An automatic brick arrangement method for two-structure masonry walls based on rule-based autonomous learning

Abstract

The present invention discloses a method for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning, including area acquisition, used to acquire the construction range of the arrangement area; vertical stratification, used to divide the two-structure masonry wall to be constructed within the construction range in the vertical direction, and divide the two-structure masonry wall from bottom to top into N layers of bricks for arrangement according to the height of the bricks used in the two-structure masonry wall and the thickness of the mortar joint; pre-processing partition, used to partition and arrange the bricks of the corresponding layer according to the partition when the partition is involved in the construction range; rule autonomous learning, used to automatically arrange each layer of bricks in the arrangement area; comprehensive arrangement, in the arrangement area, according to the rule autonomous learning, complete the brick arrangement diagram of the entire wall of the two-structure masonry wall. The present invention can automatically arrange or rearrange the bricks in the arrangement area, and has the advantage of high arrangement efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering, and in particular to an automatic brick arrangement method for a two-structure masonry wall based on rule autonomous learning. Background Art

[0002] In the process management of building structure construction, the secondary structure is usually a masonry structure, and the construction process requires the drawing of brick arrangement details in advance. The existing details algorithm requires manual research on brick arrangement rules and classification discussions for complex situations, which leads to complex programming processes, large amounts of code, complex and cumbersome logic, and the need to re-derive the algorithm formula for arrangement when facing masonry of different specifications; in addition, when the position of the opening changes, it often affects the entire structure, and all bricks need to be modified accordingly. There is a lack of a solution based on overall considerations that can automatically solve different brick sizes and arrangement types. Summary of the invention

[0003] The purpose of the present invention is to provide a method for automatic arrangement of bricks of a two-structure masonry wall based on rule-based autonomous learning, so as to solve the problem of low arrangement efficiency of the two-structure masonry wall and failure to automatically arrange the bricks when the construction scope, opening or brick size changes.

[0004] In order to solve the above technical problems, the present invention provides a method for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning, comprising:

[0005] Step 210, area acquisition, is used to obtain the construction scope of the layout area;

[0006] Step 220, vertical stratification, is used to divide the second structure masonry wall to be constructed within the construction range in the vertical direction, and divide the second structure masonry wall into N layers of bricks from bottom to top for arrangement according to the height of the bricks used in the second structure masonry wall and the thickness of the mortar joints;

[0007] Step 230, pre-processing partitions, for arranging the bricks of the corresponding layers according to the partitions when the partitions are involved in the construction scope;

[0008] Step 240, autonomous rule learning for automatically arranging each layer of bricks in the arrangement area;

[0009] Step 250, comprehensive arrangement, autonomously learning and completing the brick arrangement diagram of the entire wall of the second structure masonry wall in the arrangement area according to the rules.

[0010] Furthermore, the present invention provides a method for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning, wherein the method for automatically arranging bricks of each layer in the rule autonomous learning includes:

[0011] Step 241, dividing each layer of bricks into multiple regions, and calculating the total length of each layer of regions;

[0012] Step 242: According to formula (1), an integer is taken to estimate the number of bricks in each arrangement area, and the length of each brick in the arrangement area is recorded as X1, X2, ..., X n , as the independent variable in the autonomous learning process of rules, is optimized, and bricks are arranged in the arrangement area according to the length of each brick to form a preliminary brick arrangement diagram;

[0013] N = L / (X + L2) + C (1);

[0014] Among them, N is the number of bricks in each arrangement area, L is the total length of the area, X is the length of the brick, L2 is the length of the mortar joint, and C is a constant;

[0015] Step 243, optimizing the preliminary brick arrangement diagram according to the brick arrangement rule loss function (2), and when the total loss in the arrangement area is minimized, determining the optimized preliminary brick arrangement diagram as the arrangement plan;

[0016] Loss= k1×Loss1+ k2×Loss2+ k3×Loss3+ k4×Loss4 (2);

[0017] Among them, Loss is the total loss in the layout area, Loss1 is the total loss of the upper and lower layer staggered joints, Loss2 is the total loss of the number of non-standard bricks, Loss3 is the total loss of the total length of the area, and Loss4 is the total loss of broken bricks. k1, k2, k3, and k4 are the importance coefficients of the corresponding loss functions.

