An indoor decoration map dual-mode editing and consistent generation method, device and medium

Abstract

The application belongs to the technical field of computer-aided design, and discloses an indoor decoration drawing double-mode editing and consistent generation method, equipment and medium, which comprises the following steps: an intelligent agent is generated through a space strategy consensus draft to structure user fuzzy demand into a consensus draft, and an analysis drawing is generated and mapped for user verification, modification opinions are reversely returned to correct the text, and a stable final space strategy consensus draft is output in a closed loop; a total control routing intelligent agent flexibly routes between a local patch mode and a global replacement mode according to a current state and an editing request, outputs a candidate drawing, and performs strict syntax and availability double checking to obtain a usable two-dimensional plan; a decoration main scheme is extracted as a single fact source constraint, and a multi-intelligent agent team is called to generate multiple types of decoration drawings and a structured material list. The application solves the rework problem caused by demand understanding deviation, realizes flexible consideration of local fine tuning and global reconstruction, and ensures absolute consistency of the same source of the whole drawing data.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and computer-aided design technology, and in particular to a method, device and medium for dual-mode editing and consistent generation of interior decoration drawings based on local patching and global replacement. Background Technology

[0002] With the development of artificial intelligence, large-scale modeling technology, and digital design tools, automatically generating interior design schemes, 2D floor plans, construction drawings, and bills of materials using natural language descriptions has gradually become an important research direction in the field of intelligent design. In the interior design process, it is typically necessary to complete space planning based on user needs and further generate 2D floor plans and various types of construction drawings, such as floor layout plans, floor paving plans, ceiling plans, wall positioning plans, MEP (mechanical and electrical) integrated plans, elevation drawings, and detailed node drawings. Simultaneously, a corresponding bill of materials needs to be generated based on the design scheme to support budget assessment, material procurement, and construction implementation. In existing technologies, some methods generate design drawings through image-based model generation, while others generate 2D graphics through structured graphic codes (such as SVG and XML) to automate the interior design process.

[0003] However, despite the progress made in related technologies, the following key technical issues still exist in practical applications, leading to low system generation quality and difficulties in engineering implementation:

[0004] (1) Lack of a stable scheme alignment mechanism. Existing methods usually generate the final design drawing directly based on the user's original requirements. However, user requirements are often vague, colloquial, and incomplete, making it difficult for the system to accurately understand the user's intent. This results in a huge deviation between the directly generated drawing and the user's expectations, requiring frequent rework and modification, which seriously affects the overall design efficiency.

[0005] (2) The generation of two-dimensional floor plans lacks a flexible editing mechanism suitable for different modification scenarios. In the process of interior decoration design, there are both local small-scale adjustment needs (such as wall fine-tuning, door and window replacement, etc.) and overall structural adjustment needs. However, existing technologies usually adopt a single generation method, either regenerating the entire drawing (causing unmodified areas to be accidentally changed, resulting in loss of drawing stability), or simply performing a simple hard-coded replacement (unable to handle overall spatial structure reorganization), making it difficult to simultaneously ensure generation efficiency and result stability.

[0006] (3) Lack of a unified data source among multiple decoration drawings and bills of materials. In existing methods, different types of decoration drawings are usually generated separately by calling independent modules, and bills of materials are often compiled independently of the drawings, lacking a globally unified benchmark design basis. This fragmented generation mode is very likely to cause inconsistencies in information such as room naming, material numbering, and component status between drawings and bills of materials, directly leading to chaos in subsequent budgeting, procurement, and construction implementation. Summary of the Invention

[0007] The purpose of this invention is to provide a method, device, and medium for dual-mode editing and consistent generation of interior decoration drawings, in order to solve the problems in the prior art, such as large deviations in the generated results due to the lack of a stable alignment mechanism, the inability to balance the flexibility and stability of drawing modification due to a single editing mode, and inconsistencies between multiple drawings and bill of materials data due to the lack of a unified data source.

