiple model units based on multiple model characteristics; Arrange the abnormality rates of multiple units in descending order, and select a preset number of monitoring points to obtain a monitoring point sequence and an abnormality rate sequence.

[0023] In the embodiment of the present application, in order to achieve accurate monitoring of the construction site, it is necessary to disassemble and analyze the real-time BIM model and combine it with historical construction data to determine high-risk monitoring points, provide a clear and targeted target sequence for subsequent robot inspections, and ensure the efficiency of the monitoring process and the accuracy of anomaly identification.

[0024] Specifically, first, the real-time BIM model of the target construction body is retrieved through the BIM management platform at the construction site.

[0025] Among them, the BIM model is a digital carrier that reflects the current structural status, construction progress and other information of the construction body. It contains detailed data such as component size, material, construction stage, etc. Its real-time nature can ensure that it is updated synchronously with the on-site construction progress, providing reliable basic data for subsequent analysis.

[0026] For example, for an office building under construction, the real-time BIM model will synchronously update information such as the constructed floor structure and the location of installed pipelines.

[0027] Furthermore, the acquired real-time BIM model is structured and divided into multiple independent model units.

[0028] Specifically, the division can be based on the structural characteristics of the construction unit and the construction process, such as by floor, component type, construction area, etc. At the same time, the model features of each BIM model unit are extracted. These model features include but are not limited to the component's geometric parameters, material properties, construction progress nodes, process requirements, etc.

[0029] Among them, geometric parameters include the length, width, height, cross-sectional dimensions, shape, and position coordinates of the components; material properties include material type, material strength grade, material density, durability index, etc.; construction progress nodes include planned start time, planned completion time, actual start time, actual completion time, completion ratio, etc.; process requirements include welding process, assembly accuracy, casting process, etc.

[0030] For example, taking the frame columns of a certain floor as an example, its model features may include a cross-sectional size of 600mm×600mm, a concrete strength grade of C30, a planned pouring completion time of the 30th day, and an actual completion ratio of 70%.

[0031] Furthermore, based on the extracted multiple model features, the unit anomaly rates of multiple model units are analyzed to quantitatively assess the risk level of possible deviations of each model unit during the construction process.

[0032] In the method provided in the embodiment of the present application, the step of “analyzing the unit abnormality rates of multiple model units according to multiple model features” includes: Extracting multiple sets of model units of the same family from the historical construction data based on multiple model features, wherein each set of model units of the same family includes multiple model units of the same family that have the greatest similarity to the corresponding model features; The abnormality ratios of multiple model units of the same family are obtained by processing and obtaining the unit abnormality rates of multiple model units.

[0033] In the embodiment of the present application, in order to scientifically quantify the abnormal risk of each model unit, it is necessary to accurately match the model characteristics with historical construction data, construct a set of similar units and calculate the abnormal proportion, provide an objective numerical basis for the priority ranking of monitoring points, and ensure that subsequent monitoring resources are tilted towards high-risk units.

[0034] Specifically, based on the extracted model features, feature matching is performed within a historical construction database, which stores detailed feature data and corresponding construction records for each model unit in similar past projects, including both normal and abnormal construction cases.

[0035] At the same time, the matching process is to calculate the feature overlap between the current model unit and the historical unit to select multiple historical units with the highest similarity to form the set of model units of the same family corresponding to the model unit; Specifically, the degree of feature overlap is determined by calculating the cosine similarity of the feature vectors of the current model unit and the historical unit in each model feature dimension. First, each model feature is quantified into a vector form. For example, the cross-sectional dimensions in geometric parameters are converted into two-dimensional vectors, the strength grades in material properties are converted into numerical vectors according to standard values, the planned construction period in the construction progress node is converted into a one-dimensional vector, and the accuracy standards in the process requirements are converted into vectors according to the allowable deviation range.

[0036] Furthermore, the cosine similarity between the current unit feature vector and each historical unit feature vector is calculated using the formula " ", where A is the current unit feature vector, B is the historical unit feature vector, is the vector dot product, |A| and |B| are the vector moduli respectively.

[0037] Among them, the closer the calculated similarity value is to 1, the higher the overlap of the characteristics of the two units is. Based on this, the top N historical units with similarity ranking are screened out (N is set according to the amount of historical data, such as 50-200) to form a set of homologous model units.

[0038] Furthermore, the abnormal situations of the selected set of model units of the same family are sorted out and counted to accurately quantify the risk probability of inconsistency between the actual status of each model unit and the BIM model data during the construction process.

[0039] Among them, abnormal situations specifically refer to situations where the actual construction status in historical construction is inconsistent with the BIM model data, including but not limited to the construction progress not being updated according to the model plan, the deviation between the construction size and the model parameters exceeding the standard, the material usage not meeting the model requirements, and the process execution not meeting the model standards.

[0040] When making specific statistics, the construction records of each historical unit in the same family of model units are checked one by one, the model units with any of the above abnormal conditions are marked, and the number of abnormal units is counted.

[0041] For example, a unit set of a certain family model contains 150 historical units. After verification, 24 units are found to be abnormal, including 10 units with delayed progress, 8 units with exceeded dimensions, and 6 units with inconsistent materials. Therefore, the number of abnormal units in this set is 24.

[0042] Furthermore, the anomaly ratio is calculated to determine the unit anomaly rate. This is calculated by dividing the number of anomaly units in the same family model unit set by the total number of units in the set. The result is expressed as a percentage. The specific calculation formula is "Unit anomaly rate = (number of anomaly units in the same family model unit set / total number of units in the same family model unit set) × 100%."

[0043] For example, in the same example above, according to the formula, 24 / 150×100%=16% can be obtained, that is, the unit abnormality rate of the current model unit is 16%.

[0044] During the statistical process, the validity of historical data must be verified simultaneously to eliminate invalid data due to incomplete records or significant differences in project types to ensure the accuracy of the abnormality ratio calculation. For example, if a historical unit only records "construction abnormality" but does not specify the specific abnormality type and parameter deviation value, the unit will be excluded from the statistical scope.

