Decoration design drawing acquisition method and device, electronic equipment and storage medium

Abstract

This disclosure provides a method, apparatus, electronic device, and storage medium for acquiring interior design drawings, relating to the field of computer technology. The specific implementation includes: acquiring an interior design request; using a large language model to obtain multiple interior design tasks with an execution order based on the interior design request; determining the current task from the multiple interior design tasks according to the execution order; using the large language model to obtain a control program corresponding to the current task; executing the control program to obtain intermediate data when the control program needs to be executed in a backend environment; wherein the intermediate data is used to be invoked when processing other interior design tasks whose execution order follows the current task; and sending the control program to a terminal device when the control program needs to be executed in a frontend environment; wherein the terminal device executes the control program to display the target interior design drawing on the terminal device.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a method, apparatus, electronic device, and storage medium for obtaining interior design drawings. Background Technology

[0002] Before any building renovation, creating interior design plans is an essential step. Currently, to improve the convenience of interior design, cloud-based design software is primarily used for creating these plans. However, in practice, users need to make numerous tedious adjustments using this software, including adjusting the spatial layout, the placement of items, and lighting effects. This process is complex and time-consuming. On the one hand, it reduces the efficiency of obtaining interior design plans; on the other hand, it requires users to be familiar with the various functions of the cloud-based design software and possess extensive professional knowledge of interior design, thus diminishing the user experience. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for obtaining interior design drawings to solve or alleviate one or more technical problems in the prior art.

[0004] Firstly, this disclosure provides a method for obtaining interior design drawings, including:

[0005] Obtain a renovation design request;

[0006] Using a large language model, multiple interior design tasks with an execution order are obtained based on the interior design request;

[0007] Determine the current task from among multiple interior design tasks according to the execution order;

[0008] Using a large language model, the control program corresponding to the current task is obtained;

[0009] When the control program needs to be executed in the backend environment, the control program is executed to obtain intermediate data; the intermediate data is used when processing other interior design tasks whose execution order is after the current task.

[0010] When the control program needs to be executed in the front-end environment, the control program is sent to the terminal device; the terminal device is used to execute the control program to display the target decoration design drawing on the terminal device.

[0011] Secondly, this disclosure provides a method for obtaining interior design drawings, including:

[0012] Generate a renovation design request;

[0013] The decoration design request is sent to the service device. The service device uses a large language model to obtain multiple decoration design tasks with an execution order based on the request. Following this order, it determines the current task from among the multiple tasks and then uses the large language model to obtain the corresponding control program. This control program is executed in a backend environment to obtain intermediate data, or sent to the terminal device in a frontend environment if it needs to be executed. The intermediate data is used when processing other decoration design tasks whose execution order follows the current task.

[0014] Upon receiving the control program sent by the service device, the control program is executed to display the target decoration design drawing.

[0015] Thirdly, this disclosure provides a device for acquiring interior design drawings, including:

[0016] The request retrieval unit is used to retrieve decoration design requests;

[0017] The task planning unit is used to utilize a large language model to obtain multiple interior design tasks with an execution order based on the interior design request.

[0018] The task selection unit is used to determine the current task from multiple interior design tasks according to the execution order;

[0019] The program acquisition unit is used to obtain the control program corresponding to the current task by utilizing the large language model;

[0020] The first program processing unit is used to execute the control program to obtain intermediate data when the control program needs to be executed in the backend environment; wherein, the intermediate data is used to be called when processing other decoration design tasks whose execution order is after the current task;

[0021] The second program processing unit is used to send the control program to the terminal device when the control program needs to be executed in the front-end environment; wherein the terminal device is used to execute the control program to display the target decoration design drawing on the terminal device.

[0022] Fourthly, this disclosure provides a device for acquiring interior design drawings, comprising:

[0023] The request generation unit is used to generate decoration design requests;

[0024] The request sending unit is used to send the decoration design request to the service device. The service device uses a large language model to obtain multiple decoration design tasks with an execution order based on the decoration design request. Following this order, it determines the current task from among the multiple decoration design tasks, and then uses the large language model to obtain the control program corresponding to the current task. This control program is executed in a backend environment to obtain intermediate data, or sent to the terminal device in a frontend environment if the control program needs to be executed. The intermediate data is used when processing other decoration design tasks whose execution order follows the current task.

[0025] The program execution unit is used to execute the control program upon receiving the control program sent by the service device, so as to display the target decoration design drawing.

[0026] Fifthly, an electronic device is provided, comprising:

[0027] At least one processor; and

[0028] The memory is communicatively connected to the at least one processor; wherein,

[0029] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform any of the methods described in the present disclosure.

[0030] In a sixth aspect, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform any of the methods according to embodiments of the present disclosure.

[0031] This disclosure utilizes a large language model to fully automate the most complex and time-intensive task planning and control process acquisition. On the one hand, it leverages the highly intelligent processing capabilities of the large language model to improve the efficiency of acquiring target interior design drawings; on the other hand, it significantly reduces human intervention, ensuring that neither software operation skills nor design expertise become obstacles to acquiring target interior design drawings, thus improving the user experience.

[0032] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0033] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments provided according to this disclosure and should not be construed as limiting the scope of this disclosure.

[0034] Figure 1 A flowchart illustrating a method for obtaining interior design drawings provided in this embodiment of the present disclosure. Figure 1 ;

[0035] Figure 2 A flowchart illustrating a method for obtaining interior design drawings provided in this embodiment of the present disclosure. Figure 2 ;

[0036] Figure 3 This is a schematic diagram of the overall architecture of a decoration design drawing acquisition system provided in an embodiment of the present disclosure;

[0037] Figure 4 A schematic diagram of the algorithm flow involved in a decoration design drawing acquisition system provided in this embodiment of the disclosure;

[0038] Figure 5 A schematic structural frame of a decoration design drawing acquisition device provided in this embodiment of the present disclosure. Figure 1 ;

[0039] Figure 6 A schematic structural frame of a decoration design drawing acquisition device provided in this embodiment of the present disclosure. Figure 2 ;

[0040] Figure 7 This is a schematic structural block diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0041] The present disclosure will now be described in further detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0042] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0043] As mentioned earlier, currently, to improve the convenience of interior design, cloud-based interior design software is mainly used to create interior design drawings. However, in actual operation, users need to make numerous tedious adjustments based on the software, including adjusting the spatial layout, the placement of items, and lighting effects. This process is complex and time-consuming. On the one hand, it reduces the efficiency of obtaining interior design drawings; on the other hand, it requires users to be familiar with the various functions of the software and possess extensive design expertise, thus diminishing the user experience.