[0008] In a first aspect, the present invention provides a method for dual-mode editing and consistent generation of interior decoration drawings, comprising the following steps:

[0009] S1: Obtain the user's original requirement text, input the user's original requirement text into the spatial strategy consensus draft generation agent for structured processing, and output the initial spatial strategy consensus draft; generate an analysis diagram for scheme verification based on the initial spatial strategy consensus draft; obtain modification opinions for the analysis diagram, feed the modification opinions back to the spatial strategy consensus draft generation agent to correct the text of the initial spatial strategy consensus draft, and regenerate the analysis diagram based on the corrected spatial strategy consensus draft for verification until the preset verification pass conditions are met, and output the final spatial strategy consensus draft;

[0010] S2: Obtain the current 2D plan view and the editing request, and input the final spatial strategy consensus draft, the current 2D plan view, and the editing request into the central control routing agent; the central control routing agent selects a route between local patch mode and global replacement mode based on the modification scope of the editing request and the target fragment positioning status, and generates candidate 2D plan views based on the route selection result; performs syntax verification and usability verification on the candidate 2D plan views. If the verification passes, an usable 2D plan view is output; if the verification fails, an error message is generated and returned to the central control routing agent to re-trigger the generation process;

[0011] S3: Input the final spatial strategy consensus draft and the available two-dimensional floor plan into the main decoration scheme locking agent to extract rules and output a main decoration scheme containing multiple unified rules; input the final spatial strategy consensus draft, the available two-dimensional floor plan, and the main decoration scheme into a multi-agent team containing multiple drawing generation agents, and generate multiple decoration drawings of corresponding types under the unified data constraints of the main decoration scheme; at the same time, generate a structured bill of materials based on the final spatial strategy consensus draft and the available two-dimensional floor plan, and output the multiple decoration drawings and the structured bill of materials that are consistent with each other.

[0012] As an optional implementation of the first aspect of this application, in step S1, generating an analysis diagram for scheme verification based on the initial spatial strategy consensus draft specifically includes: inputting the initial spatial strategy consensus draft into the analysis diagram generation module; calling the overall analysis diagram generation agent to perform global feature mapping based on the initial spatial strategy consensus draft, and outputting an overall analysis diagram to express the overall goals of the project, spatial organization relationships, functional scope, and known constraints; and calling the key strategy diagram generation agent to perform key feature mapping based on the initial spatial strategy consensus draft, and outputting a key strategy diagram to express the key strategy content that should be implemented in subsequent designs.

[0013] As an optional implementation of the first aspect of this application, in step S1, obtaining modification opinions for the analysis graph and feeding the modification opinions back to the spatial strategy consensus draft generation agent to perform textual correction on the initial spatial strategy consensus draft, specifically includes: obtaining modification opinions for the overall analysis graph and the key strategy graph; blocking the modification opinions from directly modifying the overall analysis graph and the key strategy graph, and feeding the modification opinions back to the spatial strategy consensus draft generation agent as textual constraints; the spatial strategy consensus draft generation agent correcting the constraints and functional partition data in the initial spatial strategy consensus draft according to the textual constraints, and outputting an updated spatial strategy consensus draft; and calling the overall analysis graph generation agent and the key strategy graph generation agent to regenerate the target overall analysis graph and the target key strategy graph based on the updated spatial strategy consensus draft.

[0014] As an optional implementation of the first aspect of this application, in step S2, the overall control routing agent selects a route between a local patching mode and a global replacement mode based on the modification range of the editing request and the positioning status of the target fragment. Specifically, this includes: parsing the action attributes and influence range boundaries of the editing request; when it is determined that the current two-dimensional planar map does not exist, or when the action attributes of the editing request are parsed to be overall layout reconstruction, generating a global replacement routing instruction and triggering the global replacement mode to call the global replacement agent to re-output the complete candidate two-dimensional planar map; when it is determined that the current two-dimensional planar map exists, and the action attributes of the editing request are parsed to be local small-scale modification, and when it is determined that the target fragment can be reliably positioned in the current two-dimensional planar map, generating a local patching routing instruction and triggering the local patching mode to call the local patching agent to output local patch data.

[0015] As an optional implementation of the first aspect of this application, in step S2, the process of generating and applying the local patch data includes: obtaining the local patch data containing multiple patch units generated by the local patch agent; parsing each patch unit and extracting the matching field and replacement field therein; the matching field is used to represent the original graphic code fragment obtained by precise matching from the code structure of the current two-dimensional planar diagram, and the replacement field is used to represent the corrected target graphic code fragment; the replacement field of each patch unit is used to cover the area in the current two-dimensional planar diagram corresponding to the matching field according to the traversal order, and the candidate two-dimensional planar diagram is output by fusion.

[0016] As an optional implementation of the first aspect of this application, in step S2, the candidate two-dimensional planar graph is subjected to syntax verification and usability verification, specifically including: extracting the graphic code data of the candidate two-dimensional planar graph; performing syntax verification on the graphic code data, detecting whether the start and end tags of the graphic code are complete, and detecting whether there are code syntax errors during the parse tree construction process, and outputting the syntax verification result; performing usability verification on the graphic code data, detecting whether the root node of the graphic code is valid, and simulating the loading environment to detect whether the candidate two-dimensional planar graph can be loaded and rendered normally by the target editor, and outputting the usability verification result; when both the syntax verification result and the usability verification result are passed, the candidate two-dimensional planar graph is marked as the usable two-dimensional planar graph.

