Personalized home design scheme recommendation system based on bim and ai collaboration

Abstract

The application relates to the technical field of smart home design, and discloses a personalized home design scheme recommendation system based on BIM and AI cooperation, which comprises a scheme generation module, a knowledge processing module, a segmented processing module and a scheme output module. The scheme generation module generates a home candidate scheme according to a building information model and user demand; the knowledge processing module determines an installation sequence according to a cabinet main body installation process, constructs an installation dependency knowledge graph, identifies a circular installation relationship, sets a cross-cabinet closure causing the circular installation relationship as a post-installation component, and judges a post-installation blocking state; the segmented processing module divides the blocked cross-cabinet closure into multiple segmented closures according to a cabinet boundary position, and re-determines the installation sequence; and the scheme output module outputs a home recommendation scheme when there is no circular installation relationship and the segmented closures can all enter installation positions. The application improves the engineering implementability of the personalized home scheme and reduces rework and plate modification.

Description

Technical Field

[0001] This invention relates to the field of smart home design technology, and more specifically, to a personalized home design recommendation system based on the collaboration of BIM (Building Information Modeling) and AI (Artificial Intelligence). Background Technology

[0002] With the development of Building Information Modeling (BIM), artificial intelligence, and whole-house customization technologies, home design is gradually shifting from manual drafting to digital, parametric, and intelligent design. Existing systems can generate multiple candidate home furnishing solutions by combining residential space dimensions, functional zoning, furniture parameters, and user needs. They also utilize BIM to express cabinet locations, space occupancy, and layout relationships, thereby improving solution generation efficiency and personalization. For whole-house customization scenarios, the design results need further application in cabinet breakdown, processing, and on-site installation. Therefore, home design solutions must not only meet functional and spatial layout requirements but also ensure that the design results can be implemented on-site according to actual installation conditions.

[0003] In existing technologies, such as the invention patent application CN121834990A – a method and system for generating AI-driven apartment layout customization schemes based on multimodal data – artificial intelligence has been used to generate personalized apartment layout customization schemes. However, existing smart home designs typically determine the final furniture arrangement based primarily on the apartment layout, design rules, and user needs, without adequately considering the actual space occupied by cabinets during the process of moving them from the installation space to their installation positions. In scenarios with corners, ceilings, or compact arrangements of adjacent cabinets, even if the final positions of the cabinets do not directly interfere with each other after the design is completed, different cabinets may still occupy installation space during the process of moving, turning, erecting, or pushing them in. This can lead to problems such as mutual constraints on the installation sequence of the original design scheme on-site, and some furniture components being unable to be installed in the intended manner, resulting in a deviation between the design results and the actual construction conditions.

[0004] Therefore, it is necessary to design a personalized home design recommendation system based on the collaboration of BIM and AI to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a personalized home design scheme recommendation system based on BIM and AI collaboration, which aims to solve the problem that existing personalized home schemes only meet the final layout without fully considering the actual installation process of cabinets, which easily leads to mutual constraints on the installation sequence and obstacles to the subsequent installation of components.

[0006] This invention proposes a personalized home design scheme recommendation system based on BIM and AI collaboration, comprising: The scheme generation module is used to generate home furnishing candidate schemes based on building information models and user needs, using artificial intelligence. These schemes include the main body of the cabinet, cross-cabinet enclosures, and cabinet boundary locations. The knowledge processing module is used to determine the installation sequence based on the installation process of the main body of the cabinet, construct an installation dependency knowledge graph based on the installation sequence, identify cyclic installation relationships, identify cross-cabinet enclosures that cause cyclic installation relationships, set the cross-cabinet enclosures as post-installation components, re-determine the installation sequence of the main body of the cabinet and update the installation dependency knowledge graph; when the updated installation dependency knowledge graph does not contain cyclic installation relationships and the cross-cabinet enclosures cannot enter the installation position after the main body of the cabinet is fixed, it is determined that the cross-cabinet enclosures are in a post-installation obstruction state. The segmentation processing module is used to divide the cross-cabinet enclosure into multiple segmented enclosures according to the cabinet boundary position when the cross-cabinet enclosure is in a state of obstructed post-installation, redetermine the installation sequence of the multiple segmented enclosures and update the installation dependency knowledge graph. The solution output module is used to update the home furnishing candidate solutions and output the recommended home furnishing solutions when the updated installation dependency knowledge graph does not contain a cyclic installation relationship formed by multiple segmented closures, and when multiple segmented closures can all enter the installation position.

[0007] Furthermore, when the knowledge processing module determines the installation sequence, it includes: The knowledge processing module obtains the range of movement of the first cabinet body from the entrance of the installation space to the installation position; When the second cabinet body is in the installation position, if the second cabinet body occupies the movement range, then the first cabinet body is determined to be installed before the second cabinet body; if the second cabinet body does not occupy the movement range, then the installation order between the first cabinet body and the second cabinet body is not established.

[0008] Furthermore, when the knowledge processing module determines the cross-cabinet closure component, it includes: Obtain each installation sequence relationship that forms a cyclic installation relationship and the movement range used when establishing the installation sequence relationship; When the space occupied by the cross-cabinet enclosure intersects with the movement range, and the installation sequence disappears after the space occupied by the cross-cabinet enclosure is excluded, the cross-cabinet enclosure is determined to be the cross-cabinet enclosure that causes the cyclic installation relationship.

