Layout design automation system and layout design method

Abstract

To reduce the workload of a designer.SOLUTION: A layout design automation system generates routing for connection between a start point and an end point. The layout design automation system includes a point cloud data processing unit and a route generation unit. The point cloud data processing unit identifies which range in point cloud data corresponds to a range of each structure, from information indicating a building structure and interior structure composed of the point cloud data, and specifies the type of the structure, and generates three-dimensional CAD information of the specified structure. The route generation unit generates routing based on the three-dimensional CAD information generated by the point cloud data processing unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a layout design automation system and a layout design method. [Background technology]

[0002] There are technologies for automating the design work of piping routes for plants, etc. For example, Patent Document 1 describes a technology for automatically generating a piping route that satisfies numerous constraints for connecting the start and end points of the piping in layout space information. It also describes technologies for generating not only piping routes, but also supports for supporting the piping route, valves for controlling internal fluids, and aggregation of generated structures. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-86310 Summary of the Invention [Problem to be solved by the invention]

[0004] When designing the layout of an existing plant, it is necessary to proceed with the layout design rationally while taking into consideration factors such as ensuring separation from structures within the existing plant and sharing of supports. The system described in Patent Document 1 can generate piping routes based on 3D CAD information of the building. This 3D CAD information includes information composed of point cloud data acquired using a 3D laser scanner.

[0005] When designing the layout, it is necessary to identify the type of building structure (ceiling, floor, wall, etc.) and plan the route in an appropriate location depending on the type of piping being designed. It is also necessary to determine the type of indoor structure (existing piping, existing equipment, etc.). Then, it is necessary to consider arranging the route along the building structure and the distance from indoor structures.

[0006] However, point cloud data is merely a collection of points distributed in three dimensions. Therefore, it is necessary to accurately identify which areas of the point cloud data correspond to the building frame, such as ceilings, floors, and walls, and which areas correspond to indoor structures, such as piping, instrumentation piping, electrical conduits, cable racks, air conditioning ducts, and equipment such as pumps. If the type of building frame or indoor structure is incorrectly identified, it may not be possible to generate an appropriate route. Previously, designers had to perform the difficult task of identifying the building frame and indoor structures from point cloud data, which was a heavy workload.

[0007] The present invention has been made in light of the above-mentioned circumstances, and has as its object to reduce the workload of designers. [Means for solving the problem]

[0008] A layout design automation system according to an embodiment for solving the above problems is a layout design automation system that generates a route connecting a start point and an end point. The layout design automation system includes a point cloud data processing unit and a route generation unit. The point cloud data processing unit identifies the range of each structure in the point cloud data based on information indicating the building skeleton and indoor structures configured using point cloud data, identifies the type of structure, and generates 3D CAD information for the identified structure. The route generation unit generates a route based on the 3D CAD information generated by the point cloud data processing unit. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a configuration diagram of a layout design automation system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating a storage unit according to the embodiment. [Figure 3] 10 is a flowchart for generating three-dimensional CAD information from point cloud data according to an embodiment. [Figure 4] FIG. 2 is a diagram illustrating a point cloud data processing unit according to the embodiment. [Figure 5] FIG. 2 is a diagram illustrating a point cloud data processing unit according to the embodiment. [Figure 6] FIG. 2 is a diagram illustrating a point cloud data processing unit according to the embodiment. [Figure 7] FIG. 2 is a diagram illustrating a point cloud data processing unit according to the embodiment. [Figure 8] FIG. 2 is a diagram illustrating a point cloud data processing unit according to the embodiment. [Figure 9] FIG. 2 is a diagram illustrating a point cloud data processing unit according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] A layout design automation system according to an embodiment will be described below with reference to the drawings. The layout design automation system is a device that automatically generates a route by identifying structures based on point cloud data and connecting the start point and end point of the route. Routes generated by the layout design automation system include routes for piping, instrumentation piping, electrical conduits, cable racks, and air conditioning ducts in plants and buildings, but here we will describe a case in which the system is applied to the layout design of piping that carries fluids (gas, liquid, etc.) in the plant field.

