System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting, by a server, disaster information from the Internet; means for training, by the server, a generated AI model based on the collected disaster information; means for periodically distributing, by the server, the trained generated AI model to terminals; and means for providing, by the terminals, disaster information to users in offline environments by using the locally stored generated AI model.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] When communications infrastructure goes down during a large-scale disaster, victims and rescue teams cannot access the information they need, making it difficult to respond appropriately. This prevents rapid information sharing, which could hinder rescue operations, evacuation of victims, and first aid. There is a need for a system that can effectively provide disaster information even when Internet connection is unavailable, and support rapid and accurate decision-making. [Means for solving the problem]

[0005] The present invention provides a system that includes a server collecting disaster information from the Internet, a server training a generative AI model based on the collected disaster information, a server periodically distributing the trained generative AI model to a terminal, and a terminal providing disaster information to users in an offline environment using the locally stored generative AI model. This system enables disaster victims and rescue teams to obtain necessary disaster information and respond quickly and appropriately, even when communication infrastructure is down. In particular, the server distributes an updated generative AI model to the terminal once a day, and the terminal provides information on the location of evacuation shelters and first aid procedures offline, thereby improving the efficiency and effectiveness of disaster response.

[0006] The "server" is a computer system that collects disaster information from the Internet, trains a generative AI model based on that information, and distributes it to terminals.

[0007] "Disaster information" refers to information that is important in the event of a disaster, such as information about earthquakes, floods, the location of evacuation shelters, and first aid procedures.

[0008] A "generative AI model" is an artificial intelligence model trained from collected disaster information, which provides information and decision-making support to users.

[0009] "Training" is the machine learning process that uses collected data to improve the performance of a generative AI model.

[0010] "Regularly" means repeated at regular intervals, such as once a day.

[0011] "Terminal" refers to a device operated by a user to receive disaster information in an offline environment using an installed generative AI model. This includes smartphones and tablets.

[0012] "Locally stored" refers to the state in which the generative AI model is stored on the device so that it can be used even without an internet connection.

[0013] "Offline environment" refers to a situation where an internet connection is not available.

[0014] "User" refers to individuals and rescue workers who use the system to receive information during a disaster.

[0015] "Providing disaster information" means generating and displaying appropriate information in the event of a disaster in response to a user's question.

[0016] "Location of evacuation shelter" is information about designated locations for residents to evacuate to in the event of a disaster.

[0017] "First aid" refers to emergency treatment or response methods for injuries or illnesses. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] This invention provides a system that allows users to obtain necessary disaster information even when communication infrastructure is down during a disaster. This system is composed of a server and a terminal, and collects and processes disaster information on the server, and provides the information offline on the terminal.

[0040] The server first collects disaster information from the Internet. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database. The server then trains a generative AI model based on the collected disaster information. The trained generative AI model is responsible for generating appropriate information for users. This generative AI model needs to be updated, so the server retrains the model periodically (e.g., once a day) to generate the latest model.

[0041] The latest generated generative AI model is distributed from the server to the device, which then stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down.

[0042] When a user needs disaster information, they input a question into the device. For example, if the question is "Where is the nearest evacuation shelter?", the device uses a locally stored generative AI model to generate an answer to this question. As a response, the device displays the location information of the nearest evacuation shelter. In addition, in response to the question "Tell me how to provide first aid," it displays appropriate first aid instructions.

[0043] As a specific example, a server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes the trained model to devices. Even when the device is offline, when a user asks, "Where are the evacuation shelters?", the model is used to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0044] In this way, the present invention supports rapid and accurate decision-making during disasters by having the server collect disaster information, train a generative AI model, distribute the model to terminals, and have the terminals provide information in an offline environment.

[0045] The processing flow will be explained below.

[0046] Step 1:

[0047] The server collects disaster information from the Internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. The obtained data is converted into an appropriate format and stored in a database.

[0048] Step 2:

[0049] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0050] Step 3:

[0051] The server trains the generative AI model. Through training, the AI ​​model is able to generate appropriate answers to user questions based on disaster information. Once trained, the AI ​​model is saved.

[0052] Step 4:

[0053] The server distributes the trained generative AI model to the device. The server periodically, for example once a day, obtains a list of devices for each user and sends the latest model to the device.

[0054] Step 5:

[0055] The device receives the model from the server and stores it locally, allowing the device to use the generative AI model even without an internet connection.

[0056] Step 6:

[0057] The user inputs a question requesting disaster information, such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0058] Step 7:

[0059] The device uses a locally stored generative AI model to generate appropriate answers to user questions. Even in offline environments, the device can use the model to obtain useful information about the question.

[0060] Step 8:

[0061] The device will display the generated answer to the user, such as "The nearest evacuation shelter is X Park" or "As a first aid measure, first try to stop the bleeding."

[0062] This series of steps allows users to obtain the necessary disaster information in an offline environment and make quick and accurate decisions, even if the communications infrastructure goes down during a disaster.

[0063] Example 1

[0064] 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."

[0065] There is a need for a system that allows users to quickly and reliably obtain necessary disaster information even when communication infrastructure is down during a disaster. However, conventional systems require an internet connection, making it difficult to provide information when communication infrastructure is lost. Furthermore, disaster information changes over time, requiring real-time updates, but there is a lack of an efficient way to do this. To solve these problems, a system is needed that can provide the latest disaster information even in an offline environment.

[0066] 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.

[0067] In this invention, the server includes means for collecting disaster information from the Internet, means for storing the collected disaster information in a database, and means for training a generative AI model using the data. This allows disaster information to be collected from the Internet, stored in a database, and the generative AI model to be trained based on the data. Furthermore, the server includes means for periodically distributing the trained generative AI model to a terminal, and the terminal includes means for storing the distributed generative AI model in a local environment. When a user inputs a prompt to learn disaster information, the terminal includes means for providing the user with disaster information in an offline environment using the locally stored generative AI model. This makes it possible to provide the latest disaster information offline even if the communication infrastructure is down.

[0068] "Disaster information" refers to information about natural disasters such as earthquakes, floods, tsunamis, and typhoons, including the location of evacuation shelters, first aid procedures, and the current situation in disaster-stricken areas.

[0069] The "Internet" is a global communications infrastructure that connects computers and networks around the world to send and receive information.

[0070] A "server" is a computer system that provides information over a network and is responsible for collecting, processing, storing, and distributing data.

[0071] A "database" is a collection of data that is structured so that information can be efficiently stored, managed, and retrieved.

[0072] A "generative AI model" is an algorithm that has been trained from a specific dataset using machine learning algorithms and is capable of generating appropriate predictions or responses to new data.

[0073] "Training" is the process of using collected data to train a machine learning model and improve its prediction accuracy.

[0074] "Distributing" is the process of sending and receiving data and models from a server to a terminal.

[0075] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[0076] A "local environment" is a data storage and processing environment within the terminal itself, independent of any external network.

[0077] "Users" refer to people who use this system to obtain disaster information.

[0078] A "prompt sentence" is text that indicates the content of a question or request that a user enters into a terminal.

[0079] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0080] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server and a terminal, and disaster information is collected and processed on the server, and the information is provided offline on the terminal.

[0081] The server first collects disaster information from the Internet. This information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database such as MySQL or PostgreSQL. The server then uses the collected disaster information to train a generative AI model using a machine learning framework such as TensorFlow or PyTorch. This generative AI model is an important element because it generates appropriate information in response to user input. The generative AI model is updated periodically (e.g., once a day) and distributed from the server to devices.

[0082] In the process of distributing a generative AI model, the server uses a communication protocol (e.g., HTTP or FTP) to send the latest model to the device. The device receives this model and saves it in local storage. This allows the device to use the generative AI model in an offline environment, even if the communication infrastructure goes down.

[0083] When a user needs disaster information, they input a prompt into their device. Examples of prompts include "Where is the nearest evacuation shelter?" and "Tell me how to provide first aid." When a user inputs a prompt, the device uses a locally stored generative AI model to generate an answer to this question. For example, if the question "Where is the nearest evacuation shelter?" is input, the device will provide the location information of the nearest evacuation shelter. Also, in response to the question "Tell me how to provide first aid," it will display appropriate first aid instructions.

[0084] Consider the following scenario as a concrete example: A server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes it to devices. Then, even in an offline environment, when a user asks, "Where are the evacuation shelters?", the device uses the model to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0085] As described above, the present invention supports rapid and accurate decision-making during disasters by having a server collect disaster information, train a generative AI model based on that information, distribute the model to terminals, and have the terminals provide information to users in an offline environment.

[0086] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0087] Step 1:

[0088] The server collects disaster information from the Internet.

[0089] Input: API endpoint of disaster information service

[0090] Specific operation: The server sends requests to the APIs of the Japan Meteorological Agency and disaster information services to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc.

[0091] Output: JSON data of disaster information

[0092] Step 2:

[0093] The server stores the collected disaster information in a database.

[0094] Input: JSON data of disaster information

[0095] Specific operation: The server parses the received JSON data and generates and executes SQL queries to insert it into a MySQL or PostgreSQL database.

[0096] Output: Disaster information stored in a database

[0097] Step 3:

[0098] The server uses the data to train a generative AI model.

[0099] Input: Disaster information stored in the database

[0100] How it works: The server loads disaster information from a database and trains a generative AI model using machine learning frameworks such as TensorFlow and PyTorch. The training process includes data preprocessing, model building, and model training.

[0101] Output: A trained generative AI model

[0102] Step 4:

[0103] The server distributes the latest generative AI model to the device.

[0104] Input: A trained generative AI model

[0105] Specific operation: The server uses HTTP or FTP communication protocols to send the generated AI model to the terminal.

[0106] Output: The generated AI model sent to the device

[0107] Step 5:

[0108] The device stores the generated AI model locally.

[0109] Input: The generative AI model sent to the device

[0110] Specific operation: The device saves the generative AI model received from the server in a specific directory in local storage.

[0111] Output: A locally stored generative AI model

[0112] Step 6:

[0113] The user types a prompt into the terminal.

[0114] Input: The prompt text entered by the user (e.g., "Where is the nearest evacuation shelter?")

[0115] Specific operation: The user opens the application on the device, enters the information they want to know in the text box, and presses the send button.

[0116] Output: The prompt entered by the user

[0117] Step 7:

[0118] The device uses a generative AI model to generate answers to the user's questions.

[0119] Input: Locally stored generative AI model, user-entered prompt

[0120] How it works: The device application inputs a prompt sentence into the AI ​​model, which then generates an appropriate answer. This process includes interpreting the text, searching for relevant information, and generating an answer.

[0121] Output: The generated answer

[0122] Step 8:

[0123] The terminal displays the generated answer to the user.

[0124] Input: Generated answer (e.g., "The nearest evacuation shelter is the community center in XX town.")

[0125] Specific operation: The device displays the answer generated by the generative AI model on the application's user interface.

[0126] Output: The answer displayed to the user

[0127] (Application example 1)

[0128] 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."

[0129] During disasters, communication infrastructure often goes down, making it difficult for users to obtain the disaster information they need. Therefore, there is a need to provide a system that allows users to quickly and accurately obtain the disaster information they need even when communication infrastructure goes down. In particular, within factories, it is necessary to ensure the safety of workers by providing disaster information in a timely manner. In addition, there is a need for a system that can provide information via voice even in offline environments.

[0130] 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.