[0009] Furthermore, the configuration of the retrofitted component includes: when the cross-cabinet enclosure is not installed, the main body of the cabinet retains its load-bearing function, fixing function, and installation positioning function, and the knowledge processing module sets the cross-cabinet enclosure as a retrofitted component.

[0010] Furthermore, when the knowledge processing module determines the hindloading obstruction state, it includes: The knowledge processing module determines the installation direction based on the connection structure of the cross-cabinet enclosure. After the main body of the cabinet is fixed, if the movement range of the cross-cabinet enclosure component along the installation direction to the installation position interferes with the main body of the cabinet, the wall or the top surface, and the cross-cabinet enclosure component cannot reach the installation position along the installation direction, it is determined that the cross-cabinet enclosure component is in the post-installation obstruction state.

[0011] Furthermore, when the segmentation processing module divides the cross-cabinet enclosure into multiple segmented enclosures, the following steps are taken: the segmentation processing module determines the segmentation position on the cross-cabinet enclosure based on the cabinet boundary position between adjacent cabinet bodies; when the segmentation position passes through the connecting part, slot, or fixing part, the segmentation position is adjusted along the extension direction of the cross-cabinet enclosure until the segmentation position avoids the connecting part, slot, and fixing part.

[0012] Furthermore, the formation of the segmented enclosure includes: the segmentation processing module dividing the cross-cabinet enclosure according to the segmentation position, so that each segmented enclosure retains the connection part to the cabinet body, and multiple segmented enclosures together cover the installation area before the cross-cabinet enclosure is divided.

[0013] Furthermore, when the segmentation processing module determines the installation sequence of multiple segmented enclosure components, it includes: The segmented processing module obtains the movement range of each segmented enclosure as it enters the installation position; When the first segment enclosure occupies the movement range of the second segment enclosure after the first segment enclosure is installed, it is determined that the second segment enclosure is installed before the first segment enclosure. If the first segment enclosure does not occupy the movement range of the second segment enclosure after the first segment enclosure is installed, the installation sequence between the first segment enclosure and the second segment enclosure is not established.

[0014] Furthermore, the processing of the cyclic installation relationship includes: When multiple segmented closures form a cyclic installation relationship, the segmentation processing module identifies the segmented closures that form the cyclic installation relationship, adjusts the segment positions between the segmented closures that form the cyclic installation relationship, and redetermines the installation sequence of the multiple segmented closures; when the cyclic installation relationship disappears, the adjusted segment positions are retained.

[0015] Furthermore, when the solution output module updates the home furnishing candidate solutions, it includes: The solution output module obtains the segment positions and multiple segment closures determined by the segment processing module, replaces the cross-cabinet closures with the multiple segment closures, determines the processing dimensions of the segment closures based on the segment positions, determines the connection parts of the segment closures based on the connection positions between the cross-cabinet closures and the main body of the cabinet, and writes the processing dimensions, connection parts, and installation sequence into the home candidate solution.

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: By combining the cabinet layout results in the building information model with the actual cabinet installation process, it no longer relies solely on whether spatial interference occurs in the final cabinet position as the basis for judging the feasibility of the home furnishing solution. Instead, it further establishes the installation sequence relationship based on the cabinet installation process and constructs an installation dependency knowledge graph to identify cyclic installation relationships caused by cross-cabinet closure components. After adjusting the cross-cabinet closure components to be retrofitted components, it is determined whether the cross-cabinet closure components can still enter the installation position when the main body of the cabinet is fixed, thereby further discovering retrofitting obstruction problems that cannot be solved by simply adjusting the installation sequence. When retrofitting obstruction occurs, the cross-cabinet closure components are divided into multiple segmented closure components according to the cabinet boundary position, and the installation sequence relationship between the segmented closure components is re-determined. This allows the home furnishing candidate solution to eliminate mutual constraints during the installation process by adjusting the local closure structure while maintaining the original main cabinet layout and main functions. Therefore, it can reduce situations where the design is feasible but cannot be installed on-site, requiring rework, modification, or temporary cutting. This improves the consistency between personalized home furnishing design solutions and actual processing and installation conditions, and enhances the engineering feasibility of the recommended solution. Attached Figure Description

[0017] The accompanying drawings are used to illustrate the technical solutions of the present invention and do not constitute a limitation on the scope of protection of the present invention. In the drawings: Figure 1 A system block diagram of a personalized home design scheme recommendation system based on BIM and AI collaboration provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the process of handling obstructed and segmented installation of cross-cabinet enclosure components according to an embodiment of the present invention. Figure 3 This is a schematic diagram of a candidate home furnishing solution provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the segmented processing of the cross-cabinet enclosure provided in an embodiment of the present invention. Detailed Implementation

[0018] The specific embodiments of the present invention will now be described with reference to the accompanying drawings. These embodiments are used to illustrate the technical solutions of the present invention and do not limit the scope of protection of the present invention. Unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0019] In some embodiments of this application, see Figure 1-4 As shown, this application proposes a personalized home design scheme recommendation system based on BIM and AI collaboration, including: a scheme generation module, a knowledge processing module, a segmented processing module, and a scheme output module, wherein: The scheme generation module is used to generate home furnishing candidate schemes based on building information models and user needs, using artificial intelligence. These schemes include the main body of the cabinet, cross-cabinet enclosures, and cabinet boundary locations.

