An intelligent splicing method for the measurement atlas of grid-frame buildings based on BIM and SLAM positioning

Abstract

An intelligent splicing method for the measurement atlas of grid-frame buildings based on BIM and SLAM positioning, comprising the following steps: determining the self-pose of a quadruped robot by combining the BIM model of the grid-frame building with the SLAM positioning algorithm to obtain its position information; obtaining the horizontal positions of the projections of all grid-frame nodes through the BIM model of the grid-frame building; planning a path while scanning with a multi-line lidar according to the position information to reach the initial point; reaching the projection position of the initial point, measuring the distances between the grid-frame node and its adjacent nodes, comparing the measurement data with the information of the nodes in the BIM model, marking the error if there is an error, and measuring the data of the next node if there is no error, while saving the currently measured atlas data and performing splicing; repeating the steps until all nodes have been measured and the atlas data has been saved, and outputting a measurement evaluation report. The present invention not only realizes the health assessment of grid-frame buildings efficiently and precisely, but also can be adaptively applied to grid-frame buildings in complex scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of grid structure building measurement, and particularly relates to an intelligent splicing method for grid structure building measurement atlases based on BIM and SLAM positioning. Background Art

[0002] A grid structure building is a building with a structural framework as the main body. Usually, its components are connected by welding, bolt connection or riveting, and bolt connection is the mainstream connection method for grid structure buildings. During the maintenance of large grid structure buildings, a large amount of measurement data will be generated when measuring each part of the grid structure. However, due to the complex shape of the grid structure building and the limitation of the high-altitude environment, it is difficult to measure, and it is difficult to guarantee the accuracy and integrity of the measurement data. Therefore, how to efficiently and accurately obtain the measurement data of grid structure buildings and evaluate their health status has become a difficult problem that needs to be solved urgently at present.

[0003] There is no perfect solution and product for the intelligent measurement and evaluation system of grid structure buildings. The measurement methods still rely on manual measurement and semi-intelligent measurement equipment, and mainly have the following limitations: 1) It takes a long time, has large errors and low measurement efficiency; 2) It requires a large amount of manpower and energy; 3) It is impossible to conduct a comprehensive health assessment of grid structure buildings, and the existing methods are only spot checks.

[0004] The invention patent documents with Chinese application numbers CN202210681992.3 and CN202210682006.6 disclose a method for measuring the spacing of nodes of a grid structure building and a method for scanning point cloud data of nodes of a grid steel structure building. The methods proposed in these two inventions still require certain human assistance to complete the comprehensive measurement and evaluation of grid structure buildings, and cannot complete the measurement task in the scenario where the node distribution of the grid structure is uneven, and have high requirements for the applicable scenarios of grid structure buildings. Summary of the Invention

[0005] In order to overcome the above-mentioned deficiencies of the prior art, the purpose of the present invention is to provide an intelligent splicing method for grid structure building measurement atlases based on BIM and SLAM positioning. This method uses a quadruped robot as the carrying equipment, and combines the BIM model of the grid structure building and SLAM positioning technology to autonomously complete the intelligent splicing of the grid structure building measurement atlas in the grid structure building scenario. It not only realizes the health assessment of grid structure buildings efficiently and accurately, but also can be adaptively applied to grid structure buildings in complex scenarios.

[0006] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0007] An intelligent splicing method for grid structure building measurement atlases based on BIM and SLAM positioning, comprising the following steps;

[0008] S1: Use the BIM model of the grid structure building in combination with the SLAM positioning algorithm to determine the self - posture of the quadruped robot, and obtain the position information of the quadruped robot in the building at present;

[0009] S2: Obtain the horizontal positions of the projections of all grid nodes through the BIM model of the grid structure building, and then obtain the coordinate points on the horizontal plane. Select the grid node in the upper left corner as the initial measurement point;

[0010] S3: According to the position information obtained in step S1, the quadruped robot uses a multi - line lidar to scan and plan a path to reach the initial point obtained in S2 while scanning;

[0011] S4: When the quadruped robot reaches the projection position of the initial point, measure the distance between this grid node and its adjacent nodes, and compare the measurement data with the information of the nodes in the BIM model. If there is an error, mark the error. If there is no error, measure the data of the next node. At the same time, save the current measured map data and splice it;

[0012] S5: Repeat step S4 until all nodes in the grid structure building have been measured and the map data has been saved, and then output a measurement evaluation report.