[0017] As an optional implementation of the first aspect of this application, in step S3, the final spatial strategy consensus draft and the available two-dimensional plan are input into the decoration master plan locking agent for rule extraction, and a decoration master plan containing multiple unified rules is output. Specifically, this includes: extracting spatial principle data from the final spatial strategy consensus draft and geometric topology data from the available two-dimensional plan; inputting the spatial principle data and the geometric topology data into the decoration master plan locking agent; the decoration master plan locking agent extracting global constraint information from the input data and outputting the decoration master plan; the decoration master plan includes at least room naming rules, wall strategy rules, floor material zoning rules, ceiling reference rules, door and window coding rules, wet area rules, electromechanical alignment rules, and unified labeling rules for constraining the whole system.

[0018] As an optional implementation of the first aspect of this application, in step S3, multiple decoration drawings of corresponding types are generated under the unified data constraints of the main decoration scheme, including: configuring the main decoration scheme as a single fact source constraint node; the multi-agent team, based on the single fact source constraint node, calls the corresponding drawing generation agents in parallel to generate drawings respectively; calls the decoration floor plan generation agent to generate the decoration floor plan; calls the ground paving plan generation agent to generate the ground paving plan; calls the ceiling plan generation agent to generate the ceiling plan; calls the wall positioning plan generation agent to generate the wall positioning plan; calls the electromechanical integrated plan generation agent to generate the electromechanical integrated plan; calls the elevation index and interior elevation plan generation agent to generate the elevation index and interior elevation plan; calls the node detail drawing generation agent to generate the node detail drawing; the data fields of the structured bill of materials at least include the material category, material name, specifications, quantity, unit, and remarks fields aligned with the single fact source constraint node.

[0019] In a second aspect, embodiments of this application provide an electronic device, which includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the steps of the method described in the first aspect.

[0020] Thirdly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] (1) This invention, by setting up a scheme alignment mechanism, first generates a spatial strategy consensus draft, then generates an overall analysis diagram and a key strategy diagram based on the spatial strategy consensus draft, and reverse-corrects the spatial strategy consensus draft based on the user's modification suggestions for the analysis diagram, so that a more thorough understanding of requirements and confirmation of schemes can be completed before the formal generation of two-dimensional plan drawings. Compared with the existing technology of directly generating drawings based on the user's original requirements, this invention can reduce rework caused by deviations in understanding requirements and improve the accuracy and stability of subsequent design generation.

[0023] (2) This invention introduces a dual-mode editing mechanism of local patching and global replacement, and combines it with a central control routing agent that automatically selects the editing method according to different modification scenarios, enabling the two-dimensional planar graph to support both local minimum-scale modification and overall structural reconstruction. Compared with the single overall regeneration or simple replacement method in the prior art, this invention can better balance editing efficiency and result stability, and improve the flexibility of two-dimensional planar graph generation and modification.

[0024] (3) This invention sets up a unified main decoration scheme and generates multiple decoration drawings and material lists under the constraints of this main decoration scheme, so that the various decoration drawings and the drawings and material lists can maintain the same source and consistency. Compared with the existing technology of generating multiple drawings separately and calculating material lists independently, this invention can effectively avoid the problems of inconsistencies in room naming, door and window numbering, material expression and quantity statistics, thereby improving the engineering usability and implementation of the whole set of design results. Attached Figure Description

[0025] Figure 1 This invention provides a system architecture diagram for dual-mode editing and consistent generation of interior decoration drawings.

[0026] Figure 2 A flowchart illustrating a method for dual-mode editing and consistent generation of interior decoration drawings provided in an embodiment of the present invention;

[0027] Figure 3 This is a flowchart of the alignment stage of the scheme provided in the embodiments of the present invention;

[0028] Figure 4 This is a flowchart of the two-dimensional planar diagram dual-mode editing provided in this embodiment of the invention;

[0029] Figure 5 This is a flowchart of a multi-agent team generating decoration drawings based on a unified master scheme, provided in an embodiment of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0032] Please see Figure 1 This invention constructs a multi-agent collaborative dual-mode editing and consistent generation system framework for interior decoration drawings, consisting of a scheme alignment module, a two-dimensional plan generation module, and a decoration drawing and bill of materials generation module. Each agent works collaboratively based on preset input, output, and calling relationships to achieve dual-mode editing of interior decoration drawings and consistent generation of drawings and materials.

