Outer wall spraying robot automatic obstacle avoidance method based on laser radar and BIM

Abstract

The invention discloses an outer wall spraying robot automatic obstacle avoidance method based on laser radar and BIM, belongs to the technical field of outer wall spraying robots, and solves the problem that an existing outer wall spraying robot cannot automatically avoid obstacles such as windows or balconies. According to the method, the point cloud map of the outer wall is obtained through the BIM model, the point cloud map is projected into the grid map used for planning operation, the laser point cloud and the point cloud map are registered through the two-step point cloud registration method based on feature extraction, positioning of the outer wall spraying robot is achieved, meanwhile, window and balcony obstacle information is obtained, and the working efficiency is improved. The precision and the efficiency of spraying operation are improved, manual intervention is reduced, the operation risk is reduced, and the safety is greatly improved especially in high-altitude operation.

Description

Technical Field

[0001] The present invention relates to the technical field of exterior wall spraying robots, and in particular to an automatic obstacle avoidance method for an exterior wall spraying robot based on laser radar and BIM. Background Art

[0002] Manual work is the main construction method for exterior wall spraying, which is mainly achieved through high-altitude work using a liftable hanging basket. The traditional construction method is not only inefficient, prone to quality problems such as uneven spraying, but also has major safety hazards. The toxic and harmful substances volatilized during the spraying process will also endanger the occupational health and safety of construction workers.

[0003] In recent years, exterior wall spraying machines have gradually gained attention as new automated and intelligent construction equipment and have been applied to actual project construction. The application of exterior wall spraying robots can fundamentally avoid the risks of high-altitude operations and improve the quality and efficiency of spraying. However, exterior wall spraying robots still have shortcomings in actual applications. They cannot actively avoid obstacles such as exterior wall windows and balconies, and require manual intervention. This shortcoming limits the application scope and efficiency of exterior wall spraying robots, and it is difficult to perform automated spraying construction on more complex walls. Summary of the invention

[0004] In view of the above problems in the prior art, the present invention provides an automatic obstacle avoidance method for an exterior wall spraying robot based on laser radar and BIM, which solves the problem that the existing exterior wall spraying robots cannot automatically avoid obstacles such as windows or balconies.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] Provided is a method for automatic obstacle avoidance of an exterior wall spraying robot based on laser radar and BIM, comprising the following steps:

[0007] S1. Establish a BIM model of the construction wall, convert the BIM model into a point cloud map, and convert the point cloud map into a grid map, and import the point cloud map and grid map into the exterior wall spraying robot. By establishing a detailed BIM model, you can understand the specific conditions of the construction wall in advance, including the location of windows and balconies, which will help improve the accuracy of subsequent spraying operations. The grid map can clearly show the plan layout of the construction wall, especially the location of obstacles such as windows and balconies, providing an effective path planning basis for the robot. At the same time, the use of the grid map simplifies the path planning process, allowing the robot to quickly plan the optimal spraying path to avoid window and balcony obstacles.

[0008] S2. Use the laser radar on the exterior wall spraying robot to obtain the real-time laser point cloud of the construction wall, extract the main plane points and corner points of the real-time laser point cloud, and perform real-time alignment with the main plane points and corner points of the point cloud map.

[0009] S3. Generate an obstacle-avoiding operation path for the exterior wall spraying robot on the grid map through the Boustrophedon algorithm. The Boustrophedon algorithm can effectively plan a continuous and non-repetitive spraying path to ensure the uniformity and integrity of the spraying.

[0010] S4. The exterior wall spraying robot performs spraying operations along the operation path.

[0011] In this solution, considering that the construction wall is a degraded scene, there are fewer features in the range of laser radar scanning, and the corner features are provided by the protrusions and depressions that vary in depth such as windows and balconies. There will be a lack of constraints during real-time positioning, making it impossible to calculate the posture of the exterior wall spraying robot. Based on a large number of actual observations, it is found that the features of the exterior wall can be divided into two parts: the main plane and the corner points. Most of the points are concentrated on the main plane after screening, while the corner points are concentrated in windows, balconies and other parts, and most of them are linear features. Therefore, a two-step point cloud registration method based on feature extraction is adopted to register the laser point cloud with the point cloud map, which not only realizes the positioning of the exterior wall spraying robot, but also obtains the obstacle information of windows and balconies, which not only improves the accuracy and efficiency of the spraying operation, but also reduces manual intervention and reduces the risk of operation, especially in high-altitude operations, greatly improving safety.