[0045] Similarly, after completing the matching, anomaly statistics and anomaly rate calculation of the same family model unit set for all model units, the unit anomaly rate corresponding to each model unit is formed, realizing the quantitative representation of the anomaly risk of multiple model units.

[0046] Furthermore, after obtaining the unit abnormality rates of all model units, these unit abnormality rates are sorted in descending order, so that the abnormality risk degree of each model unit forms a clear priority sequence.

[0047] Among them, the sorting process is based solely on the numerical value of the unit anomaly rate. The larger the numerical value of the model unit, the higher the risk that its actual status during construction will be inconsistent with the BIM model data, and the corresponding monitoring priority will be higher.

[0048] Furthermore, according to the monitoring needs of the construction site, the inspection capabilities of the robot and the resource allocation, a preset number of monitoring points (such as 50, 100, etc.) are set, and a preset number of units are selected from the sorted model units as key monitoring objects to form a monitoring point sequence.

[0049] At the same time, the unit anomaly rates corresponding to these monitoring points are arranged in the same order to form an anomaly rate sequence that corresponds one-to-one to the monitoring point sequence.

[0050] For example, if the calculated unit abnormality rate of the model units of a project is between 5% and 30%, after sorting from large to small, the abnormality rates of the first 30 model units are 30%, 28%, 27%...18% respectively. If the preset number is set to 30, then these 30 model units constitute a monitoring point sequence, and the corresponding abnormality rate sequence is [30%, 28%, 27%,..., 18%].

[0051] Through this step, the scattered model units are converted into an ordered sequence of monitoring points and an adapted sequence of abnormality rates according to the abnormality risk level, providing clear path guidance for subsequent robots to conduct patrol monitoring according to priority, ensuring that high-risk units can be monitored first, thereby improving the accuracy and efficiency of construction site monitoring.

[0052] S120: Control the robot to perform monitoring according to the monitoring point sequence. When monitoring data of any monitoring point is obtained, extract the BIM data of the monitoring point and perform consistency verification with the inspection data to obtain an initial consistency rate. In an embodiment of the present application, in the scenario of real-time monitoring of construction site robots based on BIM, in order to achieve accurate comparison of the actual construction status of each monitoring point with the BIM model, it is necessary to use the robot to collect data in sequence and perform consistency analysis in combination with verification tools to obtain an initial matching evaluation result.

[0053] Specifically, first, according to the determined monitoring point sequence, motion control instructions are sent to the robot through the control platform to guide the robot to move to each monitoring point in sequence according to priority.

[0054] After arriving at the target monitoring point, the robot activates the onboard sensors to collect on-site data, such as component images, dimensional parameters, position information and other monitoring data, to ensure that the data can truly reflect the actual construction conditions of the monitoring point.

[0055] At the same time, while the robot is collecting monitoring data, the BIM data corresponding to the monitoring point is synchronously extracted from the real-time BIM model, including preset data such as component geometric parameters, material information, and position coordinates recorded in the model, to ensure that the BIM data used for comparison accurately corresponds to the current monitoring point.

[0056] Furthermore, a pre-trained BIM validator is obtained. This BIM validator is an intelligent analysis tool trained based on historical construction monitoring data and can quantitatively evaluate the consistency between monitoring data and BIM data.

[0057] On this basis, the monitoring data of the current monitoring point and the extracted BIM data are simultaneously input into the BIM verifier. The verifier analyzes the differences in key features between the two and outputs a quantitative consistency indicator, namely the initial consistency rate.

[0058] Among them, the closer the value of the initial consistency rate is to 100%, the higher the matching degree between the actual construction situation of the monitoring point and the BIM model; otherwise, it indicates that there is a certain deviation.

[0059] This step achieves a preliminary quantitative assessment of the construction status and model consistency of each monitoring point through the coordinated cooperation of robot sequential monitoring, precise BIM data extraction and BIM verifier analysis, providing basic data support for subsequent abnormal judgment and review and verification, and ensuring the objectivity and accuracy of the monitoring process.

[0060] Step S120 in the method provided in the embodiment of the present application includes: According to the monitoring point sequence, the robot is controlled to perform monitoring, and when monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted; Get BIM validator; The monitoring data and BIM data of the monitoring points are input into the BIM verifier, and the initial consistency rate is obtained as output.

[0061] In the embodiment of the present application, in order to achieve accurate comparison between the actual construction status of the monitoring point and the BIM model, it is necessary to use a robot to collect on-site data in sequence, match the corresponding model data, and use verification tools to perform consistency analysis to quantitatively evaluate the degree of matching between the two, providing an initial basis for subsequent abnormality judgment.

[0062] Specifically, first, based on the determined monitoring point sequence, the robot control platform sends path planning instructions to the inspection robot, guiding the robot to move to each monitoring point in order of priority from high to low abnormality rate.

[0063] For example, for the frame column monitoring point that ranks first in abnormality rate, the platform will generate control instructions including coordinate positioning, moving path, and obstacle avoidance parameters to ensure that the robot reaches the monitoring position accurately.

[0064] Furthermore, after the robot reaches the target monitoring point, it activates the onboard sensing equipment to collect on-site data.

[0065] Specifically, if the sensing device is a drone, a high-definition industrial camera can be used to capture multi-angle images of the component and a lidar can be used to scan the three-dimensional contour; if the sensing device is a ground robot, infrared sensors can be used to detect material properties and millimeter-wave radar can be used to measure installation position deviations.

[0066] In addition, the collected monitoring data includes but is not limited to the actual size, appearance, location coordinates, material characteristics, construction progress status, etc. of the components, and the data must be accompanied by a timestamp and monitoring point identification to ensure the traceability of the data.

[0067] Simultaneously with monitoring data collection, the BIM data corresponding to the monitoring point is extracted from the real-time BIM model database. The extraction process accurately associates the model unit with the unique code of the monitoring point. The acquired data includes the component geometry parameters, material information, location information, and process standards preset in the BIM model, ensuring that the model data used for comparison fully matches the current monitoring point.