[0013] In the said step S1, the following steps are included:

[0014] S11: Read the BIM model of the grid structure building, which contains the three - dimensional model of the entire grid structure building and the accurate dimensions and positions of its internal building elements. Using the quadruped robot as a carrier platform, a measurement device and a multi - line lidar device are carried on its upper part;

[0015] Building Information Modeling (BIM) is the overall process of creating and managing building asset information. BIM is based on an intelligent model supported by a remote platform, integrating structured, multi - domain data together to generate a digital representation of the asset throughout the life cycle of the building (from planning and design to construction and operation). The BIM model not only contains the mutual positions of the grid nodes in the grid structure building, but also can accurately guide the determination of the self - posture and path planning of the quadruped robot;

[0016] S12: Scan the surrounding environment of the current position of the quadruped robot through the multi - line lidar, and then combine the SLAM positioning algorithm to determine the self - posture;

[0017] S13: Compare the BIM model obtained in step S11 and the three - dimensional information of the surrounding environment of the quadruped robot obtained in step S12, and accurately determine the current position P1 of the quadruped robot in the grid structure building.

[0018] In the step S2, given the BIM model of the grid structure building, the three-dimensional information of the grid nodes it contains is horizontally projected to obtain the horizontal positions of the grid nodes after projection. The nodes distributed on the horizontal plane after projection are uneven, and all the grid nodes are covered on this horizontal projection plane. The projection coordinate points of each grid node are obtained using its projection information. The grid node at the upper left corner of this projection plane is selected as the initial measurement node, and the projection coordinates O(x0, y0) of the initial node are obtained.

[0019] In the step S3, given that the position P1 of the quadruped robot in the grid structure building is obtained in the step S1, and then the projection coordinates O(x0, y0) of the initial node are obtained in the step S2, the quadruped robot automatically plans a path to reach the projection coordinates of the initial node from P1. During the movement, the multi-line lidar scans and confirms the surrounding environment for automatic obstacle avoidance. The three-dimensional information obtained by scanning is compared with the BIM model of the building to further guide the path planning and automatic obstacle avoidance of the quadruped robot, so that it can reach the projection position of the initial node more accurately.

[0020] In the step S4, the process of the quadruped robot measuring the grid structure building and saving the atlas data includes the following steps:

[0021] S41. When the quadruped robot reaches the projection position of the grid node to be measured, the multi-line lidar scans the surrounding three-dimensional model and compares it with the data in the BIM model to ensure that the quadruped robot is located at the accurate projection position of the grid node to be measured. Since the accuracy of the multi-line lidar depends on the size of its scanning radius, the smaller the scanning radius, the higher the accuracy, and the larger the scanning radius, the relatively lower the accuracy. Therefore, it is better to make the distance between the quadruped robot and the grid node to be measured as small as possible. Also, because the shortest distance between a point and a plane is the straight line perpendicular to the plane starting from this point, it is ensured that the quadruped robot is located at the accurate projection position of the grid node to be measured on the horizontal plane, so that the accuracy of the measurement data and the generated atlas data is higher;

[0022] S42. The quadruped robot starts to measure through the measuring equipment equipped on its upper part. The distance from the quadruped robot to the grid node to be measured, the distance from the quadruped robot to the adjacent node of the grid node to be measured, and the included angle between the two are measured using a laser rangefinder and an attitude sensor (the specific calculation method refers to the application number CN202210681992.3). The distance information between the grid node to be measured and the adjacent node is calculated, and at the same time, the spacing atlas data between this section of grid nodes is saved. The atlas data of the previous adjacent node and the next adjacent node are spliced through the relative position information of the grid nodes known in the BIM model;