[0012] Furthermore, the method of converting the BIM model into a point cloud map is:

[0013] S1.1. Use Revit software to create a BIM model of the construction exterior wall data in the rvt format. The BIM model includes the window and balcony information on the construction wall.

[0014] S1.2. Export the BIM model to an FBX format file, traverse the FBX file to extract the extracted geometric information, and write the geometric information into the point cloud file to obtain the initial point cloud map.

[0015] S1.3, down-sampling and denoising the initial point cloud map to obtain an optimized point cloud map. The sampling and denoising of the point cloud map improves the accuracy and reliability of the data.

[0016] Furthermore, the real-time laser point cloud and point cloud map registration method comprises the steps of:

[0017] S2.1. Use Harris 3D corner detection algorithm to identify the main plane points and corner points in real-time laser point cloud and point cloud map; the main plane point sets in real-time laser point cloud and point cloud map are M1 and M respectively, and the corner point sets in real-time laser point cloud and point cloud map are F1 and F respectively. Harris 3D corner detection algorithm can effectively identify key feature points in point cloud, including main plane points and corner points.

[0018] S2.2. Use the SVD algorithm to perform plane fitting on the main plane points in the real-time laser point cloud acquired in each frame in real time, and obtain the normalized plane equation of the main plane of the real-time scanning laser:

[0019] A1x+B1y+C1z+D1=0

[0020] Among them, A1, B1 and C1 are all constants. The normal vector of the real-time scanning laser plane is n 1 =(A1, B1, C1); x, y and z are the three coordinate axes of the real-time scanning laser main plane, and D1 is the offset of the real-time scanning laser plane. The main plane equation of the real-time laser point cloud can be accurately calculated by plane fitting through the SVD singular value decomposition algorithm.

[0021] S2.3. Obtain the normal vector n of the main plane of the point cloud map, and calculate the preliminary transformation matrix T from the main plane of the real-time scanning laser to the main plane of the point cloud map.

[0022] S2.4, align the points in M1 to the main plane of the point cloud map through the preliminary transformation matrix T to obtain the preliminary registration plane. The application of the preliminary transformation matrix T enables the points in the real-time laser point cloud to be preliminarily aligned with the points in the point cloud map to form a preliminary registration plane, which simplifies the subsequent fine registration process and improves the speed and accuracy of the registration.

[0023] S2.5. Project the corner points in F1 onto the preliminary registration plane, calculate the covariance matrix H between the point clouds on the preliminary registration plane and on the point cloud map by the least squares method, perform singular value decomposition on the covariance matrix H to obtain the rotation matrix R and translation vector t, and synthesize the final transformation matrix T through the rotation matrix R and translation vector t f Through the least squares method and singular value decomposition, accurate rotation matrices and translation vectors can be obtained, and then the final transformation matrix can be synthesized. This ensures the high accuracy of point cloud registration and provides reliable data support for the determination of the robot's real-time posture.

[0024] S2.6, combined with the initial posture of the exterior wall spraying robot, through the final transformation matrix T f Determine the real-time pose of the exterior wall spraying robot. The final transformation matrix T fThe application of enables the exterior wall spraying robot to accurately determine its real-time position on the construction wall. It ensures the robot's high-precision positioning in a dynamic environment, provides an important reference for subsequent path planning and obstacle avoidance, and improves the efficiency and safety of the spraying operation.

[0025] Furthermore, in step S2.3, n=(A, B, C), Among them, A, B and C are all constants, A 2 +B 2 +C 2 =1,T y (θ), and T x (σ) are the transformation matrix of the real-time scanning laser main plane rotating around its y-axis, the rotation transformation matrix around the z-axis, and the translation transformation matrix along the x-axis; θ, and σ are the rotation angles around the y-axis and the z-axis respectively and the offset along the x-axis.