[0068] Furthermore, a pre-trained BIM verifier is obtained, and the feature associations between monitoring data and BIM data are mined through the verifier's deep internal learning algorithm, and the degree of consistency between the two is quantified and output.

[0069] In the method provided in the embodiment of the present application, the step of “obtaining a BIM validator” includes: Based on historical construction monitoring data, a sample monitoring data set of a sample monitoring point set is collected, and a sample BIM data set is collected. Each sample monitoring data and sample BIM data is marked for consistency to obtain a sample consistency rate set; Building a BIM validator based on twin neural networks; The sample monitoring data set, the sample BIM data set and the sample consistency rate set are used to train and optimize the BIM verifier in a supervised manner until convergence, thereby obtaining a BIM verifier.

[0070] In the embodiment of the present application, in order to build an intelligent tool that can accurately evaluate the consistency between construction site monitoring data and BIM model data, a systematic sample collection, labeling and model training process is required to enable the BIM verifier to have the ability to deeply mine the correlation between the two types of data features and quantify the output consistency rate, providing a reliable basis for the comparison of construction status and model.

[0071] Specifically, firstly, a sample monitoring data set and a sample BIM data set of a sample monitoring point set are collected based on historical construction monitoring data.

[0072] Among them, the set of sample monitoring points is selected from representative construction units in previous similar engineering projects, covering monitoring points of different structural types, different construction stages and different abnormal types to ensure the comprehensiveness of the sample.

[0073] In addition, the sample monitoring data set is actual construction data collected by the same type of robotic inspection equipment as on-site, including high-definition images, measured parameters, etc., such as the on-site image of a frame column and the measured cross-sectional size of 595mm×598mm.

[0074] In addition, the sample BIM data set is the preset data of the sample monitoring points in the historical BIM model, including the geometric parameters, material properties, location information, process standards, etc. in the model, to ensure that the two types of sample data correspond one-to-one to the monitored objects.

[0075] For example, for 1,000 sample monitoring points in a historical project, on-site inspection images (resolution 1920×1080), laser scanning dimensional data (accuracy ±1mm) and component parameters in the corresponding BIM model were collected at each point to form a sample monitoring dataset and a sample BIM dataset containing 20,000 sets of data.

[0076] Furthermore, each sample monitoring data and sample BIM data is labeled for consistency to obtain a sample consistency rate set.

[0077] The marking process is completed jointly by experienced construction management personnel and BIM engineers, and a 0-100% consistency rate is assigned based on the degree of deviation between the two in key features.

[0078] Specifically, when the actual monitoring data completely matches the BIM data, it is marked as 100%; when there is a slight deviation but within the allowable range, it is marked as 80-95%; when the deviation exceeds the allowable range, it is marked as 50-70%; when there is a serious discrepancy, it is marked as 0-40%.

[0079] At the same time, the marking results must be accompanied by detailed descriptions of the deviations, such as "size deviation 4mm, consistency rate 92%" and "material strength grade mismatch, consistency rate 60%", to form a traceable sample consistency rate set.

[0080] For example, a sample monitoring data is "actual thickness of floor slab 118mm", and the corresponding sample BIM data is "floor slab model thickness 120mm", the deviation is 2mm and is within the allowable range (±5mm), and the consistency rate of the marked samples is 96%; another sample monitoring data is "wall verticality deviation 8mm", and the sample BIM data requires "verticality allowable deviation ±5mm", the deviation exceeds the range, and the consistency rate of the marked samples is 65%.

[0081] On this basis, a BIM verifier was constructed based on a twin neural network. The twin neural network is a special deep learning model that contains two feature extraction sub-networks with identical structures and shared weights, one for processing sample monitoring data and the other for processing sample BIM data.

[0082] Specifically, one sub-network is responsible for extracting edge features, size features, material features, etc. from the monitoring data; the other sub-network is responsible for extracting corresponding features from the BIM data; finally, the similarity calculation layer outputs the degree of matching between the two, which is used as the sample consistency rate set.

[0083] Furthermore, the sample monitoring dataset, sample BIM dataset and sample consistency rate set are used to optimize the BIM verifier through supervised training until convergence.

[0084] Specifically, the sample data is first preprocessed to convert the image data in the sample monitoring data into a tensor format suitable for neural network input, for example, adjusting the resolution to 224×224 pixels and normalizing it to the range of [0, 1].

[0085] Secondly, the measured parameters are standardized using the Z-score standardization method. That is, for each measured parameter, its mean and standard deviation in the sample set are calculated, and then the parameter is converted into a distribution with a mean of 0 and a standard deviation of 1 using the formula "standardized value = (original value - mean) / standard deviation" to eliminate the impact of differences in the magnitude of different parameters on model training.

[0086] At the same time, structured data such as geometric parameters and material properties in the sample BIM data are converted into numerical vectors, and unstructured information such as process standard descriptions are encoded to ensure that the two types of data formats are unified and meet the input requirements of the twin neural network.

[0087] Furthermore, the preprocessed sample monitoring dataset and sample BIM dataset were fed into the two feature extraction sub-networks of the verifier. Through convolutional layers, pooling layers, and other network structures, the deep features of the data were extracted and corresponding feature vectors were generated. The initial prediction consistency rate was obtained by calculating the cosine similarity of the two feature vectors.

[0088] Furthermore, the labeled values ​​in the sample consistency rate set are used as supervisory signals, and the mean square error loss function is used to calculate the deviation between the predicted consistency rate and the labeled value. The network weights and bias parameters are adjusted layer by layer through the back propagation algorithm to continuously reduce the loss value.

[0089] At the same time, during the training process, the training set and validation set are divided into a preset ratio of 8:2. The validation set is used to evaluate the model performance after each round of iteration. If the validation set loss value does not decrease and tends to be stable for 10 consecutive rounds, the model is judged to have converged, and the training is stopped. Finally, the optimized BIM verifier is obtained.

[0090] On this basis, the monitoring data and BIM data of the monitoring points are input into the trained BIM verifier, and the initial consistency rate is obtained through the feature extraction and similarity calculation mechanism of the verifier.

