system

Abstract

A system is provided.SOLUTION: A system including means for acquiring past disaster information from an external database or an API, means for acquiring a floor plan of an office, evacuation route, and stock information from an in-house database, means for generating a virtual disaster scenario using a large-scale language model, means for creating an evacuation simulation video based on an actual office environment using a video generation artificial intelligence, means for integrating the generated simulation video and scenario and outputting as augmented reality content, and means for allowing a user's terminal to download the augmented reality content and view and experience.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In modern office environments, efficient and realistic disaster drills are necessary to ensure employees can evacuate quickly and safely in the event of a disaster. However, traditional disaster drills are generally static, making it difficult to recreate actual disaster scenarios. Furthermore, the training content is uniform, making it difficult to adapt specific responses to the office layout and individual evacuation routes. The present invention aims to solve these issues and provide more realistic disaster drills. [Means for solving the problem]

[0005] The present invention is a system including the following means.

[0006] A means to obtain past disaster information from an external database or API,

[0007] A means to obtain office floor plans, evacuation routes, and emergency supply information from the company database,

[0008] a means for generating virtual disaster scenarios using large-scale language models;

[0009] A means for creating an evacuation simulation video based on an actual office environment using video generation artificial intelligence;

[0010] A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content;

[0011] A means for downloading augmented reality content to a user's device, allowing the user to view and experience the content;

[0012] Includes:

[0013] In addition, by including a means for preprocessing collected data and standardizing the format, and a means for storing the generated augmented reality content in cloud storage and generating a URL that can be accessed by users, more efficient and practical disaster prevention training can be provided.

[0014] An "external database" is a database that exists on the Internet or an external server and is accessible.

[0015] "API" stands for Application Programming Interface, a standardized interface for exchanging data and functions between different software components.

[0016] An "internal database" is a database stored on a server or storage system within a company.

[0017] A floor plan is a drawing that shows the internal structure and layout of rooms in an office.

[0018] An "evacuation route" is a route designated for safely escaping from an office in the event of a disaster.

[0019] "Stockpile information" is a list of food, drinks, water, medical supplies, etc. stored in the office for use in the event of a disaster.

[0020] A "large-scale language model" is an artificial intelligence model trained from large amounts of text data using natural language processing techniques, and is capable of performing advanced language tasks such as text generation and question answering.

[0021] A "virtual disaster scenario" is a scenario of a hypothetical disaster situation that is simulated based on past disaster data and office information.

[0022] "Image generation artificial intelligence" is a technology that uses artificial intelligence to generate 3D models and CG to create realistic images.

[0023] "Evacuation simulation video" is a video that realistically recreates evacuation procedures, generated based on a hypothetical disaster scenario.

[0024] "Augmented reality content" is content that adds virtual information to real-world information and displays it visually.

[0025] A "terminal" is a device used by a user, such as a smartphone, tablet, or AR headset.

[0026] "Cloud storage" is an online storage service that allows you to store and access data over the Internet.

[0027] "URL" is an abbreviation for Uniform Resource Locator, and is an address used to specify and access resources on the Internet. [Brief explanation of the drawings]

[0028] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0029] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0030] First, the terms used in the following description will be explained.

[0031] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0032] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0033] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0034] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0035] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0036] [First embodiment]

[0037] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0038] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0039] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0040] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0041] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0042] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0043] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0044] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0045] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0046] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0047] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0048] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0049] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0050] First, the server retrieves past disaster information from an external database or API. This collects detailed data on past typhoons, earthquakes, and other events. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies from the company's internal database. This data provides the basic information needed to generate disaster scenarios.

[0051] The server preprocesses the acquired data and standardizes the format. Any data deficiencies are corrected. The preprocessed data is passed to a large-scale language model (e.g., GPT-4 (registered trademark)) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0052] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG to create a realistic evacuation simulation, allowing users to experience a realistic evacuation drill.

[0053] The server then integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0054] Users launch the application on their devices to begin disaster prevention training. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0055] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0056] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, and supports users in taking quicker and safer evacuation actions in the event of a disaster.

[0057] The processing flow will be explained below.

[0058] Step 1:

[0059] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0060] Step 2:

[0061] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0062] Step 3:

[0063] The server preprocesses the acquired data by standardizing the data format and filling in any missing information. This preprocessing includes data cleansing and normalization.

[0064] Step 4:

[0065] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0066] Step 5:

[0067] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0068] Step 6:

[0069] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0070] Step 7:

[0071] The server saves the augmented reality content in cloud storage and generates a URL that users can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and creates an access link to the file.

[0072] Step 8:

[0073] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0074] Step 9:

[0075] Users can play the downloaded augmented reality content in the application to experience a disaster prevention drill. The user's device will then visually display evacuation route guides and instructions on how to use stockpiled supplies, helping them to carry out the drill in a realistic situation.

[0076] In this way, the entire system works together to realize dynamic and realistic disaster prevention drills in real time.

[0077] Example 1

[0078] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0079] In recent years, the increasing importance of disaster prevention measures has created a demand for realistic and effective disaster prevention training. However, conventional disaster prevention training is difficult to implement and often lacks realism, making it difficult for participants to take appropriate action in the event of a real disaster. Furthermore, technology for efficiently handling large amounts of data and generating virtual scenarios has been limited. This has resulted in training that is not tailored to the actual situation on the ground and is therefore difficult to demonstrate effectiveness in the event of a real disaster. To address these issues, a system is needed that combines past disaster information with internal company data to generate realistic virtual disaster scenarios and utilizes augmented reality to provide effective disaster prevention training.

[0080] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0081] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supplies information from an internal database, means for preprocessing the collected data and standardizing the format, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on real facility environments using video generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, and means for downloading the augmented reality content to a user's device so that the user can view and experience it. This allows users to experience realistic disaster prevention training and take prompt and appropriate action in the event of an actual disaster.

[0082] "Past disaster information" refers to detailed data on natural disasters that have occurred in the past (e.g., typhoons, earthquakes, floods, etc.).

[0083] An "external database" refers to a database that exists outside an organization and is accessible via the Internet or other means.

[0084] "API" stands for Application Program Interface, and refers to a standardized interface for exchanging data and functions between different software programs.

[0085] An "internal database" refers to a database that exists within an organization and is accessible to specific users or systems.

[0086] A "facility floor plan" refers to a drawing that shows the internal structure of a building and the layout of rooms.

[0087] An "evacuation route" refers to a route that has been set up to allow people to evacuate safely in the event of a disaster.

[0088] "Stockpile information" refers to information about supplies and materials that have been prepared in advance for use in the event of a disaster.

[0089] "Preprocessing" refers to the process of cleansing and formatting raw data for data analysis and machine learning.

[0090] "Unifying formats" refers to converting data of different formats into a consistent, common format.

[0091] A "large-scale language model" refers to a natural language processing model trained on a large amount of text data.

[0092] A "virtual disaster scenario" refers to a hypothetical scenario that is created by combining information on past disasters with internal company data.

[0093] "Image generation artificial intelligence" refers to a system that uses artificial intelligence technology to generate or process images.

[0094] "Evacuation simulation video" refers to a simulation video that shows evacuation behavior, created based on a hypothetical disaster scenario.

[0095] "Augmented reality content" refers to digital content that displays virtual information overlaid on a real environment.

[0096] "User device" refers to an electronic device operated by a user (e.g., a smartphone, tablet, AR headset, etc.).

[0097] "Cloud storage" refers to an online storage service that allows you to store and manage data via the Internet.

[0098] "URL" is an abbreviation for Uniform Resource Locator, and refers to an address that points to a specific resource on the Internet.

[0099] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0100] First, the server retrieves past disaster information from an external database or API. This includes detailed data on past typhoons and earthquakes. Next, the server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database. The data collected in this way provides the basic information needed to generate disaster scenarios.

[0101] The server preprocesses the acquired data and standardizes its format. During this process, missing data is filled in. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled supplies.

[0102] The generated scenario is passed to a video generation AI system, which creates an evacuation simulation video based on the actual facility environment. 3D models and CG technology are used to generate the video, enabling a realistic simulation. This allows users to experience a realistic evacuation drill.

[0103] Next, the server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0104] Users launch a dedicated application on their device to begin disaster prevention training. The device downloads augmented reality content based on a URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0105] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0106] An example prompt is:

[0107] Generate disaster scenarios by combining past disaster information with your company's data. Provide the following information:

[0108] Detailed data on past typhoons and earthquakes

[0109] Facility floor plan, evacuation route, and emergency supplies information

[0110] Create a hypothetical disaster scenario using a specific artificial intelligence technology (e.g., GPT-4).

[0111] Using these prompts, the system provides the necessary data to generate a virtual disaster scenario that provides a realistic disaster drill.

[0112] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0113] Step 1:

[0114] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to an external API (e.g., the Japan Meteorological Agency API) and receives detailed data on past typhoons, earthquakes, etc. in JSON format.

[0115] Input: External API endpoint URL

[0116] Output: Past disaster data in JSON format

[0117] Step 2:

[0118] The server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database, using SQL queries to extract the necessary data.

[0119] Input: Internal database access information, SQL query

[0120] Output: A dataset containing facility floor plans, evacuation routes, and emergency supplies.

[0121] Step 3:

[0122] The server preprocesses the acquired data and standardizes the format, for example, by using the Pandas library to impute missing values ​​with the median and convert the data into a unified format.

[0123] Input: JSON formatted historical disaster data and datasets obtained from an internal database

[0124] Output: A consistent dataset after preprocessing and formatting

[0125] Step 4:

[0126] The server passes the preprocessed data to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The server creates a prompt sentence and sends a request to GPT-4 based on it.

[0127] Input: Dataset after preprocessing and unifying format, prompt statement

[0128] Output: Text data containing hypothetical disaster scenarios

[0129] Step 5:

[0130] The server then passes the generated virtual disaster scenario to a video generation AI, which then creates an evacuation simulation video. In this case, a video generation tool such as Blender is used.

[0131] Input: Hypothetical disaster scenario

[0132] Output: Evacuation simulation video using 3D models and CG (MP4 format)

[0133] Step 6:

[0134] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. The content is created using an AR development kit such as Unity.

[0135] Input: Virtual disaster scenario, evacuation simulation video

[0136] Output: Augmented reality content (APK or IPA format)

[0137] Step 7:

[0138] The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access. Specifically, the server uploads the content to cloud storage such as AWS (registered trademark) S3.

[0139] Input: Augmented reality content

[0140] Output: Cloud storage URL

[0141] Step 8:

[0142] The user starts the disaster prevention training by launching a dedicated application on the device. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user.

[0143] Input: Cloud storage URL

[0144] Output: Downloading augmented reality content and conducting disaster prevention drills

[0145] Based on this detailed procedure, the present invention can provide users with a realistic and effective disaster prevention training experience.

[0146] (Application example 1)

[0147] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0148] It is difficult to conduct effective and realistic disaster prevention drills in logistics centers and other facilities where many workers gather. While quick and safe evacuation actions are required in the event of a disaster, conventional training methods lack a sense of realism, limiting the effectiveness of the training. Furthermore, there was a lack of a way to manage the training progress of individual workers and visualize it as reports, leaving challenges in raising overall disaster prevention awareness and responding quickly.

[0149] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0150] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on the actual facility environment using image generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, means for downloading the augmented reality content to a user's device so that the content can be viewed and experienced, and means for tracking the user's training progress and generating reports. This allows users to experience a realistic evacuation simulation within the facility, and enables them to check their individual training progress and identify areas for improvement.

[0151] "Past disaster information" refers to detailed data from when a disaster occurred, including the location, time, damage extent, and status of rescue operations of natural disasters such as earthquakes and typhoons.

[0152] "External database or API" refers to an external data source accessible by the server, including an interface through which past disaster information can be obtained.

[0153] A "facility floor plan" refers to a drawing showing the layout of each area and room within a facility, and is used to design evacuation routes and arrange supplies.

[0154] "Evacuation routes" refer to recommended corridors and routes for safe evacuation in the event of a disaster, and are designed to allow for quick evacuation from any location within the facility.

[0155] "Stockpile information" refers to detailed data such as the types, quantities, and storage locations of supplies needed in the event of a disaster.

[0156] A "large-scale language model" is an artificial intelligence trained on massive amounts of text data, and is capable of advanced text generation and analysis through natural language processing.

[0157] A "virtual disaster scenario" refers to a sequence of hypothetical events generated to simulate a real disaster, including evacuation routes and how to use stockpiles of supplies.

[0158] "Image generation artificial intelligence" refers to software and systems that use artificial intelligence technology to generate realistic images, and perform evacuation simulations using 3D models and CG technology.

[0159] "Augmented reality content" refers to content that displays virtual information overlaid on a real environment, allowing users to experience a realistic simulation.

[0160] "User device" refers to the device used to view and experience augmented reality content, including smartphones, tablets, and AR headsets.

[0161] "Training progress tracking" refers to recording the extent to which a user has completed disaster prevention training, as well as their progress and achievement.

[0162] "Generating a report" refers to analyzing training progress and results and summarizing the results in documents, graphs, etc.

[0163] In this invention, several important hardware and software components are combined and operated to build a disaster prevention training system specialized for facilities such as logistics centers. The system is realized by the servers, terminals, and users each playing their respective roles.

[0164] First, the server uses the following means.

[0165] 1. How to obtain past disaster information: The server obtains past disaster information through an external database or API. For example, data such as the location, time, and damage status of earthquakes and typhoons is collected from the API. An example of an API used for this is "https: / / api.disasterinfo.com / get."

[0166] 2. Internal data acquisition method: The server acquires facility floor plans, evacuation routes, and emergency supplies from the internal database. This is done using the facility's internal database and storage system. For example, the server acquires data from the endpoint "https: / / warehouse.internaldb.com / data".

[0167] 3. Data preprocessing: Collected data is preprocessed and the format is standardized. Scripting languages ​​such as Python are used to fill in any deficiencies in the data and ensure consistency.

[0168] 4. Disaster scenario generation means: Communicate with a large-scale language model (e.g., GPT-4) and generate a virtual disaster scenario based on the collected data. An example of an API for the generation AI model is "https: / / api.gpt-4.com / generate_scenario."

[0169] 5. Video generation method: Using video generation AI, evacuation simulation videos based on the actual facility environment are created. In this case, 3D models and CG technology are used, and tools such as "Blender" and "Unity" are often used.

[0170] 6. Means of outputting as augmented reality content: The generated simulation video and scenario are integrated and output as augmented reality content, allowing users to enjoy realistic evacuation drill content using AR technology.

[0171] 7. Training Progress Tracking and Report Generation: The server tracks users' training progress and generates reports. A dedicated management dashboard is used to visualize training achievements and areas for improvement.

[0172] Next, the user's terminal uses the following means:

[0173] 1. Means for downloading and experiencing augmented reality content: The user's device (e.g., smartphone or AR headset) downloads the augmented reality content sent from the server, allowing them to view and experience it. When the user opens the application and presses the "Generate Scenario" button, a request is sent to the server. The generated content is saved in cloud storage, and a URL that can be accessed from the user's device is provided.

[0174] Specific examples

[0175] For example, when conducting a disaster prevention drill at a logistics center, a user can input the following prompt sentence using a smartphone:

[0176] "Please generate an earthquake scenario for a logistics center. The floor plan, evacuation routes, and stockpiled supplies are as follows... [continues data]"

[0177] Through the generated scenarios and augmented reality content, users can experience realistic evacuation simulations and learn appropriate evacuation behavior. Training progress is tracked and detailed reports are generated, allowing the effectiveness of each user's training to be evaluated.

[0178] This allows logistics center workers to learn how to evacuate quickly and safely through realistic disaster prevention training, improving overall disaster prevention awareness.

[0179] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0180] Step 1:

[0181] The server retrieves past disaster information from an external database or API. It uses the API endpoint (e.g., "https: / / api.disasterinfo.com / get") as input and obtains detailed disaster information data (e.g., location, time, and damage extent of earthquakes and typhoons) as output. Specifically, it sends an HTTP GET request and receives and parses the response in JSON format.

[0182] Step 2:

[0183] The server retrieves facility floor plans, evacuation routes, and stockpile information from an internal database. It uses the internal database URL (e.g., "https: / / warehouse.internaldb.com / data") as input and gets the floor plans, evacuation routes, and stockpile information as output. Specifically, it queries the database to retrieve the required data and formats it in the appropriate way.

[0184] Step 3:

[0185] The server preprocesses the data collected and standardizes it into a consistent format. It uses past disaster information and internal company data as input and obtains preprocessed data as output. Specifically, it checks the data type and format, complements any deficiencies, and converts them into a unified format.

[0186] Step 4:

[0187] The server generates a virtual disaster scenario using a large-scale language model (e.g., GPT-4). It uses preprocessed data as input and obtains a disaster scenario as output. Specifically, it sends a POST request to the model API and receives the generated scenario sentence as a response.

[0188] Step 5:

[0189] The server uses video generation artificial intelligence to create an evacuation simulation video based on the actual facility environment. It uses the generated disaster scenario as input and obtains the evacuation simulation video as output. Specifically, it uses 3D modeling software for video generation (e.g., Blender or Unity) to create a video based on the scenario.

[0190] Step 6:

[0191] The server integrates the generated simulation video and scenario and outputs it as augmented reality content. It uses the evacuation simulation video and disaster scenario as input and obtains augmented reality content as output. Specifically, it uses an AR toolkit (e.g., ARCore or ARKit) to prepare the content for display as augmented reality.