[0011] Here, instrumentation piping refers to piping that houses cables for transmitting information on measurement results such as temperature, pressure, and flow rate, and cables for transmitting signals to control the system, or piping that guides the fluid to be measured (gas, liquid, etc.) to a meter. Conduit refers to piping that houses cables that transmit electricity. In the following description, the building structures such as the ceiling, floor, and walls are referred to as the building frame, and structures other than the building such as piping and equipment are referred to as indoor structures. The building frame and indoor structures are sometimes collectively referred to as the structure.

[0012] Point cloud data is a collection of three-dimensional coordinate points, and can be obtained by photographing buildings, existing equipment, existing piping, etc. from multiple positions using a 3D laser scanner. Point cloud data contains information on the three-dimensional shape of the building's framework, such as the ceiling, floor, and walls, as well as indoor structures such as existing cable racks, air conditioning ducts, support members, and equipment such as tanks and pumps. However, it is unclear which areas of the point cloud data represent the framework and which areas represent which indoor structures. It is also unclear what types of structures they are.

[0013] The layout design automation system 1 is physically a computer equipped with a CPU, memory, etc. The layout design automation system operates based on layout design application software stored in the memory.

[0014] 1 is a functional configuration diagram of a layout design automation system 1. The layout design automation system 1 includes an input unit 10, a storage unit 21, a route generation unit 22, a support position determination unit 23, a piping component determination unit 24, a quantity calculation unit 25, an output unit 30, and a point cloud data processing unit 40.

[0015] The input unit 10 is composed of a touch panel, a keyboard, etc. The input unit 10 displays 3D CAD information of the building on the touch panel screen and acquires coordinate information of the start and end points of routes such as piping as specified by the designer. The input unit 10 also displays a list of the type and weight of fluid to be passed through the piping, the material of the piping, the diameter of the piping, etc. on the touch panel screen and acquires this information as specified by the designer. Furthermore, if there are multiple pipes, the input unit 10 acquires a priority order that determines which pipes should be prioritized for placement as specified by the designer.

[0016] As shown in FIG. 2, the memory unit 21 stores layout space information 21a, layout-specific space information 21b, existing structure information 21c, operation and maintenance information 21d, piping layout standard information 21e, detailed design information 21f, and training data 21g.

[0017] The layout space information 21a is 3D CAD information that indicates the structure of the building's skeleton (ceiling, floor, walls, etc.). The layout-only space information 21b is 3D CAD information that indicates the area in the building where piping and the like are to be placed. The existing structure information 21c is 3D CAD information that indicates indoor structures such as piping, instrumentation piping, electrical conduits, cable racks, air conditioning ducts, and equipment. In this embodiment, the layout space information 21a and the existing structure information 21c are information composed of point cloud data acquired using a 3D laser scanner.

[0018] The operation and maintenance information 21d is information that indicates the conditions for suitable locations for maintenance and inspection of instruments that require visual inspection, the conditions for suitable locations for operating parts that require operation, etc. Instruments that require visual inspection include flow meters, pressure gauges, thermometers, wattmeters, etc. Parts that require operation include valves such as motorized valves and manual valves, etc. The operation and maintenance information 21d includes conditions such as the height of instruments that are easy for maintenance personnel to see, the height and orientation of valves that are easy for maintenance personnel to operate, and the distance between instruments and valves, etc. and walls. The operation and maintenance information 21d also includes information on prohibited placement areas. Prohibited placement areas include areas where piping is prohibited, such as near passageways or storage areas for hazardous materials, and areas necessary for maintenance work on valves, etc.

[0019] The piping layout standard information 21e is information indicating basic design conditions for piping layout design. This piping layout standard information 21e includes conditions that specify the distance between piping and the floor, ceiling, or wall, the distance between piping and parallel piping, the distance between piping and a cable rack, the selection of bent pipes for piping, the diameter of the through-holes if any are provided in the piping, and the location of the through-holes in areas with high radiation doses in the case of a nuclear power plant. Note that a bent pipe is a component obtained by bending a straight pipe. The piping layout standard information 21e also includes safety system separation conditions. For example, the safety system separation conditions are conditions such as a configuration in which, if an accident occurs in the first system piping route, the first system piping route is switched to the second system piping route, and the first system piping route and the second system piping route are arranged separated by a wall or the like. This information is set according to the type of fluid to be passed through the piping, etc.

[0020] The detailed design information 21f is information indicating detailed design conditions. This detailed design information 21f includes conditions such as the length of straight pipes (straight sections of piping) before and after the instrument orifice, construction tolerances, piping gradients, and distances to other equipment. The detailed design information 21f also includes design conditions based on the experience and knowledge of experienced designers. Furthermore, the detailed design information 21f also includes concentrated placement conditions when there are multiple pipes.