[0091] Specifically, in an actual monitoring scenario, when the robot reaches the target monitoring point and collects monitoring data, it simultaneously extracts the BIM data corresponding to the monitoring point from the real-time BIM model. The same operations as sample preprocessing are performed on these real-time data.

[0092] Furthermore, the preprocessed monitoring data and BIM data are fed into two feature extraction subnetworks of the BIM verifier. These subnetworks, using trained and optimized convolutional and pooling layers, rapidly extract deep features from the real-time data and generate corresponding feature vectors. By calculating the cosine similarity between the two feature vectors, the matching degree between the real-time monitoring data and the BIM data is quantified, and the initial consistency rate is output.

[0093] For example, the actual cross-sectional dimensions collected by the robot at a certain frame beam monitoring point are 398mm×596mm, and the corresponding cross-sectional dimensions in the BIM data are 400mm×600mm. After preprocessing, they are input into the BIM verifier. The sub-network extracts the size features, shape features, etc. of the two and calculates the similarity. The final output initial consistency rate is 97%, indicating that the actual construction status of the monitoring point is highly consistent with the BIM model.

[0094] In addition, if the actual verticality deviation of another wall monitoring point is 7 mm, and the BIM data requires a verticality deviation of ±5 mm, the initial consistency rate output after input into the verifier is 62%, indicating that there is a certain deviation between the actual construction status of the monitoring point and the BIM model.

[0095] This step uses the BIM verifier to compare and analyze real-time monitoring data with BIM data, achieving a quantitative assessment of the consistency between the actual construction status and model data. It provides an objective initial judgment basis for subsequent review and verification based on the abnormality rate and movement abnormality rate, effectively improving the accuracy and efficiency of construction site monitoring.

[0096] S130: Predicting the movement anomaly rate based on the BIM data of the monitoring point to obtain the movement anomaly rate, configuring review parameters based on the initial consistency rate and the anomaly rate of the monitoring point, and performing review consistency verification on the monitoring point to obtain the review consistency rate; In an embodiment of the present application, in the scenario of real-time monitoring of construction site robots based on BIM, in order to avoid the impact of robot movement deviation on monitoring results and improve the reliability of consistency verification, it is necessary to predict the risk of abnormal movement, dynamically configure review parameters, and perform a secondary verification of the initial monitoring results to obtain a more accurate consistency rate assessment.

[0097] First, we predict the movement anomaly rate based on the BIM data of the monitoring points. This rate is the probability that the robot's actual monitoring location will not match the target monitoring point due to factors such as positioning error and path deviation. To achieve this prediction, we need a pre-trained movement anomaly analyzer.

[0098] Specifically, the mobility anomaly analyzer is an intelligent model trained based on a sample BIM data set and a sample mobility anomaly rate set. The sample BIM data set contains the BIM data of each monitoring point in historical projects, and the sample mobility anomaly rate set records the probability of the robot's actual mobility deviation in the corresponding scenario.

[0099] Furthermore, after the BIM data of the current monitoring point is input into the movement anomaly analyzer, the analyzer outputs the movement anomaly rate of the monitoring point by mining the correlation rules between the environmental characteristics implicit in the BIM data and the movement deviation.

[0100] Furthermore, the initial anomaly rate is calculated in combination with the initial consistency rate. Specifically, the initial anomaly rate is the difference between "1 and the initial consistency rate" and is used to quantify the degree of deviation between the actual construction and the BIM model during the initial monitoring.

[0101] Furthermore, the fusion anomaly rate is calculated based on the anomaly rate of the monitoring points, the initial anomaly rate and the moving anomaly rate.

[0102] Among them, the fusion anomaly rate comprehensively evaluates the overall deviation probability of the monitoring point by integrating three types of abnormal risks, that is, the weighted summation calculation method, which aggregates multi-dimensional abnormal risk values ​​to form a comprehensive quantitative representation of the overall deviation possibility of the monitoring point.

[0103] On this basis, the review parameters are configured based on the fusion anomaly rate. The review parameters primarily include the number of reviews, and the fusion anomaly rate is positively correlated with the number of reviews. Specifically, this can be calculated by multiplying the fusion anomaly rate by the preset review base number and rounding it up to ensure that high-risk points receive more thorough reviews.

[0104] Furthermore, according to the determined verification parameters, the robot is controlled to conduct multiple verifications and consistency checks on the monitoring point. Specifically, the robot moves to the monitoring point according to the planned path, collects new monitoring data, and simultaneously extracts the BIM data and inputs it into the BIM verifier to obtain the consistency rate for each verification. The average of these multiple verification consistency rates is then calculated as the verification consistency rate for that monitoring point.

[0105] This step effectively reduces the impact of single monitoring errors by introducing the mobile anomaly rate to assess the robot positioning risk and combines it with a multi-dimensional anomaly rate dynamic configuration review strategy, making the consistency rate assessment more in line with the actual construction status, providing more reliable data support for subsequent fusion anomaly judgment, and further improving the accuracy and reliability of construction site monitoring.

[0106] Step S130 in the method provided in the embodiment of the present application includes: Obtaining a movement anomaly analyzer, wherein the movement anomaly analyzer is trained using a sample BIM data set and a sample movement anomaly rate set, wherein the sample movement anomaly rate includes a probability of robot movement deviation; The BIM data is input into the movement anomaly analyzer, and the movement anomaly rate is obtained as an output.

[0107] Calculate and obtain an initial abnormality rate based on the initial consistency rate; The fusion anomaly rate is calculated based on the anomaly rate, initial anomaly rate and movement anomaly rate of the monitoring points; Calculate and obtain the review parameters based on the fusion abnormality rate and the preset review parameters, wherein the review parameters include the number of reviews; According to the review parameters, the monitoring points are reviewed and verified for consistency, and the mean of the consistency rates of the review verifications is calculated to obtain the review consistency rate.