[0044] Based on the above description, this disclosure proposes a method for obtaining decoration design drawings to solve or alleviate one or more technical problems in the prior art.

[0045] Specifically, Figure 1 A flowchart illustrating a method for obtaining interior design drawings provided in this embodiment of the present disclosure. Figure 1 This can be applied to service equipment. Here, service equipment can be a server, a workbench, a mainframe computer, or other similar computing devices. Furthermore, Figure 1 The methods for obtaining interior design drawings shown may include at least some of the following: Figure 1 Although a certain logical order is shown, in some cases, it can be executed in a different order. Figure 1 The steps shown or described.

[0046] like Figure 1 As shown, the methods for obtaining interior design drawings include:

[0047] Step S101: Obtain the decoration design request.

[0048] The decoration design request can be a decoration design drawing request sent by a user to a service device based on a terminal device. This request is used to request the drawing of decoration design drawings based on the original architectural drawings, according to the user's decoration design intentions. Here, the terminal device can be a conventional computer (e.g., a desktop computer, laptop computer, etc.), a vehicle-mounted computer, a tablet computer, a smartphone, a personal digital processing unit, or other similar computing device; the original architectural drawings can be the floor plan of the target building, or an initial decoration design drawing obtained based on the floor plan of the target building, and this disclosure does not limit this.

[0049] Step S102: Using the Large Language Model (LLM), based on the decoration design request, obtain multiple decoration design tasks with an execution order.

[0050] In this context, LLM can be a pre-trained neural network model that possesses general language knowledge, world knowledge, and professional knowledge in various fields (such as professional knowledge in the fields of architecture and interior design), and stores these in the form of parameters within the model.

[0051] In one example, an LLM could be an autoregressive generative model based on a Transformer architecture.

[0052] After receiving a decoration design request, the LLM can utilize its internally stored language knowledge, world knowledge, and professional knowledge in various fields to perform task planning based on the decoration design request, resulting in multiple decoration design tasks with an execution order. That is, according to the user's decoration design intention, it determines the multiple decoration design tasks that need to be executed when drawing decoration design drawings based on the original architectural drawings, as well as the execution order of the multiple decoration design tasks.

[0053] Step S103: Determine the current task from multiple interior design tasks according to the execution order.

[0054] In one example, the renovation design task that has been executed first in the execution order and has not yet been executed among multiple renovation design tasks can be designated as the current task. For instance, the multiple renovation design tasks include: Renovation Design Task A, Renovation Design Task B, and Renovation Design Task C. The execution order of the multiple renovation design tasks is: Renovation Design Task A > Renovation Design Task B > Renovation Design Task C. Assuming that the renovation design tasks that have not yet been executed among the multiple renovation design tasks include Renovation Design Task A, Renovation Design Task B, and Renovation Design Task C, then Renovation Design Task A can be designated as the current task. Further assuming that the renovation design tasks that have not yet been executed among the multiple renovation design tasks include Renovation Design Task B, then Renovation Design Task B can be designated as the current task. Finally, assuming that the renovation design tasks that have not yet been executed among the multiple renovation design tasks include Renovation Design Task C, then Renovation Design Task C can be designated as the current task.

[0055] Step S104: Use LLM to obtain the control program corresponding to the current task.

[0056] The control program can be either a data acquisition program or a display control program.

[0057] It should be noted that, in this embodiment of the disclosure, when the control program is a data acquisition program, the control program needs to be executed in the backend environment; when the control program is a display control program, the control program needs to be executed in the frontend environment. The backend environment is the operating environment provided by the service device; the frontend environment is the operating environment provided by the terminal device.

[0058] Step S105: If the control program needs to be executed in the backend environment, execute the control program to obtain intermediate data.

[0059] Intermediate data can be used when processing other interior design tasks that follow the current task in the execution order. Specifically, this data can include display resources and design parameters. Display resources can be model resources, such as furniture models, door and window models, lighting models, and bathroom fixture models; design parameters can be the positional parameters, dimensional parameters, and color parameters of each model resource.

[0060] Step S106: If the control program needs to be executed in the front-end environment, send the control program to the terminal device.

[0061] The terminal device is used to execute the control program to display the target decoration design drawing on the terminal device. Here, when the current task is not the decoration design task that is executed last in the execution order among multiple decoration design tasks, the target decoration design drawing can be further processed while processing other decoration design tasks that are executed after the current task to obtain the final decoration design drawing; when the current task is the decoration design task that is executed last in the execution order among multiple decoration design tasks, the target decoration design drawing can be used as the final decoration design drawing.

[0062] The decoration design drawing acquisition method provided in this disclosure can, after obtaining a decoration design request, utilize LLM (Limited Learning Model) to obtain multiple decoration design tasks with an execution order based on the request. Following the execution order, the current task is determined from these tasks. Then, using LLM, a control program corresponding to the current task is obtained. This control program can be executed in a backend environment to obtain intermediate data, or sent to a terminal device in a frontend environment to execute and display the target decoration design drawing. In other words, this disclosure fully automates the most complex and time-intensive task planning and control program acquisition stages using LLM. This improves the efficiency of acquiring the target decoration design drawing by leveraging the highly intelligent processing capabilities of LLM, while significantly reducing manual intervention. This eliminates barriers to acquiring the target decoration design drawing due to varying software operation skills and design expertise, thus enhancing the user experience.

[0063] In some alternative implementations, step S101, namely, "obtaining a decoration design request", may include:

[0064] Obtain the interior design request sent by the terminal device.

[0065] Using LLM, the interior design request is rewritten to obtain the rewritten interior design request.

[0066] Among them, the decoration design request can be a request for decoration design drawings sent by the user to the service device based on the terminal device. Therefore, there may be problems with unclear expression of intent.

[0067] Based on this, in this embodiment of the disclosure, after obtaining the decoration design request sent by the user via the terminal device, the decoration design request can be rewritten using LLM to achieve a clearer processing of the decoration design request, resulting in a rewritten decoration design request. The rewritten decoration design request can directly or indirectly include multiple key design elements, such as design basis, design scope, and design style. Here, the design basis can be a certain architectural drawing (e.g., the original architectural drawing); the design scope can be the overall scope represented by the design basis, or a partial scope represented by the design basis; the design style can be a medieval style, a modern minimalist style, a Chinese style, a mixed style, etc.