[0021] The training data 21g is data used by the AI ​​(Artificial Intelligence) constituting the point cloud data processing unit 40. The training data 21g is data linking a group of point cloud data (hereinafter also referred to as a point cloud group) with attribute information indicating the type of structure. A point cloud group is a group of point cloud data in which point clouds are distributed at a predetermined density. For example, training data for a "ceiling" is formed by linking data obtained by grouping point cloud data constituting the "ceiling" with attribute information called "ceiling." The training data for the "ceiling" also includes relationship information with surrounding point cloud data. For example, a predetermined range within the point cloud data constituting a room may be defined as a point cloud group. Furthermore, the training data for the "ceiling" is not a single piece of data, but is composed of multiple pieces of training data for various types of ceilings. Similarly, training data for indoor structures such as existing cable racks, air conditioning ducts, support members, tanks, pumps, and other equipment is composed of multiple pieces of data linking each group of point cloud data with its respective attribute information.

[0022] Returning to FIG. 1 , the point cloud data processing unit 40 uses information representing the building skeleton and indoor structures, which is composed of point cloud data, to identify which ranges of the point cloud data correspond to each structure, identify the type of structure, and generate 3D CAD information for the identified structure. Specifically, the point cloud data processing unit 40 uses point cloud data representing the entirety of multiple structures, including the building skeleton (e.g., ceiling, floor, and walls) and indoor structures (e.g., existing piping and existing equipment), to identify which ranges of the point cloud data correspond to each structure, such as the ceiling, floor, wall, existing piping, and existing equipment. The point cloud data processing unit 40 then identifies the type of each structure (e.g., ceiling, floor, wall, existing piping, or existing equipment) and generates 3D CAD information for the identified structure.

[0023] Specifically, the point cloud data processing unit 40 incorporates AI, and uses the AI ​​to identify the type of each structure from the point cloud data information containing information on all structures, based on training data 21g that links point cloud data groups with attribute information indicating the type of structure. The point cloud data processing unit 40 then generates 3D CAD information for the identified structures. In this way, the point cloud data processing unit 40 converts the information composed of point cloud data into 3D CAD information.

[0024] The route generation unit 22 generates a route from a coordinate indicating a start point to a coordinate indicating an end point based on the 3D CAD information generated by the point cloud data processing unit 40 and at least one of the layout space information 21a, the layout-dedicated space information 21b, the existing structure information 21c, the operation and maintenance information 21d, the piping layout standard information 21e, and the detailed design information 21f. The routes generated by the route generation unit 22 include routes for piping that carries fluids (gas, liquid, etc.), instrumentation piping, electrical conduits, cable racks, and air conditioning ducts. When generating multiple routes, the route generation unit 22 generates the multiple routes so that the placement positions of the multiple routes do not overlap (do not interfere with) each other. When generating a route that penetrates a floor, ceiling, or wall, the route generation unit 22 generates a 3D shape model indicating the positions, sizes, etc. of through-holes to be formed in the floor, ceiling, or wall, and a summary list of the through-holes.

[0025] The support position determination unit 23 determines support positions at which support members that support pipes, etc. are placed based on the constant pitch span method (also called the standard support spacing method). For example, the support position determination unit 23 determines support positions by the constant pitch span method based on a standard span (the distance between one support position and an adjacent support position) that is set in advance for straight pipe sections and bent pipe sections. The support position determination unit 23 determines support positions at which support members that support pipes, etc. are placed based on the weight of the pipes and the weight of the fluids that pass through the pipes. For example, the support position determination unit 23 determines support positions so as to support areas near parts where loads concentrate, such as the start point, end point, bends, and branches of the route.

[0026] In addition, when multiple pipes or other routes generated by the route generation unit 22 are adjacent to each other or run parallel to each other, the support position determination unit 23 determines a position where two or more pipes among the multiple pipes can be supported at one support position as a support position.

[0027] The piping component determination unit 24 determines the length of each of the multiple components that make up the piping, etc., based on the route of the piping, etc., generated by the route generation unit 22 and preset transportation conditions. For example, the piping component determination unit 24 determines the length of each of the multiple components that make up the piping, so that the length of each component is 10 m or less. Furthermore, the piping component determination unit 24 determines the length of each of the multiple components that make up the piping, so that the weight of each component is 100 kg or less.