[0192] Step 7:

[0193] The server saves the augmented reality content in cloud storage and generates a URL that the user can access. The augmented reality content is used as input, and a URL on the cloud storage is obtained as output. Specifically, the data is uploaded to a cloud storage service (e.g., Amazon S3 or Google® Cloud Storage) and the URL is obtained.

[0194] Step 8:

[0195] The user downloads, views, and experiences augmented reality content using a device. The input is a cloud storage URL, and the output is the user's ability to view the content. Specifically, the app launches, downloads the content from the provided URL, and switches to AR experience mode.

[0196] Step 9:

[0197] The server tracks training progress and generates reports. It uses the user's training data as input and obtains a training progress report as output. Specifically, it records the user's activity log in a database, periodically aggregates and analyzes it, and generates reports.

[0198] By going through the above processing steps, a realistic evacuation training system can be realized for a facility such as a logistics center.

[0199] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0200] This invention combines information on past disasters with in-house data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR), with an emotion engine that recognizes the user's emotions. This system is realized by the server, terminals, and users each playing their respective roles.

[0201] First, the server retrieves past disaster information from an external database or API. The disaster information retrieved here is specific data on typhoons, earthquakes, etc., including the date and time of occurrence, the scale of damage, and the area of ​​impact. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies information from the internal database. This provides information such as the office structure, the layout of each room, main evacuation routes, and a list of emergency supplies.

[0202] The server preprocesses the acquired data and standardizes its format. The preprocessed data is then fed into a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use emergency supplies during evacuation.

[0203] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior, allowing users to experience a realistic evacuation drill.

[0204] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0205] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0206] This is where the emotion engine, a key feature of the present invention, comes into play. The device recognizes emotions from the user's facial expressions, voice, and other data in real time and transmits them to the server. The server then dynamically adjusts the training scenario based on the received emotion data. For example, if the user is feeling anxious, the system can provide more detailed guidance and encouraging messages. The system also monitors the user's emotional state and adjusts the training to prevent overly stressful experiences.

[0207] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[0208] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, helping them to evacuate more quickly and safely in the event of a disaster. Furthermore, the addition of an emotion engine can further improve the quality of the training and the user experience.

[0209] The processing flow will be explained below.

[0210] Step 1:

[0211] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0212] Step 2:

[0213] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0214] Step 3:

[0215] The server preprocesses the acquired data and standardizes its format. Specifically, it standardizes the data format and fills in any missing information. This preprocessing includes data cleansing and normalization.

[0216] Step 4:

[0217] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0218] Step 5:

[0219] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0220] Step 6:

[0221] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0222] Step 7:

[0223] The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and generates an access link to the file.

[0224] Step 8:

[0225] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0226] Step 9:

[0227] The device uses an emotion engine to recognize emotions in real time from the user's facial expressions, voice, etc. The emotion engine analyzes data acquired through the camera and microphone to identify the user's emotional state (e.g., anxiety, relief, concentration).

[0228] Step 10:

[0229] When a user starts a disaster prevention training, the device sends emotional data to the server through the emotion engine. The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance or encouraging messages.

[0230] Step 11:

[0231] Users can play the downloaded augmented reality content in the app and experience disaster prevention drills. The device uses an emotion engine to monitor the user's emotional state in real time and sends feedback to the server, which then adjusts the content to prevent the drill from becoming too stressful.

[0232] In this way, the system combined with the emotion engine enables users to receive realistic and effective disaster prevention training, supporting them in taking quick and safe evacuation actions in the event of a disaster. The addition of the emotion engine further improves the quality of the training and the user experience.

[0233] Example 2

[0234] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0235] Conventional disaster prevention training systems are limited to simulated training that lacks realism, and users are not fully prepared to take appropriate actions in a real disaster situation. Furthermore, training does not take into account the user's emotional state, making it difficult to improve the effectiveness of the training. To solve these problems, a realistic training environment and dynamic scenario adjustments based on the user's emotional state are needed.

[0236] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0237] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating a virtual disaster scenario using a large-scale language model, means for creating an evacuation simulation video based on an actual facility environment using image recognition technology, means for integrating the generated simulation video and the scenario and outputting it as augmented reality content, means for downloading the augmented reality content to a user's device so that the user can view and experience it, and means for recognizing the user's emotions in real time using an emotion recognition engine and dynamically adjusting the training scenario based on that data. This allows users to experience a realistic disaster prevention training, and the dynamic adjustment of the training scenario according to their emotional state allows for more effective training.

[0238] "Past disaster information" refers to information such as the date and time of occurrence, scale of damage, and area of ​​impact regarding disasters such as typhoons and earthquakes, which is obtained through external databases or APIs.

[0239] An "internal database" refers to a database that stores data about the facility, such as office floor plans, evacuation routes, and information on stockpiled supplies.

[0240] A "large-scale language model" is an artificial intelligence model based on massive amounts of data that can generate and analyze text. For example, GPT-4 is one such model.

[0241] "Image recognition technology" refers to the technology that analyzes image data from cameras and sensors and creates evacuation simulation images based on a real office environment.

[0242] "Augmented reality content" refers to a form of content that integrates generated simulation footage with disaster scenarios, allowing users to experience virtual information while viewing the real world.

[0243] "Cloud storage" refers to an online storage service that stores data over the Internet and generates a URL that users can access.

[0244] An "emotion recognition engine" is an engine that recognizes emotions from a user's facial expressions and voice in real time and adjusts the system's operation based on that data.

[0245] "User's device" refers to devices used by users, such as smartphones, tablets, and AR headsets.

[0246] "Dynamic adjustment" refers to changing the scenario or system behavior appropriately in response to the user's emotional state and other real-time data.

[0247] This invention is a system that combines past disaster information with internal data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0248] First, the server retrieves past disaster information from an external database or API. The hardware used here is a high-performance server, and the software is a program that makes API calls. The external databases used include the government's disaster prevention database and the Japan Meteorological Agency's API. Specific data retrieved includes the date and time of typhoon and earthquake occurrence, the scale of damage, and the area of ​​impact.

[0249] Next, the server retrieves the facility's floor plan, evacuation routes, and stockpiled goods information from an internal database. This database uses an SQL database, and the server executes queries to retrieve the necessary information. This provides the facility's layout, the layout of each room, main evacuation routes, and a list of stockpiled goods.

[0250] The acquired data is preprocessed by the server. During the preprocessing stage, data formats are standardized, missing data is filled in, and outliers are corrected. The preprocessed data is then input into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled items needed during evacuation.

[0251] The generated scenario is passed to a video generation AI, which uses 3D modeling software (such as Blender or Unreal Engine) to create an evacuation simulation video based on the actual facility environment, allowing users to experience a realistic evacuation drill.

[0252] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content is saved in cloud storage (e.g., AWS S3) and a URL is generated that users can access.

[0253] The user's device (smartphone, tablet, AR headset, etc.) downloads the augmented reality content from this cloud storage and plays it when the user specifies. When the user launches the application and starts a disaster prevention drill, the device uses an emotion engine to recognize the user's emotions in real time. The emotion engine analyzes the user's emotional state from their facial expressions and voice and sends the data to the server.

[0254] The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance and encouraging messages. It also monitors the user's emotional state and adjusts the training to prevent the user from feeling overly stressed.

[0255] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[0256] Examples of prompts include:

[0257] "Generate evacuation scenarios in the event of flooding due to a typhoon, and create evacuation simulation videos based on floor plans and evacuation routes for designated facilities. Also, include reassuring messages if users are feeling anxious."

[0258] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0259] Program processing flow

[0260] Step 1: Obtaining past disaster information

[0261] The server obtains past disaster information through an external database or API. As input data, it receives information such as the date and time of typhoon or earthquake occurrence, the scale of damage, and the affected area from the government's disaster prevention database or the API of the Japan Meteorological Agency. As output data, it saves the obtained disaster information in JSON or CSV format.

[0262] Step 2: Acquire internal data

[0263] The server retrieves facility floor plans, evacuation routes, and stockpiled goods information from an internal database. It queries an internal SQL database to retrieve the layout of each room in the facility, the main evacuation routes, and a list of stockpiled goods. It saves this information in a standardized format as output data.

[0264] Step 3: Preprocessing the data

[0265] The server preprocesses the acquired disaster information and in-house data. The data acquired in Step 1 and Step 2 are used as input data. Preprocessing involves filling in missing data, detecting and correcting outliers, and standardizing data formats. A consistent preprocessed dataset is generated as output data.

[0266] Step 4: Generate disaster scenarios

[0267] The server inputs the preprocessed data into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. The preprocessed dataset and prompt sentences are used as input data. The data calculation generates a scenario based on the language model. The output data is a scenario that includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0268] Step 5: Generate footage

[0269] The server passes the generated scenario to a video generation AI, which creates an evacuation simulation video based on the actual facility environment. The generated scenario and a 3D model of the facility are used as input data. 3D modeling software (Blender or Unreal Engine) is used for data calculations to generate a video that realistically reproduces evacuation behavior. The evacuation simulation video is output as output data.

[0270] Step 6: Generate and serve augmented reality (AR) content

[0271] The server integrates the generated scenario and video and outputs it as augmented reality content. The scenario and video are used as input data. Data processing involves converting it into AR format. AR content is generated as output data and uploaded to cloud storage.

[0272] Step 7: Save to cloud storage and generate access URL

[0273] The server saves the generated AR content in cloud storage and generates a URL that the user can access. The AR content is used as input data. As processing, the file is uploaded to a cloud storage service (e.g., AWS S3) and an access URL is generated. As output data, an access URL that the user can use is generated.

[0274] Step 8: Download and play augmented reality content

[0275] The device downloads the augmented reality content from cloud storage and plays it when the user specifies. The generated access URL is used as input data. The downloaded content is then played on the application as processing. The augmented reality content that the user can experience is displayed as output data.

[0276] Step 9: Recognizing User Emotions

[0277] The device uses an emotion engine to recognize the user's emotions in real time. The input data is the user's facial expressions and voice data. The processing is performed using emotion recognition technology for analysis. The output data is the user's emotional state, which is sent to the server.

[0278] Step 10: Dynamically adjust the training scenario

[0279] The server dynamically adjusts the training scenario based on the received user emotional data. The emotional data received in real time is used as input data. The data calculation changes the difficulty and content of the scenario according to the emotional state. The adjusted training scenario is provided to the user as output data.

[0280] (Application example 2)

[0281] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0282] Conducting realistic and effective evacuation drills in the event of a disaster is an important challenge in industrial sites such as factories and manufacturing facilities. However, conventional evacuation drills are expensive to implement, have fixed scenarios, and often do not adequately adapt to real-world disaster situations. Furthermore, training is often conducted without taking into account the emotions and stress levels of employees, resulting in ineffective training. Furthermore, it is difficult to generate appropriate evacuation scenarios by combining disaster information and facility data, and there are limited means of providing realistic simulation images.

[0283] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0284] In this invention, the server includes: means for acquiring past disaster information from an external database or API; means for acquiring workplace floor plans, evacuation routes, and emergency supply information from an internal database; means for generating a virtual disaster scenario using a large-scale language model; means for creating an evacuation simulation video based on an actual work environment using image generation AI; means for integrating the generated simulation video and the scenario and outputting it as augmented reality content; means for downloading the augmented reality content to a user's device so that it can be viewed and experienced; means for recognizing user emotions in real time and collecting that data; and means for dynamically adjusting the training scenario based on the user's emotion data. This makes it possible to conduct realistic and effective evacuation drills and provide appropriate training content that takes into account the emotions and stress of employees.

[0285] "Disaster information" refers to data on natural and man-made disasters that have occurred in the past, including the date and time of the disaster, the scale of the damage, and the area of ​​impact.

[0286] An "internal database" is a collection of information stored within a particular facility or organization, including floor plans, evacuation routes, and emergency supply information.

[0287] A "large-scale language model" is a type of artificial intelligence model that has a very large number of parameters and performs natural language processing by learning from huge amounts of text data.

[0288] A "virtual disaster scenario" is a fictitious story or situation in which a disaster situation or evacuation route is generated by computer simulation.

[0289] "Image generation artificial intelligence" is an artificial intelligence system that can generate realistic images and simulations using 3D models and CG technology.

[0290] A "work environment" refers to the location where a specific task is performed, such as a factory or manufacturing facility, and the surrounding environment.

[0291] "Augmented reality content" means applications or content that overlay digital information onto a real-world environment, creating a visual and auditory experience.

[0292] A "terminal" is a device used by a user to access the system, and includes a smartphone, tablet, etc.

[0293] "Emotion engine" refers to technology that analyzes the user's facial expressions and voice to recognize their current emotional state.

[0294] A "training scenario" is a plan or scenario that shows the procedures and specific training content for disaster prevention and evacuation drills.

[0295] "Dynamic adjustment" means changing and adapting in real time to the situation.

[0296] "Collected Data" is a general term for information obtained from external and internal databases.

[0297] "Unifying formats" means organizing and converting data of different formats according to consistent standards.

[0298] "Cloud storage" is a storage service for saving and managing data over the Internet.

[0299] A "URL" is a uniform resource identifier used to identify resources on the Internet.

[0300] The present invention is a system for implementing effective evacuation drills at industrial sites such as factories, manufacturing facilities, etc. This system operates by having a server, terminals, and users each fulfill their respective roles.

[0301] The server first obtains past disaster information from an external database or API. The obtained disaster information includes the date and time of the disaster, the scale of damage, and the area of ​​impact. The server then obtains information on the workplace floor plan, evacuation routes, and stockpiles from an internal database. This data is then preprocessed and converted into a unified format.

[0302] The server then uses a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario, including specific disaster situations, evacuation routes, and how to use necessary supplies during evacuation. The generated scenario is then passed to a video generation AI, which creates an evacuation simulation video based on a real-world working environment.

[0303] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content is downloaded to the user's device (e.g., smartphone, head-mounted display) and can be viewed and experienced. The augmented reality content stored in cloud storage generates a URL that the user can access, making it easy to download when needed.

[0304] The device downloads augmented reality content from cloud storage at the user's specified time. When the user launches the application and begins a disaster prevention training, the device uses an emotion engine to recognize emotions in real time from the user's facial expressions and voice, and sends them to the server. The server dynamically adjusts the training scenario based on the received emotion data. For example, if the user feels anxious, the system will adjust the scenario by providing more detailed guidance and explanations.

[0305] In this way, the system of the present invention enables users to receive realistic and effective evacuation training, helping them to evacuate more quickly and safely in the event of a disaster. The addition of an emotion engine can further improve the quality of the training and the user experience.

[0306] ■Example:

[0307] For example, if a fire breaks out in a factory, the system uses past fire information to generate a virtual evacuation scenario. This scenario includes the location of fire extinguishers stored in each room, along with evacuation routes within the factory. The system monitors the user's emotions in real time, and if the user feels anxious, it provides detailed guidance on how to use a fire extinguisher and the best evacuation route.

[0308] ■Example of a prompt:

[0309] "Generate a disaster scenario. Use the following data:

[0310] Disaster information: Fire, date of occurrence, area of ​​impact

[0311] Internal factory data: floor plans, evacuation routes, and stockpile information

[0312] Required outcome: A detailed description of the evacuation scenario and the supplies that should be used.”

[0313] This enables realistic and effective evacuation drills, ensuring the safety of employees and quick evacuation.

[0314] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0315] Step 1:

[0316] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to the disaster information API, and the retrieved disaster information includes the date and time of the occurrence, the scale of damage, the area of ​​impact, etc. The API request is made as input, and disaster information data is obtained as output.

[0317] Step 2:

[0318] The server retrieves the workplace floor plan, evacuation routes, and stockpile information from an internal database. The internal database stores the factory floor plan, evacuation routes, and stockpile location information. The database query is performed as input, and the internal data is obtained as output.

[0319] Step 3:

[0320] The server preprocesses the acquired disaster information and internal data and standardizes their formats. Specifically, it converts each dataset into the same format and filters out unnecessary data. It receives disaster information data and internal data as input and outputs data in a unified format.

[0321] Step 4:

[0322] The server inputs the unified data into a large-scale language model (e.g., GPT-4) to generate a hypothetical disaster scenario. The unified data is passed to the model as input, and the disaster scenario is obtained as output. The model generates the scenario using a prompt sentence.

[0323] Step 5:

[0324] The server then passes the generated disaster scenario to a video generation AI system, which then creates an evacuation simulation video based on the actual work environment.The disaster scenario is used as input, and the simulation video, using 3D models and CG technology, is obtained as output.

[0325] Step 6:

[0326] The server integrates the simulation video and the generated scenario and outputs it as augmented reality content.The simulation video and scenario are integrated as input, and the completed augmented reality content is obtained as output.

[0327] Step 7:

[0328] Save augmented reality content in cloud storage and generate a URL that users can access. Upload augmented reality content to the cloud as input, and generate an access URL as output.

[0329] Step 8:

[0330] The user's device launches an application that downloads the augmented reality content from cloud storage and allows it to be viewed and experienced, using a URL as input and the downloaded content as output.

[0331] Step 9:

[0332] The device uses an emotion engine to recognize emotions from the user's facial expressions, voice, etc. in real time while the user is experiencing augmented reality content, and transmits the emotions to the server. The device receives emotion data collected in real time as input, and obtains emotion data to be sent to the server as output.

[0333] Step 10:

[0334] The server dynamically adjusts the training scenario based on the received emotional data. It receives the emotional data as input and outputs the adjusted training scenario in real time. For example, if the user feels anxious, it can provide more detailed guidance or encouraging messages.

