Urban facility updating method and device based on large language model, equipment and medium

Abstract

The invention provides an urban facility updating method and device based on a large language model, equipment and a medium, and relates to the technical field of urban planning, and the method comprises the steps: inputting a division cue word into a plan large language model, and obtaining a plurality of communities corresponding to a to-be-planned urban region outputted by the plan large language model; the division prompt words are used for indicating the plan big language model to divide the to-be-planned city area into a plurality of communities; the plan big language model is constructed based on a general multi-modal big language model; for each community, determining a target facility updating sub-scheme corresponding to the community through a plan big language model; and based on the target facility updating sub-schemes corresponding to all communities, determining an urban facility updating scheme corresponding to the to-be-planned urban area. According to the technical scheme, the urban facility updating scheme is constructed through the large language model, participation of experts and residents is not needed, negative influences caused by human factors are eliminated, and the efficiency and accuracy of planning the urban facility updating scheme are improved.

Description

Technical Field

[0001] The present invention relates to the field of urban planning technology, and in particular to a method, device, equipment and medium for updating urban facilities based on a large language model. Background Art

[0002] Urban facility renewal refers to the updating of facility types within urban areas to address urban issues such as functional degradation and inadequate infrastructure. In this process, participatory urban planning is often employed to fully consider the opinions of all stakeholders.

[0003] However, traditional participatory urban planning methods require multiple discussions and the involvement of planning experts and residents. This is time-consuming, labor-intensive, and inefficient, often taking weeks or even months. Furthermore, there are challenges such as low resident willingness to participate and participants' lack of planning expertise, hindering discussions and planning effectiveness, leading to less-than-perfect urban infrastructure renewal plans. Summary of the Invention

[0004] The present invention provides a method, device, equipment and medium for urban facility renewal based on a large language model, which is used to solve the defects of the existing technology that require the participation of experts and residents, have low residents' willingness to participate, and participants lack planning expertise. The technical solution of the present invention constructs an urban facility renewal plan through a large language model, does not require the participation of experts and residents, eliminates the negative impact of human factors, and improves the efficiency and accuracy of planning urban facility renewal plans.

[0005] The present invention provides a method for updating urban facilities based on a large language model, comprising the following steps.

[0006] Inputting a division prompt word into a large language model for planning engineers to obtain a plurality of communities corresponding to the urban area to be planned outputted by the large language model for planning engineers; the division prompt word is used to instruct the large language model for planning engineers to divide the urban area to be planned into the plurality of communities; the large language model for planning engineers is constructed based on a general multimodal large language model; For each of the communities, determining a target facility renewal sub-plan corresponding to the community through the planner's language model; Based on the target facility renewal sub-plans corresponding to all the communities, an urban facility renewal plan corresponding to the urban area to be planned is determined.

[0007] According to a method for urban facility renewal based on a large language model provided by the present invention, determining the target facility renewal sub-plan corresponding to the community through the planner's large language model includes: Execute the actions selected by the planner's language model in the preset action space to determine the target facility update sub-plan corresponding to the community; the actions in the preset action space include investigating residents, initiating discussions, evaluating plans, modifying plans, and ending updates.

[0008] According to the urban facility renewal method based on a large language model provided by the present invention, the urban area to be planned includes a plurality of plots, each of which corresponds to a plot number; The resident survey action includes: determining a plurality of resident numbers corresponding to the community based on at least one plot number corresponding to the community; The initiating discussion action includes: inputting all the resident numbers corresponding to the community into the planner's language model to obtain multiple target resident numbers output by the planner's language model; inputting preset discussion prompt words into the resident language model corresponding to each target resident number to obtain the discussion results corresponding to each target resident number; inputting all the discussion results into the preset language model to obtain the resident discussion summary result output by the preset language model, and inputting the resident discussion summary result into the planner's language model; the preset language model and the resident language model are both constructed based on the general language model; The evaluation scheme action includes: determining a planning requirement evaluation result corresponding to the current facility renewal sub-scheme based on preset planning requirements; determining a resident satisfaction index corresponding to the current facility renewal sub-scheme based on the initial demand facilities corresponding to each of the resident language models corresponding to the community; and inputting the planning requirement evaluation result and the resident satisfaction index into the planner language model; The modification plan action includes: determining a new facility renewal sub-plan based on the resident discussion summary results and / or the planning requirement evaluation results and the resident satisfaction index; determining the new facility renewal sub-plan as the current facility renewal sub-plan, and executing the evaluation plan action; The ending update action includes: determining the latest determined current facility update sub-plan as the target facility update sub-plan corresponding to the community.

