An intelligent home AI design optimization generation system

Abstract

The application relates to an intelligent home AI design optimization generation system and relates to the technical field of artificial intelligence, and comprises a request analysis module, a scheme generation module, a feedback receiving module and a scheme optimization module. The request analysis module is used for acquiring home environment parameters and user demand parameters of a target user; the scheme generation module is used for generating an initial design scheme; the feedback receiving module is used for receiving feedback information; and the scheme optimization module is used for generating a target home design scheme. The application solves the technical problems that the AI home design based on rules in the prior art cannot effectively utilize user feedback for learning optimization and cannot realize large-scale individual customization, and achieves the technical effects that iterative optimization is performed by combining user feedback, design efficiency is improved, and large-scale individual customization is realized.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and in particular to an AI-based smart home design optimization and generation system. Background Technology

[0002] With the rapid development of artificial intelligence (AI) technology, AI-based home design systems have begun to emerge. These systems typically generate design schemes using a rule-based approach, automatically generating schemes based on preset design rules and templates, combined with basic parameters provided by the user. However, the design process of current AI-powered home design is relatively rigid, relying mainly on predefined rule sets, lacking flexibility, and making it difficult to continuously improve based on users' individual needs, thus hindering the realization of truly large-scale personalized customization services. Summary of the Invention

[0004] The purpose of this application is to provide a smart home AI design optimization and generation system, which aims to solve the technical problems in the existing rule-based AI home design that cannot effectively utilize user feedback for learning and optimization, and is difficult to achieve large-scale personalized customization.

[0005] The smart home AI design optimization and generation system provided in this application adopts the following technical solution:

[0006] The request parsing module is used to respond to the home generation request from the terminal device, parse the home generation request, and obtain the target user's home environment parameters and user requirement parameters;

[0007] The scheme generation module is used to generate an initial design scheme based on the home environment parameters and the user requirement parameters using a pre-trained home design model.

[0008] A feedback receiving module is used to receive feedback information from the target user regarding the initial design scheme;

[0009] The scheme optimization module is used to iteratively adjust the initial design scheme based on the feedback information to generate the target home design scheme.

[0010] Optionally, the request parsing module includes:

[0011] An environment extraction unit is used to extract home space dimensions, structural layout, lighting conditions, and functional zoning from the home generation request as home environment parameters.

[0012] The requirement extraction unit is used to extract user budget, style preference, functional requirements and resident population composition from the home furnishing generation request as user requirement parameters.

[0013] Optionally, the scheme generation module includes:

[0014] Model building units are used to build home design models;

[0015] The first scheme acquisition unit is used to input the home environment parameters and the user demand parameters into the home design model and output the first design scheme.

[0016] The second scheme acquisition unit is used to match schemes in the historical home design database based on the home environment parameters and the user demand parameters to obtain a second design scheme.

[0017] The scheme fusion unit is used to fuse the first design scheme and the second design scheme to obtain the initial design scheme.

[0018] Optionally, the model building unit includes:

[0019] The sample collection subunit is used to collect multiple sample home generation requests from the historical home design database and construct a sample home generation request set. Each sample home generation request includes sample home environment parameters and sample user demand parameters.

[0020] The scheme extraction subunit is used to extract the corresponding sample home design scheme from the historical home design database based on the multiple sample home generation requests, and construct a sample home design scheme set.

[0021] The model training subunit is used to train the deep neural network model using supervised learning based on the sample home furnishing request set and the sample home furnishing design scheme set, so as to obtain the home furnishing design model.

[0022] Optionally, the second scheme acquisition unit includes:

[0023] The condition construction subunit is used to construct request query conditions based on the home environment parameters and the user demand parameters;

[0024] The request query subunit is used to perform a request query in the historical home design database based on the request query conditions to obtain multiple similar home generation requests;

[0025] The similarity scheme extraction subunit is used to extract the home design schemes corresponding to multiple similar home generation requests, and obtain multiple similar home design schemes.