[0033] It should be noted that, in this invention, an intelligent agent refers to a functional unit that performs specific functions under preset input, output, and task boundary constraints. A module is a functional set composed of one or more intelligent agents, used to complete the overall processing flow at a certain stage. Please refer to [link / reference]. Figure 1 The solution alignment module includes a spatial strategy consensus draft generation agent and an analysis diagram generation module; the analysis diagram generation module includes an overall analysis diagram generation agent and a key strategy diagram generation agent; the two-dimensional plan generation module includes a central control routing agent, a local patch agent, a global replacement agent, and a syntax and usability verification unit; the decoration drawing and bill of materials generation module includes a decoration main solution locking agent, a decoration drawing multi-agent team, and a decoration bill of materials generation agent; the decoration drawing multi-agent team includes a decoration floor plan generation agent, a ground paving plan generation agent, a ceiling plan generation agent, a wall positioning plan generation agent, an electromechanical integrated plan generation agent, an elevation index and interior elevation plan generation agent, and a node detail drawing generation agent.

[0034] Furthermore, the following description explains the intelligent agents involved in the specific implementation plan:

[0035] Spatial strategy consensus draft generation agent: It constitutes the upstream solution generation unit in the solution alignment stage. Its input is the user's original requirement text and user modification opinions, and its output is the spatial strategy consensus draft.

[0036] Overall analysis graph generation agent: This is the global analysis unit that constitutes the analysis graph generation module. Its input is the spatial policy consensus draft, and its output is the overall analysis graph.

[0037] Key strategy graph generation agent: It constitutes the strategy extraction unit in the analysis graph generation module. Its input is the spatial strategy consensus draft and its output is the key strategy graph.

[0038] The overall control routing agent constitutes the scheduling and control unit in the two-dimensional planar graph generation module. Its inputs are the spatial policy consensus draft, the current two-dimensional planar graph, the user's new editing request, and the error information when the verification fails. Its output is the local patch routing result or the global replacement routing result.

[0039] Local patch agent: It constitutes the local editing unit in the two-dimensional planar map generation module. Its inputs are the spatial strategy consensus draft, the structured graphic code corresponding to the current two-dimensional planar map, and the user's new editing request. Its output is local patch data.

[0040] Global replacement agent: constitutes the overall reconstruction unit in the two-dimensional planar graph generation module. Its inputs are the spatial policy consensus draft, the user's new editing request, and optional current two-dimensional planar graph reference information. Its output is a complete candidate two-dimensional planar graph.

[0041] The main renovation scheme locking agent is the main scheme extraction unit in the renovation drawing and bill of materials generation module. Its input is the spatial strategy consensus draft and available two-dimensional floor plan, and its output is the main renovation scheme.

[0042] The multi-agent team for interior design drawings consists of a set of parallel drawing generation units in the interior design drawing generation module. Its inputs are a consensus draft of spatial strategy, available two-dimensional floor plans, and the main interior design scheme, and its outputs are multiple interior design drawings.

[0043] The intelligent agent for generating interior floor plans is the unit that generates floor plans in the multi-agent team that makes up the interior drawings. Its inputs are the spatial strategy consensus draft, the available two-dimensional floor plan, and the main interior design scheme. Its output is the interior floor plan.

[0044] Ground paving map generation agent: The ground expression unit in the multi-agent team of decoration drawings. Its input is the spatial strategy consensus draft, the available two-dimensional floor plan and the main decoration scheme, and the output is the ground paving map;

[0045] Ceiling plan generation agent: The ceiling expression unit in the multi-agent team of decoration drawings. Its inputs are the spatial strategy consensus draft, available two-dimensional floor plan and decoration master plan, and the output is the ceiling plan.

[0046] Wall positioning map generation agent: It constitutes the wall expression unit in the multi-agent team of decoration drawings. Its input is the spatial strategy consensus draft, the available two-dimensional floor plan and the main decoration plan, and the output is the wall positioning map;

[0047] Mechanical and electrical integrated drawing generation agent: The mechanical and electrical expression unit that constitutes the multi-agent team of decoration drawings. Its inputs are the spatial strategy consensus draft, usable two-dimensional floor plan and decoration main scheme, and the output is the mechanical and electrical integrated drawing;

[0048] Elevation Index and Interior Elevation Drawing Generation Agent: This agent constitutes the elevation expression unit in the multi-agent team of decoration drawings. Its inputs are the spatial strategy consensus draft, available two-dimensional floor plan, and main decoration scheme, and its outputs are the elevation index and interior elevation drawings.