[0108] In the embodiment of the present application, in order to eliminate the interference of robot movement deviation on the monitoring results and improve the reliability of consistency verification, it is necessary to dynamically configure the review parameters by predicting the risk of movement anomaly and integrating multi-dimensional abnormal indicators, and perform a second verification on the initial monitoring results to obtain a more accurate consistency rate assessment.

[0109] Specifically, a motion anomaly analyzer is first obtained to accurately predict the probability of robot motion deviation. The motion anomaly analyzer is an intelligent prediction model trained based on a sample BIM data set and a sample motion anomaly rate set.

[0110] Among them, the sample BIM data set contains model features such as the three-dimensional coordinates of each monitoring point in the historical project, the distribution of surrounding components, and the path complexity. The sample movement anomaly rate set records the probability of positioning deviation caused by the offset probability caused by occlusion of the construction environment and uneven ground when the robot actually moves in the corresponding scenario.

[0111] Specifically, the BIM data of the current monitoring point is input into the mobile anomaly analyzer, and the built-in gradient boosting decision tree (GBDT) algorithm of the analyzer is used to mine the association rules between features and anomaly rates to output the mobile anomaly rate of the monitoring point.

[0112] Firstly, the sample BIM data set is standardized, that is, the three-dimensional coordinates are converted into relative position vectors, the distribution of surrounding components is quantified into a density index, and the path complexity is constructed into a comprehensive index through parameters such as path curvature and the number of obstacles, so that all model features are converted into quantifiable numerical vectors.

[0113] Among them, the transformation of three-dimensional coordinates adopts the establishment of a local coordinate system with the construction body reference point as the origin, and the absolute coordinates of each monitoring point are subtracted from the reference point coordinates to obtain the relative position vector; the quantification of the distribution of surrounding components is achieved by counting the number of components within a preset radius around the monitoring point and dividing it by the spatial volume within this range to obtain the density index.

[0114] In addition, the path curvature is obtained by calculating the curvature value of each curve on the robot's moving path and taking the average value. The larger the curvature value, the higher the degree of path curvature. The number of obstacles is obtained by identifying and counting the obstructions on the robot's preset path.

[0115] At the same time, the probability values ​​in the sample movement anomaly rate set are used as labels and correspond one-to-one with the feature vectors to form a training data set.

[0116] Furthermore, a mobile anomaly analyzer is constructed based on the gradient boosting decision tree (GBDT) framework.

[0117] The framework is composed of an ensemble of multiple decision trees, each trained based on the residuals of the preceding tree. Specifically, the first tree initially fits the distribution of the movement anomaly rate by splitting based on sample characteristics (for example, prioritizing splitting when the path curvature is greater than 0.8). Subsequent trees gradually correct the prediction error until the model's prediction deviation for the training set falls below a preset threshold (e.g., mean squared error ≤ 0.001).

[0118] At the same time, during the training process, 5-fold cross-validation was used to optimize hyperparameters such as the number and depth of trees to avoid overfitting.

[0119] Ultimately, the mobile anomaly analyzer can accurately capture the correlation between BIM features and robot movement anomaly rates, ensuring high accuracy when predicting BIM data of new monitoring points.

[0120] On this basis, the standardized BIM data of the current monitoring point is input into the trained motion anomaly analyzer. The analyzer uses a gradient boosting decision tree algorithm to split and combine the input features layer by layer, calling on the learned feature association rules to output the motion anomaly rate of the monitoring point.

[0121] For example, the relative position vector of a monitoring point shows that it is close to the construction elevator passage, the density index is 12 (dense components), and the path curvature is 0.9 (there are many turns). The movement anomaly analyzer combines these features and outputs a movement anomaly rate of 14%.

[0122] Furthermore, the initial anomaly rate is calculated based on the initial consistency rate. Specifically, the calculation formula for the initial anomaly rate can be expressed as "initial anomaly rate = 1 - initial consistency rate". This formula converts the initial consistency rate, an indicator reflecting the degree of consistency, into a quantitative value that directly represents the risk of deviation.

[0123] For example, if the initial consistency rate of a monitoring point is 92%, then the initial anomaly rate = 1-92% = 8%. This value intuitively reflects the possibility of deviation between the actual construction status and the BIM model in the initial monitoring.

[0124] Furthermore, the fusion anomaly rate is calculated based on the anomaly rate of the monitoring points, the initial anomaly rate and the moving anomaly rate.

[0125] Specifically, the fusion anomaly rate comprehensively evaluates the overall deviation probability of the monitoring point by integrating the three types of anomaly risks. It needs to be calculated using a weighted summation method. The specific calculation formula can be expressed as "fusion anomaly rate = anomaly rate of monitoring point × anomaly weight of monitoring point + initial anomaly rate × initial anomaly weight + moving anomaly rate × moving anomaly weight".

[0126] Among them, the allocation of each weight needs to take into account the inherent abnormal risks of the monitoring point itself, as well as the deviations generated during the initial monitoring process and the errors that may be caused by the movement of the robot, so as to make the evaluation results more comprehensive.

[0127] Therefore, the weight of the abnormality rate of the monitoring point is set to the highest (0.4), because it reflects the inherent risk of abnormalities at the monitoring point during the construction process and is the core basis of the assessment; the initial abnormality rate and the movement abnormality rate supplement the assessment deviation risk from the two aspects of the monitoring process and robot movement, respectively, and are therefore each assigned a weight of 0.3 to scientifically aggregate multi-dimensional risks and form a comprehensive quantitative representation of the overall abnormal situation of the monitoring point.

[0128] For example, if the abnormality rate of a monitoring point is 18%, the initial abnormality rate is 8%, and the moving abnormality rate is 14%, then the fusion abnormality rate = 18% × 0.4 + 8% × 0.3 + 14% × 0.3 = 7.2% + 2.4% + 4.2% = 13.8%.

[0129] Furthermore, a verification parameter is calculated based on the obtained fusion abnormality rate and a preset verification parameter. The verification parameter includes the number of verifications. The preset verification parameter includes a basic number of verifications and an amplification factor.