[0009] According to a method for urban facility renewal based on a large language model provided by the present invention, determining a planning requirement evaluation result corresponding to a current facility renewal sub-scheme based on preset planning requirements includes: For each of the plots corresponding to the community, if the facility area corresponding to the plot in the current facility renewal sub-plan is greater than the total area of ​​the plots corresponding to the plot, the plot number corresponding to the plot is determined as the first problem plot number; if the plot includes unsuitable adjacent facilities in the current facility renewal sub-plan, the plot number corresponding to the plot is determined as the second problem plot number; All of the first problem plot numbers and all of the second problem plot numbers are determined as the planning requirement evaluation results.

[0010] According to a method for urban facility renewal based on a large language model provided by the present invention, determining the resident satisfaction index corresponding to the current facility renewal sub-scheme based on the initial demand facilities corresponding to each of the large language models of the residents corresponding to the community includes: For each of the resident language models corresponding to the community, determining a target proportion of the initial demand facilities corresponding to the resident language model in the current facility update sub-plan; The average value of the target proportions corresponding to all the resident large language models is determined as the resident satisfaction index corresponding to the current facility renewal sub-plan.

[0011] According to a method for updating urban facilities based on a large language model provided by the present invention, the method further includes: Obtaining demographic data corresponding to the urban area to be planned, and generating multiple resident portraits based on the demographic data; Inputting all the resident portraits into the preset large language model to obtain the resident prompt words corresponding to each of the resident portraits output by the preset large language model; Each of the resident prompt words is input into a plurality of universal large language models respectively to obtain all the resident large language models corresponding to the urban area to be planned.

[0012] According to a method for updating urban facilities based on a large language model provided by the present invention, the method further includes: Determining a planner prompt based on a city map corresponding to the urban area to be planned, text descriptions corresponding to each block, and the preset planning requirements; The planner prompt words are input into a universal multimodal large language model to obtain the planner large language model.

[0013] The present invention also provides an urban facility renewal device based on a large language model, comprising the following modules: a partitioning module, configured to input a partitioning prompt word into a large language model for planning engineers, and obtain a plurality of communities corresponding to the urban area to be planned outputted by the large language model for planning engineers; the partitioning prompt word is used to instruct the large language model for planning engineers to partition the urban area to be planned into the plurality of communities; the large language model for planning engineers is constructed based on a general multimodal large language model; A first determination module is configured to determine, for each community, a target facility renewal sub-plan corresponding to the community using the planner's language model; The second determination module is used to determine the urban facility renewal plan corresponding to the urban area to be planned based on the target facility renewal sub-plans corresponding to all the communities.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for updating urban facilities based on a large language model as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described urban facility updating methods based on a large language model.

[0016] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned urban facility updating methods based on a large language model.

[0017] The present invention provides a large language model-based urban facility renewal method, apparatus, device, and medium. A division prompt word is input into the planner's large language model to obtain multiple communities corresponding to the urban area to be planned, as output by the planner's large language model. The division prompt word is used to instruct the planner's large language model to divide the urban area to be planned into multiple communities. The planner's large language model is constructed based on a universal multimodal large language model. For each community, the planner's large language model is used to determine the target facility renewal sub-plan corresponding to the community. Based on the target facility renewal sub-plans corresponding to all communities, the urban facility renewal plan corresponding to the urban area to be planned is determined. The technical solution of the present invention constructs an urban facility renewal plan through a large language model, eliminating the need for expert and resident participation, eliminating the negative impact of human factors, and improving the efficiency and accuracy of planning urban facility renewal plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a flow chart of the urban facility updating method based on the large language model provided by the present invention.

[0020] Figure 2 It is a schematic diagram of the construction of the planner's big language model and the resident's big language model provided by the present invention.

[0021] Figure 3 It is a flow chart of the sub-scheme for determining target facility updates provided by the present invention.

[0022] Figure 4 It is a structural diagram of the urban facility renewal device based on the large language model provided by the present invention.

[0023] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0025] In response to the above-mentioned problems in the prior art, the present invention provides a method for updating urban facilities based on a large language model. It should be noted that the execution subject of the present invention can be a computer or other electronic device, and the embodiments of the present invention are not specifically limited here. Figure 1 This is a flow chart of the urban facility renewal method based on the large language model provided by the present invention. Figure 1 As shown, the method includes the following steps 110, 120 and 130.

[0026] Step 110: Input the division prompt word into the planning engineer language model to obtain multiple communities corresponding to the urban area to be planned output by the planning engineer language model; the division prompt word is used to instruct the planning engineer language model to divide the urban area to be planned into multiple communities; the planning engineer language model is constructed based on a general multimodal language model.