[0335] Through these steps, users can receive realistic and effective evacuation training. The data processing and calculations performed at each step support more appropriate and safe evacuation behavior.

[0336] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0337] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0338] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0339] [Second embodiment]

[0340] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0341] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0342] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0343] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0344] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0345] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0346] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0347] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0348] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0349] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0350] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0351] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0352] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0353] First, the server retrieves past disaster information from an external database or API. This collects detailed data on past typhoons, earthquakes, and other events. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies from the company's internal database. This data provides the basic information needed to generate disaster scenarios.

[0354] The server preprocesses the acquired data and standardizes the format. Any inaccuracies in the data are corrected. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0355] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG to create a realistic evacuation simulation, allowing users to experience a realistic evacuation drill.

[0356] The server then integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0357] Users launch the application on their devices to begin disaster prevention training. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0358] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0359] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, and supports users in taking quicker and safer evacuation actions in the event of a disaster.

[0360] The processing flow will be explained below.

[0361] Step 1:

[0362] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0363] Step 2:

[0364] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0365] Step 3:

[0366] The server preprocesses the acquired data by standardizing the data format and filling in any missing information. This preprocessing includes data cleansing and normalization.

[0367] Step 4:

[0368] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0369] Step 5:

[0370] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0371] Step 6:

[0372] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0373] Step 7:

[0374] The server saves the augmented reality content in cloud storage and generates a URL that users can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and creates an access link to the file.

[0375] Step 8:

[0376] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0377] Step 9:

[0378] Users can play the downloaded augmented reality content in the application to experience a disaster prevention drill. The user's device will then visually display evacuation route guides and instructions on how to use stockpiled supplies, helping them to carry out the drill in a realistic situation.

[0379] In this way, the entire system works together to realize dynamic and realistic disaster prevention drills in real time.

[0380] Example 1

[0381] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0382] In recent years, the increasing importance of disaster prevention measures has created a demand for realistic and effective disaster prevention training. However, conventional disaster prevention training is difficult to implement and often lacks realism, making it difficult for participants to take appropriate action in the event of a real disaster. Furthermore, technology for efficiently handling large amounts of data and generating virtual scenarios has been limited. This has resulted in training that is not tailored to the actual situation on the ground and is therefore difficult to demonstrate effectiveness in the event of a real disaster. To address these issues, a system is needed that combines past disaster information with internal company data to generate realistic virtual disaster scenarios and utilizes augmented reality to provide effective disaster prevention training.

[0383] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0384] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supplies information from an internal database, means for preprocessing the collected data and standardizing the format, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on real facility environments using video generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, and means for downloading the augmented reality content to a user's device so that the user can view and experience it. This allows users to experience realistic disaster prevention training and take prompt and appropriate action in the event of an actual disaster.

[0385] "Past disaster information" refers to detailed data on natural disasters that have occurred in the past (e.g., typhoons, earthquakes, floods, etc.).

[0386] An "external database" refers to a database that exists outside an organization and is accessible via the Internet or other means.

[0387] "API" stands for Application Program Interface, and refers to a standardized interface for exchanging data and functions between different software programs.

[0388] An "internal database" refers to a database that exists within an organization and is accessible to specific users or systems.

[0389] A "facility floor plan" refers to a drawing that shows the internal structure of a building and the layout of rooms.

[0390] An "evacuation route" refers to a route that has been set up to allow people to evacuate safely in the event of a disaster.

[0391] "Stockpile information" refers to information about supplies and materials that have been prepared in advance for use in the event of a disaster.

[0392] "Preprocessing" refers to the process of cleansing and formatting raw data for data analysis and machine learning.

[0393] "Unifying formats" refers to converting data of different formats into a consistent, common format.

[0394] A "large-scale language model" refers to a natural language processing model trained on a large amount of text data.

[0395] A "virtual disaster scenario" refers to a hypothetical scenario that is created by combining information on past disasters with internal company data.

[0396] "Image generation artificial intelligence" refers to a system that uses artificial intelligence technology to generate or process images.

[0397] "Evacuation simulation video" refers to a simulation video that shows evacuation behavior, created based on a hypothetical disaster scenario.

[0398] "Augmented reality content" refers to digital content that displays virtual information overlaid on a real environment.

[0399] "User device" refers to an electronic device operated by a user (e.g., a smartphone, tablet, AR headset, etc.).

[0400] "Cloud storage" refers to an online storage service that allows you to store and manage data via the Internet.

[0401] "URL" is an abbreviation for Uniform Resource Locator, and refers to an address that points to a specific resource on the Internet.

[0402] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0403] First, the server retrieves past disaster information from an external database or API. This includes detailed data on past typhoons and earthquakes. Next, the server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database. The data collected in this way provides the basic information needed to generate disaster scenarios.

[0404] The server preprocesses the acquired data and standardizes its format. During this process, missing data is filled in. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled supplies.

[0405] The generated scenario is passed to a video generation AI system, which creates an evacuation simulation video based on the actual facility environment. 3D models and CG technology are used to generate the video, enabling a realistic simulation. This allows users to experience a realistic evacuation drill.

[0406] Next, the server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0407] Users launch a dedicated application on their device to begin disaster prevention training. The device downloads augmented reality content based on a URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0408] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0409] An example prompt is:

[0410] Generate disaster scenarios by combining past disaster information with your company's data. Provide the following information:

[0411] Detailed data on past typhoons and earthquakes

[0412] Facility floor plan, evacuation route, and emergency supplies information

[0413] Create a hypothetical disaster scenario using a specific artificial intelligence technology (e.g., GPT-4).

[0414] Using these prompts, the system provides the necessary data to generate a virtual disaster scenario that provides a realistic disaster drill.

[0415] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0416] Step 1:

[0417] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to an external API (e.g., the Japan Meteorological Agency API) and receives detailed data on past typhoons, earthquakes, etc. in JSON format.

[0418] Input: External API endpoint URL

[0419] Output: Past disaster data in JSON format

[0420] Step 2:

[0421] The server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database, using SQL queries to extract the necessary data.

[0422] Input: Internal database access information, SQL query

[0423] Output: A dataset containing facility floor plans, evacuation routes, and emergency supplies.

[0424] Step 3:

[0425] The server preprocesses the acquired data and standardizes the format, for example, by using the Pandas library to impute missing values ​​with the median and convert the data into a unified format.

[0426] Input: JSON formatted historical disaster data and datasets obtained from an internal database

[0427] Output: A consistent dataset after preprocessing and formatting

[0428] Step 4:

[0429] The server passes the preprocessed data to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The server creates a prompt sentence and sends a request to GPT-4 based on it.

[0430] Input: Dataset after preprocessing and unifying format, prompt statement

[0431] Output: Text data containing hypothetical disaster scenarios

[0432] Step 5:

[0433] The server then passes the generated virtual disaster scenario to a video generation AI, which then creates an evacuation simulation video. In this case, a video generation tool such as Blender is used.

[0434] Input: Hypothetical disaster scenario

[0435] Output: Evacuation simulation video using 3D models and CG (MP4 format)

[0436] Step 6:

[0437] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. The content is created using an AR development kit such as Unity.

[0438] Input: Virtual disaster scenario, evacuation simulation video

[0439] Output: Augmented reality content (APK or IPA format)

[0440] Step 7:

[0441] The server saves the generated augmented reality content in cloud storage and generates a URL that users can access. Specifically, it uploads the content to a cloud storage such as AWS S3.

[0442] Input: Augmented reality content

[0443] Output: Cloud storage URL

[0444] Step 8:

[0445] The user starts the disaster prevention training by launching a dedicated application on the device. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user.

[0446] Input: Cloud storage URL

[0447] Output: Downloading augmented reality content and conducting disaster prevention drills

[0448] Based on this detailed procedure, the present invention can provide users with a realistic and effective disaster prevention training experience.

[0449] (Application example 1)

[0450] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0451] It is difficult to conduct effective and realistic disaster prevention drills in logistics centers and other facilities where many workers gather. While quick and safe evacuation actions are required in the event of a disaster, conventional training methods lack a sense of realism, limiting the effectiveness of the training. Furthermore, there was a lack of a way to manage the training progress of individual workers and visualize it as reports, leaving challenges in raising overall disaster prevention awareness and responding quickly.

[0452] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0453] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on the actual facility environment using image generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, means for downloading the augmented reality content to a user's device so that the content can be viewed and experienced, and means for tracking the user's training progress and generating reports. This allows users to experience a realistic evacuation simulation within the facility, and enables them to check their individual training progress and identify areas for improvement.

[0454] "Past disaster information" refers to detailed data from when a disaster occurred, including the location, time, damage extent, and status of rescue operations of natural disasters such as earthquakes and typhoons.

[0455] "External database or API" refers to an external data source accessible by the server, including an interface through which past disaster information can be obtained.

[0456] A "facility floor plan" refers to a drawing showing the layout of each area and room within a facility, and is used to design evacuation routes and arrange supplies.

[0457] "Evacuation routes" refer to recommended corridors and routes for safe evacuation in the event of a disaster, and are designed to allow for quick evacuation from any location within the facility.

[0458] "Stockpile information" refers to detailed data such as the types, quantities, and storage locations of supplies needed in the event of a disaster.

[0459] A "large-scale language model" is an artificial intelligence trained on massive amounts of text data, and is capable of advanced text generation and analysis through natural language processing.

[0460] A "virtual disaster scenario" refers to a sequence of hypothetical events generated to simulate a real disaster, including evacuation routes and how to use stockpiles of supplies.

[0461] "Image generation artificial intelligence" refers to software and systems that use artificial intelligence technology to generate realistic images, and perform evacuation simulations using 3D models and CG technology.

[0462] "Augmented reality content" refers to content that displays virtual information overlaid on a real environment, allowing users to experience a realistic simulation.

[0463] "User device" refers to the device used to view and experience augmented reality content, including smartphones, tablets, and AR headsets.

[0464] "Training progress tracking" refers to recording the extent to which a user has completed disaster prevention training, as well as their progress and achievement.

[0465] "Generating a report" refers to analyzing training progress and results and summarizing the results in documents, graphs, etc.

[0466] In this invention, several important hardware and software components are combined and operated to build a disaster prevention training system specialized for facilities such as logistics centers. The system is realized by the servers, terminals, and users each playing their respective roles.

[0467] First, the server uses the following means.

[0468] 1. How to obtain past disaster information: The server obtains past disaster information through an external database or API. For example, data such as the location, time, and damage status of earthquakes and typhoons is collected from the API. An example of an API used for this is "https: / / api.disasterinfo.com / get."

[0469] 2. Internal data acquisition method: The server acquires facility floor plans, evacuation routes, and emergency supplies from the internal database. This is done using the facility's internal database and storage system. For example, the server acquires data from the endpoint "https: / / warehouse.internaldb.com / data".

[0470] 3. Data preprocessing: Collected data is preprocessed and the format is standardized. Scripting languages ​​such as Python are used to fill in any deficiencies in the data and ensure consistency.

[0471] 4. Disaster scenario generation means: Communicate with a large-scale language model (e.g., GPT-4) and generate a virtual disaster scenario based on the collected data. An example of an API for the generation AI model is "https: / / api.gpt-4.com / generate_scenario."

[0472] 5. Video generation method: Using video generation AI, evacuation simulation videos based on the actual facility environment are created. In this case, 3D models and CG technology are used, and tools such as "Blender" and "Unity" are often used.

[0473] 6. Means of outputting as augmented reality content: The generated simulation video and scenario are integrated and output as augmented reality content, allowing users to enjoy realistic evacuation drill content using AR technology.

[0474] 7. Training Progress Tracking and Report Generation: The server tracks users' training progress and generates reports. A dedicated management dashboard is used to visualize training achievements and areas for improvement.

[0475] Next, the user's terminal uses the following means:

[0476] 1. Means for downloading and experiencing augmented reality content: The user's device (e.g., smartphone or AR headset) downloads the augmented reality content sent from the server, allowing them to view and experience it. When the user opens the application and presses the "Generate Scenario" button, a request is sent to the server. The generated content is saved in cloud storage, and a URL that can be accessed from the user's device is provided.

[0477] Specific examples

[0478] For example, when conducting a disaster prevention drill at a logistics center, a user can input the following prompt sentence using a smartphone:

[0479] "Please generate an earthquake scenario for a logistics center. The floor plan, evacuation routes, and stockpiled supplies are as follows... [continues data]"

[0480] Through the generated scenarios and augmented reality content, users can experience realistic evacuation simulations and learn appropriate evacuation behavior. Training progress is tracked and detailed reports are generated, allowing the effectiveness of each user's training to be evaluated.

[0481] This allows logistics center workers to learn how to evacuate quickly and safely through realistic disaster prevention training, improving overall disaster prevention awareness.

[0482] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0483] Step 1:

[0484] The server retrieves past disaster information from an external database or API. It uses the API endpoint (e.g., "https: / / api.disasterinfo.com / get") as input and obtains detailed disaster information data (e.g., location, time, and damage extent of earthquakes and typhoons) as output. Specifically, it sends an HTTP GET request and receives and parses the response in JSON format.

[0485] Step 2:

[0486] The server retrieves facility floor plans, evacuation routes, and stockpile information from an internal database. It uses the internal database URL (e.g., "https: / / warehouse.internaldb.com / data") as input and gets the floor plans, evacuation routes, and stockpile information as output. Specifically, it queries the database to retrieve the required data and formats it in the appropriate way.

[0487] Step 3:

[0488] The server preprocesses the data collected and standardizes it into a consistent format. It uses past disaster information and internal company data as input and obtains preprocessed data as output. Specifically, it checks the data type and format, complements any deficiencies, and converts them into a unified format.

[0489] Step 4:

[0490] The server generates a virtual disaster scenario using a large-scale language model (e.g., GPT-4). It uses preprocessed data as input and obtains a disaster scenario as output. Specifically, it sends a POST request to the model API and receives the generated scenario sentence as a response.

[0491] Step 5:

[0492] The server uses video generation artificial intelligence to create an evacuation simulation video based on the actual facility environment. It uses the generated disaster scenario as input and obtains the evacuation simulation video as output. Specifically, it uses 3D modeling software for video generation (e.g., Blender or Unity) to create a video based on the scenario.

[0493] Step 6:

[0494] The server integrates the generated simulation video and scenario and outputs it as augmented reality content. It uses the evacuation simulation video and disaster scenario as input and obtains augmented reality content as output. Specifically, it uses an AR toolkit (e.g., ARCore or ARKit) to prepare the content for display as augmented reality.

[0495] Step 7:

[0496] The server saves the augmented reality content in cloud storage and generates a URL that the user can access. The augmented reality content is used as input, and a URL on the cloud storage is obtained as output. Specifically, the data is uploaded to a cloud storage service (e.g., Amazon S3 or Google Cloud Storage) and the URL is obtained.

[0497] Step 8:

[0498] The user downloads, views, and experiences augmented reality content using a device. The input is a cloud storage URL, and the output is the user's ability to view the content. Specifically, the app launches, downloads the content from the provided URL, and switches to AR experience mode.

[0499] Step 9:

[0500] The server tracks training progress and generates reports. It uses the user's training data as input and obtains a training progress report as output. Specifically, it records the user's activity log in a database, periodically aggregates and analyzes it, and generates reports.

[0501] By going through the above processing steps, a realistic evacuation training system can be realized for a facility such as a logistics center.

[0502] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0503] This invention combines information on past disasters with in-house data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR), with an emotion engine that recognizes the user's emotions. This system is realized by the server, terminals, and users each playing their respective roles.

[0504] First, the server retrieves past disaster information from an external database or API. The disaster information retrieved here is specific data on typhoons, earthquakes, etc., including the date and time of occurrence, the scale of damage, and the area of ​​impact. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies information from the internal database. This provides information such as the office structure, the layout of each room, main evacuation routes, and a list of emergency supplies.

[0505] The server preprocesses the acquired data and standardizes its format. The preprocessed data is then fed into a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use emergency supplies during evacuation.

[0506] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior, allowing users to experience a realistic evacuation drill.

[0507] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0508] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0509] This is where the emotion engine, a key feature of the present invention, comes into play. The device recognizes emotions from the user's facial expressions, voice, and other data in real time and transmits them to the server. The server then dynamically adjusts the training scenario based on the received emotion data. For example, if the user is feeling anxious, the system can provide more detailed guidance and encouraging messages. The system also monitors the user's emotional state and adjusts the training to prevent overly stressful experiences.

[0510] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[0511] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, helping them to evacuate more quickly and safely in the event of a disaster. Furthermore, the addition of an emotion engine can further improve the quality of the training and the user experience.

[0512] The processing flow will be explained below.

[0513] Step 1:

[0514] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0515] Step 2:

[0516] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0517] Step 3:

[0518] The server preprocesses the acquired data and standardizes its format. Specifically, it standardizes the data format and fills in any missing information. This preprocessing includes data cleansing and normalization.

[0519] Step 4:

[0520] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0521] Step 5:

[0522] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0523] Step 6:

[0524] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0525] Step 7:

[0526] The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and generates an access link to the file.

[0527] Step 8:

[0528] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0529] Step 9:

[0530] The device uses an emotion engine to recognize emotions in real time from the user's facial expressions, voice, etc. The emotion engine analyzes data acquired through the camera and microphone to identify the user's emotional state (e.g., anxiety, relief, concentration).

[0531] Step 10:

[0532] When a user starts a disaster prevention training, the device sends emotional data to the server through the emotion engine. The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance or encouraging messages.