[0131] In this invention, the server includes means for collecting disaster information from an information network, means for training a generative AI model based on the collected disaster information, means for periodically distributing the trained generative AI model to a terminal, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, and means for using the generative AI model installed in the robot to perform speech recognition and speech synthesis and provide appropriate disaster information by voice in response to user questions. This makes it possible to provide necessary disaster information to users in an offline environment using the locally stored model even if the communication infrastructure is down. In addition, because the robot provides information using speech recognition and speech synthesis, workers can obtain information such as safe evacuation locations and appropriate first aid methods without using their hands.

[0132] A "server" is a computing device whose role is to collect disaster information, train a generative AI model based on that information, and periodically distribute the trained model to terminals.

[0133] "Disaster information" refers to information including the latest information on natural disasters such as earthquakes, floods, and fires, the location of evacuation shelters, first aid procedures, and so on.

[0134] "Information network" means a communications infrastructure for gathering information, including the Internet and other online networks.

[0135] A "generative AI model" is an AI model trained based on collected disaster information and used to generate appropriate information in response to user questions.

[0136] "Training" is the process of using collected data to teach an AI model so that it can perform a specific task.

[0137] "Regularly" means repeated at regular time intervals, for example, once a day.

[0138] A "terminal" is a device used by a user to obtain information, and includes smartphones, tablets, PCs, etc.

[0139] "Locally stored" means that the trained generative AI model is stored in the device's internal storage and is available even in offline environments.

[0140] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0141] "User" refers to anyone who uses the system to refer to information, including ordinary people who need disaster information and workers in factories.

[0142] A "robot" is an autonomous mechanical device that uses voice recognition and voice synthesis to provide users with necessary information during disasters.

[0143] "Speech recognition" is a technology that converts speech into text and is used to recognize user questions.

[0144] "Speech synthesis" is a technology that converts text into speech and is used to provide audible responses to the user.

[0145] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server, a terminal, and a robot.

[0146] The server collects disaster information from the Internet and stores it in a database. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. Based on the collected data, the server trains a generative AI model. The generative AI model is designed to provide appropriate information in response to user questions during disasters. The trained generative AI model is updated periodically (e.g., once a day) and distributed to terminals and robots.

[0147] The device locally stores the generative AI model received from the server. This allows disaster information to be provided to users in an offline environment even if the communication infrastructure is down. When a user needs disaster information, they can input a question into the device, and the locally stored generative AI model will be used to generate an answer to that question. For example, in response to the question, "Where is the nearest evacuation shelter?", it is possible to provide the location information of the nearest evacuation shelter.

[0148] The robot utilizes speech recognition and speech synthesis technologies to provide appropriate disaster information via voice in response to voice questions from users. A generative AI model is installed in the robot, allowing it to operate in offline environments. This allows users to obtain information by voice without using their hands. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot can provide specific instructions via voice. It is also designed for use in factories, ensuring the safety of workers by quickly providing information on safe evacuation locations and emergency response methods.

[0149] The software used includes Python for processing the collected disaster information, BeautifulSoup for scraping, TensorFlow for training the model, SpeechRecognition for speech recognition, and pyttsx3 for speech synthesis, as well as PostgreSQL as the database system for managing the stored data.

[0150] Examples of prompts include:

[0151] "Where is the nearest evacuation shelter in case of an earthquake?"

[0152] "Please tell me the emergency evacuation route in case of a fire."

[0153] This system enables users and workers to quickly obtain necessary information and respond appropriately even in the event of a disaster.

[0154] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0155] Step 1:

[0156] The server collects disaster information from the Internet.

[0157] Specifically, the server uses Python requests and the BeautifulSoup library to retrieve disaster information such as earthquakes, floods, evacuation shelter locations, first aid instructions, etc. from various websites and APIs. This collected information is then stored in a PostgreSQL database.

[0158] Input: Websites and APIs that provide disaster information

[0159] Output: Disaster information stored in a PostgreSQL database

[0160] Step 2:

[0161] The server trains a generative AI model based on the collected disaster information.

[0162] Specifically, the server uses TensorFlow to train a generative AI model using collected disaster information. This model is then able to generate appropriate answers to user questions. Training data is retrieved from a database and the model's parameters are adjusted.

[0163] Input: Disaster information stored in a PostgreSQL database

[0164] Output: A trained generative AI model

[0165] Step 3:

[0166] The server periodically distributes the trained generative AI model to terminals and robots.

[0167] Specifically, the server uses the Paramiko library to transfer the trained model to each terminal and robot via SSH, with this distribution occurring once a day.

[0168] Input: A trained generative AI model

[0169] Output: Generative AI model distributed to each device and robot

[0170] Step 4:

[0171] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[0172] When a user types a question into the device, the device uses a locally stored generative AI model to generate an answer to that question, such as "Where is the nearest evacuation shelter?", providing the location of the nearest evacuation shelter.

[0173] Input: User question (e.g., "Where is the nearest evacuation shelter?")

[0174] Output: Answer to question (e.g., location of nearest evacuation shelter)

[0175] Step 5:

[0176] The robot performs speech recognition and speech synthesis, providing appropriate disaster information in voice in response to the user's questions.

[0177] Specifically, the robot uses the SpeechRecognition library to convert the user's speech into text, and then inputs that text into a generative AI model. The generated answer is then synthesized using the pyttsx3 library and provided to the user via voice. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot will provide specific instructions via voice.

[0178] Input: User's spoken question (e.g., "What is the emergency evacuation protocol in case of a fire?")

[0179] Output: Answers to questions via voice (e.g., emergency evacuation protocol instructions)

[0180] 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.

[0181] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information to users that takes their emotions into consideration.

[0182] The server first collects disaster information from the internet. It connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[0183] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[0184] Furthermore, this invention incorporates an emotion engine. The emotion engine analyzes the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine recognizes this state and provides information that will reassure the user. In this way, it becomes possible to provide information that takes into consideration the user's emotions.

[0185] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[0186] With this system, even if the communications infrastructure goes down during a disaster, users can obtain the necessary disaster information offline and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[0187] The processing flow will be explained below.

[0188] Step 1:

[0189] The server collects disaster information from the internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. This information is converted into an appropriate format and stored in a database.

[0190] Step 2:

[0191] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0192] Step 3:

[0193] The server trains the generative AI model, which learns the ability to generate appropriate answers to user questions from the collected disaster information. Once trained, the AI ​​model is saved.

[0194] Step 4:

[0195] The server periodically distributes trained generative AI models to devices. For example, the server obtains a list of each user's devices and sends it to the device to distribute the latest model once a day.

[0196] Step 5:

[0197] The device stores the generative AI model received from the server locally, allowing the device to use the generative AI model even without an internet connection.

[0198] Step 6:

[0199] The user inputs a question requiring disaster information into the terminal. For example, questions such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0200] Step 7:

[0201] The device's emotion engine recognizes the user's emotions by analyzing the user's tone of voice and facial expressions to determine whether the user is feeling anxious or impatient.

[0202] Step 8:

[0203] The device uses a locally stored generative AI model to generate appropriate answers to user questions. The device then combines the information obtained from the generative AI model with the judgment results of the emotion engine to provide the user with the most appropriate information.

[0204] Step 9:

[0205] The device displays the generated answer to the user. For example, it displays specific information such as "The nearest evacuation shelter is X Park" or first aid instructions such as "First, take a deep breath. Then, stop the bleeding."

[0206] Step 10:

[0207] The system takes prompt and appropriate action based on the information provided by the user. In addition, the system provides emotional care through an emotion engine, allowing the user to respond calmly.

[0208] This series of steps allows users to obtain the necessary disaster information in an offline environment even if communication infrastructure goes down during a disaster, and provides support based on emotion recognition, enabling quick and accurate decisions.

[0209] Example 2

[0210] 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."

[0211] During disasters, communication infrastructure is prone to downtime, making it difficult to obtain necessary disaster information in a timely manner. It is also important to provide information while responding to the user's emotional state, but no systems currently exist with this functionality. Therefore, there is a need for a system that can provide disaster information offline and that can provide information that is adapted to the user's emotional state.

[0212] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting disaster information from the Internet, a means for training a generative AI model based on the collected disaster information, and a means for periodically distributing the trained generative AI model to terminals. This makes it possible to provide disaster information in an offline environment even if the communication infrastructure is down, and further to provide information adapted to the user's emotional state.

[0213] A "server" is a device that collects, stores, processes, and provides information via the Internet to other devices that require that information.

[0214] "Disaster information" refers to data such as the latest situation, countermeasures, location information of evacuation centers, and first aid methods related to natural disasters such as earthquakes, floods, and typhoons.

[0215] The "Internet" is a global communications network that interconnects computer networks around the world and enables the exchange of information.

[0216] A "generative AI model" is an algorithmic model of artificial intelligence that is trained based on accumulated data and has the ability to perform specific tasks (e.g., question answering, information generation).

[0217] "Training" is the process of using collected data to optimize the parameters of an AI model and improve its performance.

[0218] "Terminal" means a telecommunications device that allows a user to input and receive information and view the processing results.

[0219] "Stored locally" means that data or programs are stored directly on a device outside the network, making them available without the need for a remote connection.

[0220] "Offline environment" refers to a state where there is no connection to the Internet or other networks.

[0221] "Users" refer to people who interact with the system and receive particular information or services.

[0222] A "question" is a user's inquiry for information, and is input into the terminal in text or voice format.

[0223] An "emotion analysis engine" is a software or hardware system that analyzes audio and / or video input to identify a user's emotional state.

[0224] "Emotional state" is information indicating the psychological state of the user, and includes specific emotions such as anxiety, fear, and relief.

[0225] "Analysis" is a method for examining collected data in detail and understanding its meaning and structure.

[0226] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. This system is composed of a server, a terminal, and an emotion analysis engine.

[0227] Server Operation

[0228] The server first connects to the APIs of government and disaster prevention agencies and collects disaster information via the internet, including the latest earthquake and flood information, evacuation shelter locations, and first aid procedures. The collected data is received in JSON format, converted into an appropriate format, and stored in a database. The server then uses this data to train a generative AI model. For training, it uses machine learning frameworks such as TensorFlow and PyTorch. Once the generative AI model is trained, it is distributed to devices periodically, for example, once a day.

[0229] Device behavior

[0230] The device locally stores the generative AI model received from the server, making it available offline even if the communications infrastructure goes down. When a user needs disaster information, they input a question into the device, such as "Where is the nearest evacuation shelter?" At this time, the device uses the locally stored generative AI model to generate and display an appropriate answer to the question. The device also uses the generative AI model to provide information on the location of evacuation shelters and first aid procedures.

[0231] User sentiment analysis

[0232] Furthermore, the present invention incorporates an emotion analysis engine. The emotion analysis engine uses a camera and microphone to analyze the user's voice input and facial expressions. Libraries such as OpenCV and DeepFace are used for the analysis. This analysis recognizes the user's emotional state, and if, for example, anxiety or fear is detected, the device will prioritize providing a reassuring message taking that state into consideration. For example, it will display a message such as, "First, take a deep breath. Then, do XX."

[0233] Specific examples

[0234] As a specific example, the server collects the latest information on evacuation shelters and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. If a user types "My child is injured, what should I do?" into the device, the device will use the generative AI model to provide appropriate first aid methods. It also uses an emotion analysis engine to recognize the user's sense of urgency and displays a message to calm them down. An example of a prompt sentence is "Please enter the user's question. For example, 'Where is the nearest evacuation shelter?'"

[0235] As described above, the present invention enables users to obtain necessary disaster information offline even if the communication infrastructure goes down during a disaster, and enables appropriate responses based on emotion recognition.

[0236] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0237] Step 1:

[0238] The server collects disaster information. The input is real-time disaster information received from the APIs of government and disaster prevention organizations. Specifically, the server uses the Python requests library to send HTTP requests and obtain data in JSON format. The output is the obtained disaster information in JSON format.