[0027] Specifically, a planner language model can be pre-built based on a general multimodal language model. This model can include a memory module and a planning module. The memory module can be used to store all important information in the process of determining the urban facility renewal plan, and the planning module can be used to store future facility renewal sub-plans. Both can be stored in natural language.

[0028] In one embodiment, the method further comprises: Determining a planner prompt based on a city map corresponding to the urban area to be planned, text descriptions corresponding to each block, and the preset planning requirements; The planner prompt words are input into a universal multimodal large language model to obtain the planner large language model.

[0029] Specifically, the urban area to be planned can be divided into multiple plots, and each plot has a corresponding plot number. In the process of constructing the planner's big language model, the planner's prompt words can be determined based on the city map corresponding to the urban area to be planned, the text descriptions corresponding to each plot, and the preset planning requirements. The preset planning requirements may include: 1) the minimum floor area of ​​each type of facility. For example, it can be set as 2,000 square meters for living service facilities, 10,000 square meters for office facilities, 15,000 square meters for green space, 10,000 square meters for schools, 4,000 square meters for other educational facilities, 10,000 square meters for hospitals, 4,000 square meters for medical facilities, 2,000 square meters for entertainment facilities, 2,000 square meters for catering facilities, and 2,000 square meters for shopping facilities; 2) the facility area corresponding to a plot cannot exceed the total area of ​​the plot corresponding to the plot; 3) the setting of facilities that are not suitable for being adjacent to each other. For example, schools and hospitals should not be built on the same plot, and the two can be set as facilities that are not suitable for being adjacent to each other. Figure 2 This is a schematic diagram of the construction of the planner's language model and the resident's language model provided by the present invention, as shown in FIG. Figure 2 As shown, after the planner prompt words are determined, the planner prompt words can be input into the universal multimodal large language model. Figure 2 The initial prompt word input into the planner language model is also the planner prompt word, and the universal multimodal language model that inputs the planner prompt word becomes the planner language model. The universal multimodal language model can be any existing universal multimodal language model, for example, Deepseek.

[0030] In the above embodiment, based on the city map corresponding to the urban area to be planned and the text descriptions corresponding to each plot, as well as the preset planning requirements, the planner's prompt words are determined, and then a planner's language model is constructed, which solves the problem of insufficient professional knowledge of planners in the traditional process of formulating urban facility renewal plans.

[0031] Furthermore, in order to facilitate calculation in step 110, the urban area to be planned needs to be divided into multiple communities. Therefore, a division prompt word can be pre-built. For example, the division prompt word can be "divide the plots corresponding to Q city into multiple communities." The planner's language model can divide the multiple plots in the urban area to be planned into multiple communities based on the prompt word. The urban area to be planned can be represented as , Indicates the first The set of multiple communities obtained after the planned urban area is divided can be expressed as , Indicates the communities and meet For example, the urban area to be planned corresponding to City Q includes 10 plots. After the division by the planner's language model, plots 1 to 4 are divided into community a, plot 5 is divided into community b, plots 6 and 7 are divided into community c, and plots 8 to 10 are divided into community d.

[0032] Step 120: For each of the communities, determine the target facility update sub-plan corresponding to the community through the planner's language model.

[0033] Specifically, the planner's language model can be used to determine the target facility renewal sub-plans corresponding to all communities. The target facility renewal sub-plans include the specific renewal content of the urban facilities corresponding to the community.

[0034] In one embodiment, determining the target facility renewal sub-plan corresponding to the community by using the planner's language model includes: Execute the actions selected by the planner's language model in the preset action space to determine the target facility update sub-plan corresponding to the community; the actions in the preset action space include investigating residents, initiating discussions, evaluating plans, modifying plans, and ending updates.

[0035] Specifically, the planner's language model can select actions in a preset action space, which includes investigating residents, initiating discussions, evaluating plans, modifying plans, and ending updates. Figure 3 This is a flow chart of the target facility update sub-scheme provided by the present invention, such as Figure 3As shown, the planner's language model will select one of the actions according to the logic, and then the executive body of the present invention will execute the action until the planner's language model selects to end the update action. At this time, the executive body of the present invention can determine the target facility update sub-plan corresponding to the community. Exemplarily, before each action is selected, the planner's language model can first update the memory module and the planning module based on the feedback of the previous action, and then select the action and give the input. Repeat the above operations until the planner's language model believes that the update of the current community has been completed, selects "end update action", and then enters the update process of the next community. When all community updates are completed, the final urban facility update plan is obtained.