[0023] S43. Compare the measurement data obtained in step S42 with the information in the BIM model. If there is a difference between the measured node spacing and the node spacing in the BIM model, and the difference exceeds or is less than the safety threshold of 1 cm, then an error warning message is marked for this node. If there is no difference, then measure the distance data of the next adjacent grid node;

[0024] S44. Save the distance data between the measured nodes, the error warning information, and the grid node spacing map data.

[0025] In the said step S5, after measuring all the grid nodes to be measured, the quadruped robot moves to the next grid node to be measured, and repeats step S4 until all the nodes in the grid building have been measured and the map data has been saved, and then outputs the overall measurement and evaluation report of the grid building; when there is an insurmountable obstacle at the projection position of a certain grid node to be measured, preventing the quadruped robot from reaching the precise projection position of the grid node to be measured, at this time, the quadruped robot needs to patrol around this obstacle, and during the patrol, measure the data of the current position of the quadruped robot from the grid node to be measured in real time. After the patrol ends, select the position with the smallest distance data to replace the projection position of the grid node to be measured, control the quadruped robot to reach this position, and measure the distance from the grid node to be measured to the adjacent node and the grid node map data.

[0026] Advantages of the present invention:

[0027] Under the background of the low intelligent level in the construction industry, the present invention effectively combines the BIM model and the SLAM positioning technology into the intelligent measurement and evaluation of grid buildings, so that the measurement method of large-scale grid buildings is no longer limited to manual measurement and semi-intelligent equipment. Through the mutual guidance of the BIM model and the SLAM positioning technology, with the quadruped robot as the carrying platform, measuring equipment and multi-line lidar are carried, the distance between grid nodes is measured with high precision, and the node map data of the grid building is generated. Through the comparison with the BIM model of the grid building, the grid nodes with errors are monitored and the evaluation report is output intelligently, and the measurement process is completed with high intelligence, high precision and high efficiency.

[0028] The method proposed by the present invention has high measurement accuracy for grid buildings realized by multi-line lidar, low time and labor costs for measuring grid buildings by means of being carried by a quadruped robot, and integrates three methods of the BIM model, the high-precision SLAM positioning method and being carried by a quadruped robot. High-precision measurement of grid nodes and generation of grid node maps can be achieved in large-scale grid buildings, and finally the measurement and evaluation report of the grid building is output, and the measurement and evaluation process is intelligent and efficient. Description of the drawings

[0029] Figure 1 It is a schematic flow chart of the method of the present invention.

[0030] Figure 2 This is a schematic flow chart for measuring a grid structure building according to the present invention.

[0031] Figure 3 This is a schematic diagram of the processing of a quadruped robot when there are obstacles under the grid node to be measured according to the present invention. Detailed implementation manners

[0032] The present invention will be further described in detail below with reference to the accompanying drawings.

[0033] This embodiment provides an intelligent splicing method for measuring atlas of a grid structure building based on BIM and SLAM positioning technologies. As Figure 1 shown, it includes the following steps:

[0034] S1. Use the BIM model of the grid structure building in combination with the SLAM positioning algorithm to determine the self-pose of the quadruped robot, and obtain the current position of the quadruped robot in the building.

[0035] Given the BIM model of the grid structure building, accurate information about the building structure can be obtained. Building Information Modeling (BIM) is an overall process for creating and managing building asset information. BIM is based on an intelligent model supported by a remote platform, integrating structured, multi-domain data together to generate a digital representation of the asset throughout the life cycle of the building (from planning and design to construction and operation). The BIM model not only contains the mutual positions of the grid nodes in the grid structure building, but also can accurately guide the determination of the self-pose and path planning of the quadruped robot.