[0028] Furthermore, the piping component determination unit 24 determines support members to support the piping based on the weight of the piping, the weight of the fluid passing through the piping, and the locations where the support members supporting the piping will be fixed. The piping component determination unit 24 determines the support members based on the fixing locations (floor, ceiling, wall, etc.), the number of pipes to be supported, the weight of the piping and the fluid passing through the piping, etc.

[0029] The quantity tallying unit 25 creates a quantity table that tallies the number of each component determined by the piping component determining unit 24 .

[0030] The output unit 30 is configured with a display or a printing device. The output unit 30 displays a 3D CAD image showing the route of piping, etc. generated by the route generation unit 22 and the positions of support members determined by the support position determination unit 23. The output unit 30 also outputs a material quantity table compiled by the material quantity compilation unit 25.

[0031] Next, a method for generating 3D CAD information of the building frame and indoor structures from point cloud data will be described with reference to the flowchart shown in FIG.

[0032] First, the point cloud data processing unit 40 of the layout design automation system 1 imports point cloud data from the storage unit 21 (step t01). The point cloud data is acquired by photographing the building skeleton and indoor structures from multiple positions using a 3D laser scanner, and is stored in advance in the storage unit 21. FIG. 4 is an image diagram of the acquired point cloud data. FIG. 4 is an image diagram of the point cloud data viewed from the side (+Y direction). FIG. 5 is an image diagram of the point cloud data viewed from the top (+Z direction). In the figure, the point cloud data is represented by points. The acquired point cloud data is a collection of three-dimensional points, but it is unclear which ranges of the point cloud data represent the skeleton, such as the ceiling, floor, and walls, and which ranges of the point cloud data represent indoor structures, such as piping and equipment. Furthermore, because attribute information of the skeleton and indoor structures is not assigned, it is unclear which ranges of the point cloud data represent what types of structures.

[0033] Next, the point cloud data processing unit 40 identifies a room from a spatial region of a predetermined range surrounded by the point clouds (step t02). Specifically, this processing is performed by the AI ​​constituting the point cloud data processing unit 40. In the example shown in Figures 4 and 5, the spatial region surrounded by point cloud groups T11, T12, T13, T14, T15, and T16, which are groups of point cloud data, has a predetermined size, so the point cloud data processing unit 40 identifies the spatial region T10 as a room.

[0034] Next, the point cloud data processing unit 40 identifies the ceiling, floor, and walls (step t03). Steps t02 and t03 are a structural body identification process. The training data 21g used by the AI ​​constituting the point cloud data processing unit 40 includes training data for ceilings, floors, and walls. For example, the training data for "ceiling" is assigned attribute information of "ceiling," and the point cloud group has a feature that it constitutes the upper surface of the space (the surface on the +Z side of the space) and has a predetermined area. In the example shown in FIGS. 4 and 5, the feature of the training data assigned attribute information "ceiling" is similar to the feature of the point cloud group T11, so the point cloud data processing unit 40 identifies the point cloud group T11, indicated by a thick solid line in FIG. 6, as "ceiling."

[0035] For example, the training data for "floor" is assigned attribute information of "floor," and the point cloud group has a feature that it forms the lower surface of the space (the surface on the -Z side of the space) and has a predetermined area. In the example shown in Figures 4 and 5, the feature of the training data assigned attribute information "floor" is similar to the feature of point cloud group T12, so the point cloud data processing unit 40 identifies point cloud group T12, indicated by a thick solid line in Figure 6, as "floor."

[0036] For example, the training data for "wall" is assigned attribute information of "wall," and the point cloud group has a feature that it is located on the side of the space that makes up the room and has a predetermined area. In the example shown in Figures 4 and 5, the feature of the training data assigned attribute information "wall" is similar to the feature of point cloud groups T13, T14, T15, and T16, so the point cloud data processing unit 40 identifies point cloud groups T13, T14, T15, and T16, which are indicated by thick solid lines in Figures 6 and 7, as "wall."

[0037] Next, the point cloud data processing unit 40 identifies indoor structures located within the area (room) surrounded by the identified ceiling, floor, and walls (step t04). The point cloud data processing unit 40 also identifies indoor structures based on the characteristics of the point cloud distribution connecting to adjacent rooms.