[0036] In the above embodiment, the actions selected by the planner's language model in the preset action space are executed until the modification of the plan is completed, so that the facility renewal sub-plan can be continuously optimized, and finally a better target facility renewal sub-plan corresponding to each community is obtained.

[0037] In one embodiment, the urban area to be planned includes multiple plots, each of which has a corresponding plot number. This embodiment describes all actions in the preset action space as follows: (1) The resident survey action includes: determining multiple resident numbers corresponding to the community based on at least one plot number corresponding to the community.

[0038] Specifically, at the beginning of establishing the resident language model, the resident number is assigned to the resident language model, and the plot corresponding to each resident number is determined. Therefore, after determining at least one plot number corresponding to the community, all resident numbers corresponding to the community can be determined based on the plot number. For example, the resident number corresponding to the 500-meter range around a plot can be determined as the resident number corresponding to the plot. It is easy to understand that the resident survey action only needs to be executed once, and the planner's language model will not repeatedly choose to execute the resident survey action.

[0039] (2) The action of initiating a discussion includes: inputting all the resident numbers corresponding to the community into the planner's big language model to obtain multiple target resident numbers output by the planner's big language model; inputting preset discussion prompt words into the resident big language model corresponding to each target resident number to obtain the discussion results corresponding to each target resident number; inputting all the discussion results into the preset big language model to obtain the resident discussion summary results output by the preset big language model, and inputting the resident discussion summary results into the planner's big language model; the preset big language model and the resident big language model are both constructed based on the general big language model.

[0040] Specifically, a large language model of residents can be pre-built. The method of building a large language model of residents can be illustrated by the following embodiment. In one embodiment, the method further includes: Obtaining demographic data corresponding to the urban area to be planned, and generating multiple resident portraits based on the demographic data; Inputting all the resident portraits into the preset large language model to obtain the resident prompt words corresponding to each of the resident portraits output by the preset large language model; Each of the resident prompt words is input into a plurality of universal large language models respectively to obtain all the resident large language models corresponding to the urban area to be planned.

[0041] Specifically, demographic data corresponding to the urban area to be planned can be obtained through a demographic database or a census, and then multiple resident portraits can be generated based on the demographic data. The resident portraits may include, for example, gender, age, education level, and family size. Further, all resident portraits can be input into a preset large language model, which can generate a corresponding resident prompt for each resident portrait. For example, Figure 2 As shown, Figure 2 The combination of the initial prompt word and the facility requirements input into the planner's large language model is the resident prompt word. For example, the resident prompt word could be "You are a 35-year-old male with a master's degree, married, and one child." After generating multiple resident prompt words, each resident prompt word can be input into a different universal large language model. The universal large language model that receives the resident prompt words is the resident large language model. The preset large language model is constructed based on the universal large language model.

[0042] When the planner's language model chooses to initiate a discussion action, since the number of resident numbers corresponding to a community may be large, during the execution of a discussion initiation action, only the discussion opinions output by some of the resident's language models can be selected. Therefore, all the resident numbers corresponding to the community can be input into the planner's language model, and the planner's language model selectively outputs multiple target resident numbers. Furthermore, the execution subject of the present invention can input preset discussion prompt words into the resident's language model corresponding to each target resident number, and then obtain the discussion results output by the resident's language model corresponding to each target resident number. The discussion results can be, for example, "types of facilities that you want to increase or decrease", etc., wherein the preset discussion prompt words can be set in advance as needed. Furthermore, all discussion results can be input into the preset language model to obtain the resident discussion summary results output by the preset language model, and the resident discussion summary results can be input into the planner's language model.

[0043] (3) The evaluation scheme action includes: determining the planning requirement evaluation result corresponding to the current facility renewal sub-scheme based on the preset planning requirements; determining the resident satisfaction index corresponding to the current facility renewal sub-scheme based on the initial demand facilities corresponding to each of the resident language models corresponding to the community; The planning requirement evaluation results and the resident satisfaction index are input into the planner language model.

[0044] In one embodiment, determining the planning requirement evaluation result corresponding to the current facility renewal sub-plan based on the preset planning requirement includes: For each of the plots corresponding to the community, if the facility area corresponding to the plot in the current facility renewal sub-plan is greater than the total area of ​​the plots corresponding to the plot, the plot number corresponding to the plot is determined as the first problem plot number; if the plot includes unsuitable adjacent facilities in the current facility renewal sub-plan, the plot number corresponding to the plot is determined as the second problem plot number; All of the first problem plot numbers and all of the second problem plot numbers are determined as the planning requirement evaluation results.