[0036] The process of using the BIM model of the grid structure building in combination with the SLAM positioning algorithm to determine the accurate position information of the quadruped robot includes the following steps:

[0037] S11. Read the BIM model of the grid structure building, which contains the three-dimensional model of the entire grid structure building and the accurate dimensions and positions of its internal building elements. The method proposed by the present invention uses a quadruped robot as the carrier platform, and a measuring device and a multi-line lidar device are mounted on its upper part.

[0038] S12. Scan the surrounding environment of the current position of the quadruped robot through the multi-line lidar, and then determine the self-pose in combination with the SLAM positioning algorithm.

[0039] S13. Compare the BIM model obtained in step S11 with the three-dimensional information of the surrounding environment of the quadruped robot obtained in step S12, and the current position of the quadruped robot in the grid structure building can be accurately determined.

[0040] The measurement devices carried on the quadruped robot include a laser rangefinder, a high-definition variable-focus camera, an attitude sensor, etc., which are used to measure the distances and positions of adjacent nodes of the grid structure. In addition, the multi-line lidar carried on the quadruped robot is used to scan the internal environment of the grid structure building to determine its own attitude. As an intelligent device integrating locomotion capabilities such as walking, running, jumping, and getting up from the ground, the quadruped robot has obvious advantages over wheeled and track robots in terms of obstacle crossing and passability. Therefore, it has higher and better adaptability to complex terrains and can completely replace manual measurement to collect the measurement information and node map information of the grid structure building, helping the operation and maintenance personnel to more timely and comprehensively master the health status of the grid structure building, so as to make evaluations and warnings faster and more accurately, and reduce the potential danger of collapse accidents to the lowest level.

[0041] S2. Obtain the horizontal positions of the projections of all grid nodes from the BIM model of the grid structure building, and then obtain the coordinate points on the horizontal plane. Select the grid node in the upper left corner as the initial measurement point.

[0042] Given the BIM model of the grid structure building, project the three-dimensional information of the grid nodes it contains horizontally to obtain the horizontal positions of the projected grid nodes. The node distribution projected on the horizontal plane is uneven. This horizontal projection plane covers all grid nodes. Use the projection information of each grid node to obtain its projection coordinate points, and select the grid node in the upper left corner of this projection plane as the initial measurement node to obtain the projection coordinates O(x0, y0) of the initial node.

[0043] S3. According to the position information obtained in step S1, the quadruped robot uses the multi-line lidar to scan and plan the path while moving to the initial point obtained in S2.

[0044] It is known that the position P1 of the quadruped robot in the grid structure building is obtained in step S1, and then the projection coordinates O(x0, y0) of the initial node are obtained in step S2. The quadruped robot automatically plans the path to reach the projection coordinates of the initial node from P1. During the movement, the multi-line lidar scans and confirms the surrounding environment for automatic obstacle avoidance. Compare the three-dimensional information obtained by scanning with the BIM model of the building to further guide the path planning and automatic obstacle avoidance of the quadruped robot, so that it can reach the projection position of the initial node more accurately.

[0045] S4. The quadruped robot reaches the projection position of the initial point, measures the distance between this grid node and its adjacent nodes, compares the measurement data with the information of the nodes in the BIM model. If there is an error, mark the error. If there is no error, measure the data of the next node, and at the same time save the current measured map data and splice it.