[0026] Furthermore, the expression of the covariance matrix H in step S2.5 is: f′ 1i =T·f i ;

[0027] Among them, f i is a point (x i ,y i ,z i ), f i =(x i ,y i ,z i )∈F1,(x i ,y i ,z i ) is the i-th point in F1;

[0028] f′ 1i f i The transformed point, f′ 1i =(x′ i ,y′ i ,z′ i ), (x′ i ,y′ i ,z′ i ) is (x i ,y i ,z i ) is the point transformed, i is a constant; n is the total number of corner points in F1; is the centroid of the corner point set F in the point cloud map, F′ is the initial registration plane1i The center of mass, F′ 1i f′ 1i A collection of .

[0029] Furthermore, in step S2.5, H=UΣV T 、R=VU T ,

[0030] Among them, Σ is a diagonal matrix, and the elements on the diagonal are singular values, arranged in order from large to small; U and V are obtained by singular value decomposition, U is the main direction orthogonal matrix of the corner point set F in the point cloud map; V is the preliminary registration plane corner point set F′ 1i The principal direction orthogonal matrix of .

[0031] Furthermore, in order to prevent the exterior wall spraying robot from spraying on windows and balconies and facilitate the exterior wall spraying robot to avoid obstacles, the windows and balconies are set as inaccessible areas. Specifically, considering that windows are usually recessed from the outer surface of the wall, if the robot spraying path does not consider this depth, it may mistakenly spray on the inside of the window, resulting in unnecessary waste and damage. Balconies usually protrude outward from the outer surface of the wall. If the robot spraying path does not consider this protruding distance, it may mistakenly spray on the outside of the balcony. Therefore, the method of projecting the point cloud map into a raster map in step S1 is:

[0032] S1.4. With the thickness direction of the construction wall as the Y-axis, the length direction of the construction wall as the X-axis, and the height direction of the construction wall as the Z-axis, determine the axis position Y0 in the thickness direction of the wall, obtain the depth of the window [Y0, Y1] and the depth of the balcony [Y0, Y2]. Y1 and Y2 are the sunken and raised positions of the window and balcony on the Y-axis respectively.

[0033] S1.5. Set the projection intervals U1 and U2 of the windows and balconies, where U1∈[Y0,Y1-h2], U2∈[Y0+h1,Y2], and h1 and h2 are both set projection thresholds.

[0034] S1.6, determine in turn whether the Y-axis coordinate of each point in the point cloud map falls within the projection intervals U1 and U2. If so, project the X-axis and Z-axis coordinates of the point along the Y-axis direction onto the XZ plane, otherwise ignore the point. Through this projection rule, a vacant area is formed on the grid map, which represents window and balcony obstacles. The exterior wall spraying robot will not pass through the vacant area, thereby enabling the exterior wall spraying robot to avoid window and balcony obstacles.

[0035] Furthermore, the exterior wall spraying robot uses the Boustrophedon algorithm to plan a "bow"-shaped path to cover the entire grid map and avoid the empty areas on the grid map. The path generated by the Boustrophedon algorithm is similar to the "bow"-shaped trajectory in farmland cultivation. This path planning method ensures that the robot can efficiently and evenly cover the entire construction wall surface, avoiding repeated spraying and missing areas. This path planning method ensures the continuity and smoothness of the robot's movement, reduces unnecessary turns and pauses, and improves the efficiency of the spraying operation.

[0036] Furthermore, considering that if the nozzle of the exterior wall spraying robot is not controlled, it will cause repeated spraying of the already sprayed area or spraying of the area that should not be sprayed when avoiding the window, the spraying method of the exterior wall spraying robot during the spraying process is:

[0037] S4.1. Divide the construction wall surface into multiple construction units along the operation path, each construction unit being a rectangular area with a length and a width respectively;

[0038] S4.2. Control the exterior wall spraying robot to spray the construction units in sequence along the operation path, and cumulatively calculate the X-axis displacement and Z-axis displacement of the spraying robot in each construction unit; after completing the spraying of a construction unit, clear the X-axis displacement and Z-axis displacement, and restart the cumulative calculation of displacement information to ensure that the displacement caused by avoiding windows in each construction unit can be identified and judged until the wall construction is completed.