[0020] The knowledge processing module determines the installation sequence based on the installation process of the main cabinet body, constructs an installation dependency knowledge graph based on the installation sequence, and identifies cyclic installation relationships. It identifies cross-cabinet enclosure components that cause cyclic installation relationships, sets these cross-cabinet enclosure components as subsequent installation components, re-determines the installation sequence of the main cabinet body, and updates the installation dependency knowledge graph. If the updated installation dependency knowledge graph does not contain cyclic installation relationships, and the cross-cabinet enclosure component cannot be installed in its position after the main cabinet body is fixed, it is determined that the cross-cabinet enclosure component is in a subsequent installation obstruction state.

[0021] The segmentation processing module is used to divide the cross-cabinet enclosure into multiple segmented enclosures based on the cabinet boundary position when the cross-cabinet enclosure is in a state of obstructed post-installation. It then redetermines the installation sequence of the multiple segmented enclosures and updates the installation dependency knowledge graph.

[0022] The solution output module is used to update the home furnishing candidate solutions and output the recommended home furnishing solutions when the updated installation dependency knowledge graph does not contain a cyclic installation relationship formed by multiple segmented closures, and when multiple segmented closures can all enter the installation position.

[0023] Specifically, the system first acquires the building information model (BIM) of the residential space. The BIM records at least the spatial locations of walls, floors, ceilings, installation space entrances, and fixed obstacles, as well as the effective space available for arranging cabinets. User requirements define the cabinet type, storage function, and size requirements. Artificial intelligence uses a pre-trained home layout generation model. The model input includes the boundaries of the residential space, the locations of fixed obstacles, and user requirements. The model output includes the dimensions of each cabinet body, its installation location, the location of cross-cabinet closures, and the cabinet boundary locations between adjacent cabinet bodies. The output results form candidate home layout schemes.

[0024] All output parameters are recorded using the unified spatial coordinate system of Building Information Modeling (BIM). Cabinet body parameters include at least external dimensions, installation location, and orientation. Cross-cabinet enclosure parameters include at least geometric boundaries, installation location, and connection point with the cabinet body. Cabinet boundary locations record the joint boundary positions of adjacent cabinet bodies. After the AI ​​outputs these structured parameters, they are written into the home furnishing candidate scheme. Subsequent installation sequence, cyclic installation relationships, post-installation obstruction states, and segmented processing all directly utilize the parameters from the home furnishing candidate scheme.

[0025] See Figure 3 As shown, the candidate home furnishing solutions include a first cabinet body and a second cabinet body arranged adjacent to each other along the same wall. A cabinet boundary is formed between the first and second cabinet bodies. A cross-cabinet closure is installed across the tops of both cabinet bodies along the same extension direction. The installation space entrance indicates the position the cabinet body passes through when entering the installation space; the arrow in the diagram indicates the direction of movement of the cabinet body from the installation space entrance to the installation position. The system subsequently determines the movement range based on the installation process of the first and second cabinet bodies and determines the installation sequence based on the space occupied by the cabinet body when it is in the installation position. Figure 3 This is only used to illustrate the spatial relationships and installation directions between objects in the home furnishing candidate scheme, and is not used to limit the specific shape, quantity, or arrangement angle of the main body of the cabinet.

[0026] The home layout generation model can use existing home design schemes as training samples. The data type of the training sample input is the same as that in the actual generation stage, and the training target is the cabinet layout parameters confirmed by the design. This embodiment does not focus on improving the structure or training method of the artificial intelligence model itself; the artificial intelligence is only used to generate candidate home layout schemes required for subsequent installation relationship analysis.

[0027] Specifically, the system determines the movement range based on the actual translation, turning, erection, and pushing movements the cabinet body undergoes as it moves from the installation space entrance to the installation position. For cabinet bodies that only translate, the outer contour of the cabinet body is moved step by step along the direction of movement, and the space occupied by each position is combined to form the movement range. For cabinet bodies that involve turning or erection, the outer contour of the cabinet body is rotated step by step according to the pivot point actually used during installation, and the space occupied during rotation is included in the movement range.

[0028] The movement range is determined using a discrete position merging method. When the cabinet body is translated, the translation step size is no greater than twice the installation judgment allowance. When the cabinet body is rotated, the rotation step size is limited to the displacement of the cabinet body's farthest position from the rotation axis in a single rotation not exceeding the current translation step size. When the minimum distance between the cabinet body and the wall or other fixed entity is less than six times the installation judgment allowance, a fine calculation is performed, and the translation step size is reduced to no more than half of the installation judgment allowance, with the rotation step size decreasing simultaneously. Taking an installation judgment allowance of 5mm as an example, the conventional translation step size can be 10mm, and the fine translation step size can be 2mm. For a cabinet body with a height of 2400mm and a depth of 600mm, when using the position farthest from the rotation axis as the rotation step size check position, the conventional rotation step size can be 0.2°, and the fine rotation step size can be 0.04°, ensuring that the displacement of the check position in a single rotation does not exceed the corresponding translation step size. The above adaptation method allows the accuracy of the movement range calculation to vary with the installation judgment margin and cabinet size.