[0068] In one example, an LLM (Local Language Model) can be used to interact with the user through multiple rounds of question-and-answer sessions to rewrite the interior design request, ensuring that the rewritten request includes several key design elements. In another example, an LLM can be used to invoke other language models (e.g., Generalized Questions, Generative Pre-trained Transformer (GPT)) to leverage historical search and question-and-answer data to rewrite the interior design request, ensuring that the rewritten request includes multiple key design elements. The historical search and question-and-answer data can be user search data and / or question-and-answer data collected during the use of the interior design drawing acquisition system.

[0069] In this embodiment of the present disclosure, after receiving the decoration design request sent by the terminal device, the decoration design request can be rewritten using LLM to obtain a rewritten decoration design request. This ensures that the rewritten decoration design request does not have the problem of unclear intent representation. Thus, when executing step S102, namely, "using LLM to obtain multiple decoration design tasks with execution order based on the decoration design request," multiple decoration design tasks with execution order can be obtained using LLM based on the rewritten decoration design request. This ensures the reliability of the multiple decoration design tasks, thereby ensuring a high degree of consistency between the target decoration design drawing and the user's decoration design intent, further improving the user experience.

[0070] In some optional implementations, step S102, namely, "using LLM to obtain multiple decoration design tasks with an execution order based on the decoration design request," may include:

[0071] Obtain planning constraints;

[0072] Based on the planning constraints and the decoration design request, planning instructions were obtained.

[0073] By using LLM, renovation tasks are planned based on planning instruction information, resulting in multiple renovation design tasks with an execution sequence.

[0074] Among these, planning constraints may include format requirements, language style requirements, etc., for multiple interior design tasks.

[0075] After obtaining the planning constraints, planning instructions can be derived based on the interior design request and the planning constraints. In one example, planning instructions can be derived based on the modified interior design request, according to the planning constraints.

[0076] Through the above methods, in this embodiment of the disclosure, after obtaining the planning constraints, planning instruction information can be obtained based on the decoration design request according to the planning constraints. Then, using LLM (Limited Management Model), decoration tasks are planned based on the planning instruction information to obtain multiple decoration design tasks with an execution order, ensuring that the format, language style, etc., of the multiple decoration design tasks all conform to the planning constraints. This reduces the difficulty of subsequent processing of multiple decoration design tasks. For example, it can avoid or reduce errors and rework caused by the format, language style, etc., of multiple decoration design tasks, thereby making the entire decoration design drawing acquisition process smoother and further improving the efficiency of acquiring the target decoration design drawing.

[0077] In some optional implementations, step S104, namely, "using LLM to obtain the control program corresponding to the current task", may include:

[0078] Using LLM, identify the target design tools that correspond to the current task;

[0079] Using the target design tool, the control program corresponding to the current task is obtained.

[0080] In this embodiment of the disclosure, multiple candidate design tools can be pre-set, and each candidate design tool can process one or more types of decoration design tasks through its internally encapsulated logic program. Thus, after determining the current task, LLM can be used to determine the task type of the current task, and based on the task type, a target design tool corresponding to the current task can be determined from the multiple candidate design tools.

[0081] For example, the multiple candidate design tools include a first candidate design tool, a second candidate design tool, and a third candidate design tool. The first candidate design tool is used to handle interior design tasks of the model search class; the second candidate design tool is used to handle interior design tasks of the parameter determination class; and the third candidate design tool is used to handle interior design tasks of the display control class.

[0082] Suppose the current task is "to select a coffee table from a model library based on the style of a multi-seat sofa." Then, after using LLM to determine that the task type is a model search class, a first candidate design tool can be selected from multiple candidate design tools based on this task type, serving as the target design tool for the current task. Next, suppose the current task is "to determine the placement of the coffee table in the living room area." Then, after using LLM to determine that the task type is the parameter determination class, a second candidate design tool can be selected from multiple candidate design tools based on this task type, serving as the target design tool for the current task. Finally, suppose the current task is "to add the coffee table to the living room area based on its placement." Then, after using LLM to determine that the task type is a display control class, a third candidate design tool can be selected from multiple candidate design tools based on this task type, serving as the target design tool for the current task.

[0083] After determining the target design tool corresponding to the current task, the control program corresponding to the current task can be obtained using the target design tool.

[0084] In this embodiment of the present disclosure, LLM can be used to determine the target design tool corresponding to the current task, and the control program corresponding to the current task can be obtained using the target design tool. Since the target design tool is pre-set and does not need to be configured in real-time during the decoration design drawing acquisition process, the efficiency of acquiring the target decoration design drawing can be further improved.

[0085] Furthermore, as described above, in this embodiment of the disclosure, after determining the target design tool corresponding to the current task, the control program corresponding to the current task can be obtained using the target design tool. In an optional implementation, "obtaining the control program corresponding to the current task" may include:

[0086] Identify the target Application Programming Interface (API) required to perform the current task;

[0087] Obtain the target data resources required to execute the current task;

[0088] Based on the target API and target data resources, generate a control program corresponding to the current task.

[0089] The target API can be used to connect to specific functional objects (e.g., software or services). For example, if the current task type is model search, the target API required to execute the current task connects to a first functional object. Calling the first functional object allows the model library to be searched to obtain model resources that meet the requirements for display. As another example, if the current task type is parameter determination, the target API required to execute the current task connects to a second functional object. Calling the second functional object allows the design parameters that meet the requirements to be obtained. Yet another example, if the current task type is display control, the target API required to execute the current task connects to a third functional object. Calling the third functional object allows display operations on the front-end tool interface (e.g., displaying the target interior design drawing on the front-end tool interface).

[0090] Furthermore, in this embodiment of the disclosure, the target data resource may be the result of a task execution obtained when processing other interior design tasks whose execution order precedes the current task. Here, the task execution result may be display resources and / or parameter resources, and may be stored in a memory. Based on this, in one example, "obtaining the target data resource required to execute the current task" may include:

[0091] Determine the target task execution result required to execute the current task from the execution results of multiple tasks stored in the memory bank;

[0092] Based on the results of the target task execution, the target data resources required to execute the current task are obtained.

[0093] In a specific example, after obtaining the result of the target task execution, the target task execution result can be used as the target data resource required to execute the current task.