[0039] S4.3. When the exterior wall spraying robot is spraying each construction unit, it is determined whether the Z-axis displacement generated by the X-axis displacement of the exterior wall spraying robot within the step range of a is less than b. If so, it is determined that the exterior wall spraying robot has moved to avoid obstacles and stops spraying. Otherwise, it continues spraying to complete the current construction unit.

[0040] Furthermore, the method further includes the following steps: S2.7, using a drone to perform a timed cruise on the spraying robot, the drone uses visual SLAM to locate itself and reversely resolves the position information of the exterior wall spraying robot based on its own positioning, and the exterior wall spraying robot performs posture correction by receiving the reversed position information of the drone. Through the drone's timed cruise and visual SLAM technology, the position and posture of the spraying robot can be monitored in real time, and posture deviations can be discovered and corrected in a timely manner, which not only improves the accuracy of the spraying operation, but also enhances the robustness and reliability of the system, ensuring the smooth progress of the spraying operation.

[0041] The present invention discloses an automatic obstacle avoidance method for an exterior wall spraying robot based on laser radar and BIM, and its beneficial effects are:

[0042] 1. The present invention obtains the point cloud map of the exterior wall through the BIM model, and projects the point cloud map into a raster map for planning operations. The laser point cloud is registered with the point cloud map based on a two-step point cloud registration method extracted from features, which not only realizes the positioning of the exterior wall spraying robot, but also obtains the obstacle information of windows and balconies. It not only improves the accuracy and efficiency of the spraying operation, but also reduces manual intervention and reduces the operation risk, especially in high-altitude operations, which greatly improves safety.

[0043] 2. The present invention realizes the operation route planning of the exterior wall spraying robot for window avoidance construction through the preset construction unit Boustrophedon algorithm, adopts the two-step point cloud registration and positioning algorithm to obtain the operation position of the exterior wall spraying robot, and realizes the timing correction through the drone cruise, thereby realizing the automatic window avoidance and automatic obstacle avoidance of the exterior wall spraying robot, and expanding the application scope of the exterior wall spraying robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flowchart of the automatic obstacle avoidance method of the exterior wall spraying robot based on laser radar and BIM;

[0045] Figure 2 It is a schematic diagram of the projection direction of the point cloud map and the raster map;

[0046] Figure 3 Schematic diagram of construction path planning for exterior wall spraying robot. DETAILED DESCRIPTION

[0047] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0048] refer to Figure 1 This embodiment provides an automatic obstacle avoidance method for an exterior wall spraying robot based on laser radar and BIM, comprising the steps of:

[0049] S1. Build a BIM model of the construction wall, convert the BIM model into a point cloud map, convert the point cloud map into a raster map, and import the point cloud map and raster map into the exterior wall spraying robot. By building a detailed BIM model, you can understand the specific conditions of the construction wall in advance, including the location of windows and balconies, which helps to improve the accuracy of subsequent spraying operations.

[0050] Specifically, the method of converting the BIM model into a point cloud map is:

[0051] S1.1. Use Revit software to create a BIM model of the construction exterior wall data. The format type of the BIM model is rvt. The BIM model includes the window and balcony information on the construction wall.

[0052] S1.2. Export the BIM model to an FBX format file, traverse the FBX file to extract the extracted geometric information, and write the geometric information into the point cloud file to obtain the initial point cloud map.

[0053] S1.3, down-sampling and denoising the initial point cloud map to obtain an optimized point cloud map. The sampling and denoising of the point cloud map improves the accuracy and reliability of the data.

[0054] In order to prevent the exterior wall spraying robot from spraying on windows and balconies and facilitate the exterior wall spraying robot to avoid obstacles, the windows and balconies are set as inaccessible areas. Specifically, considering that windows are usually recessed from the outer surface of the wall, if the robot spraying path does not consider this depth, it may mistakenly spray on the inside of the window, resulting in unnecessary waste and damage. Balconies usually protrude outward from the outer surface of the wall. If the robot spraying path does not consider this protruding distance, it may mistakenly spray on the outside of the balcony. Therefore, the method of projecting the point cloud map into a raster map in step S1 is:

[0055] S1.4. References Figure 2 , with the thickness direction of the construction wall as the Y axis, the length direction of the construction wall as the X axis, and the height direction of the construction wall as the Z axis, determine the axis position Y0 in the thickness direction of the wall, and obtain the depth of the window [Y0, Y1] and the depth of the balcony [Y0, Y2]. Y1 and Y2 are the sunken and raised positions of the window and balcony on the Y axis respectively.