[0130] Specifically, the basic number of reviews is the minimum number of reviews to ensure basic monitoring accuracy, usually 1; the amplification factor is the proportional coefficient used to convert the fusion abnormality rate into additional review times, usually 10; the specific calculation formula for the number of reviews is "review times = basic review times + round (fusion abnormality rate × amplification factor)", where "round" is a rounding function to achieve quantitative determination of the number of reviews.

[0131] Among them, by multiplying the fusion abnormality rate by the amplification coefficient, rounding it up, and superimposing the calculation method of the basic re-verification number, it can be ensured that the monitoring point with a higher fusion abnormality rate will obtain more re-verification times, thereby reducing the error impact of high-risk points by increasing the number of re-verifications.

[0132] For example, when the fusion abnormality rate is 13.8%, the number of rechecks = 1 + round (13.8% × 10) = 1 + 1 = 2 times; if the fusion abnormality rate is 25%, the number of rechecks = 1 + round (25% × 10) = 1 + 3 = 4 times.

[0133] On this basis, according to the obtained review parameters, the monitoring points are reviewed and verified for consistency, and the mean of the review and verification consistency rate is calculated to obtain the review consistency rate, so as to reduce the possible errors in a single monitoring and improve the reliability of the monitoring results.

[0134] Specifically, the robot re-plans the optimal path to move to the monitoring point multiple times based on the determined number of reviews. Each time it arrives at the monitoring point, it restarts the onboard sensors to collect on-site monitoring data, including the actual size, position coordinates, appearance integrity and other information of the components, to ensure that the data collected each time can truly reflect the real-time construction status of the monitoring point.

[0135] At the same time, during each review and collection of monitoring data, the latest BIM data corresponding to the monitoring point is accurately extracted from the real-time BIM model database, including preset data such as updated component geometric parameters, material properties, and process standards in the model, to ensure that the BIM data used for comparison is fully matched with the current monitoring point.

[0136] Furthermore, the monitoring data collected during each review, along with the extracted BIM data, is fed into a pre-trained BIM verifier. The verifier analyzes the degree of match between the two on key features and outputs the consistency rate for each review. After all reviews are completed, the arithmetic mean of these consistency rates is calculated to represent the consistency rate for that monitoring point.

[0137] For example, a monitoring point is reviewed twice. During the first review, the monitoring data collected by the robot and the BIM data are analyzed by the verifier, and the output consistency rate is 89%; during the second review, the output consistency rate is 91%. Then the review consistency rate of this monitoring point = (89% + 91%) / 2 = 90%.

[0138] Ultimately, by rechecking multiple times and taking the average, we effectively reduced possible accidental errors in a single monitoring, such as the robot's instantaneous positioning deviation and temporary sensor interference. This allows the final recheck consistency rate to more objectively reflect the degree of consistency between the actual construction status of the monitoring point and the BIM model, providing a reliable basis for subsequent fusion anomaly assessment.

[0139] S140: Calculate the fusion anomaly rate based on the anomaly rate, movement anomaly rate, and verification consistency rate of the monitoring points, mark the monitoring points, and obtain monitoring results.

[0140] In the embodiment of the present application, in the scenario of real-time monitoring of construction site robots based on BIM, in order to integrate multi-dimensional monitoring data to obtain the final abnormality assessment result, it is necessary to accurately mark the abnormal status of the monitoring points by integrating various abnormal indicators in the early stage to form a comprehensive and reliable monitoring result.

[0141] Specifically, the review exception rate is first calculated based on the review consistency rate. The formula for calculating the review exception rate can be expressed as "review exception rate = 1 - review consistency rate". This formula converts the review consistency rate, which reflects the degree of review consistency, into a quantitative value that represents the risk of deviation during the review stage.

[0142] Furthermore, the fusion anomaly rate is calculated by weighted summation based on the anomaly rate, movement anomaly rate and review anomaly rate of the monitoring point to comprehensively evaluate the overall anomaly probability of the monitoring point.

[0143] On this basis, the monitoring points are marked according to the fusion anomaly rate, and the monitoring results are obtained by analyzing the numerical range of the fusion anomaly rate.

[0144] Among them, labeling rules can be divided into different levels according to preset thresholds, and each level comes with specific processing suggestions.

[0145] This step forms a comprehensive judgment on the abnormal status of the monitoring point by integrating multi-dimensional abnormal indicators, and presents the monitoring results in a clear and labeled form, providing a direct and effective decision-making basis for quality management and risk control at the construction site, and further improving the integrity and practicality of the monitoring process.

[0146] Step S140 in the method provided in the embodiment of the present application includes: According to the review consistency rate, the review abnormality rate is calculated; According to the abnormality rate, movement abnormality rate and review abnormality rate of the monitoring points, the fusion abnormality rate is calculated, the monitoring points are marked and the monitoring results are obtained.

[0147] In the embodiment of the present application, in order to integrate multi-dimensional monitoring data to obtain the final abnormality assessment result, it is necessary to accurately mark the abnormal status of the monitoring point by integrating the deviation risk in the review stage and various abnormality rate indicators in the early stage to form a comprehensive and reliable monitoring result.

[0148] First, the review exception rate is calculated based on the review consistency rate. Specifically, the specific calculation formula for the review exception rate can be expressed as "review exception rate = 1 - review consistency rate". Through this transformation, the indicator reflecting the degree of consistency during the review stage is quantified into a numerical value representing the risk of deviation during this stage.

[0149] For example, if the review consistency rate of a monitoring point is 90%, then the review abnormality rate = 1-90% = 10%. This value intuitively reflects the possibility that after multiple reviews, there is still a possibility that the actual construction and the BIM model may deviate.

[0150] Furthermore, the fusion abnormality rate is calculated based on the abnormality rate of the monitoring points, the movement abnormality rate and the review abnormality rate.

[0151] Specifically, the fusion anomaly rate is obtained by weighted summation, and the specific calculation formula can be expressed as "fusion anomaly rate = monitoring point anomaly rate × monitoring point anomaly weight + movement anomaly rate × movement anomaly weight + review anomaly rate × review anomaly weight".