[0533] Step 11:

[0534] Users can play the downloaded augmented reality content in the app and experience disaster prevention drills. The device uses an emotion engine to monitor the user's emotional state in real time and sends feedback to the server, which then adjusts the content to prevent the drill from becoming too stressful.

[0535] In this way, the system combined with the emotion engine enables users to receive realistic and effective disaster prevention training, supporting them in taking quick and safe evacuation actions in the event of a disaster. The addition of the emotion engine further improves the quality of the training and the user experience.

[0536] Example 2

[0537] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0538] Conventional disaster prevention training systems are limited to simulated training that lacks realism, and users are not fully prepared to take appropriate actions in a real disaster situation. Furthermore, training does not take into account the user's emotional state, making it difficult to improve the effectiveness of the training. To solve these problems, a realistic training environment and dynamic scenario adjustments based on the user's emotional state are needed.

[0539] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0540] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating a virtual disaster scenario using a large-scale language model, means for creating an evacuation simulation video based on an actual facility environment using image recognition technology, means for integrating the generated simulation video and the scenario and outputting it as augmented reality content, means for downloading the augmented reality content to a user's device so that the user can view and experience it, and means for recognizing the user's emotions in real time using an emotion recognition engine and dynamically adjusting the training scenario based on that data. This allows users to experience a realistic disaster prevention training, and the dynamic adjustment of the training scenario according to their emotional state allows for more effective training.

[0541] "Past disaster information" refers to information such as the date and time of occurrence, scale of damage, and area of ​​impact regarding disasters such as typhoons and earthquakes, which is obtained through external databases or APIs.

[0542] An "internal database" refers to a database that stores data about the facility, such as office floor plans, evacuation routes, and information on stockpiled supplies.

[0543] A "large-scale language model" is an artificial intelligence model based on massive amounts of data that can generate and analyze text. For example, GPT-4 is one such model.

[0544] "Image recognition technology" refers to the technology that analyzes image data from cameras and sensors and creates evacuation simulation images based on a real office environment.

[0545] "Augmented reality content" refers to a form of content that integrates generated simulation footage with disaster scenarios, allowing users to experience virtual information while viewing the real world.

[0546] "Cloud storage" refers to an online storage service that stores data over the Internet and generates a URL that users can access.

[0547] An "emotion recognition engine" is an engine that recognizes emotions from a user's facial expressions and voice in real time and adjusts the system's operation based on that data.

[0548] "User's device" refers to devices used by users, such as smartphones, tablets, and AR headsets.

[0549] "Dynamic adjustment" refers to changing the scenario or system behavior appropriately in response to the user's emotional state and other real-time data.

[0550] This invention is a system that combines past disaster information with internal data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0551] First, the server retrieves past disaster information from an external database or API. The hardware used here is a high-performance server, and the software is a program that makes API calls. The external databases used include the government's disaster prevention database and the Japan Meteorological Agency's API. Specific data retrieved includes the date and time of typhoon and earthquake occurrence, the scale of damage, and the area of ​​impact.

[0552] Next, the server retrieves the facility's floor plan, evacuation routes, and stockpiled goods information from an internal database. This database uses an SQL database, and the server executes queries to retrieve the necessary information. This provides the facility's layout, the layout of each room, main evacuation routes, and a list of stockpiled goods.

[0553] The acquired data is preprocessed by the server. During the preprocessing stage, data formats are standardized, missing data is filled in, and outliers are corrected. The preprocessed data is then input into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled items needed during evacuation.

[0554] The generated scenario is passed to a video generation AI, which uses 3D modeling software (such as Blender or Unreal Engine) to create an evacuation simulation video based on the actual facility environment, allowing users to experience a realistic evacuation drill.

[0555] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content is saved in cloud storage (e.g., AWS S3) and a URL is generated that users can access.

[0556] The user's device (smartphone, tablet, AR headset, etc.) downloads the augmented reality content from this cloud storage and plays it when the user specifies. When the user launches the application and starts a disaster prevention drill, the device uses an emotion engine to recognize the user's emotions in real time. The emotion engine analyzes the user's emotional state from their facial expressions and voice and sends the data to the server.

[0557] The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance and encouraging messages. It also monitors the user's emotional state and adjusts the training to prevent the user from feeling overly stressed.

[0558] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[0559] Examples of prompts include:

[0560] "Generate evacuation scenarios in the event of flooding due to a typhoon, and create evacuation simulation videos based on floor plans and evacuation routes for designated facilities. Also, include reassuring messages if users are feeling anxious."

[0561] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0562] Program processing flow

[0563] Step 1: Obtaining past disaster information

[0564] The server obtains past disaster information through an external database or API. As input data, it receives information such as the date and time of typhoon or earthquake occurrence, the scale of damage, and the affected area from the government's disaster prevention database or the API of the Japan Meteorological Agency. As output data, it saves the obtained disaster information in JSON or CSV format.

[0565] Step 2: Acquire internal data

[0566] The server retrieves facility floor plans, evacuation routes, and stockpiled goods information from an internal database. It queries an internal SQL database to retrieve the layout of each room in the facility, the main evacuation routes, and a list of stockpiled goods. It saves this information in a standardized format as output data.

[0567] Step 3: Preprocessing the data

[0568] The server preprocesses the acquired disaster information and in-house data. The data acquired in Step 1 and Step 2 are used as input data. Preprocessing involves filling in missing data, detecting and correcting outliers, and standardizing data formats. A consistent preprocessed dataset is generated as output data.

[0569] Step 4: Generate disaster scenarios

[0570] The server inputs the preprocessed data into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. The preprocessed dataset and prompt sentences are used as input data. The data calculation generates a scenario based on the language model. The output data is a scenario that includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0571] Step 5: Generate footage

[0572] The server passes the generated scenario to a video generation AI, which creates an evacuation simulation video based on the actual facility environment. The generated scenario and a 3D model of the facility are used as input data. 3D modeling software (Blender or Unreal Engine) is used for data calculations to generate a video that realistically reproduces evacuation behavior. The evacuation simulation video is output as output data.

[0573] Step 6: Generate and serve augmented reality (AR) content

[0574] The server integrates the generated scenario and video and outputs it as augmented reality content. The scenario and video are used as input data. Data processing involves converting it into AR format. AR content is generated as output data and uploaded to cloud storage.

[0575] Step 7: Save to cloud storage and generate access URL

[0576] The server saves the generated AR content in cloud storage and generates a URL that the user can access. The AR content is used as input data. As processing, the file is uploaded to a cloud storage service (e.g., AWS S3) and an access URL is generated. As output data, an access URL that the user can use is generated.

[0577] Step 8: Download and play augmented reality content

[0578] The device downloads the augmented reality content from cloud storage and plays it when the user specifies. The generated access URL is used as input data. The downloaded content is then played on the application as processing. The augmented reality content that the user can experience is displayed as output data.

[0579] Step 9: Recognizing User Emotions

[0580] The device uses an emotion engine to recognize the user's emotions in real time. The input data is the user's facial expressions and voice data. The processing is performed using emotion recognition technology for analysis. The output data is the user's emotional state, which is sent to the server.

[0581] Step 10: Dynamically adjust the training scenario

[0582] The server dynamically adjusts the training scenario based on the received user emotional data. The emotional data received in real time is used as input data. The data calculation changes the difficulty and content of the scenario according to the emotional state. The adjusted training scenario is provided to the user as output data.

[0583] (Application example 2)

[0584] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0585] Conducting realistic and effective evacuation drills in the event of a disaster is an important challenge in industrial sites such as factories and manufacturing facilities. However, conventional evacuation drills are expensive to implement, have fixed scenarios, and often do not adequately adapt to real-world disaster situations. Furthermore, training is often conducted without taking into account the emotions and stress levels of employees, resulting in ineffective training. Furthermore, it is difficult to generate appropriate evacuation scenarios by combining disaster information and facility data, and there are limited means of providing realistic simulation images.

[0586] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0587] In this invention, the server includes: means for acquiring past disaster information from an external database or API; means for acquiring workplace floor plans, evacuation routes, and emergency supply information from an internal database; means for generating a virtual disaster scenario using a large-scale language model; means for creating an evacuation simulation video based on an actual work environment using image generation AI; means for integrating the generated simulation video and the scenario and outputting it as augmented reality content; means for downloading the augmented reality content to a user's device so that it can be viewed and experienced; means for recognizing user emotions in real time and collecting that data; and means for dynamically adjusting the training scenario based on the user's emotion data. This makes it possible to conduct realistic and effective evacuation drills and provide appropriate training content that takes into account the emotions and stress of employees.

[0588] "Disaster information" refers to data on natural and man-made disasters that have occurred in the past, including the date and time of the disaster, the scale of the damage, and the area of ​​impact.

[0589] An "internal database" is a collection of information stored within a particular facility or organization, including floor plans, evacuation routes, and emergency supply information.

[0590] A "large-scale language model" is a type of artificial intelligence model that has a very large number of parameters and performs natural language processing by learning from huge amounts of text data.

[0591] A "virtual disaster scenario" is a fictitious story or situation in which a disaster situation or evacuation route is generated by computer simulation.

[0592] "Image generation artificial intelligence" is an artificial intelligence system that can generate realistic images and simulations using 3D models and CG technology.

[0593] A "work environment" refers to the location where a specific task is performed, such as a factory or manufacturing facility, and the surrounding environment.

[0594] "Augmented reality content" means applications or content that overlay digital information onto a real-world environment, creating a visual and auditory experience.

[0595] A "terminal" is a device used by a user to access the system, and includes a smartphone, tablet, etc.

[0596] "Emotion engine" refers to technology that analyzes the user's facial expressions and voice to recognize their current emotional state.

[0597] A "training scenario" is a plan or scenario that shows the procedures and specific training content for disaster prevention and evacuation drills.

[0598] "Dynamic adjustment" means changing and adapting in real time to the situation.

[0599] "Collected Data" is a general term for information obtained from external and internal databases.

[0600] "Unifying formats" means organizing and converting data of different formats according to consistent standards.

[0601] "Cloud storage" is a storage service for saving and managing data over the Internet.

[0602] A "URL" is a uniform resource identifier used to identify resources on the Internet.

[0603] The present invention is a system for implementing effective evacuation drills at industrial sites such as factories, manufacturing facilities, etc. This system operates by having a server, terminals, and users each fulfill their respective roles.

[0604] The server first obtains past disaster information from an external database or API. The obtained disaster information includes the date and time of the disaster, the scale of damage, and the area of ​​impact. The server then obtains information on the workplace floor plan, evacuation routes, and stockpiles from an internal database. This data is then preprocessed and converted into a unified format.

[0605] The server then uses a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario, including specific disaster situations, evacuation routes, and how to use necessary supplies during evacuation. The generated scenario is then passed to a video generation AI, which creates an evacuation simulation video based on a real-world working environment.

[0606] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content is downloaded to the user's device (e.g., smartphone, head-mounted display) and can be viewed and experienced. The augmented reality content stored in cloud storage generates a URL that the user can access, making it easy to download when needed.

[0607] The device downloads augmented reality content from cloud storage at the user's specified time. When the user launches the application and begins a disaster prevention training, the device uses an emotion engine to recognize emotions in real time from the user's facial expressions and voice, and sends them to the server. The server dynamically adjusts the training scenario based on the received emotion data. For example, if the user feels anxious, the system will adjust the scenario by providing more detailed guidance and explanations.

[0608] In this way, the system of the present invention enables users to receive realistic and effective evacuation training, helping them to evacuate more quickly and safely in the event of a disaster. The addition of an emotion engine can further improve the quality of the training and the user experience.

[0609] ■Example:

[0610] For example, if a fire breaks out in a factory, the system uses past fire information to generate a virtual evacuation scenario. This scenario includes the location of fire extinguishers stored in each room, along with evacuation routes within the factory. The system monitors the user's emotions in real time, and if the user feels anxious, it provides detailed guidance on how to use a fire extinguisher and the best evacuation route.

[0611] ■Example of a prompt:

[0612] "Generate a disaster scenario. Use the following data:

[0613] Disaster information: Fire, date of occurrence, area of ​​impact

[0614] Internal factory data: floor plans, evacuation routes, and stockpile information

[0615] Required outcome: A detailed description of the evacuation scenario and the supplies that should be used.”

[0616] This enables realistic and effective evacuation drills, ensuring the safety of employees and quick evacuation.

[0617] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0618] Step 1:

[0619] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to the disaster information API, and the retrieved disaster information includes the date and time of the occurrence, the scale of damage, the area of ​​impact, etc. The API request is made as input, and disaster information data is obtained as output.

[0620] Step 2:

[0621] The server retrieves the workplace floor plan, evacuation routes, and stockpile information from an internal database. The internal database stores the factory floor plan, evacuation routes, and stockpile location information. The database query is performed as input, and the internal data is obtained as output.

[0622] Step 3:

[0623] The server preprocesses the acquired disaster information and internal data and standardizes their formats. Specifically, it converts each dataset into the same format and filters out unnecessary data. It receives disaster information data and internal data as input and outputs data in a unified format.

[0624] Step 4:

[0625] The server inputs the unified data into a large-scale language model (e.g., GPT-4) to generate a hypothetical disaster scenario. The unified data is passed to the model as input, and the disaster scenario is obtained as output. The model generates the scenario using a prompt sentence.

[0626] Step 5:

[0627] The server then passes the generated disaster scenario to a video generation AI system, which then creates an evacuation simulation video based on the actual work environment.The disaster scenario is used as input, and the simulation video, using 3D models and CG technology, is obtained as output.

[0628] Step 6:

[0629] The server integrates the simulation video and the generated scenario and outputs it as augmented reality content.The simulation video and scenario are integrated as input, and the completed augmented reality content is obtained as output.

[0630] Step 7:

[0631] Save augmented reality content in cloud storage and generate a URL that users can access. Upload augmented reality content to the cloud as input, and generate an access URL as output.

[0632] Step 8:

[0633] The user's device launches an application that downloads the augmented reality content from cloud storage and allows it to be viewed and experienced, using a URL as input and the downloaded content as output.

[0634] Step 9:

[0635] The device uses an emotion engine to recognize emotions from the user's facial expressions, voice, etc. in real time while the user is experiencing augmented reality content, and transmits the emotions to the server. The device receives emotion data collected in real time as input, and obtains emotion data to be sent to the server as output.

[0636] Step 10:

[0637] The server dynamically adjusts the training scenario based on the received emotional data. It receives the emotional data as input and outputs the adjusted training scenario in real time. For example, if the user feels anxious, it can provide more detailed guidance or encouraging messages.

[0638] Through these steps, users can receive realistic and effective evacuation training. The data processing and calculations performed at each step support more appropriate and safe evacuation behavior.

[0639] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0640] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0641] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0642] [Third embodiment]

[0643] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0644] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0645] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0646] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0647] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0648] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0649] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0650] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0651] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0652] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0653] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0654] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0655] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0656] First, the server retrieves past disaster information from an external database or API. This collects detailed data on past typhoons, earthquakes, and other events. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies from the company's internal database. This data provides the basic information needed to generate disaster scenarios.

[0657] The server preprocesses the acquired data and standardizes the format. Any inaccuracies in the data are corrected. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0658] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG to create a realistic evacuation simulation, allowing users to experience a realistic evacuation drill.

[0659] The server then integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0660] Users launch the application on their devices to begin disaster prevention training. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0661] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0662] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, and supports users in taking quicker and safer evacuation actions in the event of a disaster.

[0663] The processing flow will be explained below.

[0664] Step 1:

[0665] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0666] Step 2:

[0667] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0668] Step 3:

[0669] The server preprocesses the acquired data by standardizing the data format and filling in any missing information. This preprocessing includes data cleansing and normalization.

[0670] Step 4:

[0671] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0672] Step 5:

[0673] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0674] Step 6:

[0675] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0676] Step 7:

[0677] The server saves the augmented reality content in cloud storage and generates a URL that users can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and creates an access link to the file.

[0678] Step 8:

[0679] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0680] Step 9:

[0681] Users can play the downloaded augmented reality content in the application to experience a disaster prevention drill. The user's device will then visually display evacuation route guides and instructions on how to use stockpiled supplies, helping them to carry out the drill in a realistic situation.

[0682] In this way, the entire system works together to realize dynamic and realistic disaster prevention drills in real time.

[0683] Example 1

[0684] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0685] In recent years, the increasing importance of disaster prevention measures has created a demand for realistic and effective disaster prevention training. However, conventional disaster prevention training is difficult to implement and often lacks realism, making it difficult for participants to take appropriate action in the event of a real disaster. Furthermore, technology for efficiently handling large amounts of data and generating virtual scenarios has been limited. This has resulted in training that is not tailored to the actual situation on the ground and is therefore difficult to demonstrate effectiveness in the event of a real disaster. To address these issues, a system is needed that combines past disaster information with internal company data to generate realistic virtual disaster scenarios and utilizes augmented reality to provide effective disaster prevention training.

[0686] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0687] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supplies information from an internal database, means for preprocessing the collected data and standardizing the format, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on real facility environments using video generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, and means for downloading the augmented reality content to a user's device so that the user can view and experience it. This allows users to experience realistic disaster prevention training and take prompt and appropriate action in the event of an actual disaster.

[0688] "Past disaster information" refers to detailed data on natural disasters that have occurred in the past (e.g., typhoons, earthquakes, floods, etc.).

[0689] An "external database" refers to a database that exists outside an organization and is accessible via the Internet or other means.

[0690] "API" stands for Application Program Interface, and refers to a standardized interface for exchanging data and functions between different software programs.