[0056] S1.5. Set the projection intervals U1 and U2 of the windows and balconies, where U1∈[Y0,Y1-h2], U2∈[Y0+h1,Y2], and h1 and h2 are both set projection thresholds.

[0057] S1.6. Determine in turn whether the Y-axis coordinate of each point in the point cloud map falls within the projection intervals U1 and U2. If so, project the X-axis and Z-axis coordinates of the point along the Y-axis direction onto the XZ plane, otherwise ignore the point. This projection rule forms a vacant area on the grid map, which represents window and balcony obstacles. The exterior wall spraying robot will not pass through the vacant area, thereby enabling the exterior wall spraying robot to avoid window and balcony obstacles. S2. Use the laser radar installed on the exterior wall spraying robot to obtain the real-time laser point cloud of the construction wall, extract the main plane points and corner points of the real-time laser point cloud, and perform real-time registration with the main plane points and corner points of the point cloud map.

[0058] S3. Generate an obstacle-avoiding operation path for the exterior wall spraying robot on the grid map through the Boustrophedon algorithm. The Boustrophedon algorithm can effectively plan a continuous and non-repetitive spraying path to ensure the uniformity and integrity of the spraying.

[0059] In this embodiment, reference Figure 3 The exterior wall spraying robot uses the Boustrophedon algorithm to plan a "bow"-shaped path to cover the entire grid map and avoid the empty areas on the grid map. The path generated by the Boustrophedon algorithm is similar to the "bow"-shaped trajectory in farmland cultivation. This path planning method ensures that the robot can efficiently and evenly cover the entire construction wall surface, avoiding repeated spraying and missing areas. This path planning method ensures the continuity and smoothness of the robot's movement, reduces unnecessary turns and pauses, and improves the efficiency of the spraying operation.

[0060] S4. The exterior wall spraying robot performs spraying operations along the operation path.

[0061] Specifically, the registration method of real-time laser point cloud and point cloud map includes the following steps:

[0062] S2.1. Use Harris 3D corner detection algorithm to identify the main plane points and corner points in real-time laser point cloud and point cloud map; the main plane point sets in real-time laser point cloud and point cloud map are M1 and M respectively, and the corner point sets in real-time laser point cloud and point cloud map are F1 and F respectively. Harris 3D corner detection algorithm can effectively identify key feature points in point cloud, including main plane points and corner points.

[0063] S2.2. Use the SVD algorithm to perform plane fitting on the main plane points in the real-time laser point cloud acquired in each frame in real time, and obtain the normalized plane equation of the main plane of the real-time scanning laser:

[0064] A1x+B1y+C1z+D1=0

[0065] Among them, A1, B1 and C1 are all constants. The normal vector of the real-time scanning laser plane is n 1 =(A1, B1, C1); x, y and z are the three coordinate axes of the real-time scanning laser main plane, and D1 is the offset of the real-time scanning laser plane. The main plane equation of the real-time laser point cloud can be accurately calculated by plane fitting through the SVD singular value decomposition algorithm.

[0066] S2.3. Obtain the normal vector n of the main plane of the point cloud map, and calculate the preliminary transformation matrix T from the main plane of the real-time scanning laser to the main plane of the point cloud map.

[0067] In step S2.3, n=(A, B, C), Among them, A, B and C are all constants, A 2 +B 2 +C 2 =1,T y (θ), and T x (σ) are the transformation matrix of the real-time scanning laser main plane rotating around its y-axis, the rotation transformation matrix around the z-axis, and the translation transformation matrix along the x-axis; θ, and σ are the rotation angles around the y-axis and the z-axis respectively and the offset along the x-axis.