[0239] Step 2:

[0240] The server converts and stores the disaster information it acquires. The input is the JSON-formatted data acquired in step 1. Specifically, the server uses the Python pandas library to convert the data into an appropriate format and saves it in a database (e.g., MySQL or MongoDB). The output is the disaster information stored in the database.

[0241] Step 3:

[0242] The server trains a generative AI model based on the collected data. The input is disaster information stored in a database. Specifically, the server uses TensorFlow and PyTorch to build and train the generative AI model. The output is the fully trained generative AI model.

[0243] Step 4:

[0244] The server distributes the trained generative AI model to the terminal. The input is the generative AI model obtained in step 3. Specifically, the server uses the HTTP protocol to send the file of the generative AI model to the terminal. The output is the file of the generative AI model distributed to the terminal.

[0245] Step 5:

[0246] The device stores the generated AI model it receives locally. The input is the generated AI model file received from the server. Specifically, the device uses Python's requests library to download the generated AI model file and uses the os module to save it in a local directory. The output is the locally stored generated AI model.

[0247] Step 6:

[0248] The user inputs a question into the terminal. The input is the user's textual question (e.g., "Where is the nearest evacuation shelter?"). In concrete terms, the user inputs the question in text form through the terminal's interface. The output is the question input into the terminal.

[0249] Step 7:

[0250] The device generates an answer using a generative AI model. The input is the user's question and a locally stored generative AI model. Specifically, the device uses NLP technology to analyze the meaning of the question and generates an appropriate answer using the generative AI model. The output is the generated answer.

[0251] Step 8:

[0252] The device recognizes the user's emotional state. Input is the user's voice input and facial expression data. Specifically, the device analyzes data from the camera and microphone using an emotion analysis engine (e.g., OpenCV or DeepFace) to identify the user's emotional state. The output is the analyzed user's emotional state.

[0253] Step 9:

[0254] The device provides emotion-conscious information. The input is the answer generated in step 7 and the user's emotional state analyzed in step 8. Specifically, the device displays information by combining a message based on the user's emotional state. For example, if the device recognizes that the user is anxious, it provides an answer with a message such as "Please stay calm and take a deep breath first. Then, do XX." The output is emotion-conscious information provision.

[0255] (Application example 2)

[0256] 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."

[0257] With conventional disaster information systems, it is difficult for users to obtain the information they need when communication infrastructure goes down. Furthermore, because information is provided without consideration for users' feelings, users can become panicked. Furthermore, the provision of disaster information is limited, and information on appropriate evacuation shelters and first aid methods is often not provided promptly. This can lead to delayed responses in the event of a disaster, potentially exacerbating the damage.

[0258] 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.

[0259] In this invention, the server includes means for collecting disaster information from the Internet, means for training a generative AI model based on the disaster information, means for periodically distributing the trained generative AI model to terminals, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, means for the terminal to analyze the user's emotions using an emotion recognition engine and provide emotion-conscious information, and means for the robot to move autonomously and provide disaster information and first aid methods. This allows users to obtain necessary disaster information offline even if the communication infrastructure is down during a disaster, and the provision of emotion-conscious information prevents panic and enables quick and appropriate responses.

[0260] "Disaster information" refers to data and events related to disasters, including, for example, the location of evacuation shelters, first aid procedures, and information on earthquakes and floods.

[0261] "Means of collecting from the Internet" refers to the method by which the server obtains the necessary information in real time using various data sources and APIs on the Internet.

[0262] A "generative AI model" refers to an artificial intelligence model that generates appropriate responses to user input based on training data.

[0263] "Storing locally" means storing collected AI models and data in the device's internal storage area, making them available even in offline environments.

[0264] An "emotion recognition engine" is a system that has the ability to analyze a user's voice, facial expressions, text input, etc., and recognize and identify their emotional state (anxiety, impatience, etc.).

[0265] An "autonomous mobile robot" is a robot that can reach its destination using its own sensors and algorithms, without relying on external instructions or infrastructure.

[0266] "Offline environment" means a situation where the Internet or other network connection is unavailable and functionality must be maintained in this state.

[0267] "Location of evacuation shelters" refers to information about places where people can evacuate to ensure their safety in the event of a disaster.

[0268] "First aid methods" refers to emergency treatment procedures and measures that can be applied to injured people in disasters and emergencies.

[0269] This invention is a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information that takes emotions into consideration.

[0270] The server first collects disaster information from the internet. Specifically, it connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[0271] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[0272] Furthermore, this invention incorporates an emotion engine. The emotion engine can analyze the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine will recognize this state and provide information that will reassure the user. In this way, it becomes possible to provide information that takes the user's emotions into consideration.

[0273] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[0274] With this system, even if communication infrastructure goes down during a disaster, users can obtain necessary disaster information in an offline environment and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[0275] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0276] Step 1:

[0277] The server collects disaster information from the Internet.

[0278] Input: APIs from various data sources and government and disaster prevention agencies.

[0279] Processing: The server retrieves data from these APIs, such as the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. It converts this information into an appropriate format and stores it in a database.

[0280] Output: Organized disaster information data is stored in a database.

[0281] Step 2:

[0282] The server trains a generative AI model based on the collected disaster information.

[0283] Input: Disaster information data stored in a database.

[0284] Processing: The server uses disaster information data to train a generative AI model that is capable of generating appropriate answers to user questions. Machine learning algorithms are used for training.

[0285] Output: The trained generative AI model is saved on the server.

[0286] Step 3:

[0287] The server periodically distributes the trained generative AI model to the device.

[0288] Input: A trained generative AI model.

[0289] Processing: The server sends the trained generative AI model to the device once a day. The model is distributed via the network.

[0290] Output: A generative AI model stored locally on the device.

[0291] Step 4:

[0292] The user inputs a question requesting disaster information into the terminal.

[0293] Input: User question (e.g., "Where is the nearest shelter?").

[0294] Processing: The device receives the user's question and inputs it into a locally stored generative AI model.

[0295] Output: The answer to the question is generated.

[0296] Step 5:

[0297] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[0298] Input: User question, locally stored generative AI model.

[0299] Processing: Using the generative AI model, the device generates an appropriate answer, which is then displayed on the device screen.

[0300] Output: Disaster information provided to the user.

[0301] Step 6:

[0302] The device uses an emotion recognition engine to analyze the user's emotions.

[0303] Input: User's voice input, facial expression data.

[0304] Processing: The device uses an emotion recognition engine to analyze the user's voice and facial expressions to determine their emotional state.

[0305] Output: User's emotional state data.

[0306] Step 7:

[0307] The device provides emotionally sensitive information.

[0308] Input: User emotional state data, generated disaster information.

[0309] Processing: The emotion engine appropriately modifies the information obtained from the generative AI model based on the user's emotional state (e.g., "First, take a deep breath.") and displays it to the user.

[0310] Output: Emotionally sensitive information is provided to the user.

[0311] Step 8:

[0312] The robot moves autonomously and provides disaster information and first aid methods.

[0313] Input: Generative AI model, emotional state data, locally stored database.

[0314] Processing: The robot uses its own sensors and algorithms to move autonomously and provide the user with the necessary information. It also provides the most appropriate information depending on the user's emotional state.

[0315] Output: Disaster information and first aid instructions provided to users appropriately.

[0316] 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.

[0317] 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.

[0318] 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.

[0319] [Second embodiment]

[0320] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0321] 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.

[0322] 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).

[0323] 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.

[0324] 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.

[0325] 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).

[0326] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0327] 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.

[0328] 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.

[0329] 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.

[0330] 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.

[0331] 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."

[0332] This invention provides a system that allows users to obtain necessary disaster information even when communication infrastructure is down during a disaster. This system is composed of a server and a terminal, and collects and processes disaster information on the server, and provides the information offline on the terminal.

[0333] The server first collects disaster information from the Internet. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database. The server then trains a generative AI model based on the collected disaster information. The trained generative AI model is responsible for generating appropriate information for users. This generative AI model needs to be updated, so the server retrains the model periodically (e.g., once a day) to generate the latest model.

[0334] The latest generated generative AI model is distributed from the server to the device, which then stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down.

[0335] When a user needs disaster information, they input a question into the device. For example, if the question is "Where is the nearest evacuation shelter?", the device uses a locally stored generative AI model to generate an answer to this question. As a response, the device displays the location information of the nearest evacuation shelter. In addition, in response to the question "Tell me how to provide first aid," it displays appropriate first aid instructions.

[0336] As a specific example, a server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes the trained model to devices. Even when the device is offline, when a user asks, "Where are the evacuation shelters?", the model is used to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0337] In this way, the present invention supports rapid and accurate decision-making during disasters by having the server collect disaster information, train a generative AI model, distribute the model to terminals, and have the terminals provide information in an offline environment.

[0338] The processing flow will be explained below.

[0339] Step 1:

[0340] The server collects disaster information from the Internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. The obtained data is converted into an appropriate format and stored in a database.

[0341] Step 2:

[0342] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0343] Step 3:

[0344] The server trains the generative AI model. Through training, the AI ​​model is able to generate appropriate answers to user questions based on disaster information. Once trained, the AI ​​model is saved.

[0345] Step 4:

[0346] The server distributes the trained generative AI model to the device. The server periodically, for example once a day, obtains a list of devices for each user and sends the latest model to the device.

[0347] Step 5:

[0348] The device receives the model from the server and stores it locally, allowing the device to use the generative AI model even without an internet connection.

[0349] Step 6:

[0350] The user inputs a question requesting disaster information, such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0351] Step 7:

[0352] The device uses a locally stored generative AI model to generate appropriate answers to user questions. Even in offline environments, the device can use the model to obtain useful information about the question.

[0353] Step 8:

[0354] The device will display the generated answer to the user, such as "The nearest evacuation shelter is X Park" or "As a first aid measure, first try to stop the bleeding."

[0355] This series of steps allows users to obtain the necessary disaster information in an offline environment and make quick and accurate decisions, even if the communications infrastructure goes down during a disaster.

[0356] Example 1

[0357] 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."

[0358] There is a need for a system that allows users to quickly and reliably obtain necessary disaster information even when communication infrastructure is down during a disaster. However, conventional systems require an internet connection, making it difficult to provide information when communication infrastructure is lost. Furthermore, disaster information changes over time, requiring real-time updates, but there is a lack of an efficient way to do this. To solve these problems, a system is needed that can provide the latest disaster information even in an offline environment.

[0359] 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.

[0360] In this invention, the server includes means for collecting disaster information from the Internet, means for storing the collected disaster information in a database, and means for training a generative AI model using the data. This allows disaster information to be collected from the Internet, stored in a database, and the generative AI model to be trained based on the data. Furthermore, the server includes means for periodically distributing the trained generative AI model to a terminal, and the terminal includes means for storing the distributed generative AI model in a local environment. When a user inputs a prompt to learn disaster information, the terminal includes means for providing the user with disaster information in an offline environment using the locally stored generative AI model. This makes it possible to provide the latest disaster information offline even if the communication infrastructure is down.

[0361] "Disaster information" refers to information about natural disasters such as earthquakes, floods, tsunamis, and typhoons, including the location of evacuation shelters, first aid procedures, and the current situation in disaster-stricken areas.

[0362] The "Internet" is a global communications infrastructure that connects computers and networks around the world to send and receive information.

[0363] A "server" is a computer system that provides information over a network and is responsible for collecting, processing, storing, and distributing data.

[0364] A "database" is a collection of data that is structured so that information can be efficiently stored, managed, and retrieved.