[0152] Among them, the configuration of weights is determined by comprehensively considering the actual impact of each abnormality rate. For example, the abnormality weight of the monitoring point is set to 0.3, the mobile abnormality weight is set to 0.2, and the review abnormality weight is set to 0.5, in order to highlight the importance of the review abnormality rate in the overall assessment, because it is a more reliable deviation risk quantification value obtained after multiple reviews.

[0153] For example, if the monitoring point abnormality rate of a monitoring point is 20%, the movement abnormality rate is 15%, and the review abnormality rate is 10%, then the fusion abnormality rate = 20% × 0.3 + 15% × 0.2 + 10% × 0.5 = 6% + 3% + 5% = 14%.

[0154] On this basis, the monitoring points are marked according to the obtained fusion anomaly rate, and the monitoring points are divided into different anomaly levels by defining different anomaly threshold ranges to obtain the monitoring results.

[0155] Specifically, the labeling rules are set by a preset fusion abnormality rate threshold. For example, when the fusion abnormality rate is ≤5%, it is labeled as "normal", when it is 5%<fusion abnormality rate≤15%, it is labeled as "mild abnormality", and when it is >15%, it is labeled as "severe abnormality". Each level corresponds to different processing suggestions.

[0156] For example, when the fusion abnormality rate of a monitoring point is 14%, because 5%<14%<15%, it is marked as "minor abnormality" according to the labeling rules, and an accompanying processing suggestion "It is recommended to strengthen daily inspections of this monitoring point and pay close attention to its subsequent changes" is provided to achieve an accurate description of the abnormal status of the monitoring point and a targeted response.

[0157] This step compares the fusion anomaly rate with the preset threshold and labels it in a graded manner, combining it with specific processing suggestions to make the monitoring results more practical and instructive, providing a clear basis for action for risk management and quality assurance at the construction site.

[0158] The embodiments of the present application achieve the following technical effects through the above specific implementation methods: This application proposes a real-time monitoring method for construction site robots based on BIM. First, a real-time BIM model of the target construction body is obtained, the model units are divided and the model features are extracted, the unit anomaly rate is determined in combination with historical construction data, and high-risk units are screened to form a monitoring point sequence and anomaly rate sequence; then the robot is controlled to perform monitoring according to the monitoring point sequence, the monitoring point BIM data and inspection data are extracted, and consistency verification is performed through a BIM verifier to obtain an initial consistency rate; then the movement anomaly rate is predicted based on the monitoring point BIM data, and the fusion anomaly rate is calculated by combining the initial consistency rate and the monitoring point anomaly rate, and the review parameters are configured accordingly and review verification is performed to obtain the review consistency rate; finally, the review anomaly rate is calculated based on the review consistency rate, and the monitoring point anomaly rate, the movement anomaly rate and the review anomaly rate are integrated to obtain the final fusion anomaly rate, and the monitoring points are marked to obtain the monitoring results.

[0159] The method provided in the embodiment of the present application adopts the technical solution of "real-time BIM model monitoring point planning - robot initial monitoring and consistency verification - movement anomaly rate prediction and review parameter configuration - multi-dimensional anomaly rate fusion and result annotation", which integrates anomaly rate analysis based on historical construction data, BIM verifier constructed by twin neural networks, movement anomaly prediction of gradient boosting decision tree, multi-dimensional anomaly rate weighted fusion and other technical methods, and solves the problem of insufficient monitoring accuracy caused by BIM errors, robot recognition errors, movement error probabilities, etc. in traditional construction site monitoring, and realizes high-precision real-time monitoring of construction sites.

[0160] Example 2, as shown in the attached Figure 2 As shown, based on the inventive concept of the BIM-based construction site robot real-time monitoring method provided in Example 1, this application also provides a BIM-based construction site robot real-time monitoring system, specifically including: BIM monitoring point planning module 01 is used to obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a monitoring point sequence and an abnormality rate sequence; The robot initial verification module 02 is used to control the robot to monitor according to the monitoring point sequence. When monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted and consistency is verified with the inspection data to obtain an initial consistency rate. The movement anomaly review module 03 is used to predict the movement anomaly rate based on the BIM data of the monitoring point to obtain the movement anomaly rate, configure the review parameters based on the initial consistency rate and the anomaly rate of the monitoring point, and perform review consistency verification on the monitoring point to obtain the review consistency rate; The fusion anomaly marking module 04 is used to calculate the fusion anomaly rate based on the anomaly rate, movement anomaly rate and verification consistency rate of the monitoring point, mark the monitoring point, and obtain the monitoring result.

[0161] In one embodiment, the BIM monitoring point planning module 01 is further configured to: Obtain a real-time BIM model of a target construction body; divide the real-time BIM model to obtain multiple model units, and obtain model features of the multiple model units; analyze the unit anomaly rates of the multiple model units based on the multiple model features; arrange the multiple unit anomaly rates in descending order, and select a preset number of monitoring points to obtain a monitoring point sequence and an anomaly rate sequence.

[0162] In one embodiment, the robot initial verification module 02 is further configured to: According to the monitoring point sequence, the robot is controlled to perform monitoring, and when monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted; and a BIM verifier is obtained; The monitoring data and BIM data of the monitoring points are input into the BIM verifier, and the initial consistency rate is obtained as output.

[0163] In one embodiment, the movement anomaly review module 03 is further configured to: A movement anomaly analyzer is obtained, wherein the movement anomaly analyzer is trained using a sample BIM data set and a sample movement anomaly rate set, and the sample movement anomaly rate includes the probability of robot movement deviation; the BIM data is input into the movement anomaly analyzer, and the movement anomaly rate is obtained as an output.

[0164] Based on the initial consistency rate, the initial anomaly rate is calculated; based on the anomaly rate, initial anomaly rate and movement anomaly rate of the monitoring point, the fusion anomaly rate is calculated; based on the fusion anomaly rate and the preset review parameters, the review parameters are calculated, wherein the review parameters include the number of reviews; according to the review parameters, the monitoring points are reviewed and verified for consistency and the average of the consistency rates of the review verifications is calculated to obtain the review consistency rate.