[0018] Furthermore, the automatic brick arrangement method for a two-structure masonry wall based on rule autonomous learning provided by the present invention calculates the total amount of upper and lower layer staggered joint losses according to the upper and lower layer staggered joint loss function (3);

[0019] Loss1=Loss1_1+ Loss1_2+……+ Loss1_i+……+ Loss1_n (3);

[0020] Among them, Loss1 is the total loss of the staggered joints of the upper and lower layers, Loss1_1 is the staggered joint loss of the first brick, Loss1_2 is the staggered joint loss of the second brick, Loss1_i is the staggered joint loss of the i-th brick, and Loss1_n is the staggered joint loss of the n-th brick;

[0021] The minimum offset spacing between the upper and lower adjacent layers of masonry is set to d, and the length, width and height of the standard bricks are a, b and h respectively; when the offset is greater than d, Loss1_i is 0; when the offset is less than d, Loss1_i is the absolute value or square of the difference between the offset length and d.

[0022] Further, for the automatic brick arrangement method of the two - structure masonry wall based on rule - based autonomous learning, when X i < a, it is determined that the i - th brick is a non - standard brick, and the total loss of non - standard brick quantity is calculated according to the non - standard brick quantity loss function (4); (4);

[0023] wherein, X i is the length of the i - th brick, Loss2 is the total loss of non - standard brick quantity, F is a non - standard brick.

[0024] Further, for the automatic brick arrangement method of the two - structure masonry wall based on rule - based autonomous learning, the total loss of regional total length is calculated according to the regional total length loss function (5);

[0025] (5);

[0026] where Loss3 is the total loss of regional total length.

[0027] Further, for the automatic brick arrangement method of the two - structure masonry wall based on rule - based autonomous learning, when X i < 2d, it is determined as a broken brick, and the number of broken bricks is counted according to the broken brick quantity loss function (6);

[0028] (6);

[0029] wherein, Loss4 is the total loss of broken bricks, and z is a broken brick.

[0030] Further, for the automatic brick arrangement method of the two - structure masonry wall based on rule - based autonomous learning, the gradient descent method is used to iteratively update the parameters X1, X2, ……, X n of the brick arrangement rule loss function (2) to minimize the value of the brick arrangement rule loss function (2) and obtain the final brick arrangement diagram.

[0031] Further, for the automatic brick arrangement method of the two - structure masonry wall based on rule - based autonomous learning, the method for obtaining the final brick arrangement diagram includes:

[0032] Calculating the gradient: For the current parameter values, calculate the gradient of the brick arrangement rule loss function (2) with respect to each parameter X i ;

[0033] Updating the parameters: Use the gradient and the learning rate to update each parameter;

[0034] Repeated iteration: Repeat the process of calculating gradients and updating parameters until the brick arrangement rule loss function value drops below a certain threshold or cannot be further reduced, thus forming the final brick arrangement diagram.

[0035] Furthermore, the present invention provides a method for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning. In the comprehensive arrangement, the method for completing the brick arrangement diagram of the entire two-structure masonry wall in the arrangement area according to the rule autonomous learning includes:

[0036] For a wall, start arranging the bricks from the bottom layer. After one layer of bricks is arranged, arrange them upwards layer by layer until the entire wall is completed.

[0037] Furthermore, the present invention provides a method for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning. In the comprehensive arrangement, the method for completing the brick arrangement diagram of the entire two-structure masonry wall in the arrangement area according to the rule autonomous learning includes:

[0038] The bricks of the whole wall are regarded as a whole, and the lengths X(i, j) of all bricks in each arrangement area of ​​the whole wall are taken as independent variables. The brick rule loss function (2) of all areas of each layer of bricks in the whole wall is superimposed to obtain the total loss of the whole wall. The total loss of the whole wall is minimized by iteratively updating the value of the parameter X(i, j) using methods such as gradient descent. After the iterative update is completed, the brick arrangement diagram of the whole wall can be obtained according to X(i, j); where i in X(i, j) represents the i-th brick from left to right in each layer of bricks, and j represents the j-th brick from bottom to top in each layer of bricks.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The automatic brick arrangement method for a two-structure masonry wall based on rule autonomous learning provided by the present invention automatically completes and optimizes the brick arrangement of the entire wall of the two-structure masonry wall within the arrangement area through area acquisition, vertical stratification, pre-processing zoning, rule autonomous learning and comprehensive arrangement. When the arrangement area changes, the position of the hole changes or the size of the bricks changes, the bricks in the arrangement area can be automatically rearranged, which has the advantage of high arrangement efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of a method for automatically arranging bricks for a two-structure masonry wall based on rule-based autonomous learning;