[0365] A "generative AI model" is an algorithm that has been trained from a specific dataset using machine learning algorithms and is capable of generating appropriate predictions or responses to new data.

[0366] "Training" is the process of using collected data to train a machine learning model and improve its prediction accuracy.

[0367] "Distributing" is the process of sending and receiving data and models from a server to a terminal.

[0368] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[0369] A "local environment" is a data storage and processing environment within the terminal itself, independent of any external network.

[0370] "Users" refer to people who use this system to obtain disaster information.

[0371] A "prompt sentence" is text that indicates the content of a question or request that a user enters into a terminal.

[0372] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0373] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server and a terminal, and disaster information is collected and processed on the server, and the information is provided offline on the terminal.

[0374] The server first collects disaster information from the Internet. This information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database such as MySQL or PostgreSQL. The server then uses the collected disaster information to train a generative AI model using a machine learning framework such as TensorFlow or PyTorch. This generative AI model is an important element because it generates appropriate information in response to user input. The generative AI model is updated periodically (e.g., once a day) and distributed from the server to devices.

[0375] In the process of distributing a generative AI model, the server uses a communication protocol (e.g., HTTP or FTP) to send the latest model to the device. The device receives this model and saves it in local storage. This allows the device to use the generative AI model in an offline environment, even if the communication infrastructure goes down.

[0376] When a user needs disaster information, they input a prompt into their device. Examples of prompts include "Where is the nearest evacuation shelter?" and "Tell me how to provide first aid." When a user inputs a prompt, the device uses a locally stored generative AI model to generate an answer to this question. For example, if the question "Where is the nearest evacuation shelter?" is input, the device will provide the location information of the nearest evacuation shelter. Also, in response to the question "Tell me how to provide first aid," it will display appropriate first aid instructions.

[0377] Consider the following scenario as a concrete example: A server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes it to devices. Then, even in an offline environment, when a user asks, "Where are the evacuation shelters?", the device uses the model to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0378] As described above, the present invention supports rapid and accurate decision-making during disasters by having a server collect disaster information, train a generative AI model based on that information, distribute the model to terminals, and have the terminals provide information to users in an offline environment.

[0379] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0380] Step 1:

[0381] The server collects disaster information from the Internet.

[0382] Input: API endpoint of disaster information service

[0383] Specific operation: The server sends requests to the APIs of the Japan Meteorological Agency and disaster information services to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc.

[0384] Output: JSON data of disaster information

[0385] Step 2:

[0386] The server stores the collected disaster information in a database.

[0387] Input: JSON data of disaster information

[0388] Specific operation: The server parses the received JSON data and generates and executes SQL queries to insert it into a MySQL or PostgreSQL database.

[0389] Output: Disaster information stored in a database

[0390] Step 3:

[0391] The server uses the data to train a generative AI model.

[0392] Input: Disaster information stored in the database

[0393] How it works: The server loads disaster information from a database and trains a generative AI model using machine learning frameworks such as TensorFlow and PyTorch. The training process includes data preprocessing, model building, and model training.

[0394] Output: A trained generative AI model

[0395] Step 4:

[0396] The server distributes the latest generative AI model to the device.

[0397] Input: A trained generative AI model

[0398] Specific operation: The server uses HTTP or FTP communication protocols to send the generated AI model to the terminal.

[0399] Output: The generated AI model sent to the device

[0400] Step 5:

[0401] The device stores the generated AI model locally.

[0402] Input: The generative AI model sent to the device

[0403] Specific operation: The device saves the generative AI model received from the server in a specific directory in local storage.

[0404] Output: A locally stored generative AI model

[0405] Step 6:

[0406] The user types a prompt into the terminal.

[0407] Input: The prompt text entered by the user (e.g., "Where is the nearest evacuation shelter?")

[0408] Specific operation: The user opens the application on the device, enters the information they want to know in the text box, and presses the send button.

[0409] Output: The prompt entered by the user

[0410] Step 7:

[0411] The device uses a generative AI model to generate answers to the user's questions.

[0412] Input: Locally stored generative AI model, user-entered prompt

[0413] How it works: The device application inputs a prompt sentence into the AI ​​model, which then generates an appropriate answer. This process includes interpreting the text, searching for relevant information, and generating an answer.

[0414] Output: The generated answer

[0415] Step 8:

[0416] The terminal displays the generated answer to the user.

[0417] Input: Generated answer (e.g., "The nearest evacuation shelter is the community center in XX town.")

[0418] Specific operation: The device displays the answer generated by the generative AI model on the application's user interface.

[0419] Output: The answer displayed to the user

[0420] (Application example 1)

[0421] 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."

[0422] During disasters, communication infrastructure often goes down, making it difficult for users to obtain the disaster information they need. Therefore, there is a need to provide a system that allows users to quickly and accurately obtain the disaster information they need even when communication infrastructure goes down. In particular, within factories, it is necessary to ensure the safety of workers by providing disaster information in a timely manner. In addition, there is a need for a system that can provide information via voice even in offline environments.

[0423] 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.

[0424] In this invention, the server includes means for collecting disaster information from an information network, means for training a generative AI model based on the collected disaster information, means for periodically distributing the trained generative AI model to a terminal, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, and means for using the generative AI model installed in the robot to perform speech recognition and speech synthesis and provide appropriate disaster information by voice in response to user questions. This makes it possible to provide necessary disaster information to users in an offline environment using the locally stored model even if the communication infrastructure is down. In addition, because the robot provides information using speech recognition and speech synthesis, workers can obtain information such as safe evacuation locations and appropriate first aid methods without using their hands.

[0425] A "server" is a computing device whose role is to collect disaster information, train a generative AI model based on that information, and periodically distribute the trained model to terminals.

[0426] "Disaster information" refers to information including the latest information on natural disasters such as earthquakes, floods, and fires, the location of evacuation shelters, first aid procedures, and so on.

[0427] "Information network" means a communications infrastructure for gathering information, including the Internet and other online networks.

[0428] A "generative AI model" is an AI model trained based on collected disaster information and used to generate appropriate information in response to user questions.

[0429] "Training" is the process of using collected data to teach an AI model so that it can perform a specific task.

[0430] "Regularly" means repeated at regular time intervals, for example, once a day.

[0431] A "terminal" is a device used by a user to obtain information, and includes smartphones, tablets, PCs, etc.

[0432] "Locally stored" means that the trained generative AI model is stored in the device's internal storage and is available even in offline environments.

[0433] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0434] "User" refers to anyone who uses the system to refer to information, including ordinary people who need disaster information and workers in factories.

[0435] A "robot" is an autonomous mechanical device that uses voice recognition and voice synthesis to provide users with necessary information during disasters.

[0436] "Speech recognition" is a technology that converts speech into text and is used to recognize user questions.

[0437] "Speech synthesis" is a technology that converts text into speech and is used to provide audible responses to the user.

[0438] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server, a terminal, and a robot.

[0439] The server collects disaster information from the Internet and stores it in a database. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. Based on the collected data, the server trains a generative AI model. The generative AI model is designed to provide appropriate information in response to user questions during disasters. The trained generative AI model is updated periodically (e.g., once a day) and distributed to terminals and robots.

[0440] The device locally stores the generative AI model received from the server. This allows disaster information to be provided to users in an offline environment even if the communication infrastructure is down. When a user needs disaster information, they can input a question into the device, and the locally stored generative AI model will be used to generate an answer to that question. For example, in response to the question, "Where is the nearest evacuation shelter?", it is possible to provide the location information of the nearest evacuation shelter.

[0441] The robot utilizes speech recognition and speech synthesis technologies to provide appropriate disaster information via voice in response to voice questions from users. A generative AI model is installed in the robot, allowing it to operate in offline environments. This allows users to obtain information by voice without using their hands. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot can provide specific instructions via voice. It is also designed for use in factories, ensuring the safety of workers by quickly providing information on safe evacuation locations and emergency response methods.

[0442] The software used includes Python for processing the collected disaster information, BeautifulSoup for scraping, TensorFlow for training the model, SpeechRecognition for speech recognition, and pyttsx3 for speech synthesis, as well as PostgreSQL as the database system for managing the stored data.

[0443] Examples of prompts include:

[0444] "Where is the nearest evacuation shelter in case of an earthquake?"

[0445] "Please tell me the emergency evacuation route in case of a fire."

[0446] This system enables users and workers to quickly obtain necessary information and respond appropriately even in the event of a disaster.

[0447] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0448] Step 1:

[0449] The server collects disaster information from the Internet.

[0450] Specifically, the server uses Python requests and the BeautifulSoup library to retrieve disaster information such as earthquakes, floods, evacuation shelter locations, first aid instructions, etc. from various websites and APIs. This collected information is then stored in a PostgreSQL database.

[0451] Input: Websites and APIs that provide disaster information

[0452] Output: Disaster information stored in a PostgreSQL database

[0453] Step 2:

[0454] The server trains a generative AI model based on the collected disaster information.

[0455] Specifically, the server uses TensorFlow to train a generative AI model using collected disaster information. This model is then able to generate appropriate answers to user questions. Training data is retrieved from a database and the model's parameters are adjusted.

[0456] Input: Disaster information stored in a PostgreSQL database

[0457] Output: A trained generative AI model

[0458] Step 3:

[0459] The server periodically distributes the trained generative AI model to terminals and robots.

[0460] Specifically, the server uses the Paramiko library to transfer the trained model to each terminal and robot via SSH, with this distribution occurring once a day.

[0461] Input: A trained generative AI model

[0462] Output: Generative AI model distributed to each device and robot

[0463] Step 4:

[0464] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[0465] When a user types a question into the device, the device uses a locally stored generative AI model to generate an answer to that question, such as "Where is the nearest evacuation shelter?", providing the location of the nearest evacuation shelter.

[0466] Input: User question (e.g., "Where is the nearest evacuation shelter?")

[0467] Output: Answer to question (e.g., location of nearest evacuation shelter)

[0468] Step 5:

[0469] The robot performs speech recognition and speech synthesis, providing appropriate disaster information in voice in response to the user's questions.

[0470] Specifically, the robot uses the SpeechRecognition library to convert the user's speech into text, and then inputs that text into a generative AI model. The generated answer is then synthesized using the pyttsx3 library and provided to the user via voice. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot will provide specific instructions via voice.

[0471] Input: User's spoken question (e.g., "What is the emergency evacuation protocol in case of a fire?")

[0472] Output: Answers to questions via voice (e.g., emergency evacuation protocol instructions)

[0473] 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.

[0474] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information to users that takes their emotions into consideration.

[0475] The server first collects disaster information from the internet. It connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[0476] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[0477] Furthermore, this invention incorporates an emotion engine. The emotion engine analyzes the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine recognizes this state and provides information that will reassure the user. In this way, it becomes possible to provide information that takes into consideration the user's emotions.

[0478] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[0479] With this system, even if the communications infrastructure goes down during a disaster, users can obtain the necessary disaster information offline and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[0480] The processing flow will be explained below.

[0481] Step 1:

[0482] The server collects disaster information from the internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. This information is converted into an appropriate format and stored in a database.

[0483] Step 2:

[0484] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0485] Step 3:

[0486] The server trains the generative AI model, which learns the ability to generate appropriate answers to user questions from the collected disaster information. Once trained, the AI ​​model is saved.

[0487] Step 4:

[0488] The server periodically distributes trained generative AI models to devices. For example, the server obtains a list of each user's devices and sends it to the device to distribute the latest model once a day.

[0489] Step 5:

[0490] The device stores the generative AI model received from the server locally, allowing the device to use the generative AI model even without an internet connection.