[0029] Using six times the installation judgment margin as the switching range for fine calculation, when the regular translation step size is twice the installation judgment margin, a transition distance of no less than two regular translation calculations is reserved before entering the installation judgment margin range, thereby switching to fine calculation in advance.

[0030] To avoid inconsistencies in boundary judgment caused by on-site measurement errors, panel processing errors, and manual positioning errors, this embodiment sets a uniform installation judgment margin. The installation judgment margin is determined based on the allowable errors in residential space measurement, cabinet processing, and on-site installation positioning. The allowable error in residential space measurement is the maximum permissible absolute error specified in the measurement equipment calibration document or on-site measurement procedure; the allowable error in cabinet processing is the maximum permissible dimensional error specified in the panel processing procedure document; and the allowable error in on-site installation positioning is the maximum permissible deviation specified in the installation procedure document or acceptance standard. The installation judgment margin is the sum of these three types of permissible errors. When the measurement equipment, processing technology, or installation process changes, the installation judgment margin is re-determined based on the new permissible errors.

[0031] For example, if the allowable error for residential space measurement is 2mm, the allowable error for cabinet processing is 1mm, and the allowable error for on-site installation positioning is 2mm, then the installation judgment allowance is 5mm. When two spatial entities overlap, or the minimum distance between two spatial entities is less than the installation judgment allowance, it is treated as space occupancy. The same installation judgment allowance is used for subsequent cabinet main body installation relationships, obstructed installation of cross-cabinet closures, and segmented closure installation relationships, thus maintaining consistency in judgment criteria at each stage. When different measuring equipment or processing techniques change, only the installation judgment allowance needs to be re-obtained based on the actual allowable error; the subsequent judgment process remains unchanged.

[0032] The system analyzes the installation relationship between any two cabinet bodies. Taking the first and second cabinet bodies as examples, it obtains the movement range of the first cabinet body from the installation space entrance to the installation position. After placing the second cabinet body in the installation position, if the second cabinet body occupies the movement range of the first cabinet body, then the first cabinet body must be installed before the second cabinet body, thus establishing an installation sequence relationship where the first cabinet body is installed before the second cabinet body. If the second cabinet body does not occupy the movement range of the first cabinet body, then no installation sequence relationship is established between the two cabinet bodies based on this judgment. After performing the above judgment on each cabinet body in the candidate home furnishing scheme, an installation dependency knowledge graph is established, using the cabinet body as a knowledge node and the installation sequence relationship as the directed relationship between knowledge nodes. The installation dependency knowledge graph also records the connection relationship between the cross-cabinet closure and the cabinet body, as well as the space occupied by the cross-cabinet closure, which is used to determine the reasons for the formation of cyclic installation relationships later.

[0033] In the installation dependency knowledge graph, entity nodes are set for the main cabinet body and cross-cabinet closures. Directed connections are established between the main cabinet bodies based on their installation sequence. The connection relationship and subsequent component identifiers are recorded between the cross-cabinet closures and the main cabinet body. After the home furnishing candidate scheme is generated, cross-cabinet closures not set as subsequent components are included in the space occupied by the main cabinet body connected to them after installation, according to the assembly relationship recorded in the home furnishing candidate scheme. Once a cross-cabinet closure is set as a subsequent component, its space is no longer included when the installation sequence of the main cabinet body is re-determined.

[0034] When a cross-cabinet enclosure is designated as a post-installation component, its physical nodes and connection relationships are retained. However, the space occupied by the cross-cabinet enclosure is not considered when recalculating the installation sequence of the main cabinet body, and the recalculated installation sequence replaces the original one. If a cross-cabinet enclosure is divided into multiple segmented enclosures, the physical nodes of the original cross-cabinet enclosure are replaced with the physical nodes of the multiple segmented enclosures. The segment positions and connection points are recorded, and a new installation sequence is established based on the movement range of the multiple segmented enclosures. The identification of cyclic installation relationships is only performed along the installation sequence.

[0035] Cyclic installation relationships are identified based on the order of installation. For example, if the first cabinet body needs to be installed before the second cabinet body, and the second cabinet body needs to be installed before the first cabinet body, then a two-node cyclic installation relationship is formed. If the first cabinet body needs to be installed before the second cabinet body, the second cabinet body needs to be installed before the third cabinet body, and the third cabinet body needs to be installed before the first cabinet body, then a multi-node cyclic installation relationship is formed. In practice, each cabinet body is used as the starting node, and the search proceeds level by level along the installation order. If any search path returns to the starting node of the current search, it is determined that the search path forms a cyclic installation relationship. If none of the search paths return to their respective starting nodes, it is determined that the current installation dependency knowledge graph does not contain a cyclic installation relationship.

[0036] In this embodiment, the candidate home furnishing solutions no longer rely solely on static spatial judgment based on the final installation location of the cabinets. Instead, the actual movement of the cabinets from the entrance to the installation space is transformed into an installation sequence relationship, and the same installation judgment margin is used at different stages. Therefore, it is possible to identify home furnishing solutions during the design phase that do not have collisions in their final locations but whose actual installation sequences are mutually restrictive, reducing situations where the static building information model determines that the solution is feasible but cannot be implemented during on-site construction. At the same time, the installation relationships have a clear spatial formation basis and do not require reliance on manual experience to determine the installation sequence.