[0045] Specifically, for each plot corresponding to each community, if the facility area corresponding to the plot in the current facility update sub-plan is greater than the total area of ​​the plot corresponding to the plot, it indicates that the plan is unfeasible, and the plot number corresponding to the plot can be determined as the first problem plot number. If the plot in the current facility update sub-plan includes unsuitable adjacent facilities, the plot number corresponding to the plot can be determined as the second problem plot number. For example, in the current facility update sub-plan, plot 5 corresponding to community A includes a school and a hospital. However, since the school and hospital are unsuitable adjacent facilities, plot number 5 can be determined as the second problem plot number. If the plan modification action has not been executed, the current facility update sub-plan can be a pre-set plan based on needs or the actual facility situation of the community. If the plan modification action has been executed, the current facility update sub-plan is the most recently modified facility update sub-plan. After all first problem plot numbers and all second problem plot numbers have been determined, all first problem plot numbers and all second problem plot numbers can be determined as the planning requirement assessment results.

[0046] In the above embodiment, by presetting the restrictions of planning requirements, unreasonable plans will eventually receive lower evaluations, thereby promoting the optimization of the facility renewal sub-plans.

[0047] In one embodiment, determining the resident satisfaction index corresponding to the current facility update sub-scheme based on the initial demand facilities corresponding to each of the resident language models corresponding to the community includes: For each of the resident language models corresponding to the community, determining a target proportion of the initial demand facilities corresponding to the resident language model in the current facility update sub-plan; The average value of the target proportions corresponding to all the resident large language models is determined as the resident satisfaction index corresponding to the current facility renewal sub-plan.

[0048] Specifically, for each resident large language model corresponding to each community, the target proportion of the initial demand facilities corresponding to the resident large language model in the current facility update plan can be determined, wherein the initial demand facilities are the facility requirements used by the resident large language model to construct the resident prompt words at the beginning of its construction. For example, the facilities covering the resident large language model in the current facility update plan include schools and hospitals, and the initial demand facilities corresponding to the resident large language model are schools, hospitals and parks, then the target proportion corresponding to the resident large language model is 2 / 3. The average value of the target proportions corresponding to all resident large language models can be further determined as the resident satisfaction index corresponding to the current facility update sub-plan. It is easy to understand that the resident satisfaction index represents the proportion of resident needs being met, and its range is 0 to 1.

[0049] In the above embodiment, based on the initial demand facilities corresponding to the large language model of each resident, the needs of various stakeholders can be comprehensively considered, making the facility update sub-plan more reasonable and better able to meet the actual needs of most residents.

[0050] At the end of the evaluation plan action, after determining the planning requirement evaluation results and the resident satisfaction index, the planning requirement evaluation results and the resident satisfaction index can be input into the planner's language model.

[0051] (4) The modification plan action includes: determining a new facility renewal sub-plan based on the residents’ discussion summary results, and / or the planning requirement evaluation results and the residents’ satisfaction index; determining the new facility renewal sub-plan as the current facility renewal sub-plan, and executing the evaluation plan action.

[0052] Specifically, when executing the action of modifying the plan, if the action of initiating a discussion and the action of evaluating the plan have been executed before, a new facility update sub-plan can be determined based on the summary results of the resident discussion and the results of the planning requirement evaluation, as well as the resident satisfaction index. It is easy to understand that if only the action of initiating a discussion has been executed before, a new facility update sub-plan can be determined based on the summary results of the resident discussion. If only the action of evaluating the plan has been executed before, a new facility update sub-plan can be determined based on the results of the planning requirement evaluation and the resident satisfaction index. After obtaining the new facility update sub-plan, the new facility update sub-plan can be determined as the current facility update sub-plan, and the action of evaluating the plan can be executed again. That is, the present invention can repeatedly execute the action of initiating a discussion, the action of evaluating the plan, and the action of modifying the plan.

[0053] (5) The ending update action includes: determining the latest determined current facility update sub-plan as the target facility update sub-plan corresponding to the community.

[0054] Specifically, when the planner's language model selects the end update action, the execution subject of the present invention can determine the latest determined current facility update sub-plan as the target facility update sub-plan corresponding to the community.

[0055] In the above embodiment, the specific contents of the actions of investigating residents, initiating discussions, evaluating plans, modifying plans, and ending updates are described, so that the target facility update sub-plan can be continuously optimized and the needs of a large number of residents can be digitally taken into account, thereby satisfying the interests of all parties.

[0056] Step 130: Based on the target facility renewal sub-plans corresponding to all the communities, determine the urban facility renewal plan corresponding to the urban area to be planned.

[0057] Specifically, after determining the target facility renewal sub-plan corresponding to each community, all target facility renewal sub-plans can be integrated to obtain an urban facility renewal plan corresponding to the planned urban area.

