Urban micro-environment digital scene creation device and method

Abstract

This invention provides a device and method for creating digital scenes in urban micro-environments, belonging to the field of AR urban technology. The device is a terminal, and its system integrates various modules. The method is implemented based on these modules, including collecting heat map data, environmental data, and user data through a positioning module and an urban spatial data perception and acquisition module, and generating scene instructions based on the data; generating and encapsulating digital materials through an AIGC generation and decision-making module with a built-in generative large model; downloading digital materials through an AR interaction and dynamic rendering module and overlaying each material in real time at corresponding locations to form an AR scene, which the user interacts with through the terminal; the interactive data serves as new input to optimize the subsequently generated scene instructions, forming a self-evolving closed loop. This invention overcomes the shortcomings of traditional AR guide systems, which have fixed content and rigid formats, and can flexibly adjust content strategies according to pedestrian density and environmental changes, greatly improving the interactivity and immediacy of the scene.

Description

Technical Field

[0001] This invention belongs to the field of AR urban technology, and in particular relates to a device and method for creating digital scenes of urban micro-environments. Background Technology

[0002] With the deepening of urban digital transformation, how to revitalize street spaces at low cost and high efficiency, and enhance their commercial vitality and cultural appeal, has become a key issue in urban renewal. Currently, many urban streets, especially some historical districts or non-core business areas, generally face pain points such as a sluggish real economy, high shop vacancy rates, high costs of physical space renovation, and insufficient appeal of traditional business models to young consumers. Analysis of 24-hour heat maps of streets in cities like Hangzhou reveals that many streets suffer from insufficient foot traffic and a lack of vitality.

[0003] Existing solutions mainly fall into two categories: one is traditional physical space renovation, which is costly, time-consuming, and difficult to respond flexibly to rapidly changing business and consumer demands; the other is simple digital presentation, such as static QR codes or simple digital guides, which lack in-depth interaction and intelligent content generation capabilities, resulting in a monotonous user experience and failing to truly "awaken" the vitality of the streets. (1) AR city navigation system based on static data. This type of system usually has a fixed 3D model or multimedia content (such as building history information, merchant coupons) pre-made, which is triggered to view via mobile phone or AR glasses. Its disadvantages are that the content is fixed, lacks dynamism and intelligence, cannot generate content according to real-time street traffic or personalized user needs, and the interactive experience is monotonous.

[0004] (2) Heat map-based business analytics platforms. These platforms generate regional heat maps by collecting data such as mobile phone signaling, providing macro-level decision support for business site selection or urban planning. However, their analysis results are usually presented to managers in the form of two-dimensional charts, failing to transform the analyzed data into augmented reality scenarios that can be interacted with by ordinary citizens and tourists in real time, and lacking a public-facing interactive interface and experience.

[0005] (3) Traditional digital signage and advertising screens. This involves placing LCD screens and other equipment on the streets to display promotional content. However, content updates require manual operation, cannot intelligently interact with the street environment and pedestrians, are rigid in form, have high renovation and maintenance costs, and are difficult to personalize.

[0006] Generative large models (AIGC) technology shows great potential in content creation and natural language interaction, while augmented reality (AR) technology can seamlessly overlay digital information onto the real world. However, there is currently no mature solution to deeply integrate the two and apply them to the creation and interaction of real-time, dynamic, and personalized digital scenes in urban microenvironments (such as streets).

[0007] Therefore, there is an urgent need for an innovative device that can intelligently generate and present rich media digital content based on the real-time status of the street (such as pedestrian traffic) and user needs, thereby injecting new digital vitality into urban streets. Summary of the Invention

[0008] The purpose of this invention is to solve the problem that existing technologies cannot deeply integrate AIGC and AR technologies and apply them to the creation and interaction of real-time, dynamic and personalized digital scenes in urban micro-environments. The invention provides a digital scene creation device for urban micro-environments, including a terminal, a system integration positioning module, an urban spatial data perception and acquisition module, an AIGC generation and decision-making module, and an AR interaction and dynamic rendering module. The positioning module uses GPS, cameras, visual markers, or SLAM technology to locate the terminal's position. The urban spatial data perception and acquisition module acquires heat map data and environmental data through API interfaces and IoT sensors, and acquires user data based on the positioning module. The built-in rule engine of the terminal generates scene instructions based on the data, and completes the transformation of the data with the scene problems to be solved. The AIGC generation and decision-making module has a built-in generative big model. The AIGC generation and decision-making module packages the generated scene instructions with their context information into creation prompts, sends them to the generative big model, generates corresponding digital materials, and encapsulates them into a digital scene asset package, generating corresponding links. The terminal loads a digital scene asset package and overlays the various materials onto the user's corresponding location in real time through the AR interaction and dynamic rendering module, forming a virtual-real fusion AR scene and completing the creation of a digital scene of the urban micro-environment. Users interact with the AR scene through the terminal, and the interaction data is used as new behavioral heat map data to optimize the scene commands generated subsequently, forming a self-evolving closed loop.