[0038] Next, the point cloud data processing unit 40 identifies the type of indoor structure (step t05). Steps t04 and t05 are an indoor structure identification process. The training data 21g used by the AI ​​constituting the point cloud data processing unit 40 includes training data for indoor structures such as tanks, pumps, pipes, generators, boilers, and cable racks. For example, the training data for a "tank" is assigned attribute information of "tank," and the point cloud group has feature values ​​of a predetermined height and diameter and a cylindrical shape. In the example shown in FIGS. 4 and 5, the feature values ​​of the training data assigned attribute information "tank" are similar to the feature values ​​of the point cloud group T21, so the point cloud data processing unit 40 identifies the point cloud group T21, indicated by the solid line in FIGS. 8 and 9, as a "tank."

[0039] Furthermore, for example, the training data for "pump" is assigned attribute information of "pump," and the point cloud group has a feature of having a predetermined height and width and being rectangular or cylindrical. In the example shown in Figures 4 and 5, the feature of the training data assigned attribute information "tank" is similar to the feature of point cloud group T22, so the point cloud data processing unit 40 identifies point cloud group T22, indicated by a solid line in Figures 8 and 9, as a "pump."

[0040] Furthermore, for example, the training data for "piping" is assigned attribute information of "piping," and the point cloud group has features such as a long, thin cylindrical shape, with piping supports (supporting members) arranged at predetermined intervals, and passing through walls to connect to adjacent rooms. In the example shown in Figures 4 and 5, the feature amount of the training data assigned attribute information "piping" and the feature amount of point cloud group T23 are similar, so the point cloud data processing unit 40 identifies point cloud group T23, shown by a solid line in Figures 8 and 9, as "piping."

[0041] The point cloud data processing unit 40 generates 3D CAD information to which attribute information indicating the type of the identified structure is added, and stores the information in the storage unit 21 (step t06). Step t06 is a 3D CAD information generation step. Specifically, the point cloud data processing unit 40 identifies the type of building frame, such as ceiling, floor, or wall, and the type of indoor structure, such as existing piping or existing equipment, and stores the information in the storage unit 21 as 3D CAD information corresponding to the layout space information 21a and existing structure information 21c composed of point cloud data. Through this process, the information composed of point cloud data is converted into 3D CAD information.

[0042] The point cloud data processing unit 40 creates training data that links the point cloud groups of the building structure identified in step t02 and step t03, and the point cloud groups of the indoor structures identified in step t05 with attribute information indicating the types of the identified structures, and adds the data to the training data 21g in the storage unit 21 (step t07). Step t07 is a training data creation step.

[0043] Specifically, the point cloud data processing unit 40 adds data linking the point cloud group T11 with attribute information called "ceiling" to the teacher data 21g. The point cloud data processing unit 40 adds data linking the point cloud group T12 with attribute information called "floor" to the teacher data 21g. The point cloud data processing unit 40 adds data linking the point cloud groups T13, T14, T15, and T16 with attribute information called "wall" to the teacher data 21g. The point cloud data processing unit 40 adds data linking the space surrounded by the point cloud groups T11, T12, T13, T14, T15, and T16 with attribute information called "room" to the teacher data 21g. The point cloud data processing unit 40 adds data linking the point cloud group T21 with attribute information called "tank" to the teacher data 21g. The point cloud data processing unit 40 adds data linking the point cloud group T22 with the attribute information "pump" to the teacher data 21g. The point cloud data processing unit 40 adds data linking the point cloud group T23 with the attribute information "piping" to the teacher data 21g. The point cloud data processing unit 40 also creates teacher data for the support members.

[0044] Based on this 3D CAD information, the route generation unit 22 generates a piping route along the ceiling T11, floor T12, and walls T13 to T16, avoiding indoor structures such as the tank T21, pump T22, and pipes T23 (route generation process). Note that if the identified indoor structures are temporary or removable structures, the route generation unit 22 can also generate a route that does not avoid the removable indoor structures. Details of the generation of piping routes by the layout design automation system 1 are the same as those in Patent Document 1, and therefore will not be described here.