[0058] For example, a user wants to renovate urban facilities in a certain area of ​​London, England. This area is planned to be divided into 109 plots, of which 46 are urban areas to be planned. Using the method described above, a planner's large language model and 1,000 resident large language models are constructed (over 1% of the actual population, thus representative). The planner's large language model uses the multimodal large language model GPT-40, while the resident large language model uses the large language model GPT-3.5-turbo-0125. The temperature parameter of the large language models is set to 0. Using the large language model-based urban facility renewal method provided by the present invention, an urban facility renewal plan for this area of ​​London, England, can be obtained.

[0059] The large language model-based urban facility renewal method provided by the present invention inputs a division prompt word into the planner's large language model, and obtains multiple communities corresponding to the urban area to be planned, which are output by the planner's large language model. The division prompt word is used to instruct the planner's large language model to divide the urban area to be planned into multiple communities. The planner's large language model is constructed based on a universal multimodal large language model. For each community, the planner's large language model is used to determine the target facility renewal sub-plan corresponding to the community. Based on the target facility renewal sub-plans corresponding to all communities, the urban facility renewal plan corresponding to the urban area to be planned is determined. The technical solution of the present invention constructs an urban facility renewal plan through the large language model, eliminating the need for expert and resident participation, eliminating the negative impact of human factors, and improving the efficiency and accuracy of planning urban facility renewal plans.

[0060] The urban facility renewal device based on the large language model provided by the present invention is described below. The urban facility renewal device based on the large language model described below and the urban facility renewal method based on the large language model described above can be referenced to each other.

[0061] Figure 4 This is a schematic diagram of the structure of the urban facility renewal device based on the large language model provided by the present invention. Figure 4 As shown, the urban facility renewal device 400 based on the large language model includes the following modules: A partitioning module 410 is configured to input a partitioning prompt word into a large language model for planning engineers, and obtain a plurality of communities corresponding to the urban area to be planned, as output by the large language model; the partitioning prompt word is used to instruct the large language model for planning engineers to partition the urban area to be planned into the plurality of communities; the large language model for planning engineers is constructed based on a general multimodal large language model; A first determination module 420 is configured to determine, for each community, a target facility renewal sub-plan corresponding to the community using the planner's language model; The second determining module 430 is configured to determine an urban facility renewal plan corresponding to the urban area to be planned based on the target facility renewal sub-plans corresponding to all the communities.

[0062] In one embodiment, the first determining module 420 is specifically configured to: Execute the actions selected by the planner's language model in the preset action space to determine the target facility update sub-plan corresponding to the community; the actions in the preset action space include investigating residents, initiating discussions, evaluating plans, modifying plans, and ending updates.

[0063] In one embodiment, the urban area to be planned includes multiple plots, each of which has a corresponding plot number; the first determining module 420 is further configured to perform the following actions: The resident survey action includes: determining a plurality of resident numbers corresponding to the community based on at least one plot number corresponding to the community; The initiating discussion action includes: inputting all the resident numbers corresponding to the community into the planner's language model to obtain multiple target resident numbers output by the planner's language model; inputting preset discussion prompt words into the resident language model corresponding to each target resident number to obtain the discussion results corresponding to each target resident number; inputting all the discussion results into the preset language model to obtain the resident discussion summary result output by the preset language model, and inputting the resident discussion summary result into the planner's language model; the preset language model and the resident language model are both constructed based on the general language model; The evaluation scheme action includes: determining a planning requirement evaluation result corresponding to the current facility renewal sub-scheme based on preset planning requirements; determining a resident satisfaction index corresponding to the current facility renewal sub-scheme based on the initial demand facilities corresponding to each of the resident language models corresponding to the community; and inputting the planning requirement evaluation result and the resident satisfaction index into the planner language model; The modification plan action includes: determining a new facility renewal sub-plan based on the resident discussion summary results and / or the planning requirement evaluation results and the resident satisfaction index; determining the new facility renewal sub-plan as the current facility renewal sub-plan, and executing the evaluation plan action; The ending update action includes: determining the latest determined current facility update sub-plan as the target facility update sub-plan corresponding to the community.

[0064] In one embodiment, the first determining module 420 is further configured to: For each of the plots corresponding to the community, if the facility area corresponding to the plot in the current facility renewal sub-plan is greater than the total area of ​​the plots corresponding to the plot, the plot number corresponding to the plot is determined as the first problem plot number; if the plot includes unsuitable adjacent facilities in the current facility renewal sub-plan, the plot number corresponding to the plot is determined as the second problem plot number; All of the first problem plot numbers and all of the second problem plot numbers are determined as the planning requirement evaluation results.