[0042] Figure 2 It is a schematic diagram of the vertically layered structure;

[0043] Figure 3 It is a schematic diagram of the structure of the preprocessing partition;

[0044] Figure 4 It is a schematic diagram of the structural composition relationship of the automatic brick arrangement system for a two-structure masonry wall based on rule-based autonomous learning;

[0045] As shown in the figure:

[0046] 100. Automatic brick arrangement system for two-structure masonry walls based on rule-based autonomous learning;

[0047] 110. Area acquisition module;

[0048] 120. Vertical layered modules;

[0049] 130. Preprocessing partition module;

[0050] 140. Rules self-learning module;

[0051] 150. Comprehensive layout module. DETAILED DESCRIPTION

[0052] The present invention is described in detail below in conjunction with the accompanying drawings: The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in very simplified form and in non-precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0053] Please refer to Figure 4 The embodiment of the present invention provides a brick automatic arrangement system 100 for a two-structure masonry wall based on rule autonomous learning, comprising:

[0054] The area acquisition module 110 is used to acquire the construction scope of the layout area.

[0055] The vertical layering 120 is used to divide the second structure masonry wall to be constructed within the construction range in the vertical direction. According to the height of the bricks used in the second structure masonry wall and the thickness of the mortar joints, the second structure masonry wall is divided into N layers of bricks from bottom to top for arrangement.

[0056] The pre-processing partitioning module 130 is used to arrange the bricks of the corresponding layer in partitions according to the partitions when the partitions are involved in the construction scope.

[0057] The rule autonomous learning module 140 is used to automatically arrange each layer of bricks in the arrangement area.

[0058] The comprehensive arrangement module 150 autonomously learns and completes the brick arrangement diagram of the entire wall of the second structure masonry wall in the arrangement area according to the rules.

[0059] Please refer to Figures 1 to 3The embodiment of the present invention provides a method for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning, comprising the following steps:

[0060] Step 210, area acquisition, is used to obtain the construction scope of the layout area.

[0061] Step 220, vertical stratification, is used to divide the second structure masonry wall to be constructed within the construction scope in the vertical direction, and divide the second structure masonry wall into N layers of bricks from bottom to top for arrangement according to the height of the bricks used in the second structure masonry wall and the thickness of the mortar joints.

[0062] Step 230, pre-processing partitions, is used to arrange the bricks of the corresponding layer according to the partitions when the partitions are involved in the construction scope. The two sides of the partition are two areas, and the partitions include doors, windows, mechanical and electrical openings, structural columns, etc.

[0063] Step 240, autonomous rule learning, is used to automatically arrange each layer of bricks in the arrangement area.

[0064] Step 250, comprehensive arrangement, autonomously learning and completing the brick arrangement diagram of the entire wall of the second structure masonry wall in the arrangement area according to the rules.

[0065] In order to achieve automatic arrangement, the embodiment of the present invention provides a method and system for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning. In step 240, the method for automatically arranging each layer of bricks in the rule autonomous learning includes:

[0066] Step 241, divide each layer of bricks into multiple regions, and calculate the total length of each layer of regions. In order to improve the accuracy of the arrangement, the influence of structures such as tooth joints on the region length needs to be considered during the calculation.

[0067] Step 242: According to formula (1), an integer is taken to estimate the number of bricks in each arrangement area, and the length of each brick in the arrangement area is recorded as X1, X2, ..., X n , as the independent variable in the autonomous learning process of rules, is optimized, and the bricks are arranged in the arrangement area according to the length of each brick to form a preliminary brick arrangement diagram.

[0068] N=L / (X+L2)+C (1).

[0069] Where N is the number of bricks in each arrangement area, L is the total length of the area, X is the length of the brick, L2 is the length of the mortar joint, and C is a constant. For example, the constant C can be 5.

[0070] Step 243, optimize the preliminary brick arrangement diagram according to the brick arrangement rule loss function (2), and when the total loss in the arrangement area is minimized, determine that the optimized preliminary brick arrangement diagram is the arrangement plan.

[0071] Loss = k1×Loss1 + k2×Loss2 + k3×Loss3 + k4×Loss4 (2).