[0094] For example, if the current task is "to add the coffee table to the living room area based on its placement position in the living room area", then the first target task execution result required to execute the current task can be determined from the multiple task execution results stored in the memory bank, namely, the coffee table (which belongs to the display resource). At the same time, the second target task execution result required to execute the current task can be determined from the multiple task execution results stored in the memory bank, namely, the placement position of the coffee table in the living room area (which belongs to the parameter resource). Then, the first target task execution result and the second target task execution result are used together as the target data resource required to execute the current task.

[0095] After determining the target API required to execute the current task and obtaining the target data resources needed to execute the current task, a control program corresponding to the current task can be generated based on the target API and the target data resources. In one example, a control program corresponding to the current task can be generated based on the description information of the target API and the target data resources. The description information of the target API may include the interface parameters of the target API (e.g., query parameters, path parameters, request body parameters, and header parameters) and the interface name.

[0096] Through the above methods, in this embodiment of the disclosure, the target API required to execute the current task can be determined, and the target data resources required to execute the current task can be obtained. Then, based on the target API and target data resources, a control program corresponding to the current task can be generated. On the one hand, based on the target API and target data resources, a control program corresponding to the current task can be quickly generated to further improve the efficiency of obtaining the target decoration design drawings. On the other hand, the control program generated based on the target API and target data resources has a clear structure and well-defined logic, and can be executed quickly. Therefore, it can also further improve the efficiency of obtaining the target decoration design drawings. In addition, when obtaining the target data resources required to execute the current task, the target task execution result required to execute the current task can be determined from multiple task execution results stored in the memory bank, and the target data resources required to execute the current task can be obtained based on the target task execution result. That is to say, the target data resources are pre-stored in the memory bank, so they can be quickly obtained, thereby further improving the efficiency of obtaining the target decoration design drawings.

[0097] Based on the above description, in this embodiment of the disclosure, after executing step S105, that is, "when the control program needs to be executed in the backend environment, execute the control program to obtain intermediate data", the intermediate data can be stored in the memory as the task execution result corresponding to the current task. Alternatively, after executing step S106, that is, "when the control program needs to be executed in the frontend environment, send the control program to the terminal device" to execute the control program through the terminal device to display the target decoration design drawing on the terminal device, the target decoration design drawing can be obtained and stored in the memory as the task execution result corresponding to the current task. In this way, the purpose of expanding the memory can be achieved, thereby ensuring that the memory can have a rich data storage capacity.

[0098] In some optional implementations, step S104, namely, "using LLM to obtain the control program corresponding to the current task", may include:

[0099] Using LLM, we obtain the initial program corresponding to the current task;

[0100] Using LLM, the initial program was re-examined, and the re-examination results were obtained;

[0101] If the reflection results indicate that the initial program is an executable program, then the initial program will be used as the control program corresponding to the current task.

[0102] In one example, "using LLM to obtain the initial program corresponding to the current task" may include:

[0103] Using LLM, identify the primary objective design tool corresponding to the current task;

[0104] Using the first objective design tool, the initial program corresponding to the current task is obtained.

[0105] The process of "obtaining the initial program corresponding to the current task" can include: determining the first target API required to execute the current task, acquiring the first target data resources required to execute the current task, and then generating the initial program corresponding to the current task based on the first target API and the first target data resources. Here, "acquiring the first target data resources required to execute the current task" can further include: determining the first target task execution result required to execute the current task from multiple task execution results stored in memory, and obtaining the first target data resources required to execute the current task based on the first target task execution result.

[0106] The above process can be found in the description above regarding "using LLM to determine the target design tool corresponding to the current task; using the target design tool to obtain the control program corresponding to the current task", and will not be repeated here.

[0107] After obtaining the initial program corresponding to the current task, an LLM (Limited Language Management) can be used to reflect on the initial program and obtain a reflection result. That is, using the LLM, based on its internally stored language knowledge, world knowledge, and professional knowledge in various fields, the initial program is reflected on to obtain a reflection result, which characterizes whether the initial program is an executable program. If the reflection result indicates that the initial program is an executable program, then the initial program is used as the control program corresponding to the current task. It is understood that in this embodiment of the disclosure, if the initial program is an executable program, it means that the initial program has no logical errors and no environmental dependency issues, and therefore can be executed successfully.

[0108] Through the above methods, in this embodiment of the disclosure, an initial program corresponding to the current task can be obtained using LLM (Limited Learning Model). The initial program can then be reviewed using LLM to obtain a review result. If the review result indicates that the initial program is executable, then the initial program is used as the control program corresponding to the current task. In other words, in this embodiment of the disclosure, the initial program can only be used as the control program corresponding to the current task if the review result indicates that the initial program is executable. This avoids or reduces errors and rework caused by problems with the control program, thereby further improving the efficiency of obtaining the target decoration design drawings.

[0109] Furthermore, in this embodiment of the disclosure, after obtaining a decoration design request and using LLM to obtain multiple decoration design tasks with an execution order based on the decoration design request, and then determining the current task from the multiple decoration design tasks according to the execution order, the control program corresponding to the current task is obtained using LLM, and the control program is executed to obtain intermediate data when it needs to be executed in the backend environment, or sent to the terminal device when it needs to be executed in the frontend environment. The method for obtaining decoration design drawings may also include:

[0110] Receive progress notification text related to the current task;

[0111] Send a process notification text to the terminal device.

[0112] The terminal device is used to display process notification text.

[0113] In one example, the terminal device can display progress notification text in a user dialog box. This dialog box can also display interior design requests sent by the user via the terminal device.

[0114] In this embodiment of the present disclosure, during the processing of the current task, a progress notification text related to the current task can be obtained and sent to the terminal device so that the user can clearly understand the overall progress of the decoration design drawing acquisition process, thereby further improving the user experience.

[0115] This disclosure also proposes a method for obtaining interior design drawings, specifically, Figure 2 A flowchart illustrating a method for obtaining interior design drawings provided in this embodiment of the present disclosure. Figure 2 This can be applied to terminal devices. Here, terminal devices can be conventional computers (e.g., desktop computers, laptops, etc.), in-vehicle computers, tablets, smartphones, personal digital processors, or other similar computing devices. Furthermore, Figure 2The methods for obtaining interior design drawings shown may include at least some of the following: Figure 2 Although a certain logical order is shown, in some cases, it can be executed in a different order. Figure 2 The steps shown or described.