[0026] The weighted processing subunit is used to perform weighted processing on multiple similar home design schemes based on the similarity between multiple similar home generation requests and the home generation request, so as to obtain the second design scheme.

[0027] Optionally, the scheme optimization module includes:

[0028] The feedback parsing unit is used to parse and process the feedback information to obtain the user's modification requirements;

[0029] The first adjustment determination unit is used to determine the first part to be adjusted in the initial design scheme according to the user's modification requirements;

[0030] The first adjustment unit is used to adjust the first part to be adjusted based on the user's modification requirements, and update the initial design scheme according to the first adjustment result to obtain the first updated design scheme.

[0031] The first scheme verification unit is used to verify whether the first updated design scheme meets the feedback information based on the user's modification requirements and the initial design scheme.

[0032] The solution output unit is used to output the first updated design solution as the target home design solution when the first updated design solution meets the feedback information.

[0033] Optionally, the scheme optimization module further includes:

[0034] The second adjustment determination unit is used to determine a second part to be adjusted in the first update design scheme according to the user's modification requirements when the first update design scheme does not meet the feedback information.

[0035] The second adjustment unit is used to adjust the second part to be adjusted based on the user's modification requirements, and update the first updated design scheme according to the second adjustment result to obtain the second updated design scheme.

[0036] The second scheme verification unit is used to verify whether the second updated design scheme meets the feedback information based on the user's modification requirements and the initial design scheme.

[0037] An iterative output unit is used to iteratively execute until the Nth updated design scheme satisfies the feedback information, and then outputs the target home design scheme.

[0038] Optionally, the system further includes:

[0039] The receiving module is configured to receive the target user's confirmation signal regarding the target home design scheme.

[0040] The scheme storage module is used to associate and store the home creation request, the home environment parameters, the user requirement parameters, and the target home design scheme based on the scheme determination signal.

[0041] The beneficial effects of this application are as follows:

[0042] The request parsing module responds to home furnishing generation requests from terminal devices, parses these requests, and obtains the target user's home environment parameters and user requirement parameters. This allows for an accurate understanding of the user's specific needs and actual home environment conditions, laying the foundation for subsequent design scheme generation. The scheme generation module utilizes a pre-trained home furnishing design model to generate an initial design scheme based on the home environment parameters and user requirement parameters, achieving intelligent and automated design processes. The feedback receiving module receives feedback from the target user regarding the initial design scheme, enabling the user to participate in the design process and provide modification suggestions and personalized requirements. The scheme optimization module iteratively adjusts the initial design scheme based on feedback information, generating a target home furnishing design scheme. Through continuous optimization of the design scheme, it ultimately meets the user's personalized needs, achieving iterative design optimization based on user feedback.

[0043] Through the above technical solution, this application solves the technical problems of existing rule-based AI home design systems being unable to effectively utilize user feedback for learning and optimization, and being unable to achieve large-scale personalized customization, thereby achieving improved design efficiency and the technical effect of large-scale personalized customization. Attached Figure Description

[0044] Figure 1 This application provides a schematic diagram of the structure of an AI-powered smart home design optimization and generation system.

[0045] Figure 2 A schematic diagram of the scheme generation module is provided for the embodiments of this application.

[0046] Explanation of reference numerals in the attached figures:

[0047] The system includes a request parsing module 100, a solution generation module 200, a feedback receiving module 300, a solution optimization module 400, a model building unit 210, a first solution acquisition unit 220, a second solution acquisition unit 230, and a solution fusion unit 240. Detailed Implementation

[0048] The following is in conjunction with the appendix Figure 1 -Appendix Figure 2 This application will be described in further detail below.

[0049] Example 1: A smart home AI design optimization and generation system, referring to Figure 1 It includes a request parsing module 100, a solution generation module 200, a feedback receiving module 300, and a solution optimization module 400.