[0068] S2.4, align the points in M1 to the main plane of the point cloud map through the preliminary transformation matrix T to obtain the preliminary registration plane. The application of the preliminary transformation matrix T enables the points in the real-time laser point cloud to be preliminarily aligned with the points in the point cloud map to form a preliminary registration plane, which simplifies the subsequent fine registration process and improves the speed and accuracy of the registration.

[0069] S2.5. Project the corner points in F1 onto the preliminary registration plane, calculate the covariance matrix H between the point clouds on the preliminary registration plane and on the point cloud map by the least squares method, perform singular value decomposition on the covariance matrix H to obtain the rotation matrix R and translation vector t, and synthesize the final transformation matrix T through the rotation matrix R and translation vector t f Through the least squares method and singular value decomposition, accurate rotation matrices and translation vectors can be obtained, and then the final transformation matrix can be synthesized. This ensures the high accuracy of point cloud registration and provides reliable data support for the determination of the robot's real-time posture.

[0070] Specifically, the expression of the covariance matrix H in step S2.5 is, f′ 1i =T·f i Among them, f i is a point (x i ,y i ,z i ), f i =(x i ,y i ,z i )∈F1,(x i ,y i ,z i ) is the i-th point in F1, f′ 1i f i The transformed point, f′ 1i =(x′ i ,y′ i ,z′i ), (x′ i ,y′ i ,z′ i ) is (x i ,y i ,z i ) is the point transformed, i is a constant; n is the total number of corner points in F1; is the centroid of the corner point set F in the point cloud map, F′ is the initial registration plane 1i The center of mass, F′ 1i f′ 1i A collection of .

[0071] In step S2.5, H=UΣV T 、R=VU T , Among them, Σ is a diagonal matrix, and the elements on the diagonal are singular values, arranged in order from large to small; U and V are obtained by singular value decomposition, U is the main direction orthogonal matrix of the corner point set F in the point cloud map; V is the preliminary registration plane corner point set F′ 1i The principal direction orthogonal matrix of .

[0072] S2.6, combined with the initial posture of the exterior wall spraying robot, through the final transformation matrix T f Determine the real-time pose of the exterior wall spraying robot. The final transformation matrix T f The application of enables the exterior wall spraying robot to accurately determine its real-time position on the construction wall. It ensures the robot's high-precision positioning in a dynamic environment, provides an important reference for subsequent path planning and obstacle avoidance, and improves the efficiency and safety of the spraying operation.

[0073] As a further solution of this embodiment, S2.7, a drone is used to perform timed cruise of the spraying robot. The drone uses visual SLAM to locate itself and reversely solves the position information of the exterior wall spraying robot based on its own positioning. The exterior wall spraying robot performs posture correction by receiving the reversed position information of the drone. Through the drone's timed cruise and visual SLAM technology, the position and posture of the spraying robot can be monitored in real time, and posture deviations can be discovered and corrected in time, which not only improves the accuracy of the spraying operation, but also enhances the robustness and reliability of the system, ensuring the smooth progress of the spraying operation.

[0074] As a further solution of this embodiment, considering that if the nozzle of the exterior wall spraying robot is not controlled, it will cause repeated spraying of the already sprayed area or spraying of the area that should not be sprayed when avoiding the window, the spraying method of the exterior wall spraying robot during the spraying process is:

[0075] S4.1. Divide the construction wall surface into multiple construction units along the operation path, each construction unit being a rectangular area with a length and a width respectively;

[0076] S4.2. Control the exterior wall spraying robot to spray the construction units in sequence along the planned path, and accumulate and calculate the X-axis displacement and Z-axis displacement of the spraying robot in each construction unit; after completing the spraying of a construction unit, clear the X-axis displacement and Z-axis displacement, and restart the accumulation and calculation of displacement information to ensure that the displacement caused by avoiding windows in each construction unit can be identified and judged until the wall construction is completed.

[0077] S4.3. When the exterior wall spraying robot is spraying each construction unit, it is determined whether the Z-axis displacement generated by the X-axis displacement of the exterior wall spraying robot within the step range of a is less than b. If so, it is determined that the exterior wall spraying robot has moved to avoid obstacles and stops spraying. Otherwise, it keeps spraying to complete the construction of the current construction unit.