[0116] like Figure 2 As shown, the methods for obtaining interior design drawings include:

[0117] Step S201: Generate a decoration design request.

[0118] The decoration design request can be used to request the creation of decoration design drawings based on the original architectural drawings, according to the user's decoration design intentions. Here, the original architectural drawings can be the floor plan of the target building, or an initial decoration design drawing obtained based on the floor plan of the target building; this embodiment of the disclosure does not impose any limitations on this.

[0119] Step S202: Send the decoration design request to the service device.

[0120] The service device utilizes LLM (Limited Management Model) to obtain multiple interior design tasks with an execution order based on the interior design request. Following this order, it determines the current task from among the multiple tasks and then uses LLM to obtain the corresponding control program. This control program is executed in a backend environment to obtain intermediate data, or sent to the terminal device in a frontend environment. The intermediate data is used when processing other interior design tasks whose execution order follows the current task; specifically, it can include display resources and design parameters.

[0121] In step S203, upon receiving the control program sent by the service device, the control program is executed to display the target decoration design drawing.

[0122] Specifically, when the current task is not the last decoration design task in the execution order among multiple decoration design tasks, the target decoration design drawing can be further processed while processing other decoration design tasks that are executed after the current task to obtain the final decoration design drawing; when the current task is the last decoration design task in the execution order among multiple decoration design tasks, the target decoration design drawing can be used as the final decoration design drawing.

[0123] The decoration design drawing acquisition method provided in this disclosure can send a decoration design request to a service device after it is generated. The service device then uses an LLM (Limited Management Model) to obtain multiple decoration design tasks with an execution order based on the request. Following this order, it determines the current task and uses the LLM to obtain a control program corresponding to that task. This control program can be executed in a backend environment to obtain intermediate data, or sent to a terminal device in a frontend environment to execute and display the target decoration design drawing. In other words, this disclosure fully automates the most complex and time-intensive task planning and control program acquisition stages using an LLM. This improves the efficiency of acquiring the target decoration design drawing by leveraging the highly intelligent processing capabilities of the LLM, while significantly reducing manual intervention. This eliminates barriers to acquisition based on software operation skills or design expertise, thus enhancing the user experience.

[0124] The following, combined with Figure 3 and Figure 4 The process for obtaining decoration design drawings provided in the embodiments of this disclosure will be described.

[0125] In this embodiment of the disclosure, the decoration design drawing acquisition system may include four parts:

[0126] (1) Front-end part (that is, the part that needs to be implemented on the terminal device side)

[0127] The front-end component can include interactive tools, front-end tools, and front-end mini-programs.

[0128] The interactive tool can be a user dialog box, which can be used to receive interior design requests sent by users via terminal devices, or to display progress notification text related to the current task.

[0129] The front-end tool can provide a browsing interface where users can see the dynamic process of the front-end tool automatically performing related operations. For example, the process of a terminal device executing a control program to display a target interior design drawing.

[0130] A front-end mini-program can be an application developed based on common web development technologies. It exposes a front-end API, allowing users to perform related operations on the front-end tool interface. For example, after receiving a control program sent by a service device, the user can use the front-end API carried in the control program to call the functional object (e.g., software or service) connected to that front-end API to manipulate model objects (e.g., furniture models, door and window models, lighting models, bathroom facility models, etc.) on the front-end tool interface to obtain the target decoration design drawing to be displayed.

[0131] (2) Backend section

[0132] The backend primarily acts as a proxy, situated between the frontend and the model / algorithm section, enabling bidirectional real-time communication between the two. For example, the backend can utilize a full-duplex communication protocol (e.g., WebSocket Session communication protocol) to facilitate this bidirectional real-time communication.

[0133] (3) Model algorithm part (that is, the part that needs to be implemented on the service device side)

[0134] The model algorithm part includes multiple functional components based on LLM, such as request rewriting component, task planning component, task execution component, task reflection component and result parsing component. It may also include a memory and multiple candidate design tools (Tool component), as well as the LLM as the central driver.

[0135] The process involves a user sending a renovation design request via a terminal device, displaying it through an interactive tool, and then sending it to the backend using the request sending unit provided by the front-end mini-program. The backend then uses a first communication unit to forward the request to the model algorithm section. In the model algorithm section, a request rewriting component rewrites the renovation design request, ensuring that the rewritten request does not have unclear intent. For example, a renovation design request like "Based on the current main sofa, create a coffee table arrangement for the living room" can be rewritten as "Given a multi-seat sofa, please use this sofa as a reference style to arrange a coffee table in the living room area of ​​the original architectural drawings." Here, the design basis is the original architectural drawings; the design scope is the living room area; and the design style is the style of the multi-seat sofa. After obtaining the rewritten renovation design request, a task planning component can be used to generate multiple renovation design tasks with an execution order based on the rewritten request. In one example, planning constraints can be obtained, and based on the rewritten interior design request and according to these constraints, planning instruction information (i.e., Prompt information) can be derived. Then, using LLM (Limited Management Model), interior design tasks can be planned based on this instruction information, resulting in multiple interior design tasks with an execution order (e.g., using LLM with a historical planning database as a reference, interior design tasks can be planned based on the instruction information to obtain multiple interior design tasks with an execution order). The planning constraints may include format requirements, language style requirements, etc., for the multiple interior design tasks; the historical planning database stores historical planning cases.

[0136] For example, a rewritten interior design request might be: "A multi-seat sofa is provided. Please use this sofa as a reference style to arrange a coffee table in the living room area of ​​the original architectural drawings." The planning instructions could be: Based on this interior design request, task planning is performed to obtain multiple interior design tasks with an execution order, ensuring that each interior design task is concise and clear. Ultimately, the resulting multiple interior design tasks could include:

[0137] a. Locate a multi-seat sofa in the living room area from the original architectural drawings;

[0138] b. Determine the style of the multi-seat sofa;

[0139] c. Based on the style of the multi-seat sofa, select a coffee table from the model library;

[0140] d. Determine the placement of the coffee table in the living room area;

[0141] e. Based on the coffee table's placement in the living room area, add the coffee table to the living room area.