[0050] The request parsing module 100 is used to respond to the home generation request of the terminal device, parse the home generation request, and obtain the home environment parameters and user requirement parameters of the target user.

[0051] Specifically, when a target user sends home design requests via a mobile device, computer, or smart home control panel, the request parsing module 100 is triggered and begins operation. The request parsing module 100 employs deep semantic analysis technology to perform multi-level parsing of the user-submitted home design request, extracting home environment parameters and user requirement parameters.

[0052] The home environment parameters include, but are not limited to, the three-dimensional dimensions of the target user's home space, interior structural layout, wall locations, door and window distribution, lighting conditions, location and intensity of natural light sources, and functional zoning. These parameters constitute the basic constraints of the design scheme, ensuring that the generated scheme is physically feasible. User requirements parameters include, but are not limited to, the design budget range, style preferences (such as modern minimalist, Nordic style, traditional Chinese style, etc.), functional requirements (such as work areas, leisure spaces, storage requirements, etc.), and the composition of the resident population (such as the number of family members, age distribution, and special needs groups, etc.). These parameters determine the personalized characteristics of the design scheme.

[0053] The request parsing module 100 obtains the target user's home environment parameters and user requirement parameters to parse unstructured home generation requests, accurately extract key parameters, and lay a data foundation for subsequent design scheme generation.

[0054] The scheme generation module 200 is used to generate an initial design scheme based on the home environment parameters and the user requirement parameters using a pre-trained home design model.

[0055] Specifically, the solution generation module 200 receives home environment parameters and user requirement parameters extracted from the request parsing module 100, and generates an initial design solution based on these parameters. The solution generation module 200 uses a pre-trained home design model as an intelligent decision engine to achieve intelligent generation of home design solutions and obtain the initial design solution.

[0056] The pre-trained home design model is a deep learning model trained on a large-scale home design dataset. This model has pre-learned the spatial layout rules, style element matching relationships, functional area division principles, and design rules in a large number of home design cases. It can effectively capture the implicit rules and design expertise in home design and generate initial design schemes based on home environment parameters and user needs parameters.

[0057] The feedback receiving module 300 is used to receive feedback information from the target user regarding the initial design scheme.

[0058] Specifically, the feedback receiving module 300 is used to receive feedback information from the target user regarding the initial design scheme. The feedback receiving module 300 employs multimodal interaction technology, supporting users to provide opinions, suggestions, and modification requests for the initial design scheme through various methods, and receiving feedback information. This multimodal interaction technology includes, but is not limited to, natural language input systems, gesture recognition devices, and annotation tools. The target user can directly annotate, drag, and modify the 3D model or floor plan of the initial design scheme through the graphical user interface on the terminal device; they can also describe their modification opinions on specific areas, furniture arrangement, color schemes, or material selections through text or voice.

[0059] The scheme optimization module 400 is used to iteratively adjust the initial design scheme based on the feedback information to generate the target home design scheme.

[0060] Specifically, the scheme optimization module 400 is used to receive feedback information from the feedback receiving module 300 and to make targeted adjustments and improvements to the initial design scheme.

[0061] The solution optimization module 400 first analyzes and processes the feedback information to identify key modification requests from users. Based on these analysis results, it determines the specific elements and parameters that need adjustment in the initial design scheme, such as spatial layout, furniture selection, color matching, material application, or lighting design. Subsequently, the solution optimization module 400 iteratively adjusts the initial design scheme according to the specific elements and parameters to be adjusted. After each round of adjustments, the updated design scheme is verified to evaluate whether it meets the requirements of the user feedback. If there are any unmet needs, the next round of optimization iteration will be initiated until a target home design scheme that meets the user feedback requirements is generated.

[0062] Through the iterative adjustment mechanism of the solution optimization module 400, continuous optimization and fine-tuning of design solutions are achieved, improving the customization level of home design and user satisfaction. This overcomes the technical problem that traditional AI design systems cannot effectively utilize user feedback for learning and optimization, thereby improving design efficiency and enabling large-scale personalized customization.