[0491] Step 6:

[0492] The user inputs a question requiring disaster information into the terminal. For example, questions such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0493] Step 7:

[0494] The device's emotion engine recognizes the user's emotions by analyzing the user's tone of voice and facial expressions to determine whether the user is feeling anxious or impatient.

[0495] Step 8:

[0496] The device uses a locally stored generative AI model to generate appropriate answers to user questions. The device then combines the information obtained from the generative AI model with the judgment results of the emotion engine to provide the user with the most appropriate information.

[0497] Step 9:

[0498] The device displays the generated answer to the user. For example, it displays specific information such as "The nearest evacuation shelter is X Park" or first aid instructions such as "First, take a deep breath. Then, stop the bleeding."

[0499] Step 10:

[0500] The system takes prompt and appropriate action based on the information provided by the user. In addition, the system provides emotional care through an emotion engine, allowing the user to respond calmly.

[0501] This series of steps allows users to obtain the necessary disaster information in an offline environment even if communication infrastructure goes down during a disaster, and provides support based on emotion recognition, enabling quick and accurate decisions.

[0502] Example 2

[0503] 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."

[0504] During disasters, communication infrastructure is prone to downtime, making it difficult to obtain necessary disaster information in a timely manner. It is also important to provide information while responding to the user's emotional state, but no systems currently exist with this functionality. Therefore, there is a need for a system that can provide disaster information offline and that can provide information that is adapted to the user's emotional state.

[0505] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting disaster information from the Internet, a means for training a generative AI model based on the collected disaster information, and a means for periodically distributing the trained generative AI model to terminals. This makes it possible to provide disaster information in an offline environment even if the communication infrastructure is down, and further to provide information adapted to the user's emotional state.

[0506] A "server" is a device that collects, stores, processes, and provides information via the Internet to other devices that require that information.

[0507] "Disaster information" refers to data such as the latest situation, countermeasures, location information of evacuation centers, and first aid methods related to natural disasters such as earthquakes, floods, and typhoons.

[0508] The "Internet" is a global communications network that interconnects computer networks around the world and enables the exchange of information.

[0509] A "generative AI model" is an algorithmic model of artificial intelligence that is trained based on accumulated data and has the ability to perform specific tasks (e.g., question answering, information generation).

[0510] "Training" is the process of using collected data to optimize the parameters of an AI model and improve its performance.

[0511] "Terminal" means a telecommunications device that allows a user to input and receive information and view the processing results.

[0512] "Stored locally" means that data or programs are stored directly on a device outside the network, making them available without the need for a remote connection.

[0513] "Offline environment" refers to a state where there is no connection to the Internet or other networks.

[0514] "Users" refer to people who interact with the system and receive particular information or services.

[0515] A "question" is a user's inquiry for information, and is input into the terminal in text or voice format.

[0516] An "emotion analysis engine" is a software or hardware system that analyzes audio and / or video input to identify a user's emotional state.

[0517] "Emotional state" is information indicating the psychological state of the user, and includes specific emotions such as anxiety, fear, and relief.

[0518] "Analysis" is a method for examining collected data in detail and understanding its meaning and structure.

[0519] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. This system is composed of a server, a terminal, and an emotion analysis engine.

[0520] Server Operation

[0521] The server first connects to the APIs of government and disaster prevention agencies and collects disaster information via the internet, including the latest earthquake and flood information, evacuation shelter locations, and first aid procedures. The collected data is received in JSON format, converted into an appropriate format, and stored in a database. The server then uses this data to train a generative AI model. For training, it uses machine learning frameworks such as TensorFlow and PyTorch. Once the generative AI model is trained, it is distributed to devices periodically, for example, once a day.

[0522] Device behavior

[0523] The device locally stores the generative AI model received from the server, making it available offline even if the communications infrastructure goes down. When a user needs disaster information, they input a question into the device, such as "Where is the nearest evacuation shelter?" At this time, the device uses the locally stored generative AI model to generate and display an appropriate answer to the question. The device also uses the generative AI model to provide information on the location of evacuation shelters and first aid procedures.

[0524] User sentiment analysis

[0525] Furthermore, the present invention incorporates an emotion analysis engine. The emotion analysis engine uses a camera and microphone to analyze the user's voice input and facial expressions. Libraries such as OpenCV and DeepFace are used for the analysis. This analysis recognizes the user's emotional state, and if, for example, anxiety or fear is detected, the device will prioritize providing a reassuring message taking that state into consideration. For example, it will display a message such as, "First, take a deep breath. Then, do XX."

[0526] Specific examples

[0527] As a specific example, the server collects the latest information on evacuation shelters and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. If a user types "My child is injured, what should I do?" into the device, the device will use the generative AI model to provide appropriate first aid methods. It also uses an emotion analysis engine to recognize the user's sense of urgency and displays a message to calm them down. An example of a prompt sentence is "Please enter the user's question. For example, 'Where is the nearest evacuation shelter?'"

[0528] As described above, the present invention enables users to obtain necessary disaster information offline even if the communication infrastructure goes down during a disaster, and enables appropriate responses based on emotion recognition.

[0529] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0530] Step 1:

[0531] The server collects disaster information. The input is real-time disaster information received from the APIs of government and disaster prevention organizations. Specifically, the server uses the Python requests library to send HTTP requests and obtain data in JSON format. The output is the obtained disaster information in JSON format.

[0532] Step 2:

[0533] The server converts and stores the disaster information it acquires. The input is the JSON-formatted data acquired in step 1. Specifically, the server uses the Python pandas library to convert the data into an appropriate format and saves it in a database (e.g., MySQL or MongoDB). The output is the disaster information stored in the database.

[0534] Step 3:

[0535] The server trains a generative AI model based on the collected data. The input is disaster information stored in a database. Specifically, the server uses TensorFlow and PyTorch to build and train the generative AI model. The output is the fully trained generative AI model.

[0536] Step 4:

[0537] The server distributes the trained generative AI model to the terminal. The input is the generative AI model obtained in step 3. Specifically, the server uses the HTTP protocol to send the file of the generative AI model to the terminal. The output is the file of the generative AI model distributed to the terminal.

[0538] Step 5:

[0539] The device stores the generated AI model it receives locally. The input is the generated AI model file received from the server. Specifically, the device uses Python's requests library to download the generated AI model file and uses the os module to save it in a local directory. The output is the locally stored generated AI model.

[0540] Step 6:

[0541] The user inputs a question into the terminal. The input is the user's textual question (e.g., "Where is the nearest evacuation shelter?"). In concrete terms, the user inputs the question in text form through the terminal's interface. The output is the question input into the terminal.

[0542] Step 7:

[0543] The device generates an answer using a generative AI model. The input is the user's question and a locally stored generative AI model. Specifically, the device uses NLP technology to analyze the meaning of the question and generates an appropriate answer using the generative AI model. The output is the generated answer.

[0544] Step 8:

[0545] The device recognizes the user's emotional state. Input is the user's voice input and facial expression data. Specifically, the device analyzes data from the camera and microphone using an emotion analysis engine (e.g., OpenCV or DeepFace) to identify the user's emotional state. The output is the analyzed user's emotional state.

[0546] Step 9:

[0547] The device provides emotion-conscious information. The input is the answer generated in step 7 and the user's emotional state analyzed in step 8. Specifically, the device displays information by combining a message based on the user's emotional state. For example, if the device recognizes that the user is anxious, it provides an answer with a message such as "Please stay calm and take a deep breath first. Then, do XX." The output is emotion-conscious information provision.

[0548] (Application example 2)

[0549] 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."

[0550] With conventional disaster information systems, it is difficult for users to obtain the information they need when communication infrastructure goes down. Furthermore, because information is provided without consideration for users' feelings, users can become panicked. Furthermore, the provision of disaster information is limited, and information on appropriate evacuation shelters and first aid methods is often not provided promptly. This can lead to delayed responses in the event of a disaster, potentially exacerbating the damage.

[0551] 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.

[0552] In this invention, the server includes means for collecting disaster information from the Internet, means for training a generative AI model based on the disaster information, means for periodically distributing the trained generative AI model to terminals, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, means for the terminal to analyze the user's emotions using an emotion recognition engine and provide emotion-conscious information, and means for the robot to move autonomously and provide disaster information and first aid methods. This allows users to obtain necessary disaster information offline even if the communication infrastructure is down during a disaster, and the provision of emotion-conscious information prevents panic and enables quick and appropriate responses.

[0553] "Disaster information" refers to data and events related to disasters, including, for example, the location of evacuation shelters, first aid procedures, and information on earthquakes and floods.

[0554] "Means of collecting from the Internet" refers to the method by which the server obtains the necessary information in real time using various data sources and APIs on the Internet.

[0555] A "generative AI model" refers to an artificial intelligence model that generates appropriate responses to user input based on training data.

[0556] "Storing locally" means storing collected AI models and data in the device's internal storage area, making them available even in offline environments.

[0557] An "emotion recognition engine" is a system that has the ability to analyze a user's voice, facial expressions, text input, etc., and recognize and identify their emotional state (anxiety, impatience, etc.).

[0558] An "autonomous mobile robot" is a robot that can reach its destination using its own sensors and algorithms, without relying on external instructions or infrastructure.

[0559] "Offline environment" means a situation where the Internet or other network connection is unavailable and functionality must be maintained in this state.

[0560] "Location of evacuation shelters" refers to information about places where people can evacuate to ensure their safety in the event of a disaster.

[0561] "First aid methods" refers to emergency treatment procedures and measures that can be applied to injured people in disasters and emergencies.

[0562] This invention is a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information that takes emotions into consideration.

[0563] The server first collects disaster information from the internet. Specifically, it connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[0564] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[0565] Furthermore, this invention incorporates an emotion engine. The emotion engine can analyze the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine will recognize this state and provide information that will reassure the user. In this way, it becomes possible to provide information that takes the user's emotions into consideration.

[0566] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[0567] With this system, even if communication infrastructure goes down during a disaster, users can obtain necessary disaster information in an offline environment and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[0568] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0569] Step 1:

[0570] The server collects disaster information from the Internet.

[0571] Input: APIs from various data sources and government and disaster prevention agencies.

[0572] Processing: The server retrieves data from these APIs, such as the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. It converts this information into an appropriate format and stores it in a database.

[0573] Output: Organized disaster information data is stored in a database.

[0574] Step 2:

[0575] The server trains a generative AI model based on the collected disaster information.

[0576] Input: Disaster information data stored in a database.

[0577] Processing: The server uses disaster information data to train a generative AI model that is capable of generating appropriate answers to user questions. Machine learning algorithms are used for training.

[0578] Output: The trained generative AI model is saved on the server.

[0579] Step 3:

[0580] The server periodically distributes the trained generative AI model to the device.

[0581] Input: A trained generative AI model.

[0582] Processing: The server sends the trained generative AI model to the device once a day. The model is distributed via the network.

[0583] Output: A generative AI model stored locally on the device.

[0584] Step 4:

[0585] The user inputs a question requesting disaster information into the terminal.

[0586] Input: User question (e.g., "Where is the nearest shelter?").

[0587] Processing: The device receives the user's question and inputs it into a locally stored generative AI model.

[0588] Output: The answer to the question is generated.

[0589] Step 5:

[0590] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[0591] Input: User question, locally stored generative AI model.

[0592] Processing: Using the generative AI model, the device generates an appropriate answer, which is then displayed on the device screen.

[0593] Output: Disaster information provided to the user.

[0594] Step 6:

[0595] The device uses an emotion recognition engine to analyze the user's emotions.