[0142] Subsequently, multiple interior design tasks can be executed using the task execution component. Specifically, after determining the current task from multiple interior design tasks according to the execution order, LLM is used to determine the target design tool corresponding to the current task from multiple candidate design tools, and the control program corresponding to the current task is obtained using the target design tool. "Obtaining the control program corresponding to the current task" can include: determining the target API required to execute the current task, obtaining the target data resources required to execute the current task, and then generating the control program corresponding to the current task based on the target API and target data resources. Here, the target API required to execute the current task can be determined from multiple candidate APIs (including at least one front-end API and at least one back-end API); "obtaining the target data resources required to execute the current task" can further include: determining the target task execution result required to execute the current task from multiple task execution results stored in the memory, and obtaining the target data resources required to execute the current task based on the target task execution result.

[0143] After obtaining the control program corresponding to the current task, the task reflection component can be used to reflect on the control program. If the reflection result indicates that the control program is an executable program, the control program is taken as the final control program corresponding to the current task (the control program referred to here is the final control program). The control program is then output through the result parsing component.

[0144] Specifically, when the control program needs to be executed in the backend environment, it can be executed directly to obtain intermediate data, which is then stored in the memory. Simultaneously, process notification text related to the current task can be obtained and sent to the terminal device. For example, the second communication unit provided by the backend can be used to send the process notification text to the frontend mini-program, which then sequentially sends the text to the interactive tool via the data transmission and reception control unit, data distribution unit, and frontend display preparation unit.

[0145] When the control program needs to be executed in the front-end environment, the control program and the process notification text related to the current task can be sent to the terminal device. For example, the second communication unit provided by the back-end can be used to send the control program and the process notification text related to the current task to the front-end applet. In the front-end applet, the control program is sent to the front-end tool in sequence through the data transmission and control unit, the data distribution unit, and the control program sending unit. The process notification text is sent to the interactive tool in sequence through the data transmission and control unit, the data distribution unit, and the front-end display preparation unit. After that, the terminal device can execute the control program to display the target decoration design drawing on the front-end tool. The target decoration design drawing is then sent as the task execution result corresponding to the current task to the back-end through the result sending unit and the data transmission and control component in the front-end applet. Finally, it is sent to the model algorithm part through the first communication component provided by the back-end and stored in the memory bank.

[0146] After multiple interior design tasks have been completed, the backend can be notified to disconnect the bidirectional real-time communication between the frontend and the model algorithm to end the current interior design request.

[0147] (4) Multiple candidate APIs

[0148] Multiple candidate APIs include at least one front-end API and at least one back-end API. When the model algorithm part obtains the control program corresponding to the current task using the target design tool, it determines the target API required to execute the current task and obtains the target data resources required to execute the current task. Then, based on the target API and target data resources, it generates the control program corresponding to the current task. That is, it encapsulates the target API and target data resources to generate the control program corresponding to the current task.

[0149] Among them, the front-end API is the API of the front-end mini-program; the back-end API is the back-end API of the decoration design drawing acquisition system.

[0150] To better implement the above-described method for acquiring interior design drawings performed by service equipment, this disclosure also provides an interior design drawing acquisition device, which can be integrated into the service equipment. The following will be combined with... Figure 5 The schematic structural block diagram shown illustrates a decoration design drawing acquisition device 500 provided in the disclosed embodiment.

[0151] Decoration design drawing acquisition device 500, including:

[0152] Request retrieval unit 501 is used to retrieve decoration design requests;

[0153] Task planning unit 502 is used to generate multiple interior design tasks with an execution order based on interior design requests using LLM.

[0154] Task selection unit 503 is used to determine the current task from multiple decoration design tasks according to the execution order;

[0155] The program acquisition unit 504 is used to obtain the control program corresponding to the current task using LLM;

[0156] The first program processing unit 505 is used to execute the control program to obtain intermediate data when the control program needs to be executed in the backend environment; wherein, the intermediate data is used to be called when processing other decoration design tasks whose execution order is after the current task;

[0157] The second program processing unit 506 is used to send the control program to the terminal device when the control program needs to be executed in the front-end environment; wherein the terminal device is used to execute the control program to display the target decoration design drawing on the terminal device.

[0158] In some optional implementations, the program acquisition unit 504 is used for:

[0159] Using LLM, identify the target design tools that correspond to the current task;

[0160] Using the target design tool, the control program corresponding to the current task is obtained.

[0161] In some optional implementations, the program acquisition unit 504 is used for:

[0162] Determine the target API required to perform the current task;

[0163] Obtain the target data resources required to execute the current task;

[0164] Based on the target API and target data resources, generate a control program corresponding to the current task.

[0165] In some optional implementations, the program acquisition unit 504 is used for:

[0166] Determine the target task execution result required to execute the current task from the execution results of multiple tasks stored in the memory bank;

[0167] Based on the results of the target task execution, the target data resources required to execute the current task are obtained.

[0168] In some alternative implementations, the decoration design drawing acquisition device 500 further includes a result storage unit for one of the following:

[0169] The intermediate data is stored in the memory as the task execution result corresponding to the current task;

[0170] Obtain the target interior design drawing, and store it in memory as the result of the task execution corresponding to the current task.

[0171] In some optional implementations, the program acquisition unit 504 is used for:

[0172] Using LLM, we obtain the initial program corresponding to the current task;

[0173] Using LLM, the initial program was re-examined, and the re-examination results were obtained;

[0174] If the reflection results indicate that the initial program is an executable program, then the initial program will be used as the control program corresponding to the current task.

[0175] In some alternative implementations, the decoration design drawing acquisition device 500 further includes a process notification unit for:

[0176] Receive progress notification text related to the current task;

[0177] Send the process notification text to the terminal device; the terminal device is used to display the process notification text.

[0178] In some optional implementations, the request acquisition unit 501 is used for:

[0179] Obtain the interior design request sent by the terminal device;

[0180] Using LLM, the interior design request is rewritten to obtain the rewritten interior design request;

[0181] Using LLM, based on the interior design request, multiple interior design tasks with an execution order are obtained, including:

[0182] Using LLM, multiple interior design tasks with an execution order are obtained based on the rewritten interior design request.

[0183] In some alternative implementations, the task planning unit 502 is used for:

[0184] Obtain planning constraints;

[0185] Based on the planning constraints and the decoration design request, planning instructions were obtained.

[0186] By using LLM, renovation tasks are planned based on planning instruction information, resulting in multiple renovation design tasks with an execution sequence.

[0187] The specific functions and examples of each unit of the decoration design drawing acquisition device 500 in this embodiment can be found in the description of the relevant steps in the method embodiment corresponding to the decoration design drawing acquisition device 500, and will not be repeated here.