[0691] An "internal database" refers to a database that exists within an organization and is accessible to specific users or systems.

[0692] A "facility floor plan" refers to a drawing that shows the internal structure of a building and the layout of rooms.

[0693] An "evacuation route" refers to a route that has been set up to allow people to evacuate safely in the event of a disaster.

[0694] "Stockpile information" refers to information about supplies and materials that have been prepared in advance for use in the event of a disaster.

[0695] "Preprocessing" refers to the process of cleansing and formatting raw data for data analysis and machine learning.

[0696] "Unifying formats" refers to converting data of different formats into a consistent, common format.

[0697] A "large-scale language model" refers to a natural language processing model trained on a large amount of text data.

[0698] A "virtual disaster scenario" refers to a hypothetical scenario that is created by combining information on past disasters with internal company data.

[0699] "Image generation artificial intelligence" refers to a system that uses artificial intelligence technology to generate or process images.

[0700] "Evacuation simulation video" refers to a simulation video that shows evacuation behavior, created based on a hypothetical disaster scenario.

[0701] "Augmented reality content" refers to digital content that displays virtual information overlaid on a real environment.

[0702] "User device" refers to an electronic device operated by a user (e.g., a smartphone, tablet, AR headset, etc.).

[0703] "Cloud storage" refers to an online storage service that allows you to store and manage data via the Internet.

[0704] "URL" is an abbreviation for Uniform Resource Locator, and refers to an address that points to a specific resource on the Internet.

[0705] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0706] First, the server retrieves past disaster information from an external database or API. This includes detailed data on past typhoons and earthquakes. Next, the server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database. The data collected in this way provides the basic information needed to generate disaster scenarios.

[0707] The server preprocesses the acquired data and standardizes its format. During this process, missing data is filled in. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled supplies.

[0708] The generated scenario is passed to a video generation AI system, which creates an evacuation simulation video based on the actual facility environment. 3D models and CG technology are used to generate the video, enabling a realistic simulation. This allows users to experience a realistic evacuation drill.

[0709] Next, the server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0710] Users launch a dedicated application on their device to begin disaster prevention training. The device downloads augmented reality content based on a URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0711] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0712] An example prompt is:

[0713] Generate disaster scenarios by combining past disaster information with your company's data. Provide the following information:

[0714] Detailed data on past typhoons and earthquakes

[0715] Facility floor plan, evacuation route, and emergency supplies information

[0716] Create a hypothetical disaster scenario using a specific artificial intelligence technology (e.g., GPT-4).

[0717] Using these prompts, the system provides the necessary data to generate a virtual disaster scenario that provides a realistic disaster drill.

[0718] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0719] Step 1:

[0720] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to an external API (e.g., the Japan Meteorological Agency API) and receives detailed data on past typhoons, earthquakes, etc. in JSON format.

[0721] Input: External API endpoint URL

[0722] Output: Past disaster data in JSON format

[0723] Step 2:

[0724] The server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database, using SQL queries to extract the necessary data.

[0725] Input: Internal database access information, SQL query

[0726] Output: A dataset containing facility floor plans, evacuation routes, and emergency supplies.

[0727] Step 3:

[0728] The server preprocesses the acquired data and standardizes the format, for example, by using the Pandas library to impute missing values ​​with the median and convert the data into a unified format.

[0729] Input: JSON formatted historical disaster data and datasets obtained from an internal database

[0730] Output: A consistent dataset after preprocessing and formatting

[0731] Step 4:

[0732] The server passes the preprocessed data to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The server creates a prompt sentence and sends a request to GPT-4 based on it.

[0733] Input: Dataset after preprocessing and unifying format, prompt statement

[0734] Output: Text data containing hypothetical disaster scenarios

[0735] Step 5:

[0736] The server then passes the generated virtual disaster scenario to a video generation AI, which then creates an evacuation simulation video. In this case, a video generation tool such as Blender is used.

[0737] Input: Hypothetical disaster scenario

[0738] Output: Evacuation simulation video using 3D models and CG (MP4 format)

[0739] Step 6:

[0740] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. The content is created using an AR development kit such as Unity.

[0741] Input: Virtual disaster scenario, evacuation simulation video

[0742] Output: Augmented reality content (APK or IPA format)

[0743] Step 7:

[0744] The server saves the generated augmented reality content in cloud storage and generates a URL that users can access. Specifically, it uploads the content to a cloud storage such as AWS S3.

[0745] Input: Augmented reality content

[0746] Output: Cloud storage URL

[0747] Step 8:

[0748] The user starts the disaster prevention training by launching a dedicated application on the device. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user.

[0749] Input: Cloud storage URL

[0750] Output: Downloading augmented reality content and conducting disaster prevention drills

[0751] Based on this detailed procedure, the present invention can provide users with a realistic and effective disaster prevention training experience.

[0752] (Application example 1)

[0753] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0754] It is difficult to conduct effective and realistic disaster prevention drills in logistics centers and other facilities where many workers gather. While quick and safe evacuation actions are required in the event of a disaster, conventional training methods lack a sense of realism, limiting the effectiveness of the training. Furthermore, there was a lack of a way to manage the training progress of individual workers and visualize it as reports, leaving challenges in raising overall disaster prevention awareness and responding quickly.

[0755] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0756] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on the actual facility environment using image generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, means for downloading the augmented reality content to a user's device so that the content can be viewed and experienced, and means for tracking the user's training progress and generating reports. This allows users to experience a realistic evacuation simulation within the facility, and enables them to check their individual training progress and identify areas for improvement.

[0757] "Past disaster information" refers to detailed data from when a disaster occurred, including the location, time, damage extent, and status of rescue operations of natural disasters such as earthquakes and typhoons.

[0758] "External database or API" refers to an external data source accessible by the server, including an interface through which past disaster information can be obtained.

[0759] A "facility floor plan" refers to a drawing showing the layout of each area and room within a facility, and is used to design evacuation routes and arrange supplies.

[0760] "Evacuation routes" refer to recommended corridors and routes for safe evacuation in the event of a disaster, and are designed to allow for quick evacuation from any location within the facility.

[0761] "Stockpile information" refers to detailed data such as the types, quantities, and storage locations of supplies needed in the event of a disaster.

[0762] A "large-scale language model" is an artificial intelligence trained on massive amounts of text data, and is capable of advanced text generation and analysis through natural language processing.

[0763] A "virtual disaster scenario" refers to a sequence of hypothetical events generated to simulate a real disaster, including evacuation routes and how to use stockpiles of supplies.

[0764] "Image generation artificial intelligence" refers to software and systems that use artificial intelligence technology to generate realistic images, and perform evacuation simulations using 3D models and CG technology.

[0765] "Augmented reality content" refers to content that displays virtual information overlaid on a real environment, allowing users to experience a realistic simulation.

[0766] "User device" refers to the device used to view and experience augmented reality content, including smartphones, tablets, and AR headsets.

[0767] "Training progress tracking" refers to recording the extent to which a user has completed disaster prevention training, as well as their progress and achievement.

[0768] "Generating a report" refers to analyzing training progress and results and summarizing the results in documents, graphs, etc.

[0769] In this invention, several important hardware and software components are combined and operated to build a disaster prevention training system specialized for facilities such as logistics centers. The system is realized by the servers, terminals, and users each playing their respective roles.

[0770] First, the server uses the following means.

[0771] 1. How to obtain past disaster information: The server obtains past disaster information through an external database or API. For example, data such as the location, time, and damage status of earthquakes and typhoons is collected from the API. An example of an API used for this is "https: / / api.disasterinfo.com / get."

[0772] 2. Internal data acquisition method: The server acquires facility floor plans, evacuation routes, and emergency supplies from the internal database. This is done using the facility's internal database and storage system. For example, the server acquires data from the endpoint "https: / / warehouse.internaldb.com / data".

[0773] 3. Data preprocessing: Collected data is preprocessed and the format is standardized. Scripting languages ​​such as Python are used to fill in any deficiencies in the data and ensure consistency.

[0774] 4. Disaster scenario generation means: Communicate with a large-scale language model (e.g., GPT-4) and generate a virtual disaster scenario based on the collected data. An example of an API for the generation AI model is "https: / / api.gpt-4.com / generate_scenario."

[0775] 5. Video generation method: Using video generation AI, evacuation simulation videos based on the actual facility environment are created. In this case, 3D models and CG technology are used, and tools such as "Blender" and "Unity" are often used.

[0776] 6. Means of outputting as augmented reality content: The generated simulation video and scenario are integrated and output as augmented reality content, allowing users to enjoy realistic evacuation drill content using AR technology.

[0777] 7. Training Progress Tracking and Report Generation: The server tracks users' training progress and generates reports. A dedicated management dashboard is used to visualize training achievements and areas for improvement.

[0778] Next, the user's terminal uses the following means:

[0779] 1. Means for downloading and experiencing augmented reality content: The user's device (e.g., smartphone or AR headset) downloads the augmented reality content sent from the server, allowing them to view and experience it. When the user opens the application and presses the "Generate Scenario" button, a request is sent to the server. The generated content is saved in cloud storage, and a URL that can be accessed from the user's device is provided.

[0780] Specific examples

[0781] For example, when conducting a disaster prevention drill at a logistics center, a user can input the following prompt sentence using a smartphone:

[0782] "Please generate an earthquake scenario for a logistics center. The floor plan, evacuation routes, and stockpiled supplies are as follows... [continues data]"

[0783] Through the generated scenarios and augmented reality content, users can experience realistic evacuation simulations and learn appropriate evacuation behavior. Training progress is tracked and detailed reports are generated, allowing the effectiveness of each user's training to be evaluated.

[0784] This allows logistics center workers to learn how to evacuate quickly and safely through realistic disaster prevention training, improving overall disaster prevention awareness.

[0785] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0786] Step 1:

[0787] The server retrieves past disaster information from an external database or API. It uses the API endpoint (e.g., "https: / / api.disasterinfo.com / get") as input and obtains detailed disaster information data (e.g., location, time, and damage extent of earthquakes and typhoons) as output. Specifically, it sends an HTTP GET request and receives and parses the response in JSON format.

[0788] Step 2:

[0789] The server retrieves facility floor plans, evacuation routes, and stockpile information from an internal database. It uses the internal database URL (e.g., "https: / / warehouse.internaldb.com / data") as input and gets the floor plans, evacuation routes, and stockpile information as output. Specifically, it queries the database to retrieve the required data and formats it in the appropriate way.

[0790] Step 3:

[0791] The server preprocesses the data collected and standardizes it into a consistent format. It uses past disaster information and internal company data as input and obtains preprocessed data as output. Specifically, it checks the data type and format, complements any deficiencies, and converts them into a unified format.

[0792] Step 4:

[0793] The server generates a virtual disaster scenario using a large-scale language model (e.g., GPT-4). It uses preprocessed data as input and obtains a disaster scenario as output. Specifically, it sends a POST request to the model API and receives the generated scenario sentence as a response.

[0794] Step 5:

[0795] The server uses video generation artificial intelligence to create an evacuation simulation video based on the actual facility environment. It uses the generated disaster scenario as input and obtains the evacuation simulation video as output. Specifically, it uses 3D modeling software for video generation (e.g., Blender or Unity) to create a video based on the scenario.

[0796] Step 6:

[0797] The server integrates the generated simulation video and scenario and outputs it as augmented reality content. It uses the evacuation simulation video and disaster scenario as input and obtains augmented reality content as output. Specifically, it uses an AR toolkit (e.g., ARCore or ARKit) to prepare the content for display as augmented reality.

[0798] Step 7:

[0799] The server saves the augmented reality content in cloud storage and generates a URL that the user can access. The augmented reality content is used as input, and a URL on the cloud storage is obtained as output. Specifically, the data is uploaded to a cloud storage service (e.g., Amazon S3 or Google Cloud Storage) and the URL is obtained.

[0800] Step 8:

[0801] The user downloads, views, and experiences augmented reality content using a device. The input is a cloud storage URL, and the output is the user's ability to view the content. Specifically, the app launches, downloads the content from the provided URL, and switches to AR experience mode.

[0802] Step 9:

[0803] The server tracks training progress and generates reports. It uses the user's training data as input and obtains a training progress report as output. Specifically, it records the user's activity log in a database, periodically aggregates and analyzes it, and generates reports.

[0804] By going through the above processing steps, a realistic evacuation training system can be realized for a facility such as a logistics center.

[0805] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0806] This invention combines information on past disasters with in-house data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR), with an emotion engine that recognizes the user's emotions. This system is realized by the server, terminals, and users each playing their respective roles.

[0807] First, the server retrieves past disaster information from an external database or API. The disaster information retrieved here is specific data on typhoons, earthquakes, etc., including the date and time of occurrence, the scale of damage, and the area of ​​impact. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies information from the internal database. This provides information such as the office structure, the layout of each room, main evacuation routes, and a list of emergency supplies.

[0808] The server preprocesses the acquired data and standardizes its format. The preprocessed data is then fed into a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use emergency supplies during evacuation.

[0809] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior, allowing users to experience a realistic evacuation drill.

[0810] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0811] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0812] This is where the emotion engine, a key feature of the present invention, comes into play. The device recognizes emotions from the user's facial expressions, voice, and other data in real time and transmits them to the server. The server then dynamically adjusts the training scenario based on the received emotion data. For example, if the user is feeling anxious, the system can provide more detailed guidance and encouraging messages. The system also monitors the user's emotional state and adjusts the training to prevent overly stressful experiences.

[0813] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[0814] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, helping them to evacuate more quickly and safely in the event of a disaster. Furthermore, the addition of an emotion engine can further improve the quality of the training and the user experience.

[0815] The processing flow will be explained below.

[0816] Step 1:

[0817] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0818] Step 2:

[0819] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0820] Step 3:

[0821] The server preprocesses the acquired data and standardizes its format. Specifically, it standardizes the data format and fills in any missing information. This preprocessing includes data cleansing and normalization.

[0822] Step 4:

[0823] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0824] Step 5:

[0825] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0826] Step 6:

[0827] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0828] Step 7:

[0829] The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and generates an access link to the file.

[0830] Step 8:

[0831] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0832] Step 9:

[0833] The device uses an emotion engine to recognize emotions in real time from the user's facial expressions, voice, etc. The emotion engine analyzes data acquired through the camera and microphone to identify the user's emotional state (e.g., anxiety, relief, concentration).

[0834] Step 10:

[0835] When a user starts a disaster prevention training, the device sends emotional data to the server through the emotion engine. The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance or encouraging messages.

[0836] Step 11:

[0837] Users can play the downloaded augmented reality content in the app and experience disaster prevention drills. The device uses an emotion engine to monitor the user's emotional state in real time and sends feedback to the server, which then adjusts the content to prevent the drill from becoming too stressful.

[0838] In this way, the system combined with the emotion engine enables users to receive realistic and effective disaster prevention training, supporting them in taking quick and safe evacuation actions in the event of a disaster. The addition of the emotion engine further improves the quality of the training and the user experience.

[0839] Example 2

[0840] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0841] Conventional disaster prevention training systems are limited to simulated training that lacks realism, and users are not fully prepared to take appropriate actions in a real disaster situation. Furthermore, training does not take into account the user's emotional state, making it difficult to improve the effectiveness of the training. To solve these problems, a realistic training environment and dynamic scenario adjustments based on the user's emotional state are needed.

[0842] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0843] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating a virtual disaster scenario using a large-scale language model, means for creating an evacuation simulation video based on an actual facility environment using image recognition technology, means for integrating the generated simulation video and the scenario and outputting it as augmented reality content, means for downloading the augmented reality content to a user's device so that the user can view and experience it, and means for recognizing the user's emotions in real time using an emotion recognition engine and dynamically adjusting the training scenario based on that data. This allows users to experience a realistic disaster prevention training, and the dynamic adjustment of the training scenario according to their emotional state allows for more effective training.

[0844] "Past disaster information" refers to information such as the date and time of occurrence, scale of damage, and area of ​​impact regarding disasters such as typhoons and earthquakes, which is obtained through external databases or APIs.

[0845] An "internal database" refers to a database that stores data about the facility, such as office floor plans, evacuation routes, and information on stockpiled supplies.

[0846] A "large-scale language model" is an artificial intelligence model based on massive amounts of data that can generate and analyze text. For example, GPT-4 is one such model.

[0847] "Image recognition technology" refers to the technology that analyzes image data from cameras and sensors and creates evacuation simulation images based on a real office environment.

[0848] "Augmented reality content" refers to a form of content that integrates generated simulation footage with disaster scenarios, allowing users to experience virtual information while viewing the real world.

[0849] "Cloud storage" refers to an online storage service that stores data over the Internet and generates a URL that users can access.

[0850] An "emotion recognition engine" is an engine that recognizes emotions from a user's facial expressions and voice in real time and adjusts the system's operation based on that data.

[0851] "User's device" refers to devices used by users, such as smartphones, tablets, and AR headsets.

[0852] "Dynamic adjustment" refers to changing the scenario or system behavior appropriately in response to the user's emotional state and other real-time data.

[0853] This invention is a system that combines past disaster information with internal data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0854] First, the server retrieves past disaster information from an external database or API. The hardware used here is a high-performance server, and the software is a program that makes API calls. The external databases used include the government's disaster prevention database and the Japan Meteorological Agency's API. Specific data retrieved includes the date and time of typhoon and earthquake occurrence, the scale of damage, and the area of ​​impact.