[0065] In one embodiment, the first determining module 420 is further configured to: For each of the resident language models corresponding to the community, determining a target proportion of the initial demand facilities corresponding to the resident language model in the current facility update sub-plan; The average value of the target proportions corresponding to all the resident large language models is determined as the resident satisfaction index corresponding to the current facility renewal sub-plan.

[0066] In one embodiment, the urban facility renewal device based on the large language model further includes a large language model construction module, which is specifically used to: Obtaining demographic data corresponding to the urban area to be planned, and generating multiple resident portraits based on the demographic data; Inputting all the resident portraits into the preset large language model to obtain the resident prompt words corresponding to each of the resident portraits output by the preset large language model; Each of the resident prompt words is input into a plurality of universal large language models respectively to obtain all the resident large language models corresponding to the urban area to be planned.

[0067] In one embodiment, the large language model construction module is further configured to: Determining a planner prompt based on a city map corresponding to the urban area to be planned, text descriptions corresponding to each block, and the preset planning requirements; The planner prompt words are input into a universal multimodal large language model to obtain the planner large language model.

[0068] The large language model-based urban facility renewal device provided by the present invention inputs a division prompt word into the planner's large language model, and obtains multiple communities corresponding to the urban area to be planned, which are output by the planner's large language model. The division prompt word is used to instruct the planner's large language model to divide the urban area to be planned into multiple communities. The planner's large language model is constructed based on a universal multimodal large language model. For each community, the planner's large language model is used to determine the target facility renewal sub-plan corresponding to the community. Based on the target facility renewal sub-plans corresponding to all communities, the urban facility renewal plan corresponding to the urban area to be planned is determined. The technical solution of the present invention constructs an urban facility renewal plan using the large language model, eliminating the need for expert and resident participation, eliminating the negative impact of human factors, and improving the efficiency and accuracy of planning urban facility renewal plans.

[0069] Figure 5 An example of a physical structure diagram of an electronic device is shown below. Figure 5As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communications interface 520, and the memory 530 communicate with each other via the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the urban facility updating method based on the large language model, which includes: Inputting a division prompt word into a large language model for planning engineers to obtain a plurality of communities corresponding to the urban area to be planned outputted by the large language model for planning engineers; the division prompt word is used to instruct the large language model for planning engineers to divide the urban area to be planned into the plurality of communities; the large language model for planning engineers is constructed based on a general multimodal large language model; For each of the communities, determining a target facility renewal sub-plan corresponding to the community through the planner's language model; Based on the target facility renewal sub-plans corresponding to all the communities, an urban facility renewal plan corresponding to the urban area to be planned is determined.

[0070] Furthermore, the logic instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0071] On the other hand, the present invention further provides a computer program product, comprising a computer program, which may be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the urban facility updating method based on a large language model provided by each of the above methods, the method comprising: Inputting a division prompt word into a large language model for planning engineers to obtain a plurality of communities corresponding to the urban area to be planned outputted by the large language model for planning engineers; the division prompt word is used to instruct the large language model for planning engineers to divide the urban area to be planned into the plurality of communities; the large language model for planning engineers is constructed based on a general multimodal large language model; For each of the communities, determining a target facility renewal sub-plan corresponding to the community through the planner's language model; Based on the target facility renewal sub-plans corresponding to all the communities, an urban facility renewal plan corresponding to the urban area to be planned is determined.

[0072] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the urban facility updating method based on a large language model provided by the above methods, the method comprising: Inputting a division prompt word into a large language model for planning engineers to obtain a plurality of communities corresponding to the urban area to be planned outputted by the large language model for planning engineers; the division prompt word is used to instruct the large language model for planning engineers to divide the urban area to be planned into the plurality of communities; the large language model for planning engineers is constructed based on a general multimodal large language model; For each of the communities, determining a target facility renewal sub-plan corresponding to the community through the planner's language model; Based on the target facility renewal sub-plans corresponding to all the communities, an urban facility renewal plan corresponding to the urban area to be planned is determined.

[0073] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0074] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for updating urban facilities based on a large language model, characterized in that: include: Inputting the division prompt words into the planning language model to obtain multiple communities corresponding to the urban area to be planned output by the planning language model; The division prompt word is used to instruct the planner language model to divide the urban area to be planned into a plurality of communities; the planner language model is constructed based on a general multimodal language model; For each of the communities, determining a target facility renewal sub-plan corresponding to the community through the planner's language model; Based on the target facility renewal sub-plans corresponding to all the communities, an urban facility renewal plan corresponding to the urban area to be planned is determined.