[0037] This embodiment, based on the installation dependency knowledge graph established in the above embodiments, further processes the reasons for the formation of cyclic installation relationships. For each installation sequence in the cyclic installation relationship, the system reads the movement range of the cabinet body used when establishing the installation sequence relationship and obtains the space occupied by the cross-cabinet enclosure connected to the relevant cabinet body. The cross-cabinet enclosure referred to in this embodiment is a plate that spans two adjacent cabinet bodies and is used to close the installation gap between adjacent cabinet bodies or between the cabinet body and the building interface; specifically, it can be a top sealing component, a decorative enclosure, or a finishing component.

[0038] For any cross-cabinet enclosure, first determine whether the space occupied by the enclosure intersects with the movement range used when establishing the installation sequence. If there is no intersection, the enclosure is not the cause of the current installation sequence. If there is an intersection, the space occupied by the enclosure is temporarily ignored in the installation process analysis of the building information model, and the installation sequence between the cabinet bodies is re-determined according to the same judgment rules as in the above embodiment. If the original installation sequence still holds, it means that the cabinet body itself or other components are still causing spatial obstruction, and the enclosure is not considered a component causing the current cyclic installation relationship. When the original installation sequence disappears, the enclosure is determined to be the enclosure causing the current cyclic installation relationship. By using two consecutive conditions—spatial intersection and the disappearance of the relationship after removal—non-critical components can be avoided from being mistakenly identified as processing objects based on only one spatial intersection.

[0039] After identifying the cross-cabinet enclosure causing the cyclical installation relationship, the system further determines whether the cross-cabinet enclosure is suitable as a retrofit component. This determination involves reading the connection structure between the cross-cabinet enclosure and the main cabinet body, as well as the stress and positioning relationships of the main cabinet body. If, without the cross-cabinet enclosure, the main cabinet body can still bear its own load through the base plate, side panels, back panel, or frame, can still be fixed to the wall or adjacent cabinets through existing fixing structures, and still retains the positioning structure used to determine the installation location, then the main cabinet body is considered to retain its load-bearing, fixing, and installation positioning functions, and the cross-cabinet enclosure is designated as a retrofit component. If any of these functions cannot be maintained, the current cross-cabinet enclosure will not enter the retrofit processing flow, and the home furnishing candidate solution will not adopt the cyclical release result based on the current cross-cabinet enclosure. The determination of load-bearing, fixing, and installation positioning functions can be directly obtained from the load-bearing connections, fixing component positions, and positioning surfaces in the cabinet structural model, without requiring subjective scoring.

[0040] After the cross-cabinet enclosure is set as a post-installation component, the installation sequence of the main cabinet body is re-determined and the installation dependency knowledge graph is updated. If the updated installation dependency knowledge graph still contains the original cyclic installation relationship, it means that the current post-installation process cannot remove the cyclic installation relationship. If the cyclic installation relationship disappears, the main cabinet body is fixed according to the new installation sequence. After the main cabinet body is fixed, it is determined whether the post-installation component can enter the installation position. The installation direction is determined according to the connection structure of the cross-cabinet enclosure. For example, for a plug-in structure, the installation direction is determined according to the plug-in axis; for a sliding groove structure, the installation direction is determined according to the sliding groove extension direction; and for a front-pressed structure, the installation direction is determined according to the normal direction of the connection surface. When the cross-cabinet enclosure enters the installation position along the installation direction, the movement range is checked according to the same installation judgment margin as in the above embodiment. If the movement range interferes with the fixed main cabinet body, wall, or top surface, and the cross-cabinet enclosure cannot reach the installation position along the installation direction, it is determined that the cross-cabinet enclosure is in a post-installation obstruction state.

[0041] The initial installation position of the cross-cabinet enclosure is obtained by moving the enclosure in the opposite direction of the installation direction from its original installation position until there is no space occupied between the enclosure and the fixed cabinet body, walls, and top surface, and the minimum distance is not less than the installation judgment allowance. Subsequently, the enclosure is moved gradually from its initial installation position along the installation direction, and each position is checked using the same translation discretization rule as the cabinet body's movement range. If any position is considered to be occupied according to the installation judgment allowance, and the enclosure cannot continue to reach the installation position along the installation direction allowed by the connecting structure, the enclosure is determined to be unable to enter the installation position. If, after moving in the opposite direction of the installation direction to the installation space boundary recorded in the building information model, a starting position satisfying the above spacing conditions is still not obtained, it is determined that there is no executable path for the enclosure to enter, and the enclosure is identified as being in an obstructed installation state.

[0042] When a post-installation obstruction occurs, the segmentation module does not change the position of the main cabinet body. Instead, it determines the initial segmentation position on the cross-cabinet enclosure based on the existing cabinet boundary positions between adjacent main cabinet bodies. To avoid damaging the original connection structure during segmentation, the system reads the connection parts, slots, and fixing parts in the cross-cabinet enclosure and establishes a prohibited segmentation zone. The prohibited segmentation zone can be defined by adding a machining allowance to the actual boundaries of the connection parts, slots, and fixing parts. The machining allowance is read from the furniture processing technology library, including the minimum edge distance of the connector, the minimum edge distance of the slot, and the width required for the edge banding process, and the largest value is taken as the machining allowance for this operation. For example, if the minimum edge distance of the connector is 8mm, the minimum edge distance of the slot is 10mm, and the edge banding width is 6mm, the machining allowance is 10mm. When the initial segmentation position enters the prohibited segmentation zone, the segmentation position moves along the extension direction of the cross-cabinet enclosure until the segmentation position exits the prohibited segmentation zone.