[0063] Furthermore, the request parsing module 100 includes an environment extraction unit and a requirement extraction unit.

[0064] The environment extraction unit is used to extract the home space size, structural layout, lighting conditions and functional zoning from the home generation request as the home environment parameters;

[0065] The requirement extraction unit is used to extract user budget, style preference, functional requirements and resident population composition from the home furnishing generation request as user requirement parameters.

[0066] In one optional implementation, the environment extraction unit obtains the user's text description, uploaded floor plan, or 3D scan data from the home creation request, and accurately extracts the home space dimensions (such as room length, width, height, and area), structural layout (such as the location of load-bearing walls, door and window distribution, and pipework), lighting conditions (such as window orientation, natural light angle and intensity, and duration of sunshine), and functional zoning (such as the division of areas like bedrooms, living rooms, kitchens, and bathrooms, and their relative positions). These objective physical parameters constitute the home environment parameters of the design scheme, ensuring that the generated design scheme can be realized in the actual physical space.

[0067] The requirement extraction unit is responsible for extracting information related to the user's subjective preferences and actual needs from the home furnishing request, serving as user requirement parameters. Through natural language processing and user preference mining, the requirement extraction unit extracts the user's budget (including the overall budget range and the budget proportion for each functional area or furniture category), style preferences (such as modern minimalist, Nordic, traditional Chinese, industrial, and other design styles and their combinations), functional requirements (such as specific functional requirements like special workspace requirements, entertainment area settings, storage space needs, and smart home integration), and resident demographics (such as the number of family members, age distribution, and specific requirements for special groups such as the elderly, children, or people with disabilities). These subjective requirement parameters determine the personalized characteristics of the design scheme, ensuring that the generated scheme meets the user's actual living needs and aesthetic preferences.

[0068] Through the collaborative work of the environment extraction unit and the demand extraction unit, the request parsing module 100 can comprehensively and accurately parse the home generation request, providing complete parameter input for the solution generation module 200, thereby achieving accurate generation of the initial design solution.

[0069] Furthermore, such as Figure 2 As shown, the scheme generation module 200 includes a model building unit 210, a first scheme acquisition unit 220, a second scheme acquisition unit 230, and a scheme fusion unit 240.

[0070] Among them, the model building unit 210 is used to build a home design model;

[0071] The first scheme acquisition unit 220 is used to input the home environment parameters and the user demand parameters into the home design model and output the first design scheme;

[0072] The second scheme acquisition unit 230 is used to perform scheme matching in the historical home design database based on the home environment parameters and the user demand parameters to obtain a second design scheme;

[0073] The scheme fusion unit 240 is used to fuse the first design scheme and the second design scheme to obtain the initial design scheme.

[0074] Specifically, model building unit 210 is responsible for constructing the home design model. Based on a deep neural network architecture, this unit trains on large-scale home design sample data to build a home design model capable of understanding spatial relationships, functional requirements, and stylistic characteristics. The constructed home design model possesses multi-dimensional design capabilities, including space planning, style matching, and functional layout, and can automatically generate suitable home design solutions based on input parameters.

[0075] The first scheme acquisition unit 220 is used to input home environment parameters and user demand parameters into the home design model, and output the first design scheme through the model inference process. This unit formats the parameters obtained from the request parsing module 100 into input vectors that the model can recognize, and then generates a complete design scheme including elements such as spatial layout, furniture configuration, color matching, and material selection through forward inference calculation of the home design model, which serves as the first design scheme. The first design scheme reflects the design concept of the AI ​​model based on data learning.

[0076] The second design acquisition unit 230, based on the principle of case-based reasoning, matches design options in a historical home design database according to home environment parameters and user needs parameters to obtain a second design option. This unit uses a similarity calculation algorithm to retrieve the design option from historical successful cases that most closely matches the current needs parameters, and makes appropriate adjustments to form the second design option. The second design option reflects the practicality and reliability of mature design cases.