[0855] Next, the server retrieves the facility's floor plan, evacuation routes, and stockpiled goods information from an internal database. This database uses an SQL database, and the server executes queries to retrieve the necessary information. This provides the facility's layout, the layout of each room, main evacuation routes, and a list of stockpiled goods.

[0856] The acquired data is preprocessed by the server. During the preprocessing stage, data formats are standardized, missing data is filled in, and outliers are corrected. The preprocessed data is then input into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled items needed during evacuation.

[0857] The generated scenario is passed to a video generation AI, which uses 3D modeling software (such as Blender or Unreal Engine) to create an evacuation simulation video based on the actual facility environment, allowing users to experience a realistic evacuation drill.

[0858] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content is saved in cloud storage (e.g., AWS S3) and a URL is generated that users can access.

[0859] The user's device (smartphone, tablet, AR headset, etc.) downloads the augmented reality content from this cloud storage and plays it when the user specifies. When the user launches the application and starts a disaster prevention drill, the device uses an emotion engine to recognize the user's emotions in real time. The emotion engine analyzes the user's emotional state from their facial expressions and voice and sends the data to the server.

[0860] The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance and encouraging messages. It also monitors the user's emotional state and adjusts the training to prevent the user from feeling overly stressed.

[0861] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[0862] Examples of prompts include:

[0863] "Generate evacuation scenarios in the event of flooding due to a typhoon, and create evacuation simulation videos based on floor plans and evacuation routes for designated facilities. Also, include reassuring messages if users are feeling anxious."

[0864] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0865] Program processing flow

[0866] Step 1: Obtaining past disaster information

[0867] The server obtains past disaster information through an external database or API. As input data, it receives information such as the date and time of typhoon or earthquake occurrence, the scale of damage, and the affected area from the government's disaster prevention database or the API of the Japan Meteorological Agency. As output data, it saves the obtained disaster information in JSON or CSV format.

[0868] Step 2: Acquire internal data

[0869] The server retrieves facility floor plans, evacuation routes, and stockpiled goods information from an internal database. It queries an internal SQL database to retrieve the layout of each room in the facility, the main evacuation routes, and a list of stockpiled goods. It saves this information in a standardized format as output data.

[0870] Step 3: Preprocessing the data

[0871] The server preprocesses the acquired disaster information and in-house data. The data acquired in Step 1 and Step 2 are used as input data. Preprocessing involves filling in missing data, detecting and correcting outliers, and standardizing data formats. A consistent preprocessed dataset is generated as output data.

[0872] Step 4: Generate disaster scenarios

[0873] The server inputs the preprocessed data into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. The preprocessed dataset and prompt sentences are used as input data. The data calculation generates a scenario based on the language model. The output data is a scenario that includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0874] Step 5: Generate footage

[0875] The server passes the generated scenario to a video generation AI, which creates an evacuation simulation video based on the actual facility environment. The generated scenario and a 3D model of the facility are used as input data. 3D modeling software (Blender or Unreal Engine) is used for data calculations to generate a video that realistically reproduces evacuation behavior. The evacuation simulation video is output as output data.

[0876] Step 6: Generate and serve augmented reality (AR) content

[0877] The server integrates the generated scenario and video and outputs it as augmented reality content. The scenario and video are used as input data. Data processing involves converting it into AR format. AR content is generated as output data and uploaded to cloud storage.

[0878] Step 7: Save to cloud storage and generate access URL

[0879] The server saves the generated AR content in cloud storage and generates a URL that the user can access. The AR content is used as input data. As processing, the file is uploaded to a cloud storage service (e.g., AWS S3) and an access URL is generated. As output data, an access URL that the user can use is generated.

[0880] Step 8: Download and play augmented reality content

[0881] The device downloads the augmented reality content from cloud storage and plays it when the user specifies. The generated access URL is used as input data. The downloaded content is then played on the application as processing. The augmented reality content that the user can experience is displayed as output data.

[0882] Step 9: Recognizing User Emotions

[0883] The device uses an emotion engine to recognize the user's emotions in real time. The input data is the user's facial expressions and voice data. The processing is performed using emotion recognition technology for analysis. The output data is the user's emotional state, which is sent to the server.

[0884] Step 10: Dynamically adjust the training scenario

[0885] The server dynamically adjusts the training scenario based on the received user emotional data. The emotional data received in real time is used as input data. The data calculation changes the difficulty and content of the scenario according to the emotional state. The adjusted training scenario is provided to the user as output data.

[0886] (Application example 2)

[0887] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0888] Conducting realistic and effective evacuation drills in the event of a disaster is an important challenge in industrial sites such as factories and manufacturing facilities. However, conventional evacuation drills are expensive to implement, have fixed scenarios, and often do not adequately adapt to real-world disaster situations. Furthermore, training is often conducted without taking into account the emotions and stress levels of employees, resulting in ineffective training. Furthermore, it is difficult to generate appropriate evacuation scenarios by combining disaster information and facility data, and there are limited means of providing realistic simulation images.

[0889] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0890] In this invention, the server includes: means for acquiring past disaster information from an external database or API; means for acquiring workplace floor plans, evacuation routes, and emergency supply information from an internal database; means for generating a virtual disaster scenario using a large-scale language model; means for creating an evacuation simulation video based on an actual work environment using image generation AI; means for integrating the generated simulation video and the scenario and outputting it as augmented reality content; means for downloading the augmented reality content to a user's device so that it can be viewed and experienced; means for recognizing user emotions in real time and collecting that data; and means for dynamically adjusting the training scenario based on the user's emotion data. This makes it possible to conduct realistic and effective evacuation drills and provide appropriate training content that takes into account the emotions and stress of employees.

[0891] "Disaster information" refers to data on natural and man-made disasters that have occurred in the past, including the date and time of the disaster, the scale of the damage, and the area of ​​impact.

[0892] An "internal database" is a collection of information stored within a particular facility or organization, including floor plans, evacuation routes, and emergency supply information.

[0893] A "large-scale language model" is a type of artificial intelligence model that has a very large number of parameters and performs natural language processing by learning from huge amounts of text data.

[0894] A "virtual disaster scenario" is a fictitious story or situation in which a disaster situation or evacuation route is generated by computer simulation.

[0895] "Image generation artificial intelligence" is an artificial intelligence system that can generate realistic images and simulations using 3D models and CG technology.

[0896] A "work environment" refers to the location where a specific task is performed, such as a factory or manufacturing facility, and the surrounding environment.

[0897] "Augmented reality content" means applications or content that overlay digital information onto a real-world environment, creating a visual and auditory experience.

[0898] A "terminal" is a device used by a user to access the system, and includes a smartphone, tablet, etc.

[0899] "Emotion engine" refers to technology that analyzes the user's facial expressions and voice to recognize their current emotional state.

[0900] A "training scenario" is a plan or scenario that shows the procedures and specific training content for disaster prevention and evacuation drills.

[0901] "Dynamic adjustment" means changing and adapting in real time to the situation.

[0902] "Collected Data" is a general term for information obtained from external and internal databases.

[0903] "Unifying formats" means organizing and converting data of different formats according to consistent standards.

[0904] "Cloud storage" is a storage service for saving and managing data over the Internet.

[0905] A "URL" is a uniform resource identifier used to identify resources on the Internet.

[0906] The present invention is a system for implementing effective evacuation drills at industrial sites such as factories, manufacturing facilities, etc. This system operates by having a server, terminals, and users each fulfill their respective roles.

[0907] The server first obtains past disaster information from an external database or API. The obtained disaster information includes the date and time of the disaster, the scale of damage, and the area of ​​impact. The server then obtains information on the workplace floor plan, evacuation routes, and stockpiles from an internal database. This data is then preprocessed and converted into a unified format.

[0908] The server then uses a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario, including specific disaster situations, evacuation routes, and how to use necessary supplies during evacuation. The generated scenario is then passed to a video generation AI, which creates an evacuation simulation video based on a real-world working environment.

[0909] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content is downloaded to the user's device (e.g., smartphone, head-mounted display) and can be viewed and experienced. The augmented reality content stored in cloud storage generates a URL that the user can access, making it easy to download when needed.

[0910] The device downloads augmented reality content from cloud storage at the user's specified time. When the user launches the application and begins a disaster prevention training, the device uses an emotion engine to recognize emotions in real time from the user's facial expressions and voice, and sends them to the server. The server dynamically adjusts the training scenario based on the received emotion data. For example, if the user feels anxious, the system will adjust the scenario by providing more detailed guidance and explanations.

[0911] In this way, the system of the present invention enables users to receive realistic and effective evacuation training, helping them to evacuate more quickly and safely in the event of a disaster. The addition of an emotion engine can further improve the quality of the training and the user experience.

[0912] ■Example:

[0913] For example, if a fire breaks out in a factory, the system uses past fire information to generate a virtual evacuation scenario. This scenario includes the location of fire extinguishers stored in each room, along with evacuation routes within the factory. The system monitors the user's emotions in real time, and if the user feels anxious, it provides detailed guidance on how to use a fire extinguisher and the best evacuation route.

[0914] ■Example of a prompt:

[0915] "Generate a disaster scenario. Use the following data:

[0916] Disaster information: Fire, date of occurrence, area of ​​impact

[0917] Internal factory data: floor plans, evacuation routes, and stockpile information

[0918] Required outcome: A detailed description of the evacuation scenario and the supplies that should be used.”

[0919] This enables realistic and effective evacuation drills, ensuring the safety of employees and quick evacuation.

[0920] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0921] Step 1:

[0922] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to the disaster information API, and the retrieved disaster information includes the date and time of the occurrence, the scale of damage, the area of ​​impact, etc. The API request is made as input, and disaster information data is obtained as output.

[0923] Step 2:

[0924] The server retrieves the workplace floor plan, evacuation routes, and stockpile information from an internal database. The internal database stores the factory floor plan, evacuation routes, and stockpile location information. The database query is performed as input, and the internal data is obtained as output.

[0925] Step 3:

[0926] The server preprocesses the acquired disaster information and internal data and standardizes their formats. Specifically, it converts each dataset into the same format and filters out unnecessary data. It receives disaster information data and internal data as input and outputs data in a unified format.

[0927] Step 4:

[0928] The server inputs the unified data into a large-scale language model (e.g., GPT-4) to generate a hypothetical disaster scenario. The unified data is passed to the model as input, and the disaster scenario is obtained as output. The model generates the scenario using a prompt sentence.

[0929] Step 5:

[0930] The server then passes the generated disaster scenario to a video generation AI system, which then creates an evacuation simulation video based on the actual work environment.The disaster scenario is used as input, and the simulation video, using 3D models and CG technology, is obtained as output.

[0931] Step 6:

[0932] The server integrates the simulation video and the generated scenario and outputs it as augmented reality content.The simulation video and scenario are integrated as input, and the completed augmented reality content is obtained as output.

[0933] Step 7:

[0934] Save augmented reality content in cloud storage and generate a URL that users can access. Upload augmented reality content to the cloud as input, and generate an access URL as output.

[0935] Step 8:

[0936] The user's device launches an application that downloads the augmented reality content from cloud storage and allows it to be viewed and experienced, using a URL as input and the downloaded content as output.

[0937] Step 9:

[0938] The device uses an emotion engine to recognize emotions from the user's facial expressions, voice, etc. in real time while the user is experiencing augmented reality content, and transmits the emotions to the server. The device receives emotion data collected in real time as input, and obtains emotion data to be sent to the server as output.

[0939] Step 10:

[0940] The server dynamically adjusts the training scenario based on the received emotional data. It receives the emotional data as input and outputs the adjusted training scenario in real time. For example, if the user feels anxious, it can provide more detailed guidance or encouraging messages.

[0941] Through these steps, users can receive realistic and effective evacuation training. The data processing and calculations performed at each step support more appropriate and safe evacuation behavior.

[0942] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0943] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0944] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0945] [Fourth embodiment]

[0946] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0947] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0948] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0949] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0950] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0951] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0952] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0953] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0954] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0955] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0956] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0957] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0958] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0959] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[0960] First, the server retrieves past disaster information from an external database or API. This collects detailed data on past typhoons, earthquakes, and other events. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies from the company's internal database. This data provides the basic information needed to generate disaster scenarios.

[0961] The server preprocesses the acquired data and standardizes the format. Any inaccuracies in the data are corrected. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[0962] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG to create a realistic evacuation simulation, allowing users to experience a realistic evacuation drill.

[0963] The server then integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[0964] Users launch the application on their devices to begin disaster prevention training. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[0965] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[0966] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, and supports users in taking quicker and safer evacuation actions in the event of a disaster.

[0967] The processing flow will be explained below.

[0968] Step 1:

[0969] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[0970] Step 2:

[0971] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[0972] Step 3:

[0973] The server preprocesses the acquired data by standardizing the data format and filling in any missing information. This preprocessing includes data cleansing and normalization.

[0974] Step 4:

[0975] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[0976] Step 5:

[0977] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[0978] Step 6:

[0979] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[0980] Step 7:

[0981] The server saves the augmented reality content in cloud storage and generates a URL that users can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and creates an access link to the file.

[0982] Step 8:

[0983] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[0984] Step 9:

[0985] Users can play the downloaded augmented reality content in the application to experience a disaster prevention drill. The user's device will then visually display evacuation route guides and instructions on how to use stockpiled supplies, helping them to carry out the drill in a realistic situation.

[0986] In this way, the entire system works together to realize dynamic and realistic disaster prevention drills in real time.

[0987] Example 1

[0988] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0989] In recent years, the increasing importance of disaster prevention measures has created a demand for realistic and effective disaster prevention training. However, conventional disaster prevention training is difficult to implement and often lacks realism, making it difficult for participants to take appropriate action in the event of a real disaster. Furthermore, technology for efficiently handling large amounts of data and generating virtual scenarios has been limited. This has resulted in training that is not tailored to the actual situation on the ground and is therefore difficult to demonstrate effectiveness in the event of a real disaster. To address these issues, a system is needed that combines past disaster information with internal company data to generate realistic virtual disaster scenarios and utilizes augmented reality to provide effective disaster prevention training.

[0990] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0991] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supplies information from an internal database, means for preprocessing the collected data and standardizing the format, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on real facility environments using video generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, and means for downloading the augmented reality content to a user's device so that the user can view and experience it. This allows users to experience realistic disaster prevention training and take prompt and appropriate action in the event of an actual disaster.

[0992] "Past disaster information" refers to detailed data on natural disasters that have occurred in the past (e.g., typhoons, earthquakes, floods, etc.).

[0993] An "external database" refers to a database that exists outside an organization and is accessible via the Internet or other means.

[0994] "API" stands for Application Program Interface, and refers to a standardized interface for exchanging data and functions between different software programs.

[0995] An "internal database" refers to a database that exists within an organization and is accessible to specific users or systems.

[0996] A "facility floor plan" refers to a drawing that shows the internal structure of a building and the layout of rooms.

[0997] An "evacuation route" refers to a route that has been set up to allow people to evacuate safely in the event of a disaster.

[0998] "Stockpile information" refers to information about supplies and materials that have been prepared in advance for use in the event of a disaster.

[0999] "Preprocessing" refers to the process of cleansing and formatting raw data for data analysis and machine learning.

[1000] "Unifying formats" refers to converting data of different formats into a consistent, common format.

[1001] A "large-scale language model" refers to a natural language processing model trained on a large amount of text data.

[1002] A "virtual disaster scenario" refers to a hypothetical scenario that is created by combining information on past disasters with internal company data.

[1003] "Image generation artificial intelligence" refers to a system that uses artificial intelligence technology to generate or process images.

[1004] "Evacuation simulation video" refers to a simulation video that shows evacuation behavior, created based on a hypothetical disaster scenario.

[1005] "Augmented reality content" refers to digital content that displays virtual information overlaid on a real environment.

[1006] "User device" refers to an electronic device operated by a user (e.g., a smartphone, tablet, AR headset, etc.).

[1007] "Cloud storage" refers to an online storage service that allows you to store and manage data via the Internet.

[1008] "URL" is an abbreviation for Uniform Resource Locator, and refers to an address that points to a specific resource on the Internet.

[1009] This invention is a system that combines past disaster information with internal company data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[1010] First, the server retrieves past disaster information from an external database or API. This includes detailed data on past typhoons and earthquakes. Next, the server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database. The data collected in this way provides the basic information needed to generate disaster scenarios.

[1011] The server preprocesses the acquired data and standardizes its format. During this process, missing data is filled in. The preprocessed data is passed to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled supplies.

[1012] The generated scenario is passed to a video generation AI system, which creates an evacuation simulation video based on the actual facility environment. 3D models and CG technology are used to generate the video, enabling a realistic simulation. This allows users to experience a realistic evacuation drill.

[1013] Next, the server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[1014] Users launch a dedicated application on their device to begin disaster prevention training. The device downloads augmented reality content based on a URL sent from the server and provides it to the user. Through the downloaded content, users learn how to check evacuation routes and use stockpiled supplies, enabling them to act quickly and safely in the event of a real disaster.

[1015] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. The user can then experience the downloaded content and conduct a realistic evacuation drill.

[1016] An example prompt is:

[1017] Generate disaster scenarios by combining past disaster information with your company's data. Provide the following information:

[1018] Detailed data on past typhoons and earthquakes

[1019] Facility floor plan, evacuation route, and emergency supplies information

[1020] Create a hypothetical disaster scenario using a specific artificial intelligence technology (e.g., GPT-4).

[1021] Using these prompts, the system provides the necessary data to generate a virtual disaster scenario that provides a realistic disaster drill.