[0043] Search for available segmentation locations along both extension directions of the cross-cabinet enclosure. If available segmentation locations exist in both directions, select the location with the smaller distance from the initial segmentation location. If the distances are the same, select the location that minimizes the difference in machining dimensions between the two segmented enclosures. The minimum allowable values ​​for connector edge distance, hole / slot edge distance, and edge sealing width in the machining clearance allowance are read from the minimum allowable values ​​specified in the connector installation process, drilling / slotting process, and edge sealing process, respectively. If no available segmentation locations satisfying the clearance requirements for prohibited segmentation areas exist along either extension direction of the cross-cabinet enclosure, the current segmentation result is not adopted, and segmentation processing based on the current cross-cabinet enclosure is stopped.

[0044] See Figure 4 As shown, before segmentation, the cross-cabinet enclosure spans between adjacent cabinet bodies. If there are connecting parts or slots near the cabinet boundary, directly dividing according to the cabinet boundary might damage the original connection structure of the cross-cabinet enclosure. Therefore, the segmentation module uses the cabinet boundary as the initial reference for the segmentation position. When the segmentation position passes through connecting parts, slots, or fixing parts, the segmentation position is adjusted along the extension direction of the cross-cabinet enclosure. After the adjusted segmentation position avoids connecting parts, slots, and fixing parts, the cross-cabinet enclosure is divided into a first segment enclosure and a second segment enclosure. The first and second segment enclosures retain their connecting parts to the cabinet body, and their installation directions are determined according to their respective connection structures, so that subsequent determinations can be made regarding whether the first and second segment enclosures can be installed in their respective positions.

[0045] After dividing the cross-cabinet enclosures according to the final segment positions, each segment enclosure retains at least one connection point to the main cabinet body. Multiple segment enclosures, once installed, collectively cover the installation area before the cross-cabinet enclosure division. Then, the installation direction is determined based on the connection structure of each segment enclosure, and the movement range to the installation position is recalculated. When the first segment enclosure occupies the movement range of the second segment enclosure after installation, an installation sequence relationship is established, indicating that the second segment enclosure was installed before the first. The same rules are used to establish the installation sequence relationship between all segment enclosures, and the installation dependency knowledge graph is updated. If multiple segment enclosures still form a cyclic installation relationship, the segment positions of the segment enclosures participating in the cyclic installation relationship are adjusted, and the movement range judgment is re-executed. If multiple segment enclosures do not form a cyclic installation relationship, but some segment enclosures cannot enter the installation position, the current segment position is still not considered a valid segment position. A new segment position is selected according to the aforementioned segment position selection rules, and the movement range and installation sequence relationship of multiple segment enclosures are re-determined. When the cyclic installation relationship disappears and each segment closure can enter the installation position, the current segment result is confirmed to be valid.

[0046] This embodiment does not directly change the entire furniture layout after discovering cabinet installation conflicts. Instead, it first proves that the cross-cabinet closure is indeed the cause of the cyclical installation relationship, and then verifies that the cross-cabinet closure can be installed retrofitted from the main structure. When retrofitting creates new installation obstacles, only the manufacturing composition of the cross-cabinet closure is changed. This allows for the continuous removal of installation constraints between cabinet bodies and retrofitting constraints of the cross-cabinet closure itself without altering the cabinet body's installation position, functional layout, or main structure. This avoids regenerating the entire furniture solution to solve local installation problems and reduces dimensional inconsistencies and connection damage caused by on-site cutting of cross-cabinet closures.

[0047] This embodiment uses a floor-to-ceiling cabinet adjacent to another unit in a bedroom as an example. The net dimensions of the residential space are 3600mm × 3200mm, with a net height of 2500mm and a net width of 850mm at the entrance to the installation space. The Building Information Model (BIM) sets up a first cabinet body and a second cabinet body. The first cabinet body is 900mm wide, 600mm deep, and 2400mm high, while the second cabinet body is 880mm wide, 600mm deep, and 2400mm high. The two cabinet bodies are arranged adjacent to each other along the same wall after installation. The candidate home furnishing solution also includes a cabinet-crossing closure that spans the two cabinet bodies. The cabinet-crossing closure has a length of 1780mm, a width of 100mm, and a thickness of 18mm. The cabinet boundary between adjacent cabinet bodies is located 900mm from one end along the length of the cabinet-crossing closure. The allowable error for on-site laser measurement is 2mm, the allowable error for panel processing is 1mm, and the allowable error for installation positioning is 2mm. Therefore, this embodiment uses 5mm as the installation margin.

[0048] The artificial intelligence system first generates the aforementioned candidate home furnishing solutions based on the boundaries of the residential space and the user's proposed floor-to-ceiling storage needs. The system then calculates the movement range of the two main cabinet units according to the predetermined installation process. The first cabinet unit needs to enter the bedroom from the installation space entrance and turn around; the second cabinet unit, once installed, occupies the movement range of the first cabinet unit during its turning process. Therefore, an installation sequence is established where the first cabinet unit is installed before the second. Conversely, once the first cabinet unit is fixed, it occupies the movement range of the second cabinet unit during its movement towards its installation position. Therefore, an installation sequence is established where the second cabinet unit is installed before the first. The installation dependency knowledge graph thus forms a cyclical installation relationship between the first and second cabinet units.