[0045] As described above, the layout design automation system 1 according to the embodiment includes a point cloud data processing unit 40 that, based on information indicating the building skeleton and indoor structures formed by point cloud data, identifies which ranges of the point cloud data correspond to each structure, identifies the type of structure, and generates 3D CAD information for the identified structure. The point cloud data processing unit 40 uses AI to identify the type of each structure from the point cloud data information containing information on all structures, based on training data that links point cloud data groups with attribute information indicating the type of structure. Therefore, the designer does not need to identify which ranges of the point cloud data correspond to which types of structures and convert the point cloud data into 3D CAD information for the corresponding structures. Therefore, the layout design automation system 1 according to the embodiment reduces the burden on the designer.

[0046] Furthermore, the point cloud data processing unit 40 can accurately identify the type of structure. The route generation unit 22 generates a route based on the 3D CAD information generated by the point cloud data processing unit 40. Therefore, the layout design automation system 1 according to the embodiment can improve the design quality of the layout design automation system using point cloud data.

[0047] Furthermore, as explained in the processing of step t07, the point cloud data processing unit 40 of the layout design automation system 1 according to the embodiment accumulates teacher data 21g that links attribute information indicating the type of structure identified for a group of point cloud data with that group of point cloud data. Therefore, the more the layout design automation system 1 is used, the more teacher data 21g is accumulated, and the design quality achieved by the layout design automation system 1 is further improved.

[0048] Although the above description has been given of the case where piping is arranged inside a building, the routes generated by the route generation unit 22 of the layout design automation system 1 include routes for piping, instrumentation piping, electrical conduits, cable racks, and air conditioning ducts. The layout design automation system 1 can also be applied to generating piping routes outside a building. In this case, 3D CAD information showing the structure on the site is used as the layout space information 21a.

[0049] In the above, a method for generating 3D CAD information of a building frame, indoor structures, etc. from point cloud data has been described using the flowchart shown in FIG. 3. However, this flowchart is merely an example, and the present invention is not limited to this. For example, the processes of steps t02 and t03 may be reversed. Since the training data for the ceiling, floor, and walls also includes information about the surroundings, the point cloud data processing unit 40 can identify each of the ceiling, floor, and walls from the point cloud data of the entire structure. The point cloud data processing unit 40 can identify the space surrounded by the identified ceiling, floor, and walls as a room.

[0050] Although the embodiments of the present invention have been described above, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments may be embodied in various other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the inventions and their equivalents as set forth in the claims. [Explanation of symbols]

[0051] 1. Layout design automation system 10...Input section 21...Storage section 21a…location space information 21b…Placement space information 21c... Existing structure information 21d...Operation and Maintenance Information 21e...Piping arrangement standard information 21f…Detailed design information 22... Route generation section 23...Support position determining section 24...Piping material determination unit 25...Quantity Collection Department 30...Output section 40...Point cloud data processing unit

Claims

1. A layout design automation system that generates a route connecting a start point and an end point, a point cloud data processing unit that identifies which range of the point cloud data corresponds to each structure based on information indicating the building skeleton and indoor structures configured from point cloud data, identifies the type of the structure, and generates 3D CAD information for the identified structure; a route generation unit that generates the route based on the three-dimensional CAD information generated by the point cloud data processing unit.

2. the point cloud data processing unit uses AI to identify the type of each structure from the information of the point cloud data including information of all structures, based on training data linking groups of point cloud data with attribute information indicating the type of structure; The layout design automation system according to claim 1 .

3. the point cloud data processing unit creates training data linking attribute information indicating the type of the identified structure with a group of point cloud data indicating the identified structure, and stores the training data in a storage unit; The layout design automation system according to claim 2 .

4. The route generated by the route generation unit includes at least one of a route for piping, an instrumentation piping, an electrical conduit, a cable rack, and an air conditioning duct. The layout design automation system according to claim 1 .

5. a structural body identification step of identifying a structural body including a ceiling, a floor, and a wall; an indoor structure identification step of identifying the type of indoor structure located in the area surrounded by the identified ceiling, floor, and wall; a three-dimensional CAD information generating step of generating three-dimensional CAD information of the identified building frame and the identified indoor structure; and a route generation step of generating a route for at least one of piping, instrumentation piping, electrical conduit, cable rack, and air conditioning duct based on the generated three-dimensional CAD information.

6. a training data creation step of creating training data linking a group of point cloud data indicating the identified structure with attribute information indicating the type of the identified structure; The layout design method according to claim 5 .

Images