[0072] Among them, Loss is the total loss within the layout area, Loss1 is the total loss of the vertical and horizontal staggered joints, Loss2 is the total loss of the number of non-standard bricks, Loss3 is the total loss of the total length of the area, Loss4 is the total loss of broken bricks, and k1, k2, k3, and k4 are the importance coefficients of the corresponding loss functions. These importance coefficients are determined before optimization and can be used to modify the constraint degree of each rule through the importance coefficients.

[0073] In step 243, calculate the total loss of the vertical and horizontal staggered joints according to the vertical and horizontal staggered joint loss function (3).

[0074] Loss1 = Loss1_1 + Loss1_2 + …… + Loss1_i + …… + Loss1_n (3).

[0075] Among them, Loss1 is the total loss of the vertical and horizontal staggered joints, Loss1_1 is the staggered joint loss of the first brick, Loss1_2 is the staggered joint loss of the second brick, Loss1_i is the staggered joint loss of the i-th brick, and Loss1_n is the staggered joint loss of the n-th brick;

[0076] Set the minimum value of the staggered joint spacing between the upper and lower adjacent masonry bricks as d, and the length, width, and height of the standard bricks are a, b, and h respectively; when the staggered joint is greater than d, Loss1_i takes 0; when the staggered joint is less than d, Loss1_i takes the absolute value or square of the difference between the staggered joint length and d.

[0077] Among them, the vertical and horizontal staggered joint loss function (3) can be used to measure whether the staggered joint with the lower layer of bricks meets the requirements.

[0078] In step 243, when X i < a, determine that the i-th brick is a non-standard brick, and calculate the total loss of the number of non-standard bricks according to the non-standard brick number loss function (4).

[0079] (4).

[0080] Among them, X i is the length of the i-th brick, Loss2 is the total loss of the number of non-standard bricks, F is a non-standard brick, and when this brick is a non-standard brick F take 1, otherwise F take 0.

[0081] The non-standard brick quantity loss function (4) can be used to measure whether the non-standard bricks are small enough and whether they are conducive to construction.

[0082] In step 243, the total area length loss is calculated according to the area total length loss function (5).

[0083] (5).

[0084] Where Loss3 is the total loss of the total length of the region.

[0085] The total length loss function (5) can be used to measure whether the total length of a layer of bricks is consistent with the actual situation.

[0086] In step 243, when X i When <2d, it is judged as broken bricks, and the number of broken bricks is counted according to the broken brick number loss function (6);

[0087] (6);

[0088] Among them, Loss4 is the total amount of broken brick loss, z is a broken brick, and z is 1 if the brick is a broken brick, otherwise z is 0.

[0089] The broken brick quantity loss function (6) can be used to measure whether there are bricks that are less than 2d and do not meet the specifications.

[0090] Adding the values ​​of the four loss functions together gives the brick arrangement rule loss function (2). Since Loss1, Loss2, Loss3, and Loss4 are the sum of X1, X2, …, X n Determined, so Loss1, Loss2, Loss3, Loss4 are all X1, X2, ..., X n The function of loss = .

[0091] In order to improve the accuracy of the optimized preliminary brick arrangement diagram as an arrangement scheme, the automatic brick arrangement method and system for a two-structure masonry wall based on rule autonomous learning provided by the embodiment of the present invention use a gradient descent method to iteratively update the parameters X1, X2, ..., X n The value of is used to minimize the value of the brick arrangement rule loss function (2) to obtain the final brick arrangement diagram; the specific method may include:

[0092] Compute the gradient: For the current parameter value, calculate the brick layout rule loss function (2) with respect to each parameter X i gradient.

[0093] Update parameters: Use gradient and learning rate to update each parameter.

[0094] Repeated iteration: Repeat the process of calculating gradients and updating parameters until the brick arrangement rule loss function value drops below a certain threshold or cannot be further reduced, thus forming the final brick arrangement diagram.

[0095] Among them, the iterative process is to learn the arrangement rules of the length of the bricks. After the iteration is completed, the autonomous learning process of the arrangement rules is completed, and the obtained arrangement result is the optimal arrangement method.