[0049] The system further analyzes the reasons for the formation of the two installation sequence relationships. In the installation sequence where the first cabinet body is installed before the second cabinet body, the space occupied by the cross-cabinet enclosure intersects with the turning and moving range of the first cabinet body. After temporarily ignoring the space occupied by the cross-cabinet enclosure in the installation process calculations, the aforementioned installation sequence relationship between the first and second cabinet bodies no longer holds. Therefore, it is determined that the cross-cabinet enclosure participates in forming a cyclical installation relationship. The cross-cabinet enclosure only serves as the upper enclosure and decorative connection between the two cabinet bodies; it does not participate in the load-bearing capacity of the cabinet body, does not act as a fixing component between the cabinet body and the wall, and does not undertake the installation positioning function of the two cabinet bodies. Therefore, the cross-cabinet enclosure is set as a later-installed component.

[0050] Without the cross-cabinet enclosure, the installation sequence is recalculated. The turning range of the first cabinet body is no longer obstructed by the cross-cabinet enclosure on one side of the second cabinet body. The original cyclical installation relationship disappears, and the two cabinet bodies can be fixed sequentially according to the newly determined installation order. After the two cabinet bodies are fixed, the cross-cabinet enclosure needs to enter the installation position along the horizontal direction defined by the connecting structure. Since the cross-cabinet enclosure is 1780mm long, the minimum remaining space between the cross-cabinet enclosure and the wall and the fixed cabinet body during the entry process is less than 5mm. According to the aforementioned installation margin judgment rule, this is considered as spatial interference, and the cross-cabinet enclosure cannot continue to reach the installation position in the horizontal direction. Therefore, it is determined that the cross-cabinet enclosure is in a state of obstructed installation.

[0051] The system then uses the cabinet boundary position, 900mm from one end of the cross-cabinet enclosure, between the two main cabinet bodies as the initial segmentation position. The Building Information Model (BIM) shows that the connecting slot is located on the side of the cabinet boundary facing the second main cabinet body, and the distance between the edge of the connecting slot near the cabinet boundary and the initial segmentation position is 6mm. The machining process library records a minimum clearance distance of 10mm for the connecting slot; therefore, the initial segmentation position is located in the prohibited segmentation area. The segmentation module moves the segmentation position 4mm along one side of the first main cabinet body, making the distance between the adjusted segmentation position and the edge of the connecting slot 10mm. The adjusted segmentation position is located 896mm from one end of the cross-cabinet enclosure. The designed net lengths of the first and second segment enclosures are 896mm and 884mm, respectively. The actual cutting dimensions are determined based on the above designed net lengths, according to the existing saw cut and edge banding compensation processes.

[0052] The saw kerf compensation amount is read from the cutting equipment process documents, the edge banding compensation amount is read from the edge banding material and edge banding process documents, and the scheme output module generates the processing dimensions of the segmented closure part based on the design net length and the read process compensation amount.

[0053] The movement range is redefined based on the actual geometric dimensions of the two segmented enclosure components. The first segmented enclosure component can enter the installation position from the front of the first cabinet body, and the second segmented enclosure component can enter the installation position from the front of the second cabinet body. The movement range of both segmented enclosure components maintains an effective installation allowance of no less than 5mm with the fixed cabinet body, wall, and top surface. Furthermore, the installation of the first segmented enclosure component does not encroach on the movement range of the second segmented enclosure component, and vice versa; therefore, the two segmented enclosure components do not form a cyclical installation relationship.

[0054] After confirming the validity of the current segmentation results, the solution output module replaces the original cross-cabinet closure with the first and second segment closures. Based on the adjusted segmentation positions, it re-obtains the processing dimensions of the two segment closures and uses the existing connection points between the original cross-cabinet closure and the two main cabinet bodies to determine the connection points of the two segment closures. The processing dimensions, connection points, and installation sequence of the two segment closures are then written into the home furnishing candidate solution. The updated home furnishing candidate solution maintains the original installation positions and storage layout of the two main cabinet bodies, but the cross-cabinet closure has been converted into two separate, customizable panels, forming the recommended home furnishing solution.