[0049] Node detail drawing generation agent: It constitutes the detailed expression unit in the multi-agent team of decoration drawings. Its input is the spatial strategy consensus draft, available two-dimensional floor plan and decoration main scheme, and its output is the node detail drawing;

[0050] The intelligent agent for generating decoration material lists consists of the list generation unit in the decoration drawing and material list generation module. Its inputs are the spatial strategy consensus draft and the available two-dimensional floor plan, and its output is a structured material list.

[0051] Please see Figure 2 The flowchart below illustrates a method for dual-mode editing and consistent generation of interior decoration drawings provided in an embodiment of the present invention. The method provided in this embodiment mainly includes a scheme alignment and consensus generation stage (corresponding to the scheme alignment module), a dual-mode editing stage of two-dimensional floor plans (corresponding to the two-dimensional floor plan generation module), and a consistent generation stage of drawings and bill of materials (corresponding to the decoration drawings and bill of materials generation module).

[0052] Please see Figure 3 During the scheme alignment and consensus generation phase:

[0053] The first objective of this invention is to provide a solution alignment method for interior decoration needs. This method can first understand the user's needs as a whole, extract key strategies, and accumulate textual consensus before formally generating two-dimensional floor plans and decoration drawings, thereby forming a stable and executable upstream design basis.

[0054] Principle: In interior design scenarios, users' original needs are often colloquial, vague, and incomplete. Directly generating two-dimensional floor plans or design drawings based on these needs can easily lead to misunderstandings by the system, resulting in inconsistencies between the generated results and user expectations. Therefore, this invention first uses a spatial strategy consensus draft generation agent to organize and summarize the user's original needs, forming a text-based spatial strategy consensus draft. Then, an overall analysis diagram generation agent and a key strategy diagram generation agent execute in parallel based on this spatial strategy consensus draft, generating two analysis diagrams for scheme verification. User suggestions for modification to the analysis diagrams do not directly affect the diagrams themselves but are fed back to the spatial strategy consensus draft generation agent to revise the consensus draft. The analysis diagrams are then regenerated based on the updated consensus draft. This "scheme-driven diagram, diagram-backward verification of scheme" approach allows for sufficient scheme alignment before formally generating two-dimensional floor plans, improving the accuracy and stability of subsequent drawing generation.

[0055] Specifically, the solution to achieve the first objective of this invention is: the user's original requirement text input scheme alignment module is executed according to the following process.

[0056] Step 1: Generation of Spatial Strategy Consensus Draft

[0057] The user's original requirements text is input into the spatial strategy consensus draft to generate an intelligent agent, thus generating an initial spatial strategy consensus draft. This spatial planning consensus draft includes at least the project objectives, spatial scope, functional zoning, style direction, constraints, supplementary system assumptions, and unified principles that should be followed in subsequent design.

[0058] Step 2: Generating the Analysis Chart

[0059] The spatial strategy consensus draft is input into the analysis diagram generation module. The overall analysis diagram generation agent generates the overall analysis diagram, and the key strategy diagram generation agent generates the key strategy diagram. The overall analysis diagram expresses the overall project objectives, spatial organization relationships, functional scope, and known constraints, while the key strategy diagram expresses the strategies that should be prioritized in subsequent design.

[0060] Step 3: User Inspection

[0061] The overall analysis diagram and key strategy diagram are provided to users for review to determine whether the current spatial strategy consensus draft meets user expectations.

[0062] Step 4: Revise the plan

[0063] If the user is not satisfied, suggestions for modification are proposed and re-inputted into the spatial strategy consensus draft generation agent to update the spatial strategy consensus draft. Then, the overall analysis graph generation agent and the key strategy graph generation agent regenerate the overall analysis graph and key strategy graph based on the updated spatial strategy consensus draft until the user is satisfied.

[0064] Step 5: Output the final spatial strategy consensus draft

[0065] When the user is satisfied, the final space strategy consensus draft is output, which serves as the common input basis for subsequent generation of 2D floor plans, decoration drawings, and bill of materials.

[0066] Please see Figure 4 In the dual-model editing stage of the two-dimensional planar diagram:

[0067] The second objective of this invention is to provide a dual-mode editing method for two-dimensional floor plans of interior decoration, which combines local patching and global replacement. This method can automatically select either local patching mode or global replacement mode according to different modification scenarios, thereby balancing local modification efficiency and overall reconstruction capability.