[0078] In summary, this solution takes into account that the construction wall is a degraded scene, and there are fewer features in the range of laser radar scanning. The corner features are provided by the protrusions and depressions that vary in depth such as windows and balconies. There will be a lack of constraints during real-time positioning, making it impossible to calculate the posture of the exterior wall spraying robot. Based on a large number of actual observations, it is found that the features of the exterior wall can be divided into two parts: the main plane and the corner points. Most of the points are concentrated on the main plane after screening, while the corner points are concentrated in windows, balconies and other parts, and most of them are linear features. Therefore, this solution adopts a two-step point cloud registration method based on feature extraction to register the laser point cloud with the point cloud map, which not only realizes the positioning of the exterior wall spraying robot, but also obtains the obstacle information of windows and balconies, which not only improves the accuracy and efficiency of the spraying operation, but also reduces manual intervention and reduces the risk of operation, especially in high-altitude operations, greatly improving safety.

[0079] Although the specific implementation of the invention is described in detail in conjunction with the drawings, it should not be understood as limiting the scope of protection of this patent. Within the scope described in the claims, various modifications and variations that can be made by those skilled in the art without creative work still fall within the scope of protection of this patent.

Claims

1. Automatic obstacle avoidance method for exterior wall spraying robot based on laser radar and BIM, characterized in that: Includes steps: S1. Establish a BIM model of the construction wall, convert the BIM model into a point cloud map, convert the point cloud map into a raster map, and import the point cloud map and the raster map into the exterior wall spraying robot; S2. Use the laser radar on the exterior wall spraying robot to obtain the real-time laser point cloud of the construction wall, extract the main plane points and corner points of the real-time laser point cloud, and perform real-time registration with the main plane points and corner points of the point cloud map; S3, generating an operation path for the exterior wall spraying robot to avoid obstacles on the grid map through the Boustrophedon algorithm; S4. The exterior wall spraying robot performs spraying operations along the operation path.

2. The automatic obstacle avoidance method of the exterior wall spraying robot according to claim 1 is characterized in that: The method of converting BIM model into point cloud map is: S1.

1. Use Revit software to create a BIM model of the construction exterior wall data in rvt format, which includes the window and balcony information on the construction wall; S1.2, export the BIM model to an FBX format file, traverse the FBX file to extract the extracted geometric information, and write the geometric information into the point cloud file to obtain an initial point cloud map; S1.

3. Down-sampling and denoising the initial point cloud map are performed to obtain an optimized point cloud map.

3. The automatic obstacle avoidance method of the exterior wall spraying robot according to claim 1, characterized in that: The registration method of real-time laser point cloud and point cloud map includes the following steps: S2.

1. Use Harris 3D corner point detection algorithm to identify the main plane points and corner points in the real-time laser point cloud and point cloud map; wherein the main plane point sets in the real-time laser point cloud and point cloud map are M1 and M respectively, and the corner point sets in the real-time laser point cloud and point cloud map are F1 and F respectively; S2.

2. Use the SVD algorithm to perform plane fitting on the main plane points in the real-time laser point cloud acquired in each frame in real time, and obtain the normalized plane equation of the main plane of the real-time scanning laser: A1x+B1y+C1z+D1=0 Among them, A1, B1 and C1 are all constants. The normal vector of the real-time scanning laser plane is n 1 =(A1, B1, C1); x, y and z are the three coordinate axes of the real-time scanning laser main plane, and D1 is the offset of the real-time scanning laser plane; S2.3, obtain the normal vector n of the main plane of the point cloud map, and calculate the preliminary transformation matrix T from the main plane of the real-time scanning laser to the main plane of the point cloud map; S2.4, aligning the points in M1 to the main plane of the point cloud map through the preliminary transformation matrix T to obtain the preliminary registration plane; S2.

5. Project the corner points in F1 onto the preliminary registration plane, calculate the covariance matrix H between the point clouds on the preliminary registration plane and on the point cloud map by the least squares method, perform singular value decomposition on the covariance matrix H to obtain the rotation matrix R and translation vector t, and synthesize the final transformation matrix T through the rotation matrix R and translation vector t f ; S2.6, combined with the initial posture of the exterior wall spraying robot, through the final transformation matrix T f Determine the real-time pose of the exterior wall painting robot.