[0596] Input: User's voice input, facial expression data.

[0597] Processing: The device uses an emotion recognition engine to analyze the user's voice and facial expressions to determine their emotional state.

[0598] Output: User's emotional state data.

[0599] Step 7:

[0600] The device provides emotionally sensitive information.

[0601] Input: User emotional state data, generated disaster information.

[0602] Processing: The emotion engine appropriately modifies the information obtained from the generative AI model based on the user's emotional state (e.g., "First, take a deep breath.") and displays it to the user.

[0603] Output: Emotionally sensitive information is provided to the user.

[0604] Step 8:

[0605] The robot moves autonomously and provides disaster information and first aid methods.

[0606] Input: Generative AI model, emotional state data, locally stored database.

[0607] Processing: The robot uses its own sensors and algorithms to move autonomously and provide the user with the necessary information. It also provides the most appropriate information depending on the user's emotional state.

[0608] Output: Disaster information and first aid instructions provided to users appropriately.

[0609] 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.

[0610] 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.

[0611] 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.

[0612] [Third embodiment]

[0613] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0614] 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.

[0615] 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).

[0616] 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.

[0617] 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.

[0618] 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).

[0619] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0620] 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.

[0621] 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.

[0622] 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.

[0623] 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.

[0624] 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."

[0625] This invention provides a system that allows users to obtain necessary disaster information even when communication infrastructure is down during a disaster. This system is composed of a server and a terminal, and collects and processes disaster information on the server, and provides the information offline on the terminal.

[0626] The server first collects disaster information from the Internet. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database. The server then trains a generative AI model based on the collected disaster information. The trained generative AI model is responsible for generating appropriate information for users. This generative AI model needs to be updated, so the server retrains the model periodically (e.g., once a day) to generate the latest model.

[0627] The latest generated generative AI model is distributed from the server to the device, which then stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down.

[0628] When a user needs disaster information, they input a question into the device. For example, if the question is "Where is the nearest evacuation shelter?", the device uses a locally stored generative AI model to generate an answer to this question. As a response, the device displays the location information of the nearest evacuation shelter. In addition, in response to the question "Tell me how to provide first aid," it displays appropriate first aid instructions.

[0629] As a specific example, a server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes the trained model to devices. Even when the device is offline, when a user asks, "Where are the evacuation shelters?", the model is used to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0630] In this way, the present invention supports rapid and accurate decision-making during disasters by having the server collect disaster information, train a generative AI model, distribute the model to terminals, and have the terminals provide information in an offline environment.

[0631] The processing flow will be explained below.

[0632] Step 1:

[0633] The server collects disaster information from the Internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. The obtained data is converted into an appropriate format and stored in a database.

[0634] Step 2:

[0635] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0636] Step 3:

[0637] The server trains the generative AI model. Through training, the AI ​​model is able to generate appropriate answers to user questions based on disaster information. Once trained, the AI ​​model is saved.

[0638] Step 4:

[0639] The server distributes the trained generative AI model to the device. The server periodically, for example once a day, obtains a list of devices for each user and sends the latest model to the device.

[0640] Step 5:

[0641] The device receives the model from the server and stores it locally, allowing the device to use the generative AI model even without an internet connection.

[0642] Step 6:

[0643] The user inputs a question requesting disaster information, such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0644] Step 7:

[0645] The device uses a locally stored generative AI model to generate appropriate answers to user questions. Even in offline environments, the device can use the model to obtain useful information about the question.

[0646] Step 8:

[0647] The device will display the generated answer to the user, such as "The nearest evacuation shelter is X Park" or "As a first aid measure, first try to stop the bleeding."

[0648] This series of steps allows users to obtain the necessary disaster information in an offline environment and make quick and accurate decisions, even if the communications infrastructure goes down during a disaster.

[0649] Example 1

[0650] 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."

[0651] There is a need for a system that allows users to quickly and reliably obtain necessary disaster information even when communication infrastructure is down during a disaster. However, conventional systems require an internet connection, making it difficult to provide information when communication infrastructure is lost. Furthermore, disaster information changes over time, requiring real-time updates, but there is a lack of an efficient way to do this. To solve these problems, a system is needed that can provide the latest disaster information even in an offline environment.

[0652] 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.

[0653] In this invention, the server includes means for collecting disaster information from the Internet, means for storing the collected disaster information in a database, and means for training a generative AI model using the data. This allows disaster information to be collected from the Internet, stored in a database, and the generative AI model to be trained based on the data. Furthermore, the server includes means for periodically distributing the trained generative AI model to a terminal, and the terminal includes means for storing the distributed generative AI model in a local environment. When a user inputs a prompt to learn disaster information, the terminal includes means for providing the user with disaster information in an offline environment using the locally stored generative AI model. This makes it possible to provide the latest disaster information offline even if the communication infrastructure is down.

[0654] "Disaster information" refers to information about natural disasters such as earthquakes, floods, tsunamis, and typhoons, including the location of evacuation shelters, first aid procedures, and the current situation in disaster-stricken areas.

[0655] The "Internet" is a global communications infrastructure that connects computers and networks around the world to send and receive information.

[0656] A "server" is a computer system that provides information over a network and is responsible for collecting, processing, storing, and distributing data.

[0657] A "database" is a collection of data that is structured so that information can be efficiently stored, managed, and retrieved.

[0658] A "generative AI model" is an algorithm that has been trained from a specific dataset using machine learning algorithms and is capable of generating appropriate predictions or responses to new data.

[0659] "Training" is the process of using collected data to train a machine learning model and improve its prediction accuracy.

[0660] "Distributing" is the process of sending and receiving data and models from a server to a terminal.

[0661] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[0662] A "local environment" is a data storage and processing environment within the terminal itself, independent of any external network.

[0663] "Users" refer to people who use this system to obtain disaster information.

[0664] A "prompt sentence" is text that indicates the content of a question or request that a user enters into a terminal.

[0665] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0666] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server and a terminal, and disaster information is collected and processed on the server, and the information is provided offline on the terminal.

[0667] The server first collects disaster information from the Internet. This information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database such as MySQL or PostgreSQL. The server then uses the collected disaster information to train a generative AI model using a machine learning framework such as TensorFlow or PyTorch. This generative AI model is an important element because it generates appropriate information in response to user input. The generative AI model is updated periodically (e.g., once a day) and distributed from the server to devices.

[0668] In the process of distributing a generative AI model, the server uses a communication protocol (e.g., HTTP or FTP) to send the latest model to the device. The device receives this model and saves it in local storage. This allows the device to use the generative AI model in an offline environment, even if the communication infrastructure goes down.

[0669] When a user needs disaster information, they input a prompt into their device. Examples of prompts include "Where is the nearest evacuation shelter?" and "Tell me how to provide first aid." When a user inputs a prompt, the device uses a locally stored generative AI model to generate an answer to this question. For example, if the question "Where is the nearest evacuation shelter?" is input, the device will provide the location information of the nearest evacuation shelter. Also, in response to the question "Tell me how to provide first aid," it will display appropriate first aid instructions.

[0670] Consider the following scenario as a concrete example: A server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes it to devices. Then, even in an offline environment, when a user asks, "Where are the evacuation shelters?", the device uses the model to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0671] As described above, the present invention supports rapid and accurate decision-making during disasters by having a server collect disaster information, train a generative AI model based on that information, distribute the model to terminals, and have the terminals provide information to users in an offline environment.

[0672] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0673] Step 1:

[0674] The server collects disaster information from the Internet.

[0675] Input: API endpoint of disaster information service

[0676] Specific operation: The server sends requests to the APIs of the Japan Meteorological Agency and disaster information services to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc.

[0677] Output: JSON data of disaster information

[0678] Step 2:

[0679] The server stores the collected disaster information in a database.

[0680] Input: JSON data of disaster information

[0681] Specific operation: The server parses the received JSON data and generates and executes SQL queries to insert it into a MySQL or PostgreSQL database.

[0682] Output: Disaster information stored in a database

[0683] Step 3:

[0684] The server uses the data to train a generative AI model.

[0685] Input: Disaster information stored in the database

[0686] How it works: The server loads disaster information from a database and trains a generative AI model using machine learning frameworks such as TensorFlow and PyTorch. The training process includes data preprocessing, model building, and model training.

[0687] Output: A trained generative AI model

[0688] Step 4:

[0689] The server distributes the latest generative AI model to the device.

[0690] Input: A trained generative AI model

[0691] Specific operation: The server uses HTTP or FTP communication protocols to send the generated AI model to the terminal.

[0692] Output: The generated AI model sent to the device

[0693] Step 5:

[0694] The device stores the generated AI model locally.

[0695] Input: The generative AI model sent to the device

[0696] Specific operation: The device saves the generative AI model received from the server in a specific directory in local storage.

[0697] Output: A locally stored generative AI model

[0698] Step 6:

[0699] The user types a prompt into the terminal.

[0700] Input: The prompt text entered by the user (e.g., "Where is the nearest evacuation shelter?")

[0701] Specific operation: The user opens the application on the device, enters the information they want to know in the text box, and presses the send button.

[0702] Output: The prompt entered by the user

[0703] Step 7:

[0704] The device uses a generative AI model to generate answers to the user's questions.

[0705] Input: Locally stored generative AI model, user-entered prompt

[0706] How it works: The device application inputs a prompt sentence into the AI ​​model, which then generates an appropriate answer. This process includes interpreting the text, searching for relevant information, and generating an answer.

[0707] Output: The generated answer

[0708] Step 8:

[0709] The terminal displays the generated answer to the user.

[0710] Input: Generated answer (e.g., "The nearest evacuation shelter is the community center in XX town.")

[0711] Specific operation: The device displays the answer generated by the generative AI model on the application's user interface.

[0712] Output: The answer displayed to the user

[0713] (Application example 1)

[0714] 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."

[0715] During disasters, communication infrastructure often goes down, making it difficult for users to obtain the disaster information they need. Therefore, there is a need to provide a system that allows users to quickly and accurately obtain the disaster information they need even when communication infrastructure goes down. In particular, within factories, it is necessary to ensure the safety of workers by providing disaster information in a timely manner. In addition, there is a need for a system that can provide information via voice even in offline environments.

[0716] 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.

[0717] In this invention, the server includes means for collecting disaster information from an information network, means for training a generative AI model based on the collected disaster information, means for periodically distributing the trained generative AI model to a terminal, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, and means for using the generative AI model installed in the robot to perform speech recognition and speech synthesis and provide appropriate disaster information by voice in response to user questions. This makes it possible to provide necessary disaster information to users in an offline environment using the locally stored model even if the communication infrastructure is down. In addition, because the robot provides information using speech recognition and speech synthesis, workers can obtain information such as safe evacuation locations and appropriate first aid methods without using their hands.

[0718] A "server" is a computing device whose role is to collect disaster information, train a generative AI model based on that information, and periodically distribute the trained model to terminals.

[0719] "Disaster information" refers to information including the latest information on natural disasters such as earthquakes, floods, and fires, the location of evacuation shelters, first aid procedures, and so on.

[0720] "Information network" means a communications infrastructure for gathering information, including the Internet and other online networks.

[0721] A "generative AI model" is an AI model trained based on collected disaster information and used to generate appropriate information in response to user questions.

[0722] "Training" is the process of using collected data to teach an AI model so that it can perform a specific task.

[0723] "Regularly" means repeated at regular time intervals, for example, once a day.

[0724] A "terminal" is a device used by a user to obtain information, and includes smartphones, tablets, PCs, etc.

[0725] "Locally stored" means that the trained generative AI model is stored in the device's internal storage and is available even in offline environments.