[0009] Furthermore, the terminal is a mobile phone or glasses. The mobile phone or glasses use its preset camera to capture the real environment in which the user is located. This data is then used by the subsequent AR interaction and dynamic rendering modules to overlay the digital materials with the real environment to form an AI scene.

[0010] A method for creating digital scenes of urban micro-environments, using an urban micro-environment digital scene creation device, includes the following steps: S1: The terminal system collects data through API interfaces and IoT sensors; S2: The rule engine analyzes the collected data to generate various scenario instructions; S3: The system connects each scene instruction with its context information and packages them into creation prompts; S4: Generative large model receives various prompts and generates various digital materials; S5: The AIGC engine encapsulates various digital materials into digital scene asset packages and generates corresponding links; S6: The terminal uses the positioning module to locate its specific position in real time and loads a digital scene asset package, binding the content corresponding to various digital materials with specific coordinates in the real world; S7: The AI ​​rendering engine overlays various digital materials onto specific bound coordinates in real time to form AR images; S8: Users interact with AR visuals via voice and gestures through the terminal; S9: Feeds user interaction data as new behavior heatmap data back to the terminal system, optimizes subsequent generated scene commands, and forms a self-evolving closed loop.

[0011] Furthermore, in S1, the data includes heat map data, environmental data, and user data. The heat map data of the target street's 24-hour population is pulled from the city's big data platform in real time or at regular intervals via API interface. This data is presented in grid or vector form and is used to identify the absolute pedestrian flow and density change trends at different locations on the street at different times. Environmental data is acquired through IoT sensors or public weather APIs, including time, weather, temperature, and light intensity. Once the terminal is turned on, user data is uploaded through its positioning module, including the user's anonymity, location coordinates, and device orientation.

[0012] Furthermore, in S2, the rule engine analyzes heatmap data to identify relatively low-activity and high-activity areas, and generates various scenario instructions based on trigger conditions, including: A: When it is identified that the real-time pedestrian density of a street segment is lower than the preset threshold and the duration is greater than 30 minutes, then the street segment is a low-activity area, triggering the activity wake-up scenario and generating scenario instructions; B: When it is identified that the real-time pedestrian density of a street segment is greater than the preset threshold and the duration is greater than 30 minutes, the street segment is a high-activity area, triggering the traffic diversion and guidance scenario, generating scenario instructions, and guiding users to the nearby low-activity area through the terminal; C: When the rule engine obtains that the current weather data is rainy, it will add a rainy day label to all generated scene instructions, affecting the content generation style of subsequent steps.

[0013] Furthermore, in S4, digital assets include scene scripts, AI character settings and dialogue libraries, visual asset descriptions, and UI or UX copy. The scene scripts are used to determine the core flow of the AR experience; the AI ​​character settings and dialogue libraries are used to create the AI's character settings, knowledge base, and opening remarks; the visual asset descriptions are used to generate the calling parameters for the image model; and the UI or UX copy is used to generate the prompt text in the AR interface.

[0014] Furthermore, in S7, the AI ​​rendering engine, based on specific coordinates, uses AR interaction and dynamic rendering modules to overlay various generated scene materials in real time onto the real scene captured by the user's terminal camera with correct perspective and lighting relationships, forming an AR image that blends the virtual and real worlds.

[0015] Furthermore, in S9, the duration of a user's stay at their location, their interactive behavior, and their sharing actions are anonymously recorded and fed back to the terminal system as new behavioral heatmap data. This new behavioral heatmap data is used to evaluate the scene's effectiveness and serves as input for new scene commands, optimizing subsequently generated scene commands and forming a self-evolving technological closed loop.