4. The automatic obstacle avoidance method of the exterior wall spraying robot according to claim 3 is characterized in that: In step S2.3, n=(A, B, C), Among them, A, B and C are all constants, A 2 +B 2 +C 2 =1,T y (θ), and T x (σ) are the transformation matrix of the real-time scanning laser main plane rotating around its y-axis, the rotation transformation matrix around the z-axis, and the translation transformation matrix along the x-axis; θ, and σ are the rotation angles around the y-axis and the z-axis respectively and the offset along the x-axis.

5. The automatic obstacle avoidance method for an exterior wall spraying robot according to claim 3, characterized in that: The expression of the covariance matrix H in step S2.5 is, Among them, f i is a point (x i ,y i ,z i ), f i =(x i ,y i ,z i )∈F1,(x i ,y i ,z i ) is the i-th point in F1, f1 ′ i f i The transformed point, f1 ′ i =(x i ′ ,y i ′ ,x i ′ ), (x i ′ ,y i ′ ,z i ′ ) is (x i ,y i ,z i ) is the point transformed, i is a constant; n is the total number of corner points in F1; is the centroid of the corner point set F in the point cloud map, F′ is the initial registration plane 1i The center of mass, F′ 1i f′ 1i A collection of .

6. The automatic obstacle avoidance method for an exterior wall spraying robot according to claim 3, characterized in that: In step S2.5, H=UΣV T 、R=VU T , Among them, Σ is a diagonal matrix, and the elements on the diagonal are singular values, arranged in order from large to small; U and V are obtained by singular value decomposition, U is the main direction orthogonal matrix of the corner point set F in the point cloud map; V is the preliminary registration plane corner point set F1 ′ i The principal direction orthogonal matrix of .

7. The automatic obstacle avoidance method for an exterior wall spraying robot according to claim 1, characterized in that: The method for projecting the point cloud map into a raster map in step S1 is: S1.

4. With the thickness direction of the construction wall as the Y axis, the length direction of the construction wall as the X axis, and the height direction of the construction wall as the Z axis, determine the axis position Y0 in the thickness direction of the wall, obtain the depth of the window [Y0, Y1] and the depth of the balcony [Y0, Y2], where Y1 and Y2 are the sunken position and convex position of the window and balcony on the Y axis respectively; S1.

5. Set the projection intervals U1 and U2 of the windows and balconies, where U1∈[Y0,Y1-h2], U2∈[Y0+h1,Y2], and h1 and h2 are both set projection thresholds; S1.

6. Determine in turn whether the Y-axis coordinate of each point in the point cloud map falls within the projection intervals U1 and U2. If so, project the X-axis and Z-axis coordinates of the point along the Y-axis direction onto the XZ plane; otherwise, ignore the point.

8. The automatic obstacle avoidance method for an exterior wall spraying robot according to claim 7, characterized in that: The exterior wall spraying robot uses the Boustrophedon algorithm to plan a "bow"-shaped path to cover the entire grid map and avoid empty areas on the grid map.

9. The automatic obstacle avoidance method for an exterior wall spraying robot according to claim 8, characterized in that: The spraying method of the exterior wall spraying robot during the spraying process is: S4.

1. Divide the construction wall surface into multiple construction units along the operation path, each construction unit being a rectangular area with a length and a width respectively; S4.2, control the exterior wall spraying robot to spray the construction units in sequence along the operation path, and cumulatively calculate the X-axis displacement and Z-axis displacement of the spraying robot in each construction unit; S4.3, when the exterior wall spraying robot is spraying each construction unit, it is determined whether the Z-axis displacement generated by the exterior wall spraying robot within the step length range of a is less than b. If so, spraying is stopped; otherwise, spraying is continued to complete the current construction unit.

10. The automatic obstacle avoidance method for an exterior wall spraying robot according to claim 1, characterized in that: Also includes the steps: S2.

7. Use a drone to perform timed patrol of the spraying robot. The drone uses visual SLAM to locate itself and reversely solves the position information of the exterior wall spraying robot based on its own positioning. The exterior wall spraying robot corrects its posture by receiving the reversed position information from the drone.

Images