[0046] The process of the quadruped robot measuring the grid structure building and saving the map data is as Figure 2As shown in the figure, it includes the following steps:

[0047] S41. After the quadruped robot reaches the projection position of the node to be measured on the grid, the multi-line lidar scans the surrounding 3D model and compares the data with that in the BIM model to ensure that the quadruped robot is located at the precise projection position of the node to be measured on the grid. Since the accuracy of the multi-line lidar depends on the size of its scanning radius, the smaller the scanning radius, the higher the accuracy, and the larger the scanning radius, the relatively lower the accuracy. Thus, it is better that the distance between the quadruped robot and the node to be measured on the grid is as small as possible. Also, because the shortest distance between a point and a plane is the straight line perpendicular to the plane starting from this point, it is ensured that the quadruped robot is located at the precise projection position of the node to be measured on the grid on the horizontal plane, so that the accuracy of the measurement data and the spectral data obtained by scanning is high;

[0048] S42. The quadruped robot starts to measure through the measuring equipment equipped on its upper part. The distance from the quadruped robot to the node to be measured, the distance from the quadruped robot to the adjacent node of the node to be measured, and the included angle between the two are measured by using a laser rangefinder and an attitude sensor. The distance information between the node to be measured and the adjacent node can be obtained through calculation, and at the same time, the spectral data of the spacing between the grid nodes in this section is saved. The spectral data of the previous adjacent node and the next adjacent node can be spliced through the relative position information of the known grid nodes in the BIM model;

[0049] S43. Compare the measurement data obtained in step S42 with the information in the BIM model. If there is a difference between the measured node spacing and the node spacing in the BIM model, and it exceeds or is less than the safety threshold of 1 cm, then this node is marked with error warning information. If there is no difference, then measure the distance data of the next adjacent grid node;

[0050] S44. Save the distance data between the measured nodes, the error warning information, and the spectral data of the spacing between the grid nodes.

[0051] S5. Repeat step S4 until all the nodes in the grid building have been measured and the spectral data is saved, and then output a measurement evaluation report.

[0052] After the measurement of the node to be measured on the grid is completed, the quadruped robot moves to the next node to be measured on the grid and repeats step S4 until all the nodes in the grid building have been measured and the spectral data is saved, and then output an overall measurement evaluation report of this grid building. It should be noted that when there are obstacles that cannot be crossed at the projection position of a certain node to be measured on the grid, preventing the quadruped robot from reaching the precise projection position of this node to be measured on the grid, at this time, the quadruped robot needs to patrol around this obstacle, and during the patrol, the data of the distance from the current position of the quadruped robot to the adjacent node of the node to be measured on the grid is measured in real time, such as Figure 3As shown, after the inspection tour ends, select the node position with the smallest distance data to replace the projection position of the node to be measured in the grid framework, control the quadruped robot to reach this position, and measure the distance from the node to be measured in the grid framework to the adjacent node and the grid node atlas data.

Claims

1. An intelligent splicing method for the measurement atlas of grid-frame buildings based on BIM and SLAM positioning, characterized in that, Including the following steps; S1: Determine the self - posture of the quadruped robot by combining the BIM model of the grid - framed building with the SLAM positioning algorithm, and obtain the position information of the quadruped robot in the building; S2: Obtain the horizontal positions of the projections of all grid nodes through the BIM model of the grid - framed building, and then obtain the coordinate points on the horizontal plane. Select the grid node in the upper - left corner as the initial measurement point; S3: According to the position information obtained in step S1, the quadruped robot uses a multi - line lidar to scan and plan the path while moving to the initial point obtained in S2; S4: The quadruped robot reaches the projection position of the initial point, measures the distance between the grid node and its adjacent nodes, compares the measurement data with the information of the nodes in the BIM model. If there is an error, mark the error. If there is no error, measure the data of the next node, and at the same time save the current measured atlas data and splice it; S5: Repeat step S4 until all nodes in the grid - framed building have been measured and the atlas data has been saved, and output a measurement evaluation report.

2. The intelligent splicing method of the grid building measurement atlas based on BIM and SLAM positioning according to claim 1, wherein In step S1, the following steps are included: S11: Read the BIM model of the grid - framed building, which contains the three - dimensional model of the entire grid - framed building and the accurate dimensions and positions of its internal building elements. Using the quadruped robot as a carrier platform, a measurement device and a multi - line lidar device are mounted on its upper part; S12: Scan the surrounding environment of the current position of the quadruped robot through the multi - line lidar, and then determine the self - posture in combination with the SLAM positioning algorithm; S13: Compare the BIM model obtained in step S11 with the three - dimensional information of the surrounding environment of the quadruped robot obtained in step S12, and accurately determine the current position P1 of the quadruped robot in the grid - framed building.