[0016] Compared with the prior art, the beneficial effects of the present invention are mainly reflected in: It enables real-time dynamic generation and personalized experience of urban micro-environment scenes: This invention integrates AIGC generation and decision-making modules, enabling the creation of digital materials (such as dynamic scrolls and AI character dialogues) tailored to the current scene using a generative large model based on real-time urban heat map data, environmental data (such as weather), and user data. This overcomes the shortcomings of traditional AR tour guide systems, which have fixed content and rigid formats, and can flexibly adjust content strategies (such as "attracting traffic" or "awakening") according to crowd density and environmental changes, greatly improving the interactivity and immediacy of the scene.

[0017] This enables the system to continuously evaluate scene effects and automatically optimize subsequent scene commands. This invention not only outputs content unidirectionally, but also collects data such as user dwell time and interaction behavior through AR interaction and dynamic rendering modules, and transforms this data into new behavioral heatmap data to feed back to the system. This data-driven closed-loop mechanism enables the system to continuously evaluate the scene effect and automatically optimize subsequent scene commands, thereby continuously improving the user experience and stimulating street vitality, solving the problem of traditional physical transformation or simple digital presentation lacking a long-term iterative mechanism. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the structure and method of a digital scene creation device for urban micro-environments according to the present invention.

[0019] Figure 2This is a schematic diagram of urban spatial multi-source data acquisition and heat map analysis in an embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram illustrating the virtual-real fusion effect of AR digital content based on a real-world scene in an embodiment of the present invention. Detailed Implementation

[0021] The following is a more detailed description of a digital scene creation device and method for urban micro-environment according to the present invention, with reference to the schematic diagrams, which illustrate preferred embodiments of the present invention. It should be understood that those skilled in the art can modify the present invention described herein while still achieving the advantageous effects of the present invention. Therefore, the following description should be understood as being of general knowledge to those skilled in the art and is not intended to limit the present invention.

[0022] A digital scene creation device and method for urban micro-environments, such as Figure 1 As shown, the device is a terminal, which integrates a positioning module, an urban spatial data sensing and acquisition module, an AIGC generation and decision-making module, and an AR interaction and dynamic rendering module. The method is implemented based on each module, as detailed below: I. Urban spatial data perception and acquisition module, positioning module.

[0023] Step 1: Multi-source data acquisition.

[0024] Core data source (heat map): The system retrieves 24-hour population heat map data for the target street from the city's big data platform periodically (e.g., every 15 minutes) or in real time via API. This data is presented in grid or vector form, indicating the absolute pedestrian flow and density trends at different locations on the street over different time periods.

[0025] Environmental data: Obtain data such as time, weather, temperature, and light intensity from IoT sensors or public weather APIs.

[0026] User data: Based on the positioning module, when a user enters the service area and starts the AR application, the application will upload their anonymous location coordinates and device orientation (view direction).

[0027] Step 2: Data analysis and scenario triggering (“switch” logic).

[0028] The system's built-in rule engine analyzes heatmap data to identify "low-activity areas" and "high-activity areas." Triggering conditions include, for example: Condition A (Activate Low Vitality Zone): When the real-time pedestrian density of a certain street segment A is less than the preset threshold and the duration is greater than 30 minutes, the "Vitality Awakening" scene generation command is triggered.

[0029] Condition B (Guidance and Diversion): When the real-time pedestrian density of street segment B exceeds the preset threshold (which may cause congestion), the "Guidance and Diversion" scenario generation command is triggered to guide users to the nearby low-activity area.

[0030] Condition C (Environment Adaptation): When the weather data is "rainy", add the "rainy day" label to all scene generation instructions, affecting the style of subsequent content generation.

[0031] At this point, the abstract street data has been transformed into concrete, unresolved "scenario problems," completing the "driving" process. For example... Figure 2 As shown, the system performs visual analysis on the collected multi-source data (such as pedestrian density, depth information, etc.) to identify different activity areas in the street, providing data support for subsequent scene generation.

[0032] II. AIGC Generation and Decision Module.

[0033] Step 3: Generate instructions and build the context.

[0034] The system packages the scene instructions generated in step 2 (such as "generate a vitality awakening scene for street segment A"), along with rich contextual information, into a detailed "creation prompt" and sends it to the generative large model.

[0035] Example prompt: "You are a city street vitality designer. It is 3 pm on a sunny Sunday afternoon. Located at the Okinoba Square of the College of Architecture and Urban Planning, Tongji University, please design an AR interactive scene for this. Requirements: 1. The theme is a public art installation, aiming to enhance the fun of the site; 2. The content should be in line with the interests of today's young people, such as anime and cartoon character IPs; 3. Accompany it with an eye-catching and mood-enhancing copy." Step 4: Multimodal content generation (“materials” creation).