[0077] The scheme fusion unit 240 is responsible for merging the first and second design schemes, combining the advantages of both to obtain the initial design scheme. This unit, while maintaining a rational spatial layout, integrates the innovation of the first design scheme and the practicality of the second, balancing aesthetics, functionality, and cost-effectiveness to generate a more comprehensive and balanced initial design scheme.

[0078] Through the coordinated operation of the above four functional units, the solution generation module 200 can generate an initial design solution that is both AI-innovative and practically feasible, providing a good foundation for subsequent user feedback and solution optimization.

[0079] Furthermore, the model building unit 210 includes a sample acquisition subunit, a scheme extraction subunit, and a model training subunit.

[0080] The sample collection subunit is used to collect multiple sample home generation requests from the historical home design database and construct a sample home generation request set. Each sample home generation request includes sample home environment parameters and sample user demand parameters.

[0081] The scheme extraction subunit is used to extract the corresponding sample home design scheme from the historical home design database based on the multiple sample home generation requests, and construct a sample home design scheme set.

[0082] The model training subunit is used to train the deep neural network model using supervised learning based on the sample home furnishing request set and the sample home furnishing design scheme set, so as to obtain the home furnishing design model.

[0083] Specifically, the sample collection subunit selects representative home furnishing requests from the historical home design database, ensuring the collected samples are diverse and comprehensive in terms of spatial dimensions, structural layout, functional requirements, and style preferences. Each sample home furnishing request includes sample home environment parameters (such as house area, floor plan, and lighting conditions) and sample user demand parameters (such as budget range, style preferences, and functional requirements), which constitute the input dataset for model training. The scheme extraction subunit extracts sample home furnishing design schemes that are associated with multiple sample home furnishing requests from the historical home design database. These schemes are then aggregated to form a sample home furnishing design scheme set, which serves as the target output dataset for model training. Subsequently, a deep neural network structure suitable for home design tasks is designed, including a feature extraction layer, a spatial relationship modeling layer, and a design generation layer. Then, the model training subunit uses the sample home furnishing request set as model input and the sample home furnishing design scheme set as training target. The model parameters are optimized using the backpropagation algorithm, enabling the model to learn the mapping relationship between home environment parameters, user demand parameters, and design schemes, thereby obtaining the home design model.

[0084] Through the collaborative work of the above three sub-units, the model building unit 210 can build a home design model with home design capabilities, providing a reliable AI design engine for the first solution acquisition unit 220, thereby supporting the efficient operation of the entire solution generation module 200.

[0085] Furthermore, the second scheme acquisition unit 230 includes a condition construction subunit, a request query subunit, a similar scheme extraction subunit, and a weighted processing subunit.

[0086] The condition construction subunit is used to construct request query conditions based on the home environment parameters and the user demand parameters.

[0087] The request query subunit is used to perform a request query in the historical home design database based on the request query conditions to obtain multiple similar home generation requests;

[0088] The similarity scheme extraction subunit is used to extract home design schemes corresponding to multiple similar home generation requests, and obtain multiple similar home design schemes;

[0089] The weighted processing subunit is used to perform weighted processing on multiple similar home design schemes based on the similarity between multiple similar home generation requests and the home generation request, so as to obtain the second design scheme.

[0090] Specifically, the condition construction sub-unit performs feature analysis and weight labeling on home environment parameters and user demand parameters, extracting key search attributes such as space size range, main structural features, functional requirement types, and style preference categories. Then, through parameter normalization and vectorization, these attributes are converted into structured request query conditions, facilitating efficient and accurate similar case retrieval in the historical home design database. The request query sub-unit employs multi-dimensional index retrieval technology to quickly locate historical home design requests with high similarity to the current request query conditions in the historical home design database. A similarity calculation algorithm then filters out multiple similar home design requests that meet threshold requirements, serving as the basis for subsequent solution extraction.