[0068] Principle: In the process of generating and modifying 2D interior design plans, there are both localized small-scale adjustment needs, such as minor wall adjustments, door and window replacements, and partial furniture layout updates, and overall spatial structure reconstruction needs. If all modifications are generated in full, unmodified areas may be changed along with the plan, affecting the stability of the drawings; if all modifications are replaced locally, it is difficult to handle overall adjustments. Therefore, this invention sets up a central control routing agent that automatically routes between a local patching agent and a global replacement agent based on whether the current 2D plan exists, the scope of the user's modification request, and whether the target content can be reliably located. The local patching mode is used to make minimal modifications to the existing 2D plan, while the global replacement mode is used to regenerate the complete 2D plan. Regardless of the mode used, the candidate 2D plan must undergo syntax and usability checks. If the checks fail, the error information is fed back to the central control routing agent, re-triggering the corresponding generation process, thus forming a closed loop of "generation—verification—correction".

[0069] Specifically, the solution to achieve the second objective of this invention is to input the spatial strategy consensus draft, the current two-dimensional plan (if it exists), and the user's new editing request into the two-dimensional plan generation module, and execute the following process.

[0070] Step 1: Central Control Route Determination

[0071] The spatial strategy consensus draft, the current 2D plan view, and the user's new edit request are input into the central control routing agent. When no available 2D plan view exists, or the request involves overall layout reconstruction, a global replacement route is output; when a available 2D plan view exists, the request involves minor local modifications, and the target fragment can be reliably located in the existing 2D plan view, a local patch route is output.

[0072] Step 2: Generating Local Patches

[0073] When the central control routing determines that the mode is local patching, it invokes the local patching agent to output local patch data. This local patch data includes at least multiple patch units, each containing at least a matching field and a replacement field. The matching field represents the original graphic code fragment precisely matched from the current 2D planar graph, and the replacement field represents the target graphic code fragment after replacement. Subsequently, each patch unit is applied sequentially to the current 2D planar graph to obtain candidate 2D planar graphs.

[0074] Step 3: Global Replacement Generation

[0075] When the central control routing determines that the mode is global replacement, it invokes the global replacement agent to output a complete candidate 2D planar graph. This complete 2D planar graph is used to replace the current 2D planar graph and is suitable for scenarios where a 2D planar graph does not yet exist or requires a complete redraw.

[0076] Step 4: Syntax and Usability Validation

[0077] Perform syntax and usability checks on candidate 2D planar graphs, including but not limited to:

[0078] (1) Check if the start and end labels of the graphic code are complete;

[0079] (2) Check for syntax errors during the parsing process;

[0080] (3) Check if the root node is valid;

[0081] (4) Check whether the candidate two-dimensional planar graph can be loaded normally by the editor.

[0082] If the verification passes, a usable 2D plan view will be output; if the verification fails, an error message and the code for the 2D plan view that failed the verification will be generated.

[0083] Step 5: Error reflow and regeneration

[0084] The error message and the failed 2D planar diagram code are fed back to the central control routing agent, which then determines whether to call the local patch agent or the global replacement agent to generate candidate 2D planar diagrams again, until a usable 2D planar diagram that passes the verification is obtained.

[0085] Please see Figure 5 During the stage where drawings and bills of materials are generated from the same source:

[0086] The third objective of this invention is to provide a method for generating decoration drawings and bills of materials based on a unified master scheme. This method can ensure consistency between multiple decoration drawings and between the drawings and the bill of materials.

[0087] Principle: Two-dimensional floor plans are a crucial foundation for generating subsequent decoration drawings and bills of materials. However, generating various drawings solely based on two-dimensional floor plans can easily lead to discrepancies in room naming, material strategies, door and window numbering, and component status. To address this issue, this invention first extracts and locks a master decoration plan based on a consensus draft of the spatial strategy and available two-dimensional floor plans, serving as the single source of fact for generating subsequent decoration drawings. Then, a multi-agent team for decoration drawings generates various types of decoration drawings. Simultaneously, a bill of materials generation agent generates a bill of materials based on the same consensus draft of the spatial strategy and available two-dimensional floor plans, thus achieving consistency between the drawings and the bill of materials.

[0088] Specifically, the technical solution to achieve the third objective of this invention is: inputting the spatial strategy consensus draft and the available two-dimensional floor plan into the decoration drawings and bill of materials generation module, and executing the following process.

[0089] Step 1: Finalize the main renovation plan

[0090] The consensus draft of the spatial strategy and the available two-dimensional floor plan are input into the smart agent of the main decoration scheme, and the main decoration scheme is extracted and locked. The main decoration scheme includes at least room naming, wall strategy, floor material zoning, ceiling benchmark, door and window coding rules, wet area rules, electromechanical alignment rules, and unified labeling rules, and serves as the single source of facts for the generation of subsequent decoration drawings.