[0036] After receiving the prompt words, the generative large model begins to generate various digital materials in parallel: Scene script (based on the example above): Determine the core flow of the AR experience, such as "User enters the AR interactive installation site → triggers the AR public art installation → displays the public art installation and text description".

[0037] AI Character Setup and Dialogue Library: Create character settings, knowledge bases, and opening lines for virtual IP characters (such as AI guides or virtual curators). For example: "I am the virtual art guide for this neighborhood. Are you interested in the renovation history of this square or the current AR art installations?" (Simultaneously generating a series of question-and-answer pairs about the site's history, design concepts, or artistic interpretations).

[0038] Visual material description: The calling parameters for generating image models, such as "generate a sequence of images showcasing a futuristic urban public space scene".

[0039] UI / UX copywriting: Generate prompts in the AR interface, such as "Click on the art installation to view details" or "Slide to explore more renovation options".

[0040] Step 5: Content structuring and encapsulation.

[0041] The AIGC engine encapsulates the generated text, images (or image generation parameters), 3D model calling instructions, etc., into a complete "digital scene asset package" according to a predefined format (such as JSON), and prepares the corresponding resource links.

[0042] III. AR Interaction and Dynamic Rendering Module.

[0043] Step 6: Space registration and anchoring.

[0044] The user's AR terminal (phone / glasses) uses GPS, visual markers or SLAM technology to achieve precise positioning in street segment A.

[0045] Based on the scenario settings, the system binds the virtual content in the "digital scene asset package" to specific coordinates (spatial anchor points) in the physical world.

[0046] Step 7: Content rendering and overlay.

[0047] The user's AR terminal downloads or streams the "digital scene asset package" from the server.

[0048] The AR rendering engine, based on spatial anchor points, overlays generated street scenes, UI text, etc., onto the real wall surface captured by the user's camera in real time, with correct perspective and lighting relationships, creating a seamless blend of reality and virtuality. For example... Figure 3 As shown, when a user observes a real street through a terminal (such as AR glasses), the system has accurately anchored the virtual geometry, operation interface, and prompts (such as "Scheme generation in progress" and "Renovation scheme A") into the physical space according to the generated scheme, thus achieving seamless integration of digital content and the real environment.

[0049] Step 8: Natural interaction and closed-loop feedback.

[0050] After seeing the AR scene, users can interact with the elements within it: Voice dialogue: The user speaks directly to the device: "Please introduce the design concept of this art installation." → The voice is recognized as text → sent to the generative big model → The big model generates a response in real time based on the virtual character settings and scene knowledge base: "This installation aims to enhance the fun of space use, and its flowing form is inspired by..." → The response is played out through speech synthesis technology.

[0051] Gesture interaction: Users can click or touch virtual devices in the AR screen to trigger more detailed multimedia introductions.

[0052] Step 9: New Data Generation: User dwell time, interaction behaviors (clicks, number of conversations), sharing actions, etc., are anonymously recorded and fed back to the system as new "behavioral heatmap" data (this data is transmitted and updated through the urban spatial data sensing and collection module integrated into the terminal). This data can be used to evaluate the scene's effectiveness and serve as input for the next round of "data-driven" processing, thereby optimizing subsequent content generation and forming a complete, self-evolving technological closed loop.

[0053] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the present invention. Any equivalent substitutions or modifications made by those skilled in the art to the technical solutions and content disclosed in the present invention without departing from the scope of the present invention shall be deemed to have remained within the protection scope of the present invention.

Claims

1. A digital scene creation device for urban micro-environments, characterized in that, The terminal includes a system integration positioning module, an urban spatial data perception and acquisition module, an AIGC generation and decision-making module, and an AR interaction and dynamic rendering module. The positioning module uses GPS, a camera, visual markers, or SLAM technology to locate the terminal's position. The urban spatial data sensing and acquisition module acquires heat map data and environmental data through API interfaces and IoT sensors, and acquires user data based on the positioning module. The rule engine built into the terminal generates scene instructions based on each data, and completes the transformation of each data with the scene problem to be solved. The AIGC generation and decision-making module has a built-in generative big model. The AIGC generation and decision-making module packages the generated scene instructions with their context information into creation prompts, sends them to the generative big model, generates corresponding digital materials, encapsulates them into digital scene asset packages, and generates corresponding links. The terminal loads a digital scene asset package and overlays the various materials onto the user's corresponding location in real time through the AR interaction and dynamic rendering module, forming a virtual-real fusion AR scene to complete the creation of a digital scene of the urban micro-environment. The user interacts with the AR scene through the terminal, and uses the interaction data as new behavioral heat map data to optimize the scene commands generated subsequently, forming a self-evolving closed loop.