[0091] The similarity scheme extraction subunit retrieves the design schemes corresponding to each similar home furnishing request from the historical home design database through the request-scheme mapping relationship, resulting in multiple similar home furnishing design schemes and providing a foundation for subsequent scheme fusion processing. Subsequently, the weighted processing subunit calculates the similarity score between each similar request and the current request, and then assigns weight coefficients to the corresponding design schemes based on these scores. Next, a weighted fusion algorithm selectively synthesizes the various components of the multiple similar home furnishing design schemes, prioritizing the retention of features from schemes with high similarity while filtering out elements from schemes with low similarity, ultimately generating a second design scheme that integrates the advantages of multiple similar home furnishing design schemes.

[0092] Furthermore, the scheme optimization module 400 includes a feedback analysis unit, a first adjustment determination unit, a first scheme adjustment unit, a first scheme verification unit, and a scheme output unit.

[0093] The feedback parsing unit is used to parse and process the feedback information to obtain the user's modification requirements.

[0094] The first adjustment determination unit is used to determine the first part to be adjusted in the initial design scheme according to the user's modification requirements;

[0095] The first adjustment unit is used to adjust the first part to be adjusted based on the user's modification requirements, and update the initial design scheme according to the first adjustment result to obtain the first updated design scheme;

[0096] The first scheme verification unit is used to verify whether the first updated design scheme meets the feedback information based on the user's modification requirements and the initial design scheme.

[0097] The solution output unit is used to output the first updated design solution as the target home design solution when the first updated design solution meets the feedback information.

[0098] Specifically, the feedback parsing unit receives feedback information from the feedback receiving module 300, processes the feedback information, and extracts specific and clear user modification requests from the feedback information. For example, the user may request specific modifications such as "moving the living room sofa to the window" or "changing the kitchen wall color to light gray." The first adjustment determination unit, based on the user's modification requests, locates the specific part that needs adjustment in the initial design scheme and determines it as the first part to be adjusted. For example, if the user requests to adjust the position of the living room sofa, this unit will determine the living room sofa and its surrounding area as the first part to be adjusted in the initial design scheme.

[0099] The first adjustment unit performs specific adjustments to the first part to be adjusted, such as moving furniture, changing the color scheme, and adjusting functional areas, to obtain the first adjustment result. After obtaining the first adjustment result, the first adjustment unit applies it to the initial design scheme to generate the first updated design scheme. Next, the first scheme verification unit verifies the first updated design scheme to check if it meets all the modification requirements raised by the user in the feedback information. When the first updated design scheme is verified to meet all the requirements in the feedback information, the scheme output unit outputs the first updated design scheme as the final target home design scheme, completing the optimization process. If the first updated design scheme fails to meet all feedback requirements, further iterative optimization is required.

[0100] Through an iterative optimization mechanism, the solution optimization module 400 can effectively adjust the initial design scheme based on user feedback until a target home design scheme that meets the user's needs is generated.

[0101] Furthermore, the scheme optimization module 400 also includes a second adjustment determination unit, a second scheme adjustment unit, a second scheme verification unit, and an iterative output unit.

[0102] The second adjustment determination unit is used to determine a second part to be adjusted in the first update design scheme according to the user's modification requirements when the first update design scheme does not meet the feedback information.

[0103] The second adjustment unit is used to adjust the second part to be adjusted based on the user's modification requirements, and update the first updated design scheme according to the second adjustment result to obtain the second updated design scheme;

[0104] The second scheme verification unit is used to verify whether the second updated design scheme meets the feedback information based on the user's modification requirements and the initial design scheme.

[0105] The iterative output unit is used for iterative execution until the Nth updated design scheme satisfies the feedback information, and then outputs the target home design scheme.