3. The intelligent splicing method of the grid building measurement atlas based on BIM and SLAM positioning according to claim 1, characterized in that In step S2, given the BIM model of the grid - framed building, project the three - dimensional information of the grid nodes it contains horizontally to obtain the horizontal positions of the projected grid nodes. The nodes projected on the horizontal plane are unevenly distributed. This horizontal projection plane covers all grid nodes. Use the projection information of each grid node to obtain its projection coordinate points, and select the grid node in the upper - left corner of this projection plane as the initial measurement node to obtain the projection coordinates O(x0,y0) of the initial node.

4. A method for intelligent splicing of grid building measurement atlases based on BIM and SLAM positioning according to claim 1, characterized in that In step S3, given that the position P1 of the quadruped robot in the grid - framed building is obtained in step S1, and then the projection coordinates O(x0,y0) of the initial node are obtained in step S2. The quadruped robot automatically plans a path from P1 to the projection coordinates of the initial node. During the movement, the multi - line lidar scans and confirms the surrounding environment for automatic obstacle avoidance. Compare the scanned three - dimensional information with the BIM model of the building to further guide the path planning and automatic obstacle avoidance of the quadruped robot, so that it can reach the projection position of the initial node more accurately.

5. A method for intelligent splicing of a grid building measurement atlas based on BIM and SLAM positioning according to claim 1, characterized in that, In step S4, the process of the quadruped robot measuring the grid - framed building and saving the atlas data includes the following steps: S41: When the quadruped robot reaches the projection position of the grid node to be measured, the multi - line lidar scans the surrounding three - dimensional model and compares it with the data in the BIM model to ensure that the quadruped robot is located at the accurate projection position of the grid node to be measured; S42. The quadruped robot starts measuring via the measuring equipment equipped on its upper part. The distance from the quadruped robot to the node to be measured, the distance from the quadruped robot to the adjacent node of the node to be measured, and the included angle between the two are measured using a laser rangefinder and an attitude sensor. The distance information between the node to be measured and the adjacent node is calculated, and meanwhile, the spacing map data between the nodes of this section of the grid structure is saved. The map data of the previous adjacent node and the next adjacent node are spliced through the relative position information of the known grid structure nodes in the BIM model; S43. Compare the measurement data obtained in step S42 with the information in the BIM model. If there is a difference between the measured node spacing and the node spacing in the BIM model, and the difference exceeds or is less than the safety threshold of 1 cm, then an error warning information is marked for this node. If there is no difference, then measure the distance data of the next adjacent grid structure node; S44. Save the distance data between the measured nodes, the error warning information, and the spacing map data of the grid structure nodes.

6. The intelligent splicing method of the grid building measurement atlas based on BIM and SLAM positioning according to claim 1, characterized in that, In step S5, after the measurement of the grid structure nodes to be measured is completed, the quadruped robot moves to the next grid structure node to be measured, and repeats step S4 until all the nodes in the grid structure building have been measured and the map data is saved, and then outputs the overall measurement and evaluation report of this grid structure building; when there is an insurmountable obstacle at the projection position of a certain grid structure node to be measured, preventing the quadruped robot from reaching the accurate projection position of this grid structure node to be measured, at this time, the quadruped robot needs to patrol around this obstacle. During the patrol, the data of the distance between the current position of the quadruped robot and the grid structure node to be measured is measured in real time. After the patrol is completed, select the position with the smallest distance data to replace the projection position of the grid structure node to be measured, control the quadruped robot to reach this position, and measure the distance from this grid structure node to the adjacent node and the grid structure node map data.

Images