[0165] In one embodiment, the fusion anomaly annotation module 04 is further configured to: Based on the review consistency rate, the review anomaly rate is calculated; based on the anomaly rate, movement anomaly rate and review anomaly rate of the monitoring points, the fusion anomaly rate is calculated, the monitoring points are marked, and the monitoring results are obtained.

[0166] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0167] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0168] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A real-time monitoring method for construction site robots based on BIM, characterized in that: The method comprises: Obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a monitoring point sequence and an abnormality rate sequence; According to the monitoring point sequence, the robot is controlled to perform monitoring. When monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted and consistency is verified with the inspection data to obtain an initial consistency rate. Based on the BIM data of the monitoring points, the movement anomaly rate is predicted to obtain the movement anomaly rate. The initial consistency rate and the anomaly rate of the monitoring points are combined to configure the review parameters, and the monitoring points are reviewed and verified for consistency to obtain the review consistency rate. According to the abnormality rate, movement abnormality rate and verification consistency rate of the monitoring points, the fusion abnormality rate is calculated, the monitoring points are marked and the monitoring results are obtained.

2. The method for real-time monitoring of a construction site robot based on BIM according to claim 1, characterized in that: Obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a sequence of inspection points and anomaly rate sequence, including: Obtain the real-time BIM model of the target construction body; Dividing the real-time BIM model to obtain a plurality of model units, and acquiring model features of the plurality of model units; Analyze the unit abnormality rates of multiple model units based on multiple model characteristics; Arrange the abnormality rates of multiple units in descending order, and select a preset number of monitoring points to obtain a monitoring point sequence and an abnormality rate sequence.

3. The method for real-time monitoring of a construction site robot based on BIM according to claim 1, characterized in that: Analyze the unit anomaly rates of multiple model units based on multiple model features, including: Extracting multiple sets of model units of the same family from the historical construction data based on multiple model features, wherein each set of model units of the same family includes multiple model units of the same family that have the greatest similarity to the corresponding model features; The abnormality ratios of multiple model units of the same family are obtained by processing and obtaining the unit abnormality rates of multiple model units.

4. The method for real-time monitoring of a construction site robot based on BIM according to claim 1, characterized in that: According to the monitoring point sequence, the robot is controlled to perform monitoring. When monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted and consistency is verified with the inspection data to obtain an initial consistency rate, including: According to the monitoring point sequence, the robot is controlled to perform monitoring, and when monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted; Get BIM validator; The monitoring data and BIM data of the monitoring points are input into the BIM verifier, and the initial consistency rate is obtained as output.

5. The method for real-time monitoring of a construction site robot based on BIM according to claim 4, characterized in that: Get BIM Validator, including: Based on historical construction monitoring data, a sample monitoring data set of a sample monitoring point set is collected, and a sample BIM data set is collected. Each sample monitoring data and sample BIM data is marked for consistency to obtain a sample consistency rate set; Building a BIM validator based on twin neural networks; The sample monitoring data set, the sample BIM data set and the sample consistency rate set are used to train and optimize the BIM verifier in a supervised manner until convergence, thereby obtaining a BIM verifier.

6. The method for real-time monitoring of a construction site robot based on BIM according to claim 1, characterized in that: Based on the BIM data of the monitoring points, the movement abnormality rate is predicted to obtain the movement abnormality rate, including: Obtaining a movement anomaly analyzer, wherein the movement anomaly analyzer is trained using a sample BIM data set and a sample movement anomaly rate set, wherein the sample movement anomaly rate includes a probability of robot movement deviation; The BIM data is input into the movement anomaly analyzer, and the movement anomaly rate is obtained as an output.

7. The method for real-time monitoring of a construction site robot based on BIM according to claim 1, characterized in that: Based on the initial consistency rate and the abnormality rate of the monitoring point, the review parameters are configured, and the monitoring point is reviewed and verified for consistency to obtain the review consistency rate, including: Calculate and obtain an initial abnormality rate based on the initial consistency rate; The fusion anomaly rate is calculated based on the anomaly rate, initial anomaly rate and movement anomaly rate of the monitoring points; Calculate and obtain the review parameters based on the fusion abnormality rate and the preset review parameters, wherein the review parameters include the number of reviews; According to the review parameters, the monitoring points are reviewed and verified for consistency, and the mean of the consistency rates of the review verifications is calculated to obtain the review consistency rate.

8. The method for real-time monitoring of a construction site robot based on BIM according to claim 1, characterized in that: Based on the abnormality rate, movement abnormality rate and verification consistency rate of the monitoring points, the fusion abnormality rate is calculated, the monitoring points are marked, and the monitoring results are obtained, including: According to the review consistency rate, the review abnormality rate is calculated; According to the abnormality rate, movement abnormality rate and review abnormality rate of the monitoring points, the fusion abnormality rate is calculated, the monitoring points are marked and the monitoring results are obtained.

9. The BIM-based construction site robot real-time monitoring system is characterized by: The system is used to execute the BIM-based construction site robot real-time monitoring method according to any one of claims 1 to 8, and the system includes: A BIM monitoring point planning module is used to obtain a real-time BIM model of the target construction body, plan monitoring points based on the real-time BIM model, and obtain a monitoring point sequence and an abnormality rate sequence; The robot initial verification module is used to control the robot to monitor according to the monitoring point sequence. When monitoring data of any monitoring point is obtained, the BIM data of the monitoring point is extracted and consistency verification is performed with the inspection data to obtain an initial consistency rate. A movement anomaly review module is used to predict the movement anomaly rate based on the BIM data of the monitoring point, obtain the movement anomaly rate, configure the review parameters based on the initial consistency rate and the anomaly rate of the monitoring point, and perform review consistency verification on the monitoring point to obtain the review consistency rate; The fusion anomaly labeling module is used to calculate the fusion anomaly rate based on the anomaly rate, movement anomaly rate and review consistency rate of the monitoring points, label the monitoring points and obtain the monitoring results.

Images