[0091] Step 2: The multi-agent team generates the renovation blueprints.

[0092] The spatial strategy consensus draft, available 2D floor plan, and main decoration scheme are input into the multi-agent team responsible for decoration drawings. Multiple drawing generation agents then generate corresponding drawings in parallel under the constraints of a unified main decoration scheme. The generated drawings include:

[0093] (1) Generate the decoration floor plan from the intelligent agent;

[0094] (2) The ground paving map is generated by the intelligent agent;

[0095] (3) Generate ceiling plan from intelligent agent based on ceiling plan;

[0096] (4) Generate the wall positioning map from the intelligent agent;

[0097] (5) Generate an intelligent agent from the electromechanical integrated diagram to generate the electromechanical integrated diagram;

[0098] (6) Generate the smart body's elevation index and interior elevation drawing from the elevation index and interior elevation drawing;

[0099] (7) Generate node details from the node details diagram.

[0100] All drawing-generated intelligent entities are constrained by the same main decoration plan, thereby ensuring consistency in room naming, component relationships, material representation, and numbering system.

[0101] Step 3: Bill of Materials Generation

[0102] The spatial strategy consensus draft and available 2D floor plan are input into the decoration material list to generate an intelligent agent, and a structured material list is output. The material list includes at least the fields of material category, material name, specifications, quantity, unit, and remarks, and is consistent with the spatial planning and 2D floor plan representation.

[0103] Step 4: Output drawings – material consistency results

[0104] The various decoration drawings and corresponding material lists will be output uniformly as the basis for subsequent budget evaluation, material procurement and construction implementation.

[0105] Optionally, embodiments of this application also provide an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the various processes of the above-described embodiment of the dual-mode editing and consistent generation method for interior decoration drawings, and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0106] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described embodiment of a dual-mode editing and consistent generation method for interior decoration drawings, and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0107] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0108] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0109] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0110] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for dual-mode editing and consistent generation of interior decoration drawings, characterized in that, Includes the following steps: S1: Obtain the user's original demand text, input the user's original demand text into the spatial policy consensus draft generation agent for structured processing, and output the initial spatial policy consensus draft; An analysis diagram for scheme verification is generated based on the initial spatial strategy consensus draft; modification opinions for the analysis diagram are obtained, and the modification opinions are fed back to the spatial strategy consensus draft generating agent to correct the text of the initial spatial strategy consensus draft, and the analysis diagram is regenerated based on the corrected spatial strategy consensus draft for verification until the preset verification pass conditions are met, and the final spatial strategy consensus draft is output. S2: Obtain the current 2D plan view and the editing request, and input the final spatial strategy consensus draft, the current 2D plan view, and the editing request into the central control routing agent; the central control routing agent selects a route between local patch mode and global replacement mode based on the modification scope of the editing request and the target fragment positioning status, and generates candidate 2D plan views based on the route selection result; performs syntax verification and usability verification on the candidate 2D plan views. If the verification passes, an usable 2D plan view is output; if the verification fails, an error message is generated and returned to the central control routing agent to re-trigger the generation process; The central control routing agent selects between local patch mode and global replacement mode based on the modification scope of the editing request and the target fragment location status. Specifically, this includes parsing the action attributes and influence scope boundaries of the editing request. When it is determined that the current 2D planar image does not exist, or when the action attribute of the edit request is parsed to be an overall layout reconstruction, a global replacement route instruction is generated, and the global replacement mode is triggered to call the global replacement agent to re-output the complete candidate 2D planar image; when it is determined that the current 2D planar image exists, and the action attribute of the edit request is parsed to be a local small-scale modification, and it is determined that the target segment to be modified can be reliably located in the current 2D planar image, a local patch route instruction is generated, and the local patch mode is triggered to call the local patch agent to output local patch data; S3: Input the final spatial strategy consensus draft and the available two-dimensional floor plan into the main decoration scheme locking agent to extract rules and output a main decoration scheme containing multiple unified rules; input the final spatial strategy consensus draft, the available two-dimensional floor plan, and the main decoration scheme into a multi-agent team containing multiple drawing generation agents, and generate multiple decoration drawings of corresponding types under the unified data constraints of the main decoration scheme; at the same time, generate a structured bill of materials based on the final spatial strategy consensus draft and the available two-dimensional floor plan, and output the multiple decoration drawings and the structured bill of materials that are consistent with each other.