[0096] In order to realize the comprehensive arrangement of the whole wall, the embodiment of the present invention provides a method and system for automatically arranging bricks of a two-structure masonry wall based on rule autonomous learning. In the comprehensive arrangement, the method for completing the brick arrangement diagram of the whole wall of the two-structure masonry wall according to the rule autonomous learning in the arrangement area includes the following two methods:

[0097] The first method is: for a wall, start arranging the bricks from the bottom layer, and after one layer of bricks is arranged, arrange them layer by layer upwards until the whole wall is arranged. Each layer of bricks undergoes a process of autonomous learning of rules, and the brick arrangement process of the whole wall is completed, that is, a brick arrangement diagram of a two-structure masonry wall that can be used to guide actual use is obtained.

[0098] The second method is:

[0099] The bricks of the whole wall are regarded as a whole, and the lengths X(i, j) of all bricks in each arrangement area of ​​the whole wall are taken as independent variables. The brick rule loss function (2) of all areas of each layer of bricks in the whole wall is superimposed to obtain the total loss of the whole wall. The total loss of the whole wall is a function of the lengths X(i, j) of all bricks; in this way, bricks are no longer arranged layer by layer, but each brick in the wall is iterated at the same time during the optimization process. Among them, i in X(i, j) represents the i-th brick from left to right in each layer of bricks, and j represents the j-th brick from bottom to top in each layer of bricks. Using methods such as gradient descent, the total loss of the whole wall is minimized by iteratively updating the value of the parameter X(i, j); after the iterative update is completed, the brick arrangement diagram of the whole wall can be obtained according to X(i, j).

[0100] The automatic brick arrangement method and system for a two-structure masonry wall based on rule-based autonomous learning provided in an embodiment of the present invention automatically complete and optimize the brick arrangement of the entire wall of the two-structure masonry wall within the arrangement area through area acquisition, vertical stratification, pre-processing partitioning, rule-based autonomous learning and comprehensive arrangement. When the arrangement area changes, the position of the opening changes or the size of the bricks changes, the bricks in the arrangement area can be automatically rearranged, which has the advantage of high arrangement efficiency.

[0101] The automatic brick arrangement method and system for a two-structure masonry wall based on rule-based autonomous learning provided in an embodiment of the present invention utilizes the rule-based autonomous learning method and can transform the arrangement process into a mathematical optimization problem by automatically constructing and optimizing four loss functions. This method can adapt to any masonry size and arrangement interval and has the advantage of strong arrangement adaptability.

[0102] The automatic brick arrangement method and system for a two-structure masonry wall based on rule-based autonomous learning provided in an embodiment of the present invention can arrange the entire wall together, consider the arrangement of bricks from an overall perspective, and when boundary conditions change due to situations such as the displacement of mechanical and electrical pipeline openings, rearrangement can be achieved at a relatively low cost, thereby having the advantage of an integral wall.

[0103] The present invention is not limited to the above-mentioned specific implementation modes. Obviously, the above-mentioned embodiments are only some embodiments of the embodiments of the present invention, but not all embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. In this way, if these modifications and changes of the present invention fall within the scope of the claims of the present invention, the present invention is also intended to include these changes and changes.

Claims

1. A method for automatically arranging bricks for a two-structure masonry wall based on rule-based autonomous learning, characterized in that: include: Step 210, area acquisition, is used to obtain the construction scope of the layout area; Step 220, vertical stratification, is used to divide the second structure masonry wall to be constructed within the construction range in the vertical direction, and divide the second structure masonry wall into N layers of bricks from bottom to top for arrangement according to the height of the bricks used in the second structure masonry wall and the thickness of the mortar joints; Step 230, pre-processing partitions, for arranging the bricks of the corresponding layers according to the partitions when the partitions are involved in the construction scope; Step 240, autonomous rule learning, for automatically arranging each layer of bricks in the arrangement area; the method for automatically arranging each layer of bricks in the autonomous rule learning includes: Step 241, dividing each layer of bricks into multiple regions, and calculating the total length of each layer of regions; Step 242: According to formula (1), an integer is taken to estimate the number of bricks in each arrangement area, and the length of each brick in the arrangement area is recorded as X1, X2, ..., X n , as the independent variable in the autonomous learning process of rules, is optimized, and bricks are arranged in the arrangement area according to the length of each brick to form a preliminary brick arrangement diagram; N = L / (X + L2) + C (1); Among them, N is the number of bricks in each arrangement area, L is the total length of the area, X is the length of the brick, L2 is the length of the mortar joint, and C is a constant; Step 243, optimizing the preliminary brick arrangement diagram according to the brick arrangement rule loss function (2), and when the total loss in the arrangement area is minimized, determining the optimized preliminary brick arrangement diagram as the arrangement plan; Loss= k1×Loss1+ k2×Loss2+ k3×Loss3+ k4×Loss4 (2); Among them, Loss is the total loss in the layout area, Loss1 is the total loss of the upper and lower layer staggered joints, Loss2 is the total loss of the number of non-standard bricks, Loss3 is the total loss of the total length of the area, and Loss4 is the total loss of broken bricks. k1, k2, k3, and k4 are the importance coefficients of the corresponding loss functions; Step 250, comprehensive arrangement, autonomously learning and completing the brick arrangement diagram of the entire wall of the second structure masonry wall in the arrangement area according to the rules.

2. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 1 is characterized in that: Calculate the total amount of upper and lower layer staggered loss according to the upper and lower layer staggered loss function (3); Loss1=Loss1_1+ Loss1_2+……+ Loss1_i+……+ Loss1_n (3); Among them, Loss1 is the total loss of the staggered joints of the upper and lower layers, Loss1_1 is the staggered joint loss of the first brick, Loss1_2 is the staggered joint loss of the second brick, Loss1_i is the staggered joint loss of the i-th brick, and Loss1_n is the staggered joint loss of the n-th brick; The minimum offset spacing between the upper and lower adjacent layers of masonry is set to d, and the length, width and height of the standard bricks are a, b and h respectively; when the offset is greater than d, Loss1_i is 0; when the offset is less than d, Loss1_i is the absolute value or square of the difference between the offset length and d.

3. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 2 is characterized in that: When X i <is less than a, it is determined that the i-th brick is a non-standard brick, and the total loss of the number of non-standard bricks is calculated according to the non-standard brick quantity loss function (4); (4); Among them, X i is the length of the i-th brick, Loss2 is the total loss of non-standard bricks, F For non-standard bricks.

4. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 3 is characterized in that: Calculate the total area length loss according to the area total length loss function (5); （5）； Where Loss3 is the total loss of the total length of the region.

5. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 4 is characterized in that: When X i When <2d, it is judged as broken bricks, and the number of broken bricks is counted according to the broken brick number loss function (6); （6）； Among them, Loss4 is the total amount of broken bricks lost, and z is the broken bricks.

6. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 5 is characterized in that: Using the gradient descent method, we iteratively update the brick layout rule loss function (2) parameters X1, X2, ..., X n The value of is used to minimize the value of the brick arrangement rule loss function (2) and obtain the final brick arrangement diagram.

7. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 6 is characterized in that: The methods for obtaining the final brick layout diagram include: Compute the gradient: For the current parameter value, calculate the brick layout rule loss function (2) with respect to each parameter X i The gradient of Update parameters: Use gradient and learning rate to update each parameter; Repeated iteration: Repeat the process of calculating gradients and updating parameters until the brick arrangement rule loss function value drops below a certain threshold or cannot be further reduced, thus forming the final brick arrangement diagram.

8. The automatic brick arrangement method for a two-structure masonry wall based on rule-based autonomous learning according to claim 7, characterized in that: In the comprehensive arrangement, the method of autonomously learning and completing the brick arrangement diagram of the entire wall of the two-structure masonry wall in the arrangement area according to the rules includes: For a wall, start arranging the bricks from the bottom layer. After one layer of bricks is arranged, arrange them upwards layer by layer until the entire wall is completed.

9. The automatic brick arrangement method for building a two-structure masonry wall based on rule-based autonomous learning according to claim 7, characterized in that: In the comprehensive arrangement, the method of autonomously learning and completing the brick arrangement diagram of the entire wall of the two-structure masonry wall in the arrangement area according to the rules includes: The bricks of the whole wall are regarded as a whole; the lengths X(i, j) of all bricks in each arrangement area of ​​the whole wall are taken as independent variables, and the brick rule loss function (2) of all areas of each layer of bricks in the whole wall is superimposed to obtain the total loss of the whole wall; the gradient descent method is used to iteratively update the value of the parameter X(i, j) to minimize the total loss of the whole wall. After the iterative update is completed, the brick arrangement diagram of the whole wall can be obtained according to X(i, j); wherein i in X(i, j) represents the i-th brick from left to right in each layer of bricks, and j represents the j-th brick from bottom to top in each layer of bricks.

Images