[0726] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0727] "User" refers to anyone who uses the system to refer to information, including ordinary people who need disaster information and workers in factories.

[0728] A "robot" is an autonomous mechanical device that uses voice recognition and voice synthesis to provide users with necessary information during disasters.

[0729] "Speech recognition" is a technology that converts speech into text and is used to recognize user questions.

[0730] "Speech synthesis" is a technology that converts text into speech and is used to provide audible responses to the user.

[0731] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server, a terminal, and a robot.

[0732] The server collects disaster information from the Internet and stores it in a database. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. Based on the collected data, the server trains a generative AI model. The generative AI model is designed to provide appropriate information in response to user questions during disasters. The trained generative AI model is updated periodically (e.g., once a day) and distributed to terminals and robots.

[0733] The device locally stores the generative AI model received from the server. This allows disaster information to be provided to users in an offline environment even if the communication infrastructure is down. When a user needs disaster information, they can input a question into the device, and the locally stored generative AI model will be used to generate an answer to that question. For example, in response to the question, "Where is the nearest evacuation shelter?", it is possible to provide the location information of the nearest evacuation shelter.

[0734] The robot utilizes speech recognition and speech synthesis technologies to provide appropriate disaster information via voice in response to voice questions from users. A generative AI model is installed in the robot, allowing it to operate in offline environments. This allows users to obtain information by voice without using their hands. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot can provide specific instructions via voice. It is also designed for use in factories, ensuring the safety of workers by quickly providing information on safe evacuation locations and emergency response methods.

[0735] The software used includes Python for processing the collected disaster information, BeautifulSoup for scraping, TensorFlow for training the model, SpeechRecognition for speech recognition, and pyttsx3 for speech synthesis, as well as PostgreSQL as the database system for managing the stored data.

[0736] Examples of prompts include:

[0737] "Where is the nearest evacuation shelter in case of an earthquake?"

[0738] "Please tell me the emergency evacuation route in case of a fire."

[0739] This system enables users and workers to quickly obtain necessary information and respond appropriately even in the event of a disaster.

[0740] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0741] Step 1:

[0742] The server collects disaster information from the Internet.

[0743] Specifically, the server uses Python requests and the BeautifulSoup library to retrieve disaster information such as earthquakes, floods, evacuation shelter locations, first aid instructions, etc. from various websites and APIs. This collected information is then stored in a PostgreSQL database.

[0744] Input: Websites and APIs that provide disaster information

[0745] Output: Disaster information stored in a PostgreSQL database

[0746] Step 2:

[0747] The server trains a generative AI model based on the collected disaster information.

[0748] Specifically, the server uses TensorFlow to train a generative AI model using collected disaster information. This model is then able to generate appropriate answers to user questions. Training data is retrieved from a database and the model's parameters are adjusted.

[0749] Input: Disaster information stored in a PostgreSQL database

[0750] Output: A trained generative AI model

[0751] Step 3:

[0752] The server periodically distributes the trained generative AI model to terminals and robots.

[0753] Specifically, the server uses the Paramiko library to transfer the trained model to each terminal and robot via SSH, with this distribution occurring once a day.

[0754] Input: A trained generative AI model

[0755] Output: Generative AI model distributed to each device and robot

[0756] Step 4:

[0757] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[0758] When a user types a question into the device, the device uses a locally stored generative AI model to generate an answer to that question, such as "Where is the nearest evacuation shelter?", providing the location of the nearest evacuation shelter.

[0759] Input: User question (e.g., "Where is the nearest evacuation shelter?")

[0760] Output: Answer to question (e.g., location of nearest evacuation shelter)

[0761] Step 5:

[0762] The robot performs speech recognition and speech synthesis, providing appropriate disaster information in voice in response to the user's questions.

[0763] Specifically, the robot uses the SpeechRecognition library to convert the user's speech into text, and then inputs that text into a generative AI model. The generated answer is then synthesized using the pyttsx3 library and provided to the user via voice. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot will provide specific instructions via voice.

[0764] Input: User's spoken question (e.g., "What is the emergency evacuation protocol in case of a fire?")

[0765] Output: Answers to questions via voice (e.g., emergency evacuation protocol instructions)

[0766] 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.

[0767] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information to users that takes their emotions into consideration.

[0768] The server first collects disaster information from the internet. It connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[0769] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[0770] Furthermore, this invention incorporates an emotion engine. The emotion engine analyzes the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine recognizes this state and provides information that will reassure the user. In this way, it becomes possible to provide information that takes into consideration the user's emotions.

[0771] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[0772] With this system, even if the communications infrastructure goes down during a disaster, users can obtain the necessary disaster information offline and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[0773] The processing flow will be explained below.

[0774] Step 1:

[0775] The server collects disaster information from the internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. This information is converted into an appropriate format and stored in a database.

[0776] Step 2:

[0777] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0778] Step 3:

[0779] The server trains the generative AI model, which learns the ability to generate appropriate answers to user questions from the collected disaster information. Once trained, the AI ​​model is saved.

[0780] Step 4:

[0781] The server periodically distributes trained generative AI models to devices. For example, the server obtains a list of each user's devices and sends it to the device to distribute the latest model once a day.

[0782] Step 5:

[0783] The device stores the generative AI model received from the server locally, allowing the device to use the generative AI model even without an internet connection.

[0784] Step 6:

[0785] The user inputs a question requiring disaster information into the terminal. For example, questions such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0786] Step 7:

[0787] The device's emotion engine recognizes the user's emotions by analyzing the user's tone of voice and facial expressions to determine whether the user is feeling anxious or impatient.

[0788] Step 8:

[0789] The device uses a locally stored generative AI model to generate appropriate answers to user questions. The device then combines the information obtained from the generative AI model with the judgment results of the emotion engine to provide the user with the most appropriate information.

[0790] Step 9:

[0791] The device displays the generated answer to the user. For example, it displays specific information such as "The nearest evacuation shelter is X Park" or first aid instructions such as "First, take a deep breath. Then, stop the bleeding."

[0792] Step 10:

[0793] The system takes prompt and appropriate action based on the information provided by the user. In addition, the system provides emotional care through an emotion engine, allowing the user to respond calmly.

[0794] This series of steps allows users to obtain the necessary disaster information in an offline environment even if communication infrastructure goes down during a disaster, and provides support based on emotion recognition, enabling quick and accurate decisions.

[0795] Example 2

[0796] 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."

[0797] During disasters, communication infrastructure is prone to downtime, making it difficult to obtain necessary disaster information in a timely manner. It is also important to provide information while responding to the user's emotional state, but no systems currently exist with this functionality. Therefore, there is a need for a system that can provide disaster information offline and that can provide information that is adapted to the user's emotional state.

[0798] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting disaster information from the Internet, a means for training a generative AI model based on the collected disaster information, and a means for periodically distributing the trained generative AI model to terminals. This makes it possible to provide disaster information in an offline environment even if the communication infrastructure is down, and further to provide information adapted to the user's emotional state.

[0799] A "server" is a device that collects, stores, processes, and provides information via the Internet to other devices that require that information.

[0800] "Disaster information" refers to data such as the latest situation, countermeasures, location information of evacuation centers, and first aid methods related to natural disasters such as earthquakes, floods, and typhoons.

[0801] The "Internet" is a global communications network that interconnects computer networks around the world and enables the exchange of information.

[0802] A "generative AI model" is an algorithmic model of artificial intelligence that is trained based on accumulated data and has the ability to perform specific tasks (e.g., question answering, information generation).

[0803] "Training" is the process of using collected data to optimize the parameters of an AI model and improve its performance.

[0804] "Terminal" means a telecommunications device that allows a user to input and receive information and view the processing results.

[0805] "Stored locally" means that data or programs are stored directly on a device outside the network, making them available without the need for a remote connection.

[0806] "Offline environment" refers to a state where there is no connection to the Internet or other networks.

[0807] "Users" refer to people who interact with the system and receive particular information or services.

[0808] A "question" is a user's inquiry for information, and is input into the terminal in text or voice format.

[0809] An "emotion analysis engine" is a software or hardware system that analyzes audio and / or video input to identify a user's emotional state.

[0810] "Emotional state" is information indicating the psychological state of the user, and includes specific emotions such as anxiety, fear, and relief.

[0811] "Analysis" is a method for examining collected data in detail and understanding its meaning and structure.

[0812] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. This system is composed of a server, a terminal, and an emotion analysis engine.

[0813] Server Operation

[0814] The server first connects to the APIs of government and disaster prevention agencies and collects disaster information via the internet, including the latest earthquake and flood information, evacuation shelter locations, and first aid procedures. The collected data is received in JSON format, converted into an appropriate format, and stored in a database. The server then uses this data to train a generative AI model. For training, it uses machine learning frameworks such as TensorFlow and PyTorch. Once the generative AI model is trained, it is distributed to devices periodically, for example, once a day.

[0815] Device behavior

[0816] The device locally stores the generative AI model received from the server, making it available offline even if the communications infrastructure goes down. When a user needs disaster information, they input a question into the device, such as "Where is the nearest evacuation shelter?" At this time, the device uses the locally stored generative AI model to generate and display an appropriate answer to the question. The device also uses the generative AI model to provide information on the location of evacuation shelters and first aid procedures.

[0817] User sentiment analysis

[0818] Furthermore, the present invention incorporates an emotion analysis engine. The emotion analysis engine uses a camera and microphone to analyze the user's voice input and facial expressions. Libraries such as OpenCV and DeepFace are used for the analysis. This analysis recognizes the user's emotional state, and if, for example, anxiety or fear is detected, the device will prioritize providing a reassuring message taking that state into consideration. For example, it will display a message such as, "First, take a deep breath. Then, do XX."

[0819] Specific examples

[0820] As a specific example, the server collects the latest information on evacuation shelters and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. If a user types "My child is injured, what should I do?" into the device, the device will use the generative AI model to provide appropriate first aid methods. It also uses an emotion analysis engine to recognize the user's sense of urgency and displays a message to calm them down. An example of a prompt sentence is "Please enter the user's question. For example, 'Where is the nearest evacuation shelter?'"

[0821] As described above, the present invention enables users to obtain necessary disaster information offline even if the communication infrastructure goes down during a disaster, and enables appropriate responses based on emotion recognition.

[0822] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0823] Step 1:

[0824] The server collects disaster information. The input is real-time disaster information received from the APIs of government and disaster prevention organizations. Specifically, the server uses the Python requests library to send HTTP requests and obtain data in JSON format. The output is the obtained disaster information in JSON format.

[0825] Step 2:

[0826] The server converts and stores the disaster information it acquires. The input is the JSON-formatted data acquired in step 1. Specifically, the server uses the Python pandas library to convert the data into an appropriate format and saves it in a database (e.g., MySQL or MongoDB). The output is the disaster information stored in the database.

[0827] Step 3:

[0828] The server trains a generative AI model based on the collected data. The input is disaster information stored in a database. Specifically, the server uses TensorFlow and PyTorch to build and train the generative AI model. The output is the fully trained generative AI model.

[0829] Step 4:

[0830] The server distributes the trained generative AI model to the terminal. The input is the generative AI model obtained in step 3. Specifically, the server uses the HTTP protocol to send the file of the generative AI model to the terminal. The output is the file of the generative AI model distributed to the terminal.

[0831] Step 5:

[0832] The device stores the generated AI model it receives locally. The input is the generated AI model file received from the server. Specifically, the device uses Python's requests library to download the generated AI model file and uses the os module to save it in a local directory. The output is the locally stored generated AI model.