[1022] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1023] Step 1:

[1024] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to an external API (e.g., the Japan Meteorological Agency API) and receives detailed data on past typhoons, earthquakes, etc. in JSON format.

[1025] Input: External API endpoint URL

[1026] Output: Past disaster data in JSON format

[1027] Step 2:

[1028] The server retrieves facility floor plans, evacuation routes, and emergency supplies from an internal database, using SQL queries to extract the necessary data.

[1029] Input: Internal database access information, SQL query

[1030] Output: A dataset containing facility floor plans, evacuation routes, and emergency supplies.

[1031] Step 3:

[1032] The server preprocesses the acquired data and standardizes the format, for example, by using the Pandas library to impute missing values ​​with the median and convert the data into a unified format.

[1033] Input: JSON formatted historical disaster data and datasets obtained from an internal database

[1034] Output: A consistent dataset after preprocessing and formatting

[1035] Step 4:

[1036] The server passes the preprocessed data to a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The server creates a prompt sentence and sends a request to GPT-4 based on it.

[1037] Input: Dataset after preprocessing and unifying format, prompt statement

[1038] Output: Text data containing hypothetical disaster scenarios

[1039] Step 5:

[1040] The server then passes the generated virtual disaster scenario to a video generation AI, which then creates an evacuation simulation video. In this case, a video generation tool such as Blender is used.

[1041] Input: Hypothetical disaster scenario

[1042] Output: Evacuation simulation video using 3D models and CG (MP4 format)

[1043] Step 6:

[1044] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. The content is created using an AR development kit such as Unity.

[1045] Input: Virtual disaster scenario, evacuation simulation video

[1046] Output: Augmented reality content (APK or IPA format)

[1047] Step 7:

[1048] The server saves the generated augmented reality content in cloud storage and generates a URL that users can access. Specifically, it uploads the content to a cloud storage such as AWS S3.

[1049] Input: Augmented reality content

[1050] Output: Cloud storage URL

[1051] Step 8:

[1052] The user starts the disaster prevention training by launching a dedicated application on the device. The device downloads the augmented reality content based on the URL sent from the server and provides it to the user.

[1053] Input: Cloud storage URL

[1054] Output: Downloading augmented reality content and conducting disaster prevention drills

[1055] Based on this detailed procedure, the present invention can provide users with a realistic and effective disaster prevention training experience.

[1056] (Application example 1)

[1057] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1058] It is difficult to conduct effective and realistic disaster prevention drills in logistics centers and other facilities where many workers gather. While quick and safe evacuation actions are required in the event of a disaster, conventional training methods lack a sense of realism, limiting the effectiveness of the training. Furthermore, there was a lack of a way to manage the training progress of individual workers and visualize it as reports, leaving challenges in raising overall disaster prevention awareness and responding quickly.

[1059] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1060] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating virtual disaster scenarios using a large-scale language model, means for creating evacuation simulation videos based on the actual facility environment using image generation AI, means for integrating the generated simulation videos and scenarios and outputting them as augmented reality content, means for downloading the augmented reality content to a user's device so that the content can be viewed and experienced, and means for tracking the user's training progress and generating reports. This allows users to experience a realistic evacuation simulation within the facility, and enables them to check their individual training progress and identify areas for improvement.

[1061] "Past disaster information" refers to detailed data from when a disaster occurred, including the location, time, damage extent, and status of rescue operations of natural disasters such as earthquakes and typhoons.

[1062] "External database or API" refers to an external data source accessible by the server, including an interface through which past disaster information can be obtained.

[1063] A "facility floor plan" refers to a drawing showing the layout of each area and room within a facility, and is used to design evacuation routes and arrange supplies.

[1064] "Evacuation routes" refer to recommended corridors and routes for safe evacuation in the event of a disaster, and are designed to allow for quick evacuation from any location within the facility.

[1065] "Stockpile information" refers to detailed data such as the types, quantities, and storage locations of supplies needed in the event of a disaster.

[1066] A "large-scale language model" is an artificial intelligence trained on massive amounts of text data, and is capable of advanced text generation and analysis through natural language processing.

[1067] A "virtual disaster scenario" refers to a sequence of hypothetical events generated to simulate a real disaster, including evacuation routes and how to use stockpiles of supplies.

[1068] "Image generation artificial intelligence" refers to software and systems that use artificial intelligence technology to generate realistic images, and perform evacuation simulations using 3D models and CG technology.

[1069] "Augmented reality content" refers to content that displays virtual information overlaid on a real environment, allowing users to experience a realistic simulation.

[1070] "User device" refers to the device used to view and experience augmented reality content, including smartphones, tablets, and AR headsets.

[1071] "Training progress tracking" refers to recording the extent to which a user has completed disaster prevention training, as well as their progress and achievement.

[1072] "Generating a report" refers to analyzing training progress and results and summarizing the results in documents, graphs, etc.

[1073] In this invention, several important hardware and software components are combined and operated to build a disaster prevention training system specialized for facilities such as logistics centers. The system is realized by the servers, terminals, and users each playing their respective roles.

[1074] First, the server uses the following means.

[1075] 1. How to obtain past disaster information: The server obtains past disaster information through an external database or API. For example, data such as the location, time, and damage status of earthquakes and typhoons is collected from the API. An example of an API used for this is "https: / / api.disasterinfo.com / get."

[1076] 2. Internal data acquisition method: The server acquires facility floor plans, evacuation routes, and emergency supplies from the internal database. This is done using the facility's internal database and storage system. For example, the server acquires data from the endpoint "https: / / warehouse.internaldb.com / data".

[1077] 3. Data preprocessing: Collected data is preprocessed and the format is standardized. Scripting languages ​​such as Python are used to fill in any deficiencies in the data and ensure consistency.

[1078] 4. Disaster scenario generation means: Communicate with a large-scale language model (e.g., GPT-4) and generate a virtual disaster scenario based on the collected data. An example of an API for the generation AI model is "https: / / api.gpt-4.com / generate_scenario."

[1079] 5. Video generation method: Using video generation AI, evacuation simulation videos based on the actual facility environment are created. In this case, 3D models and CG technology are used, and tools such as "Blender" and "Unity" are often used.

[1080] 6. Means of outputting as augmented reality content: The generated simulation video and scenario are integrated and output as augmented reality content, allowing users to enjoy realistic evacuation drill content using AR technology.

[1081] 7. Training Progress Tracking and Report Generation: The server tracks users' training progress and generates reports. A dedicated management dashboard is used to visualize training achievements and areas for improvement.

[1082] Next, the user's terminal uses the following means:

[1083] 1. Means for downloading and experiencing augmented reality content: The user's device (e.g., smartphone or AR headset) downloads the augmented reality content sent from the server, allowing them to view and experience it. When the user opens the application and presses the "Generate Scenario" button, a request is sent to the server. The generated content is saved in cloud storage, and a URL that can be accessed from the user's device is provided.

[1084] Specific examples

[1085] For example, when conducting a disaster prevention drill at a logistics center, a user can input the following prompt sentence using a smartphone:

[1086] "Please generate an earthquake scenario for a logistics center. The floor plan, evacuation routes, and stockpiled supplies are as follows... [continues data]"

[1087] Through the generated scenarios and augmented reality content, users can experience realistic evacuation simulations and learn appropriate evacuation behavior. Training progress is tracked and detailed reports are generated, allowing the effectiveness of each user's training to be evaluated.

[1088] This allows logistics center workers to learn how to evacuate quickly and safely through realistic disaster prevention training, improving overall disaster prevention awareness.

[1089] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1090] Step 1:

[1091] The server retrieves past disaster information from an external database or API. It uses the API endpoint (e.g., "https: / / api.disasterinfo.com / get") as input and obtains detailed disaster information data (e.g., location, time, and damage extent of earthquakes and typhoons) as output. Specifically, it sends an HTTP GET request and receives and parses the response in JSON format.

[1092] Step 2:

[1093] The server retrieves facility floor plans, evacuation routes, and stockpile information from an internal database. It uses the internal database URL (e.g., "https: / / warehouse.internaldb.com / data") as input and gets the floor plans, evacuation routes, and stockpile information as output. Specifically, it queries the database to retrieve the required data and formats it in the appropriate way.

[1094] Step 3:

[1095] The server preprocesses the data collected and standardizes it into a consistent format. It uses past disaster information and internal company data as input and obtains preprocessed data as output. Specifically, it checks the data type and format, complements any deficiencies, and converts them into a unified format.

[1096] Step 4:

[1097] The server generates a virtual disaster scenario using a large-scale language model (e.g., GPT-4). It uses preprocessed data as input and obtains a disaster scenario as output. Specifically, it sends a POST request to the model API and receives the generated scenario sentence as a response.

[1098] Step 5:

[1099] The server uses video generation artificial intelligence to create an evacuation simulation video based on the actual facility environment. It uses the generated disaster scenario as input and obtains the evacuation simulation video as output. Specifically, it uses 3D modeling software for video generation (e.g., Blender or Unity) to create a video based on the scenario.

[1100] Step 6:

[1101] The server integrates the generated simulation video and scenario and outputs it as augmented reality content. It uses the evacuation simulation video and disaster scenario as input and obtains augmented reality content as output. Specifically, it uses an AR toolkit (e.g., ARCore or ARKit) to prepare the content for display as augmented reality.

[1102] Step 7:

[1103] The server saves the augmented reality content in cloud storage and generates a URL that the user can access. The augmented reality content is used as input, and a URL on the cloud storage is obtained as output. Specifically, the data is uploaded to a cloud storage service (e.g., Amazon S3 or Google Cloud Storage) and the URL is obtained.

[1104] Step 8:

[1105] The user downloads, views, and experiences augmented reality content using a device. The input is a cloud storage URL, and the output is the user's ability to view the content. Specifically, the app launches, downloads the content from the provided URL, and switches to AR experience mode.

[1106] Step 9:

[1107] The server tracks training progress and generates reports. It uses the user's training data as input and obtains a training progress report as output. Specifically, it records the user's activity log in a database, periodically aggregates and analyzes it, and generates reports.

[1108] By going through the above processing steps, a realistic evacuation training system can be realized for a facility such as a logistics center.

[1109] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1110] This invention combines information on past disasters with in-house data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR), with an emotion engine that recognizes the user's emotions. This system is realized by the server, terminals, and users each playing their respective roles.

[1111] First, the server retrieves past disaster information from an external database or API. The disaster information retrieved here is specific data on typhoons, earthquakes, etc., including the date and time of occurrence, the scale of damage, and the area of ​​impact. Next, the server retrieves office floor plans, evacuation routes, and emergency supplies information from the internal database. This provides information such as the office structure, the layout of each room, main evacuation routes, and a list of emergency supplies.

[1112] The server preprocesses the acquired data and standardizes its format. The preprocessed data is then fed into a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use emergency supplies during evacuation.

[1113] The generated scenario is passed to a video generation AI, which creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior, allowing users to experience a realistic evacuation drill.

[1114] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content can be viewed and experienced on the user's device (e.g., smartphone, AR headset). The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access.

[1115] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[1116] This is where the emotion engine, a key feature of the present invention, comes into play. The device recognizes emotions from the user's facial expressions, voice, and other data in real time and transmits them to the server. The server then dynamically adjusts the training scenario based on the received emotion data. For example, if the user is feeling anxious, the system can provide more detailed guidance and encouraging messages. The system also monitors the user's emotional state and adjusts the training to prevent overly stressful experiences.

[1117] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[1118] In this way, the system of the present invention enables users to receive realistic and effective disaster prevention training, helping them to evacuate more quickly and safely in the event of a disaster. Furthermore, the addition of an emotion engine can further improve the quality of the training and the user experience.

[1119] The processing flow will be explained below.

[1120] Step 1:

[1121] The server retrieves past disaster information from an external database or API. Specifically, it sends an HTTP request to the API endpoint and receives data on typhoons and earthquakes from the past 10 years in JSON format. This data includes details such as the date and time of occurrence, the scale of damage, and the affected area.

[1122] Step 2:

[1123] The server connects to an internal database and executes SQL queries to retrieve office floor plans, evacuation routes, and stockpiled supplies. Information retrieved from the database includes the office layout, the purpose of each room, primary evacuation routes, and a list of stockpiled supplies.

[1124] Step 3:

[1125] The server preprocesses the acquired data and standardizes its format. Specifically, it standardizes the data format and fills in any missing information. This preprocessing includes data cleansing and normalization.

[1126] Step 4:

[1127] The server uses the preprocessed data to input a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario. The input text includes collected disaster data, office layout, evacuation routes, and emergency supplies. The model then simulates a realistic disaster situation based on this information and generates a detailed scenario for the user.

[1128] Step 5:

[1129] The server then passes the generated scenario to a video generation AI that creates an evacuation simulation video based on the actual office environment. The video generation AI uses 3D models and CG technology to recreate realistic evacuation behavior.

[1130] Step 6:

[1131] The server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content, which can be viewed and experienced on the user's smartphone or AR headset.

[1132] Step 7:

[1133] The server saves the generated augmented reality content in cloud storage and generates a URL that the user can access. Specifically, the server uploads a file to a cloud storage service (e.g., Amazon S3) and generates an access link to the file.

[1134] Step 8:

[1135] The device downloads the augmented reality content from the cloud storage at the time specified by the user. When the user launches the application and starts the disaster prevention training, the device accesses the URL sent from the server and downloads the necessary files.

[1136] Step 9:

[1137] The device uses an emotion engine to recognize emotions in real time from the user's facial expressions, voice, etc. The emotion engine analyzes data acquired through the camera and microphone to identify the user's emotional state (e.g., anxiety, relief, concentration).

[1138] Step 10:

[1139] When a user starts a disaster prevention training, the device sends emotional data to the server through the emotion engine. The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance or encouraging messages.

[1140] Step 11:

[1141] Users can play the downloaded augmented reality content in the app and experience disaster prevention drills. The device uses an emotion engine to monitor the user's emotional state in real time and sends feedback to the server, which then adjusts the content to prevent the drill from becoming too stressful.

[1142] In this way, the system combined with the emotion engine enables users to receive realistic and effective disaster prevention training, supporting them in taking quick and safe evacuation actions in the event of a disaster. The addition of the emotion engine further improves the quality of the training and the user experience.

[1143] Example 2

[1144] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1145] Conventional disaster prevention training systems are limited to simulated training that lacks realism, and users are not fully prepared to take appropriate actions in a real disaster situation. Furthermore, training does not take into account the user's emotional state, making it difficult to improve the effectiveness of the training. To solve these problems, a realistic training environment and dynamic scenario adjustments based on the user's emotional state are needed.

[1146] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1147] In this invention, the server includes means for acquiring past disaster information from an external database or API, means for acquiring facility floor plans, evacuation routes, and emergency supply information from an internal database, means for generating a virtual disaster scenario using a large-scale language model, means for creating an evacuation simulation video based on an actual facility environment using image recognition technology, means for integrating the generated simulation video and the scenario and outputting it as augmented reality content, means for downloading the augmented reality content to a user's device so that the user can view and experience it, and means for recognizing the user's emotions in real time using an emotion recognition engine and dynamically adjusting the training scenario based on that data. This allows users to experience a realistic disaster prevention training, and the dynamic adjustment of the training scenario according to their emotional state allows for more effective training.

[1148] "Past disaster information" refers to information such as the date and time of occurrence, scale of damage, and area of ​​impact regarding disasters such as typhoons and earthquakes, which is obtained through external databases or APIs.

[1149] An "internal database" refers to a database that stores data about the facility, such as office floor plans, evacuation routes, and information on stockpiled supplies.

[1150] A "large-scale language model" is an artificial intelligence model based on massive amounts of data that can generate and analyze text. For example, GPT-4 is one such model.

[1151] "Image recognition technology" refers to the technology that analyzes image data from cameras and sensors and creates evacuation simulation images based on a real office environment.

[1152] "Augmented reality content" refers to a form of content that integrates generated simulation footage with disaster scenarios, allowing users to experience virtual information while viewing the real world.

[1153] "Cloud storage" refers to an online storage service that stores data over the Internet and generates a URL that users can access.

[1154] An "emotion recognition engine" is an engine that recognizes emotions from a user's facial expressions and voice in real time and adjusts the system's operation based on that data.

[1155] "User's device" refers to devices used by users, such as smartphones, tablets, and AR headsets.

[1156] "Dynamic adjustment" refers to changing the scenario or system behavior appropriately in response to the user's emotional state and other real-time data.

[1157] This invention is a system that combines past disaster information with internal data, generates virtual disaster scenarios using artificial intelligence technology, and provides realistic disaster prevention training content using augmented reality (AR). This system is realized by the server, terminals, and users each playing their respective roles.

[1158] First, the server retrieves past disaster information from an external database or API. The hardware used here is a high-performance server, and the software is a program that makes API calls. The external databases used include the government's disaster prevention database and the Japan Meteorological Agency's API. Specific data retrieved includes the date and time of typhoon and earthquake occurrence, the scale of damage, and the area of ​​impact.

[1159] Next, the server retrieves the facility's floor plan, evacuation routes, and stockpiled goods information from an internal database. This database uses an SQL database, and the server executes queries to retrieve the necessary information. This provides the facility's layout, the layout of each room, main evacuation routes, and a list of stockpiled goods.

[1160] The acquired data is preprocessed by the server. During the preprocessing stage, data formats are standardized, missing data is filled in, and outliers are corrected. The preprocessed data is then input into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. This scenario includes specific disaster situations, evacuation routes, and how to use stockpiled items needed during evacuation.