2. The method according to claim 1, characterized in that, In step S1, an analysis diagram for scheme verification is generated based on the initial spatial strategy consensus draft, specifically including: Input the initial spatial strategy consensus draft into the analysis graph generation module; The intelligent agent generates a global feature mapping based on the initial spatial strategy consensus draft by calling the overall analysis graph, and outputs an overall analysis graph that expresses the overall goals of the project, spatial organization relationships, functional scope and known constraints. The agent generates a key strategy graph by calling the key strategy graph and performing key feature mapping based on the initial spatial strategy consensus draft. The output is a key strategy graph that expresses the key strategy content that should be implemented in subsequent designs.

3. The method according to claim 2, characterized in that, In step S1, modification suggestions for the analysis graph are obtained, and these suggestions are fed back to the spatial policy consensus draft generating agent to perform text correction on the initial spatial policy consensus draft. Specifically, this includes: Obtain feedback on the overall analysis diagram and the key strategy diagram; The modification opinions are blocked from directly modifying the overall analysis diagram and the key strategy diagram. Instead, the modification opinions are fed back as text constraints into the spatial strategy consensus draft generation agent. The spatial strategy consensus draft generating agent corrects the constraints and functional partition data in the initial spatial strategy consensus draft according to the text constraints, and outputs the updated spatial strategy consensus draft. The overall analysis graph generation agent and the key strategy graph generation agent are invoked to regenerate the target overall analysis graph and the target key strategy graph based on the updated spatial strategy consensus draft.

4. The method according to claim 1, characterized in that, In step S2, the process of generating and applying the local patch data includes: Obtain the local patch data, which contains multiple patch units, generated by the local patching agent; Each patch unit is parsed, and the matching field and replacement field are extracted therein; the matching field is used to represent the original graphic code fragment that is accurately matched from the code structure of the current two-dimensional planar diagram, and the replacement field is used to represent the corrected target graphic code fragment; The replacement fields of each patch unit are overwritten with the areas in the current two-dimensional planar graph corresponding to the matching fields according to the traversal order, and the candidate two-dimensional planar graph is then output.

5. The method according to claim 1, characterized in that, In step S2, syntax verification and usability verification are performed on the candidate two-dimensional planar graph, specifically including: Extract the graphic code data of the candidate two-dimensional planar image; Perform syntax validation on the graphic code data, check whether the start and end tags of the graphic code are complete, and check whether there are any code syntax errors during the parse tree construction process, and output the syntax validation results; Availability verification is performed on the graphic code data to check whether the root node of the graphic code is valid, and the loading environment is simulated to check whether the candidate two-dimensional planar image can be loaded and rendered normally by the target editor, and the availability verification result is output. When both the syntax verification result and the usability verification result pass, the candidate two-dimensional planar graph is marked as the usable two-dimensional planar graph.

6. The method according to claim 1, characterized in that, In step S3, the final spatial strategy consensus draft and the available two-dimensional floor plan are input into the main decoration scheme locking agent for rule extraction, and the main decoration scheme containing multiple unified rules is output, specifically including: Extract the spatial principle data from the final spatial strategy consensus draft and the geometric topology data from the available two-dimensional planar graph; The spatial principle data and the geometric topology data are input into the smart body that locks the main decoration scheme. The main renovation scheme locks the intelligent agent to extract global constraint information from the input data and outputs the main renovation scheme; the main renovation scheme includes at least room naming rules, wall strategy rules, floor material zoning rules, ceiling benchmark rules, door and window coding rules, wet area rules, electromechanical alignment rules, and unified labeling rules for global constraints.

7. The method according to claim 6, characterized in that, In step S3, under the unified data constraints of the main decoration scheme, multiple decoration drawings of corresponding types are generated, including: Configure the main renovation plan as a single fact source constraint node; The multi-agent team, based on the single fact source constraint node, calls the corresponding drawing generation agents in parallel to generate drawings respectively. The system generates an intelligent entity by calling the interior design floor plan; it generates an intelligent entity by calling the floor paving plan; it generates an intelligent entity by calling the ceiling plan; it generates an intelligent entity by calling the wall positioning plan; it generates an intelligent entity by calling the electromechanical integrated plan; it generates an intelligent entity by calling the elevation index and interior elevation plan; and it generates an intelligent entity by calling the node detail plan. The structured bill of materials data fields include at least the material category, material name, specifications, quantity, unit, and remarks fields aligned to the single fact source constraint node.

8. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and capable of running on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the dual-mode editing and consistent generation method for interior decoration drawings as described in any one of claims 1-7.

9. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a processor, implement the steps of the dual-mode editing and consistent generation method for interior decoration drawings as described in any one of claims 1-7.

Images