[0188] To better implement the above-described method for acquiring interior design drawings executed by a terminal device, this disclosure also provides an interior design drawing acquisition device, which can be integrated into the terminal device. The following will be combined with... Figure 6 The schematic structural block diagram shown illustrates a decoration design drawing acquisition device 600 provided in the disclosed embodiment.

[0189] Decoration design drawing acquisition device 600, including:

[0190] Request generation unit 601 is used to generate decoration design requests;

[0191] The request sending unit 602 is used to send a decoration design request to a service device. The service device uses an LLM (Limited Management Model) to obtain multiple decoration design tasks with an execution order based on the decoration design request. Following this order, it determines the current task from among the multiple decoration design tasks, and then uses the LLM to obtain a control program corresponding to the current task. This control program is executed in a backend environment to obtain intermediate data, or it is sent to a terminal device in a frontend environment if it needs to be executed. The intermediate data is used when processing other decoration design tasks whose execution order follows that of the current task.

[0192] The program execution unit 603 is used to execute the control program upon receiving the control program sent by the service device, so as to display the target decoration design drawing.

[0193] The specific functions and examples of each unit of the decoration design drawing acquisition device 600 in this embodiment can be found in the description of the relevant steps in the method embodiment corresponding to the decoration design drawing acquisition device, and will not be repeated here.

[0194] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0195] Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of this disclosure. Figure 7As shown, the electronic device includes a memory 701 and a processor 702. The memory 701 stores a computer program that can run on the processor 702. The number of memories 701 and processors 702 can be one or more. The memory 701 can store one or more computer programs, which, when executed by the electronic device, cause the electronic device to perform the methods provided in the above-described method embodiments. The electronic device may also include a communication interface 703 for communicating with external devices and performing data exchange and transmission.

[0196] If the memory 701, processor 702, and communication interface 703 are implemented independently, they can be interconnected via a bus to communicate with each other. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 7 The symbol is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0197] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0198] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors can be microprocessors or any conventional processor. It is worth noting that the processor can be a processor supporting Advanced Reduced Instruction Set Machines (ARM) architecture.

[0199] Further, optionally, the aforementioned memory may include read-only memory and random access memory, and may also include non-volatile random access memory. The memory may be volatile or non-volatile, or may include both. Non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which serves as an external cache. Many forms of RAM are available by way of example, but not limitation. Examples include Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct RAMBUS RAM (DR RAM).

[0200] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line, DSL) or wireless (e.g., infrared, Bluetooth, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer, or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)). It is worth noting that the computer-readable storage media mentioned in this disclosure can be non-volatile storage media; in other words, it can be non-transient storage media.

[0201] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0202] In the description of the embodiments of this disclosure, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0203] In the description of the embodiments disclosed herein, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0204] In the description of embodiments of this disclosure, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include one or more of that feature. In the description of embodiments of this disclosure, unless otherwise stated, "a plurality of" means two or more.

[0205] The above description is merely an exemplary embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A method for obtaining interior design drawings, comprising: Obtain a decoration design request; wherein, the decoration design request is a request for decoration design drawing sent by the user to the service device based on the terminal device; Using a large language model, multiple interior design tasks with an execution order are obtained based on the interior design request; The current task is determined from the plurality of interior design tasks according to the execution order; Using the large language model, a control program corresponding to the current task is obtained; wherein, if the control program is a data acquisition program, it is determined that the control program needs to be executed in a backend environment; if the control program is a display control program, it is determined that the control program needs to be executed in a frontend environment; the backend environment is the operating environment provided by the service device; the frontend environment is the operating environment provided by the terminal device. If the control program needs to be executed in the backend environment, the control program is executed to obtain intermediate data; wherein, the intermediate data is used to retrieve other interior design tasks whose execution order is after the current task. When the control program needs to be executed in the front-end environment, the control program is sent to the terminal device; wherein, the terminal device is used to execute the control program to display the target decoration design drawing on the terminal device; when the current task is not the decoration design task with the latest execution order among the multiple decoration design tasks, the target decoration design drawing is further processed while processing other decoration design tasks with the latest execution order to obtain the final decoration design drawing; when the current task is the decoration design task with the latest execution order among the multiple decoration design tasks, the target decoration design drawing is used as the final decoration design drawing; The step of using a large language model to obtain the control program corresponding to the current task includes: Using the large language model, the task type of the current task is determined, and based on the task type of the current task, a target design tool corresponding to the current task is determined from multiple candidate design tools; wherein, each of the multiple candidate design tools can process at least one type of decoration design task through its internally encapsulated logic program; the multiple candidate design tools include a first candidate design tool, a second candidate design tool, and a third candidate design tool, wherein the first candidate design tool is used to process decoration design tasks of the model search type, the second candidate design tool is used to process decoration design tasks of the parameter determination type, and the third candidate design tool is used to process decoration design tasks of the display control type; Using the target design tool, a control program corresponding to the current task is obtained.

2. The method according to claim 1, wherein, The process of obtaining the control program corresponding to the current task includes: Determine the target application programming interface required to perform the current task; Obtain the target data resources required to execute the current task; Based on the target application programming interface and the target data resources, a control program corresponding to the current task is generated.

3. The method according to claim 2, wherein, The acquisition of the target data resources required to execute the current task includes: From the execution results of multiple tasks stored in the memory bank, determine the target task execution result required to execute the current task; Based on the execution result of the target task, the target data resources required to execute the current task are obtained.

4. The method according to claim 3, further comprising one of the following: The intermediate data is stored in the memory bank as the task execution result corresponding to the current task. Obtain the target interior design drawing, and store the target interior design drawing as the task execution result corresponding to the current task in the memory.

5. The method according to claim 1, wherein, The step of using the large language model to obtain the control program corresponding to the current task includes: Using the large language model, an initial program corresponding to the current task is obtained; Using the large language model, the initial program is reflected upon to obtain the reflection results; If the reflection results indicate that the initial program is an executable program, then the initial program is used as the control program corresponding to the current task.

6. The method according to claim 1, further comprising: Receive the process notification text related to the current task; The process notification text is sent to the terminal device; wherein the terminal device is used to display the process notification text.

7. The method according to claim 1, wherein, The process of obtaining the interior design request includes: Obtain the interior design request sent by the terminal device; Using the large language model, the decoration design request is rewritten to obtain the rewritten decoration design request; The method of utilizing a large language model to obtain multiple interior design tasks with an execution order based on the interior design request includes: Using a large language model, multiple interior design tasks with an execution order are obtained based on the rewritten interior design request.