[0106] Specifically, when the first updated design scheme is found to be unable to fully meet all the modification requirements in the user feedback information, the scheme optimization module 400 will start a second round of adjustment process to further optimize the design scheme.

[0107] The second adjustment determination unit, based on the first updated design scheme, continues to analyze unmet user modification needs and identifies the second part to be adjusted within the first updated design scheme. By comparing the first updated design scheme with feedback information, the second adjustment determination unit identifies design elements or areas requiring further adjustment as the second part to be adjusted. This step-by-step adjustment strategy effectively avoids a global restructuring of the design scheme and improves optimization efficiency.

[0108] The second adjustment unit of the scheme performs precise adjustments to the second part to be adjusted, obtains the second adjustment result, and integrates the second adjustment result into the first updated design scheme to form the second updated design scheme. During the adjustment process, the second adjustment unit of the scheme will consider the results achieved in the first round of adjustments to ensure that the second round of adjustments will not destroy or reverse the already met design requirements, while addressing the remaining unmet needs in a targeted manner.

[0109] The second verification unit conducts a comprehensive verification of the second updated design scheme, checking whether it fully meets all the modification requirements raised by the user in the feedback information. The verification process uses the same technical means and standards as the first verification unit to ensure the quality and completeness of the design scheme.

[0110] The iterative output unit implements a loop iterative mechanism for scheme optimization. When the second updated design scheme still does not fully meet the feedback information, the subsequent adjustment-verification loop will continue to be executed, forming the third, fourth, and even the Nth updated design scheme, until an updated design scheme fully meets all the requirements in the user feedback information. At this point, the iterative output unit outputs the final updated design scheme that meets the conditions as the target home design scheme, completing the entire scheme optimization process.

[0111] Through a multi-round iterative optimization mechanism, the design scheme can be adjusted gradually and precisely to effectively address complex user needs, ensuring that the final target home design scheme can fully meet the user's personalized requirements and improve user satisfaction.

[0112] Furthermore, the system also includes a receiving module and a scheme storage module.

[0113] The receiving module is used to receive the target user's confirmation signal for the target home design scheme;

[0114] The scheme storage module is used to determine the signal based on the scheme and associate and store the home creation request, the home environment parameters, the user requirement parameters and the target home design scheme.

[0115] Specifically, after the scheme optimization module 400 completes the design scheme optimization and outputs the target home design scheme, it displays the final design scheme result to the target user through the user interface and provides a confirmation option. The target user can send a scheme confirmation signal by clicking the confirmation button or other interactive methods, indicating acceptance and approval of the target home design scheme. The confirmation receiving module responds in real time and receives the scheme confirmation signal. After receiving the scheme confirmation signal, the scheme storage module initiates the data storage process, establishing a correlation mapping between the initial home generation request, the parsed home environment parameters, user requirement parameters, and the final determined target home design scheme, storing it as a complete design case data package in the historical home design database. During storage, a unique identifier is assigned to this data package, and a multi-dimensional index is established to facilitate subsequent retrieval and utilization.

[0116] By defining the receiving module and the solution storage module, a complete closed loop of the design process is achieved. This not only outputs design solutions that meet user needs but also continuously accumulates successful cases into a historical database, providing richer reference resources for subsequent solution generation. This data recycling mechanism enables the system to continuously learn from successful cases and optimize the design model, improving the overall performance and design level of the AI ​​design system. Simultaneously, the associative storage function of the solution storage module facilitates source analysis and modification tracking of design solutions, helping to understand the evolution of design preferences and changing needs of different types of users, providing data support for long-term system optimization.

[0117] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Identical components are represented by the same reference numerals. Therefore, all equivalent changes made to the structure, shape, and principle of this application should be covered within the scope of protection of this application.