2. The urban micro-environment digital scene creation device according to claim 1, characterized in that, The terminal is a mobile phone or glasses. The mobile phone or glasses use its preset camera to capture the real environment in which the user is located. This data is then used by the AR interaction and dynamic rendering module to overlay the digital materials with the real environment to form an AI scene.

3. A method for creating digital scenes of urban micro-environments, employing the digital scene creation device for urban micro-environments as described in claim 1 or 2, characterized in that, Includes the following steps: S1: The terminal system collects data through API interfaces and IoT sensors; S2: The rule engine analyzes the collected data to generate various scenario instructions; S3: The system connects each scene instruction with its context information and packages them into creation prompts; S4: Generative large model receives various prompts and generates various digital materials; S5: The AIGC engine encapsulates various digital materials into digital scene asset packages and generates corresponding links; S6: The terminal uses the positioning module to locate its specific position in real time and loads a digital scene asset package, binding the content corresponding to various digital materials with specific coordinates in the real world; S7: The AI ​​rendering engine overlays various digital materials onto specific bound coordinates in real time to form AR images; S8: Users interact with AR visuals via voice and gestures through the terminal; S9: Feeds user interaction data as new behavior heatmap data back to the terminal system, optimizes subsequent generated scene commands, and forms a self-evolving closed loop.

4. The method for creating digital scenes of urban micro-environments according to claim 3, characterized in that, In S1, the data includes heat map data, environmental data, and user data. The heat map data of the target street’s 24-hour population is pulled from the city’s big data platform in real time or at regular intervals through an API interface. This data is presented in grid or vector form and is used to identify the absolute flow and density change trends of different locations on the street at different times. Environmental data is acquired through IoT sensors or public weather APIs, including time, weather, temperature, and light intensity. Once the terminal is turned on, user data is uploaded through its positioning module, including the user's anonymity, location coordinates, and device orientation.

5. The method for creating digital scenes of urban micro-environments according to claim 3, characterized in that, In step S2, the rule engine analyzes heatmap data to identify relatively low-activity and high-activity areas, and generates various scenario instructions based on triggering conditions, including: A: When it is identified that the real-time pedestrian density of a street segment is lower than the preset threshold and the duration is greater than 30 minutes, then the street segment is a low-activity area, triggering the activity wake-up scenario and generating scenario instructions; B: When it is identified that the real-time pedestrian density of a street segment is greater than the preset threshold and the duration is greater than 30 minutes, the street segment is a high-activity area, triggering the traffic diversion and guidance scenario, generating scenario instructions, and guiding users to the nearby low-activity area through the terminal; C: When the rule engine obtains that the current weather data is rainy, it will add a rainy day label to all generated scene instructions, affecting the content generation style of subsequent steps.

6. The method for creating digital scenes of urban micro-environments according to claim 3, characterized in that, In step S4, the digital materials include a scene script, an AI character setting and dialogue library, visual material descriptions, and UI or UX copy. The scene script is used to determine the core flow of the AR experience; the AI ​​character setting and dialogue library is used to create the AI's character settings, knowledge base, and opening remarks; the visual material descriptions are used to generate the calling parameters of the image model; and the UI or UX copy is used to generate the prompt text in the AR interface.

7. The method for creating digital scenes of urban micro-environments according to claim 3, characterized in that, In S7, the AI ​​rendering engine, based on specific coordinates, uses the AR interaction and dynamic rendering module to overlay various scene materials generated with correct perspective and lighting relationships onto the real wall captured by the user terminal's camera in real time, forming an AR image that blends the virtual and real worlds.

8. The method for creating digital scenes of urban micro-environments according to claim 3, characterized in that, In S9, the user's dwell time, interactive behavior, and sharing actions at their location are anonymously recorded and fed back to the terminal system as new behavioral heatmap data. The new behavioral heatmap data is used to evaluate the scene effect and as input for new scene commands, optimizing the subsequently generated scene commands and forming a self-evolving technical closed loop.

Images