8. The method according to claim 1 or 7, wherein, The method of utilizing a large language model to obtain multiple interior design tasks with an execution order based on the interior design request includes: Obtain planning constraints; Based on the planning constraints and the decoration design request, planning instruction information is obtained. Using a large language model, decoration tasks are planned based on the planning instruction information to obtain multiple decoration design tasks with an execution order.

9. A method for obtaining interior design drawings, comprising: Generate a renovation design request; Based on the terminal device, the decoration design request is sent to the service device. The service device uses a large language model to obtain multiple decoration design tasks with an execution order based on the decoration design request. Following this execution order, it determines the current task from among the multiple decoration design tasks and then uses the large language model to obtain the control program corresponding to the current task. If the control program is determined to be a data acquisition program and therefore needs to be executed in a backend environment, the control program is executed to obtain intermediate data. Alternatively, if the control program is determined to be a display control program and therefore needs to be executed in a frontend environment, the control program is sent to the terminal device. The intermediate data is used when processing other decoration design tasks whose execution order follows the current task. The backend environment is the runtime environment provided by the service device; the frontend environment is the runtime environment provided by the terminal device. Upon receiving a control program sent by the service device, the control program is executed to display the target decoration design drawing; wherein, when the current task is not the decoration design task with the latest execution order among the multiple decoration design tasks, the target decoration design drawing is further processed while processing other decoration design tasks with the latest execution order, to obtain the final decoration design drawing; when the current task is the decoration design task with the latest execution order among the multiple decoration design tasks, the target decoration design drawing is used as the final decoration design drawing; The step of using a large language model to obtain the control program corresponding to the current task includes: Using the large language model, the task type of the current task is determined, and based on the task type of the current task, a target design tool corresponding to the current task is determined from multiple candidate design tools; wherein, each of the multiple candidate design tools can process at least one type of decoration design task through its internally encapsulated logic program; the multiple candidate design tools include a first candidate design tool, a second candidate design tool, and a third candidate design tool, wherein the first candidate design tool is used to process decoration design tasks of the model search type, the second candidate design tool is used to process decoration design tasks of the parameter determination type, and the third candidate design tool is used to process decoration design tasks of the display control type; Using the target design tool, a control program corresponding to the current task is obtained.

10. A device for acquiring interior design drawings, comprising: The request acquisition unit is used to acquire decoration design requests; wherein, the decoration design request is a request for drawing decoration design drawings sent by the user to the service device based on the terminal device; the task planning unit is used to use a large language model to obtain multiple decoration design tasks with execution order based on the decoration design request. The task selection unit is used to determine the current task from the plurality of decoration design tasks according to the execution order; A program acquisition unit is used to obtain a control program corresponding to the current task using the large language model; wherein, when the control program is a data acquisition program, it is determined that the control program needs to be executed in a backend environment; when the control program is a display control program, it is determined that the control program needs to be executed in a frontend environment; the backend environment is the operating environment provided by the service device; the frontend environment is the operating environment provided by the terminal device; a first program processing unit is used to execute the control program to obtain intermediate data when the control program needs to be executed in the backend environment; wherein, the intermediate data is used to be invoked when processing other decoration design tasks whose execution order is after the current task; The second program processing unit is used to send the control program to the terminal device when the control program needs to be executed in the front-end environment; wherein the terminal device is used to execute the control program to display the target decoration design drawing on the terminal device; when the current task is not the decoration design task with the latest execution order among the multiple decoration design tasks, the target decoration design drawing is further processed while processing other decoration design tasks with the latest execution order after the current task to obtain the final decoration design drawing; when the current task is the decoration design task with the latest execution order among the multiple decoration design tasks, the target decoration design drawing is used as the final decoration design drawing; The program acquisition unit is specifically used for: Using the large language model, the task type of the current task is determined, and based on the task type of the current task, a target design tool corresponding to the current task is determined from multiple candidate design tools; wherein, each of the multiple candidate design tools can process at least one type of decoration design task through its internally encapsulated logic program; the multiple candidate design tools include a first candidate design tool, a second candidate design tool, and a third candidate design tool, wherein the first candidate design tool is used to process decoration design tasks of the model search type, the second candidate design tool is used to process decoration design tasks of the parameter determination type, and the third candidate design tool is used to process decoration design tasks of the display control type; Using the target design tool, a control program corresponding to the current task is obtained.

11. A device for acquiring interior design drawings, comprising: The request generation unit is used to generate decoration design requests; A request sending unit is configured to send the interior design request to a service device based on a terminal device. The service device uses a large language model to obtain multiple interior design tasks with an execution order based on the request. Following this execution order, it determines the current task from among the multiple interior design tasks and then uses the large language model to obtain a control program corresponding to the current task. If the control program is determined to be a data acquisition program and therefore needs to be executed in a backend environment, the control program is executed to obtain intermediate data. Alternatively, if the control program is determined to be a display control program and therefore needs to be executed in a frontend environment, the control program is sent to the terminal device. The intermediate data is used to process other interior design tasks whose execution order follows that of the current task. The backend environment is the runtime environment provided by the service device. The frontend environment is the runtime environment provided by the terminal device. The program execution unit is configured to execute the control program sent by the service device upon receiving the control program, in order to display the target decoration design drawing; wherein, when the current task is not the decoration design task with the latest execution order among the plurality of decoration design tasks, the target decoration design drawing is further processed while processing other decoration design tasks with the latest execution order, to obtain the final decoration design drawing; when the current task is the decoration design task with the latest execution order among the plurality of decoration design tasks, the target decoration design drawing is used as the final decoration design drawing; The step of using a large language model to obtain the control program corresponding to the current task includes: Using the large language model, the task type of the current task is determined, and based on the task type of the current task, a target design tool corresponding to the current task is determined from multiple candidate design tools; wherein, each of the multiple candidate design tools can process at least one type of decoration design task through its internally encapsulated logic program; the multiple candidate design tools include a first candidate design tool, a second candidate design tool, and a third candidate design tool, wherein the first candidate design tool is used to process decoration design tasks of the model search type, the second candidate design tool is used to process decoration design tasks of the parameter determination type, and the third candidate design tool is used to process decoration design tasks of the display control type; Using the target design tool, a control program corresponding to the current task is obtained.

12. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 9.

13. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 9.

Images