2. The urban facility renewal method based on a large language model according to claim 1, characterized in that: The determining of the target facility renewal sub-plan corresponding to the community by the planner's language model includes: Execute the actions selected by the planner's language model in the preset action space to determine the target facility update sub-plan corresponding to the community; the actions in the preset action space include investigating residents, initiating discussions, evaluating plans, modifying plans, and ending updates.

3. The urban facility renewal method based on a large language model according to claim 2 is characterized in that: The urban area to be planned includes a plurality of plots, each of which has a corresponding plot number; The resident survey action includes: determining a plurality of resident numbers corresponding to the community based on at least one plot number corresponding to the community; The initiating discussion action includes: inputting all the resident numbers corresponding to the community into the planner's language model to obtain multiple target resident numbers output by the planner's language model; inputting preset discussion prompt words into the resident language model corresponding to each target resident number to obtain the discussion results corresponding to each target resident number; inputting all the discussion results into the preset language model to obtain the resident discussion summary result output by the preset language model, and inputting the resident discussion summary result into the planner's language model; the preset language model and the resident language model are both constructed based on the general language model; The evaluation scheme action includes: determining a planning requirement evaluation result corresponding to the current facility renewal sub-scheme based on preset planning requirements; determining a resident satisfaction index corresponding to the current facility renewal sub-scheme based on the initial demand facilities corresponding to each of the resident language models corresponding to the community; and inputting the planning requirement evaluation result and the resident satisfaction index into the planner language model; The modification plan action includes: determining a new facility renewal sub-plan based on the resident discussion summary results and / or the planning requirement evaluation results and the resident satisfaction index; determining the new facility renewal sub-plan as the current facility renewal sub-plan, and executing the evaluation plan action; The ending update action includes: determining the latest determined current facility update sub-plan as the target facility update sub-plan corresponding to the community.

4. The urban facility renewal method based on a large language model according to claim 3 is characterized in that: The determining of the planning requirement evaluation result corresponding to the current facility renewal sub-plan based on the preset planning requirements includes: For each of the plots corresponding to the community, if the facility area corresponding to the plot in the current facility renewal sub-plan is greater than the total area of ​​the plots corresponding to the plot, the plot number corresponding to the plot is determined as the first problem plot number; if the plot includes unsuitable adjacent facilities in the current facility renewal sub-plan, the plot number corresponding to the plot is determined as the second problem plot number; All of the first problem plot numbers and all of the second problem plot numbers are determined as the planning requirement evaluation results.

5. The urban facility renewal method based on a large language model according to claim 3 is characterized in that: The determining of the resident satisfaction index corresponding to the current facility update sub-scheme based on the initial demand facilities respectively corresponding to the large language models of the residents corresponding to the community includes: For each of the resident language models corresponding to the community, determining a target proportion of the initial demand facilities corresponding to the resident language model in the current facility update sub-plan; The average value of the target proportions corresponding to all the resident large language models is determined as the resident satisfaction index corresponding to the current facility renewal sub-plan.

6. The urban facility renewal method based on a large language model according to any one of claims 3 to 5, characterized in that: The method further comprises: Obtaining demographic data corresponding to the urban area to be planned, and generating multiple resident portraits based on the demographic data; Inputting all the resident portraits into the preset large language model to obtain the resident prompt words corresponding to each of the resident portraits output by the preset large language model; Each of the resident prompt words is input into a plurality of universal large language models respectively to obtain all the resident large language models corresponding to the urban area to be planned.

7. The urban facility renewal method based on a large language model according to any one of claims 1 to 5, characterized in that: The method further comprises: Determining a planner prompt based on a city map corresponding to the urban area to be planned, text descriptions corresponding to each block, and the preset planning requirements; The planner prompt words are input into a universal multimodal large language model to obtain the planner large language model.

8. An urban facility renewal device based on a large language model, characterized in that: include: a partitioning module, configured to input a partitioning prompt word into a planning language model, and obtain a plurality of communities corresponding to the urban area to be planned outputted by the planning language model; The division prompt word is used to instruct the planner language model to divide the urban area to be planned into a plurality of communities; the planner language model is constructed based on a general multimodal language model; A first determination module is configured to determine, for each community, a target facility renewal sub-plan corresponding to the community using the planner's language model; The second determination module is used to determine the urban facility renewal plan corresponding to the urban area to be planned based on the target facility renewal sub-plans corresponding to all the communities.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the urban facility updating method based on the large language model as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for updating urban facilities based on a large language model as described in any one of claims 1 to 7 is implemented.

Images