[1161] The generated scenario is passed to a video generation AI, which uses 3D modeling software (such as Blender or Unreal Engine) to create an evacuation simulation video based on the actual facility environment, allowing users to experience a realistic evacuation drill.

[1162] Next, the server integrates the generated scenario with the simulation video and outputs it as augmented reality (AR) content. This augmented reality content is saved in cloud storage (e.g., AWS S3) and a URL is generated that users can access.

[1163] The user's device (smartphone, tablet, AR headset, etc.) downloads the augmented reality content from this cloud storage and plays it when the user specifies. When the user launches the application and starts a disaster prevention drill, the device uses an emotion engine to recognize the user's emotions in real time. The emotion engine analyzes the user's emotional state from their facial expressions and voice and sends the data to the server.

[1164] The server dynamically adjusts the training scenario based on the received emotional data. For example, if the user feels anxious, the system will provide more detailed guidance and encouraging messages. It also monitors the user's emotional state and adjusts the training to prevent the user from feeling overly stressed.

[1165] As a concrete example, when a user starts a disaster drill and selects "Generate Scenario" in the application, the device sends a request to the server. The server collects the necessary data and generates a scenario. The generated scenario and video are integrated as augmented reality content, stored in cloud storage, and then made available for download from the user's device. When the user starts the drill, the device uses an emotion engine to recognize the user's emotions in real time and sends the data to the server. The server dynamically adjusts the training scenario based on the emotion data, helping the user receive optimal training.

[1166] Examples of prompts include:

[1167] "Generate evacuation scenarios in the event of flooding due to a typhoon, and create evacuation simulation videos based on floor plans and evacuation routes for designated facilities. Also, include reassuring messages if users are feeling anxious."

[1168] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1169] Program processing flow

[1170] Step 1: Obtaining past disaster information

[1171] The server obtains past disaster information through an external database or API. As input data, it receives information such as the date and time of typhoon or earthquake occurrence, the scale of damage, and the affected area from the government's disaster prevention database or the API of the Japan Meteorological Agency. As output data, it saves the obtained disaster information in JSON or CSV format.

[1172] Step 2: Acquire internal data

[1173] The server retrieves facility floor plans, evacuation routes, and stockpiled goods information from an internal database. It queries an internal SQL database to retrieve the layout of each room in the facility, the main evacuation routes, and a list of stockpiled goods. It saves this information in a standardized format as output data.

[1174] Step 3: Preprocessing the data

[1175] The server preprocesses the acquired disaster information and in-house data. The data acquired in Step 1 and Step 2 are used as input data. Preprocessing involves filling in missing data, detecting and correcting outliers, and standardizing data formats. A consistent preprocessed dataset is generated as output data.

[1176] Step 4: Generate disaster scenarios

[1177] The server inputs the preprocessed data into a large-scale language model (such as GPT-4) to generate a virtual disaster scenario. The preprocessed dataset and prompt sentences are used as input data. The data calculation generates a scenario based on the language model. The output data is a scenario that includes specific disaster situations, evacuation routes, and how to use stockpiles of supplies needed during evacuation.

[1178] Step 5: Generate footage

[1179] The server passes the generated scenario to a video generation AI, which creates an evacuation simulation video based on the actual facility environment. The generated scenario and a 3D model of the facility are used as input data. 3D modeling software (Blender or Unreal Engine) is used for data calculations to generate a video that realistically reproduces evacuation behavior. The evacuation simulation video is output as output data.

[1180] Step 6: Generate and serve augmented reality (AR) content

[1181] The server integrates the generated scenario and video and outputs it as augmented reality content. The scenario and video are used as input data. Data processing involves converting it into AR format. AR content is generated as output data and uploaded to cloud storage.

[1182] Step 7: Save to cloud storage and generate access URL

[1183] The server saves the generated AR content in cloud storage and generates a URL that the user can access. The AR content is used as input data. As processing, the file is uploaded to a cloud storage service (e.g., AWS S3) and an access URL is generated. As output data, an access URL that the user can use is generated.

[1184] Step 8: Download and play augmented reality content

[1185] The device downloads the augmented reality content from cloud storage and plays it when the user specifies. The generated access URL is used as input data. The downloaded content is then played on the application as processing. The augmented reality content that the user can experience is displayed as output data.

[1186] Step 9: Recognizing User Emotions

[1187] The device uses an emotion engine to recognize the user's emotions in real time. The input data is the user's facial expressions and voice data. The processing is performed using emotion recognition technology for analysis. The output data is the user's emotional state, which is sent to the server.

[1188] Step 10: Dynamically adjust the training scenario

[1189] The server dynamically adjusts the training scenario based on the received user emotional data. The emotional data received in real time is used as input data. The data calculation changes the difficulty and content of the scenario according to the emotional state. The adjusted training scenario is provided to the user as output data.

[1190] (Application example 2)

[1191] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1192] Conducting realistic and effective evacuation drills in the event of a disaster is an important challenge in industrial sites such as factories and manufacturing facilities. However, conventional evacuation drills are expensive to implement, have fixed scenarios, and often do not adequately adapt to real-world disaster situations. Furthermore, training is often conducted without taking into account the emotions and stress levels of employees, resulting in ineffective training. Furthermore, it is difficult to generate appropriate evacuation scenarios by combining disaster information and facility data, and there are limited means of providing realistic simulation images.

[1193] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1194] In this invention, the server includes: means for acquiring past disaster information from an external database or API; means for acquiring workplace floor plans, evacuation routes, and emergency supply information from an internal database; means for generating a virtual disaster scenario using a large-scale language model; means for creating an evacuation simulation video based on an actual work environment using image generation AI; means for integrating the generated simulation video and the scenario and outputting it as augmented reality content; means for downloading the augmented reality content to a user's device so that it can be viewed and experienced; means for recognizing user emotions in real time and collecting that data; and means for dynamically adjusting the training scenario based on the user's emotion data. This makes it possible to conduct realistic and effective evacuation drills and provide appropriate training content that takes into account the emotions and stress of employees.

[1195] "Disaster information" refers to data on natural and man-made disasters that have occurred in the past, including the date and time of the disaster, the scale of the damage, and the area of ​​impact.

[1196] An "internal database" is a collection of information stored within a particular facility or organization, including floor plans, evacuation routes, and emergency supply information.

[1197] A "large-scale language model" is a type of artificial intelligence model that has a very large number of parameters and performs natural language processing by learning from huge amounts of text data.

[1198] A "virtual disaster scenario" is a fictitious story or situation in which a disaster situation or evacuation route is generated by computer simulation.

[1199] "Image generation artificial intelligence" is an artificial intelligence system that can generate realistic images and simulations using 3D models and CG technology.

[1200] A "work environment" refers to the location where a specific task is performed, such as a factory or manufacturing facility, and the surrounding environment.

[1201] "Augmented reality content" means applications or content that overlay digital information onto a real-world environment, creating a visual and auditory experience.

[1202] A "terminal" is a device used by a user to access the system, and includes a smartphone, tablet, etc.

[1203] "Emotion engine" refers to technology that analyzes the user's facial expressions and voice to recognize their current emotional state.

[1204] A "training scenario" is a plan or scenario that shows the procedures and specific training content for disaster prevention and evacuation drills.

[1205] "Dynamic adjustment" means changing and adapting in real time to the situation.

[1206] "Collected Data" is a general term for information obtained from external and internal databases.

[1207] "Unifying formats" means organizing and converting data of different formats according to consistent standards.

[1208] "Cloud storage" is a storage service for saving and managing data over the Internet.

[1209] A "URL" is a uniform resource identifier used to identify resources on the Internet.

[1210] The present invention is a system for implementing effective evacuation drills at industrial sites such as factories, manufacturing facilities, etc. This system operates by having a server, terminals, and users each fulfill their respective roles.

[1211] The server first obtains past disaster information from an external database or API. The obtained disaster information includes the date and time of the disaster, the scale of damage, and the area of ​​impact. The server then obtains information on the workplace floor plan, evacuation routes, and stockpiles from an internal database. This data is then preprocessed and converted into a unified format.

[1212] The server then uses a large-scale language model (e.g., GPT-4) to generate a virtual disaster scenario, including specific disaster situations, evacuation routes, and how to use necessary supplies during evacuation. The generated scenario is then passed to a video generation AI, which creates an evacuation simulation video based on a real-world working environment.

[1213] The server integrates the generated simulation video with the scenario and outputs it as augmented reality content. This augmented reality content is downloaded to the user's device (e.g., smartphone, head-mounted display) and can be viewed and experienced. The augmented reality content stored in cloud storage generates a URL that the user can access, making it easy to download when needed.

[1214] The device downloads augmented reality content from cloud storage at the user's specified time. When the user launches the application and begins a disaster prevention training, the device uses an emotion engine to recognize emotions in real time from the user's facial expressions and voice, and sends them to the server. The server dynamically adjusts the training scenario based on the received emotion data. For example, if the user feels anxious, the system will adjust the scenario by providing more detailed guidance and explanations.

[1215] In this way, the system of the present invention enables users to receive realistic and effective evacuation training, helping them to evacuate more quickly and safely in the event of a disaster. The addition of an emotion engine can further improve the quality of the training and the user experience.

[1216] ■Example:

[1217] For example, if a fire breaks out in a factory, the system uses past fire information to generate a virtual evacuation scenario. This scenario includes the location of fire extinguishers stored in each room, along with evacuation routes within the factory. The system monitors the user's emotions in real time, and if the user feels anxious, it provides detailed guidance on how to use a fire extinguisher and the best evacuation route.

[1218] ■Example of a prompt:

[1219] "Generate a disaster scenario. Use the following data:

[1220] Disaster information: Fire, date of occurrence, area of ​​impact

[1221] Internal factory data: floor plans, evacuation routes, and stockpile information

[1222] Required outcome: A detailed description of the evacuation scenario and the supplies that should be used.”

[1223] This enables realistic and effective evacuation drills, ensuring the safety of employees and quick evacuation.

[1224] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1225] Step 1:

[1226] The server retrieves past disaster information from an external database or API. Specifically, the server sends a request to the disaster information API, and the retrieved disaster information includes the date and time of the occurrence, the scale of damage, the area of ​​impact, etc. The API request is made as input, and disaster information data is obtained as output.

[1227] Step 2:

[1228] The server retrieves the workplace floor plan, evacuation routes, and stockpile information from an internal database. The internal database stores the factory floor plan, evacuation routes, and stockpile location information. The database query is performed as input, and the internal data is obtained as output.

[1229] Step 3:

[1230] The server preprocesses the acquired disaster information and internal data and standardizes their formats. Specifically, it converts each dataset into the same format and filters out unnecessary data. It receives disaster information data and internal data as input and outputs data in a unified format.

[1231] Step 4:

[1232] The server inputs the unified data into a large-scale language model (e.g., GPT-4) to generate a hypothetical disaster scenario. The unified data is passed to the model as input, and the disaster scenario is obtained as output. The model generates the scenario using a prompt sentence.

[1233] Step 5:

[1234] The server then passes the generated disaster scenario to a video generation AI system, which then creates an evacuation simulation video based on the actual work environment.The disaster scenario is used as input, and the simulation video, using 3D models and CG technology, is obtained as output.

[1235] Step 6:

[1236] The server integrates the simulation video and the generated scenario and outputs it as augmented reality content.The simulation video and scenario are integrated as input, and the completed augmented reality content is obtained as output.

[1237] Step 7:

[1238] Save augmented reality content in cloud storage and generate a URL that users can access. Upload augmented reality content to the cloud as input, and generate an access URL as output.

[1239] Step 8:

[1240] The user's device launches an application that downloads the augmented reality content from cloud storage and allows it to be viewed and experienced, using a URL as input and the downloaded content as output.

[1241] Step 9:

[1242] The device uses an emotion engine to recognize emotions from the user's facial expressions, voice, etc. in real time while the user is experiencing augmented reality content, and transmits the emotions to the server. The device receives emotion data collected in real time as input, and obtains emotion data to be sent to the server as output.

[1243] Step 10:

[1244] The server dynamically adjusts the training scenario based on the received emotional data. It receives the emotional data as input and outputs the adjusted training scenario in real time. For example, if the user feels anxious, it can provide more detailed guidance or encouraging messages.

[1245] Through these steps, users can receive realistic and effective evacuation training. The data processing and calculations performed at each step support more appropriate and safe evacuation behavior.

[1246] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1247] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1248] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1249] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1250] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1251] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1252] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1253] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1254] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1255] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1256] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1257] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1258] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1259] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1260] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1261] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1262] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1263] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1264] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1265] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1266] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1267] The following is further disclosed regarding the above embodiment.

[1268] (Claim 1)

[1269] A means to obtain past disaster information from an external database or API,

[1270] A means to obtain office floor plans, evacuation routes, and emergency supply information from the company database,

[1271] a means for generating virtual disaster scenarios using large-scale language models;

[1272] A means for creating an evacuation simulation video based on an actual office environment using video generation artificial intelligence;

[1273] A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content;

[1274] A means for downloading augmented reality content to a user's device, allowing the user to view and experience the content;

[1275] A system including:

[1276] (Claim 2)

[1277] 10. The system of claim 1, further comprising means for preprocessing and formatting the collected data.

[1278] (Claim 3)

[1279] 10. The system of claim 1, further comprising means for storing the generated augmented reality content in cloud storage and generating a URL accessible by the user.

[1280] "Example 1"

[1281] (Claim 1)

[1282] A means to obtain past disaster information from an external database or API,

[1283] A means of obtaining facility floor plans, evacuation routes, and stockpile information from an internal database;

[1284] A means of preprocessing the collected data and standardizing the format;

[1285] a means for generating virtual disaster scenarios using large-scale language models;

[1286] A means for creating an evacuation simulation video based on a real facility environment using video generation artificial intelligence;

[1287] A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content;

[1288] A means for downloading augmented reality content to a user's device, allowing the user to view and experience the content;

[1289] A system including:

[1290] (Claim 2)

[1291] 10. The system of claim 1, further comprising means for storing the generated augmented reality content in cloud storage and generating a URL accessible by the user.

[1292] (Claim 3)

[1293] The system according to claim 1, further comprising means for generating evacuation routes and emergency supplies based on the generated virtual disaster scenario.

[1294] "Application Example 1"

[1295] (Claim 1)

[1296] A means to obtain past disaster information from an external database or API,

[1297] A means of obtaining facility floor plans, evacuation routes, and stockpile information from the company's internal database;

[1298] a means for generating virtual disaster scenarios using large-scale language models;

[1299] A means for creating an evacuation simulation video based on an actual facility environment using video generation artificial intelligence;

[1300] A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content;

[1301] A means for downloading augmented reality content to a user's device, allowing the user to view and experience the content;

[1302] means for tracking a user's training progress and generating reports;

[1303] A system including:

[1304] (Claim 2)

[1305] 10. The system of claim 1, further comprising means for preprocessing and formatting the collected data.

[1306] (Claim 3)

[1307] 10. The system of claim 1, further comprising means for storing the generated augmented reality content in cloud storage and generating a URL accessible by the user.

[1308] "Example 2: Combining Emotion Engines"

[1309] (Claim 1)

[1310] A means to obtain past disaster information from an external database or API,

[1311] A means of obtaining facility floor plans, evacuation routes, and stockpile information from an internal database;

[1312] a means for generating virtual disaster scenarios using large-scale language models;

[1313] A means for creating evacuation simulation images based on the actual facility environment using image recognition technology;

[1314] A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content;

[1315] A means for downloading augmented reality content to a user's device so that the content can be viewed and experienced;

[1316] a means for recognizing user emotions in real time using an emotion recognition engine and dynamically adjusting training scenarios based on that data;

[1317] A system including:

[1318] (Claim 2)

[1319] 10. The system of claim 1, further comprising means for preprocessing and formatting the collected data.

[1320] (Claim 3)

[1321] 10. The system of claim 1, further comprising means for storing the generated augmented reality content in cloud storage and generating a URL accessible by the user.

[1322] "Application example 2 when combining emotion engines"

[1323] (Claim 1)

[1324] A means to obtain past disaster information from an external database or API,

[1325] A means of obtaining information on workplace floor plans, evacuation routes, and emergency supplies from an internal database;

[1326] a means for generating virtual disaster scenarios using large-scale language models;

[1327] A means for creating an evacuation simulation video based on an actual work environment using video generation artificial intelligence;

[1328] A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content;

[1329] A means for downloading augmented reality content to a user's device, allowing the user to view and experience the content;

[1330] A means for recognizing and collecting data on user emotions in real time;

[1331] means for dynamically adjusting the training scenario based on the user's emotional data;

[1332] A system including:

[1333] (Claim 2)

[1334] 10. The system of claim 1, further comprising means for preprocessing and formatting the collected data.

[1335] (Claim 3)

[1336] 10. The system of claim 1, further comprising means for storing the generated augmented reality content in cloud storage and generating a URL accessible by the user. [Explanation of symbols]

[1337] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means to obtain past disaster information from an external database or API, A means to obtain office floor plans, evacuation routes, and emergency supply information from the company database, a means for generating virtual disaster scenarios using large-scale language models; A means for creating an evacuation simulation video based on an actual office environment using video generation artificial intelligence; A means for integrating the generated simulation video with the scenario and outputting it as augmented reality content; A means for downloading augmented reality content to a user's device, allowing the user to view and experience the content; A system including:

2. 2. The system of claim 1, further comprising means for preprocessing and formatting the collected data.

3. The system of claim 1 , further comprising means for storing the generated augmented reality content in cloud storage and generating a URL accessible to a user.

Images