[0055] In this embodiment, within a complete dimensional scenario, the system can distinguish between the mutual constraints of the cabinet main body installation and the additional constraints caused by cross-cabinet enclosure components. Furthermore, it discovers that post-installation measures originally intended to resolve the initial installation cycle still create new spatial constraints due to the fixed cabinets. By adjusting the actual panel composition using existing cabinet boundary positions, segmented results suitable for factory cutting and on-site installation can be obtained without modifying the cabinet main body dimensions and installation positions. This embodiment further demonstrates the ability to directly convert installation problems identified during the design phase into panel processing parameters, thereby reducing the need for on-site trial assembly followed by factory rework or temporary manual cutting, while maintaining the original cabinet main body design unchanged.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A personalized home design scheme recommendation system based on BIM and AI collaboration, characterized in that, include: The scheme generation module is used to generate home furnishing candidate schemes based on building information models and user needs, using artificial intelligence. These schemes include the main body of the cabinet, cross-cabinet enclosures, and cabinet boundary locations. The knowledge processing module is used to determine the installation sequence based on the installation process of the main body of the cabinet, construct an installation dependency knowledge graph based on the installation sequence, and identify cyclic installation relationships. Identify the cross-cabinet enclosure that causes the cyclic installation relationship, set the cross-cabinet enclosure as a post-installation component, redetermine the installation sequence of the cabinet body and update the installation dependency knowledge graph; if the updated installation dependency knowledge graph does not contain the cyclic installation relationship and the cross-cabinet enclosure cannot enter the installation position after the cabinet body is fixed, determine that the cross-cabinet enclosure is in a post-installation obstruction state. The segmentation processing module is used to divide the cross-cabinet enclosure into multiple segmented enclosures according to the cabinet boundary position when the cross-cabinet enclosure is in a state of obstructed post-installation, redetermine the installation sequence of the multiple segmented enclosures and update the installation dependency knowledge graph. The solution output module is used to update the home furnishing candidate solutions and output the recommended home furnishing solutions when the updated installation dependency knowledge graph does not contain a cyclic installation relationship formed by multiple segmented closures, and when multiple segmented closures can all enter the installation position.

2. The system according to claim 1, characterized in that, When the knowledge processing module determines the installation order, it includes: The knowledge processing module obtains the range of movement of the first cabinet body from the entrance of the installation space to the installation position; When the second cabinet body is in the installation position, if the second cabinet body occupies the movement range, then the first cabinet body is determined to be installed before the second cabinet body; if the second cabinet body does not occupy the movement range, then no installation order relationship is established between the first cabinet body and the second cabinet body.

3. The system according to claim 2, characterized in that, When the knowledge processing module determines the cross-cabinet enclosure, it includes: Obtain each installation sequence relationship that forms a cyclic installation relationship and the movement range used when establishing the installation sequence relationship; When the space occupied by the cross-cabinet enclosure intersects with the movement range, and the installation sequence disappears after the space occupied by the cross-cabinet enclosure is excluded, the cross-cabinet enclosure is determined to be the cross-cabinet enclosure that causes the cyclic installation relationship.

4. The system according to claim 1, characterized in that, The configuration of the retrofitted component includes: when the cross-cabinet enclosure is not installed, the main body of the cabinet maintains its load-bearing function, fixing function and installation positioning function, and the knowledge processing module sets the cross-cabinet enclosure as a retrofitted component.

5. The system according to claim 1, characterized in that, When the knowledge processing module determines the post-installation obstruction state, it includes: The knowledge processing module determines the installation direction based on the connection structure of the cross-cabinet enclosure. After the main body of the cabinet is fixed, if the movement range of the cross-cabinet enclosure component along the installation direction to the installation position interferes with the main body of the cabinet, the wall or the top surface, and the cross-cabinet enclosure component cannot reach the installation position along the installation direction, it is determined that the cross-cabinet enclosure component is in the post-installation obstruction state.

6. The system according to claim 1, characterized in that, When the segmentation processing module divides the cross-cabinet enclosure into multiple segmented enclosures, the following steps are taken: the segmentation processing module determines the segmentation position on the cross-cabinet enclosure based on the cabinet boundary position between adjacent cabinet bodies; when the segmentation position passes through the connecting part, slot, or fixing part, the segmentation position is adjusted along the extension direction of the cross-cabinet enclosure until the segmentation position avoids the connecting part, slot, and fixing part.

7. The system according to claim 6, characterized in that, The formation of the segmented enclosure includes: the segmentation processing module dividing the cross-cabinet enclosure according to the segmentation position, so that each segmented enclosure retains the connection part to the cabinet body, and multiple segmented enclosures together cover the installation area before the cross-cabinet enclosure is divided.

8. The system according to claim 7, characterized in that, When the segmented processing module determines the installation sequence of multiple segmented enclosure components, it includes: The segmented processing module obtains the movement range of each segmented enclosure as it enters the installation position; When the first segment enclosure occupies the movement range of the second segment enclosure after the first segment enclosure is installed, it is determined that the second segment enclosure is installed before the first segment enclosure. If the first segment enclosure does not occupy the movement range of the second segment enclosure after the first segment enclosure is installed, the installation sequence between the first segment enclosure and the second segment enclosure is not established.

9. The system according to claim 8, characterized in that, The processing of the cyclic installation relationship includes: When multiple segmented closures form a cyclic installation relationship, the segmentation processing module identifies the segmented closures that form the cyclic installation relationship, adjusts the segment positions between the segmented closures that form the cyclic installation relationship, and redetermines the installation sequence of the multiple segmented closures; when the cyclic installation relationship disappears, the adjusted segment positions are retained.

10. The system according to claim 1, characterized in that, When the solution output module updates the home furnishing candidate solutions, it includes: The solution output module obtains the segment positions and multiple segment closures determined by the segment processing module, replaces the cross-cabinet closures with the multiple segment closures, determines the processing dimensions of the segment closures based on the segment positions, determines the connection parts of the segment closures based on the connection positions between the cross-cabinet closures and the main body of the cabinet, and writes the processing dimensions, connection parts, and installation sequence into the home candidate solution.

Images