Claims

1. A smart home AI design optimization and generation system, characterized in that, The system includes: The request parsing module is used to respond to the home generation request from the terminal device, parse the home generation request, and obtain the target user's home environment parameters and user requirement parameters; The scheme generation module is used to generate an initial design scheme based on the home environment parameters and the user requirement parameters using a pre-trained home design model. A feedback receiving module is used to receive feedback information from the target user regarding the initial design scheme; The scheme optimization module is used to iteratively adjust the initial design scheme based on the feedback information to generate the target home design scheme.

2. The system according to claim 1, characterized in that, The request parsing module includes: An environment extraction unit is used to extract home space dimensions, structural layout, lighting conditions, and functional zoning from the home generation request as home environment parameters. The requirement extraction unit is used to extract user budget, style preference, functional requirements and resident population composition from the home furnishing generation request as user requirement parameters.

3. The system according to claim 1, characterized in that, The scheme generation module includes: Model building units are used to build home design models; The first scheme acquisition unit is used to input the home environment parameters and the user demand parameters into the home design model and output the first design scheme; The second scheme acquisition unit is used to match schemes in the historical home design database based on the home environment parameters and the user demand parameters to obtain a second design scheme. The scheme fusion unit is used to fuse the first design scheme and the second design scheme to obtain the initial design scheme.

4. The system according to claim 3, characterized in that, The model building unit includes: The sample collection subunit is used to collect multiple sample home generation requests from the historical home design database and construct a sample home generation request set. Each sample home generation request includes sample home environment parameters and sample user demand parameters. The scheme extraction subunit is used to extract the corresponding sample home design scheme from the historical home design database based on the multiple sample home generation requests, and construct a sample home design scheme set. The model training subunit is used to train the deep neural network model using supervised learning based on the sample home furnishing request set and the sample home furnishing design scheme set, so as to obtain the home furnishing design model.

5. The system according to claim 4, characterized in that, The second scheme acquisition unit includes: The condition construction subunit is used to construct request query conditions based on the home environment parameters and the user demand parameters; The request query subunit is used to perform a request query in the historical home design database based on the request query conditions to obtain multiple similar home generation requests; The similarity scheme extraction subunit is used to extract the home design schemes corresponding to multiple similar home generation requests, and obtain multiple similar home design schemes. The weighted processing subunit is used to perform weighted processing on multiple similar home design schemes based on the similarity between multiple similar home generation requests and the home generation request, so as to obtain the second design scheme.

6. The system according to claim 1, characterized in that, The scheme optimization module includes: The feedback parsing unit is used to parse and process the feedback information to obtain the user's modification requirements; The first adjustment determination unit is used to determine the first part to be adjusted in the initial design scheme according to the user's modification requirements; The first adjustment unit is used to adjust the first part to be adjusted based on the user's modification requirements, and update the initial design scheme according to the first adjustment result to obtain the first updated design scheme. The first scheme verification unit is used to verify whether the first updated design scheme meets the feedback information based on the user's modification requirements and the initial design scheme. The solution output unit is used to output the first updated design solution as the target home design solution when the first updated design solution meets the feedback information.

7. The system according to claim 6, characterized in that, The scheme optimization module also includes: The second adjustment determination unit is used to determine a second part to be adjusted in the first update design scheme according to the user's modification requirements when the first update design scheme does not meet the feedback information. The second adjustment unit is used to adjust the second part to be adjusted based on the user's modification requirements, and update the first updated design scheme according to the second adjustment result to obtain the second updated design scheme. The second scheme verification unit is used to verify whether the second updated design scheme meets the feedback information based on the user's modification requirements and the initial design scheme. An iterative output unit is used to iteratively execute until the Nth updated design scheme satisfies the feedback information, and then outputs the target home design scheme.

8. The system according to claim 1, characterized in that, The system also includes: The receiving module is configured to receive the target user's confirmation signal regarding the target home design scheme. The scheme storage module is used to associate and store the home creation request, the home environment parameters, the user requirement parameters, and the target home design scheme based on the scheme determination signal.

Images