[0833] Step 6:

[0834] The user inputs a question into the terminal. The input is the user's textual question (e.g., "Where is the nearest evacuation shelter?"). In concrete terms, the user inputs the question in text form through the terminal's interface. The output is the question input into the terminal.

[0835] Step 7:

[0836] The device generates an answer using a generative AI model. The input is the user's question and a locally stored generative AI model. Specifically, the device uses NLP technology to analyze the meaning of the question and generates an appropriate answer using the generative AI model. The output is the generated answer.

[0837] Step 8:

[0838] The device recognizes the user's emotional state. Input is the user's voice input and facial expression data. Specifically, the device analyzes data from the camera and microphone using an emotion analysis engine (e.g., OpenCV or DeepFace) to identify the user's emotional state. The output is the analyzed user's emotional state.

[0839] Step 9:

[0840] The device provides emotion-conscious information. The input is the answer generated in step 7 and the user's emotional state analyzed in step 8. Specifically, the device displays information by combining a message based on the user's emotional state. For example, if the device recognizes that the user is anxious, it provides an answer with a message such as "Please stay calm and take a deep breath first. Then, do XX." The output is emotion-conscious information provision.

[0841] (Application example 2)

[0842] 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."

[0843] With conventional disaster information systems, it is difficult for users to obtain the information they need when communication infrastructure goes down. Furthermore, because information is provided without consideration for users' feelings, users can become panicked. Furthermore, the provision of disaster information is limited, and information on appropriate evacuation shelters and first aid methods is often not provided promptly. This can lead to delayed responses in the event of a disaster, potentially exacerbating the damage.

[0844] 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.

[0845] In this invention, the server includes means for collecting disaster information from the Internet, means for training a generative AI model based on the disaster information, means for periodically distributing the trained generative AI model to terminals, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, means for the terminal to analyze the user's emotions using an emotion recognition engine and provide emotion-conscious information, and means for the robot to move autonomously and provide disaster information and first aid methods. This allows users to obtain necessary disaster information offline even if the communication infrastructure is down during a disaster, and the provision of emotion-conscious information prevents panic and enables quick and appropriate responses.

[0846] "Disaster information" refers to data and events related to disasters, including, for example, the location of evacuation shelters, first aid procedures, and information on earthquakes and floods.

[0847] "Means of collecting from the Internet" refers to the method by which the server obtains the necessary information in real time using various data sources and APIs on the Internet.

[0848] A "generative AI model" refers to an artificial intelligence model that generates appropriate responses to user input based on training data.

[0849] "Storing locally" means storing collected AI models and data in the device's internal storage area, making them available even in offline environments.

[0850] An "emotion recognition engine" is a system that has the ability to analyze a user's voice, facial expressions, text input, etc., and recognize and identify their emotional state (anxiety, impatience, etc.).

[0851] An "autonomous mobile robot" is a robot that can reach its destination using its own sensors and algorithms, without relying on external instructions or infrastructure.

[0852] "Offline environment" means a situation where the Internet or other network connection is unavailable and functionality must be maintained in this state.

[0853] "Location of evacuation shelters" refers to information about places where people can evacuate to ensure their safety in the event of a disaster.

[0854] "First aid methods" refers to emergency treatment procedures and measures that can be applied to injured people in disasters and emergencies.

[0855] This invention is a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information that takes emotions into consideration.

[0856] The server first collects disaster information from the internet. Specifically, it connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[0857] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[0858] Furthermore, this invention incorporates an emotion engine. The emotion engine can analyze the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine will recognize this state and provide information that will reassure the user. In this way, it becomes possible to provide information that takes the user's emotions into consideration.

[0859] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[0860] With this system, even if communication infrastructure goes down during a disaster, users can obtain necessary disaster information in an offline environment and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[0861] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0862] Step 1:

[0863] The server collects disaster information from the Internet.

[0864] Input: APIs from various data sources and government and disaster prevention agencies.

[0865] Processing: The server retrieves data from these APIs, such as the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. It converts this information into an appropriate format and stores it in a database.

[0866] Output: Organized disaster information data is stored in a database.

[0867] Step 2:

[0868] The server trains a generative AI model based on the collected disaster information.

[0869] Input: Disaster information data stored in a database.

[0870] Processing: The server uses disaster information data to train a generative AI model that is capable of generating appropriate answers to user questions. Machine learning algorithms are used for training.

[0871] Output: The trained generative AI model is saved on the server.

[0872] Step 3:

[0873] The server periodically distributes the trained generative AI model to the device.

[0874] Input: A trained generative AI model.

[0875] Processing: The server sends the trained generative AI model to the device once a day. The model is distributed via the network.

[0876] Output: A generative AI model stored locally on the device.

[0877] Step 4:

[0878] The user inputs a question requesting disaster information into the terminal.

[0879] Input: User question (e.g., "Where is the nearest shelter?").

[0880] Processing: The device receives the user's question and inputs it into a locally stored generative AI model.

[0881] Output: The answer to the question is generated.

[0882] Step 5:

[0883] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[0884] Input: User question, locally stored generative AI model.

[0885] Processing: Using the generative AI model, the device generates an appropriate answer, which is then displayed on the device screen.

[0886] Output: Disaster information provided to the user.

[0887] Step 6:

[0888] The device uses an emotion recognition engine to analyze the user's emotions.

[0889] Input: User's voice input, facial expression data.

[0890] Processing: The device uses an emotion recognition engine to analyze the user's voice and facial expressions to determine their emotional state.

[0891] Output: User's emotional state data.

[0892] Step 7:

[0893] The device provides emotionally sensitive information.

[0894] Input: User emotional state data, generated disaster information.

[0895] Processing: The emotion engine appropriately modifies the information obtained from the generative AI model based on the user's emotional state (e.g., "First, take a deep breath.") and displays it to the user.

[0896] Output: Emotionally sensitive information is provided to the user.

[0897] Step 8:

[0898] The robot moves autonomously and provides disaster information and first aid methods.

[0899] Input: Generative AI model, emotional state data, locally stored database.

[0900] Processing: The robot uses its own sensors and algorithms to move autonomously and provide the user with the necessary information. It also provides the most appropriate information depending on the user's emotional state.

[0901] Output: Disaster information and first aid instructions provided to users appropriately.

[0902] 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.

[0903] 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.

[0904] 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.

[0905] [Fourth embodiment]

[0906] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0907] 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.

[0908] 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).

[0909] 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.

[0910] 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.

[0911] 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).

[0912] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0913] 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.

[0914] 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.

[0915] 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.

[0916] 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.

[0917] 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.

[0918] 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."

[0919] This invention provides a system that allows users to obtain necessary disaster information even when communication infrastructure is down during a disaster. This system is composed of a server and a terminal, and collects and processes disaster information on the server, and provides the information offline on the terminal.

[0920] The server first collects disaster information from the Internet. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database. The server then trains a generative AI model based on the collected disaster information. The trained generative AI model is responsible for generating appropriate information for users. This generative AI model needs to be updated, so the server retrains the model periodically (e.g., once a day) to generate the latest model.

[0921] The latest generated generative AI model is distributed from the server to the device, which then stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down.

[0922] When a user needs disaster information, they input a question into the device. For example, if the question is "Where is the nearest evacuation shelter?", the device uses a locally stored generative AI model to generate an answer to this question. As a response, the device displays the location information of the nearest evacuation shelter. In addition, in response to the question "Tell me how to provide first aid," it displays appropriate first aid instructions.

[0923] As a specific example, a server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes the trained model to devices. Even when the device is offline, when a user asks, "Where are the evacuation shelters?", the model is used to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0924] In this way, the present invention supports rapid and accurate decision-making during disasters by having the server collect disaster information, train a generative AI model, distribute the model to terminals, and have the terminals provide information in an offline environment.

[0925] The processing flow will be explained below.

[0926] Step 1:

[0927] The server collects disaster information from the Internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. The obtained data is converted into an appropriate format and stored in a database.

[0928] Step 2:

[0929] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[0930] Step 3:

[0931] The server trains the generative AI model. Through training, the AI ​​model is able to generate appropriate answers to user questions based on disaster information. Once trained, the AI ​​model is saved.

[0932] Step 4:

[0933] The server distributes the trained generative AI model to the device. The server periodically, for example once a day, obtains a list of devices for each user and sends the latest model to the device.

[0934] Step 5:

[0935] The device receives the model from the server and stores it locally, allowing the device to use the generative AI model even without an internet connection.

[0936] Step 6:

[0937] The user inputs a question requesting disaster information, such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[0938] Step 7:

[0939] The device uses a locally stored generative AI model to generate appropriate answers to user questions. Even in offline environments, the device can use the model to obtain useful information about the question.

[0940] Step 8:

[0941] The device will display the generated answer to the user, such as "The nearest evacuation shelter is X Park" or "As a first aid measure, first try to stop the bleeding."

[0942] This series of steps allows users to obtain the necessary disaster information in an offline environment and make quick and accurate decisions, even if the communications infrastructure goes down during a disaster.

[0943] Example 1

[0944] 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."

[0945] There is a need for a system that allows users to quickly and reliably obtain necessary disaster information even when communication infrastructure is down during a disaster. However, conventional systems require an internet connection, making it difficult to provide information when communication infrastructure is lost. Furthermore, disaster information changes over time, requiring real-time updates, but there is a lack of an efficient way to do this. To solve these problems, a system is needed that can provide the latest disaster information even in an offline environment.

[0946] 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.

[0947] In this invention, the server includes means for collecting disaster information from the Internet, means for storing the collected disaster information in a database, and means for training a generative AI model using the data. This allows disaster information to be collected from the Internet, stored in a database, and the generative AI model to be trained based on the data. Furthermore, the server includes means for periodically distributing the trained generative AI model to a terminal, and the terminal includes means for storing the distributed generative AI model in a local environment. When a user inputs a prompt to learn disaster information, the terminal includes means for providing the user with disaster information in an offline environment using the locally stored generative AI model. This makes it possible to provide the latest disaster information offline even if the communication infrastructure is down.

[0948] "Disaster information" refers to information about natural disasters such as earthquakes, floods, tsunamis, and typhoons, including the location of evacuation shelters, first aid procedures, and the current situation in disaster-stricken areas.

[0949] The "Internet" is a global communications infrastructure that connects computers and networks around the world to send and receive information.

[0950] A "server" is a computer system that provides information over a network and is responsible for collecting, processing, storing, and distributing data.

[0951] A "database" is a collection of data that is structured so that information can be efficiently stored, managed, and retrieved.

[0952] A "generative AI model" is an algorithm that has been trained from a specific dataset using machine learning algorithms and is capable of generating appropriate predictions or responses to new data.

[0953] "Training" is the process of using collected data to train a machine learning model and improve its prediction accuracy.

[0954] "Distributing" is the process of sending and receiving data and models from a server to a terminal.

[0955] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[0956] A "local environment" is a data storage and processing environment within the terminal itself, independent of any external network.

[0957] "Users" refer to people who use this system to obtain disaster information.

[0958] A "prompt sentence" is text that indicates the content of a question or request that a user enters into a terminal.

[0959] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[0960] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server and a terminal, and disaster information is collected and processed on the server, and the information is provided offline on the terminal.

[0961] The server first collects disaster information from the Internet. This information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. The collected disaster information is stored in a database such as MySQL or PostgreSQL. The server then uses the collected disaster information to train a generative AI model using a machine learning framework such as TensorFlow or PyTorch. This generative AI model is an important element because it generates appropriate information in response to user input. The generative AI model is updated periodically (e.g., once a day) and distributed from the server to devices.

[0962] In the process of distributing a generative AI model, the server uses a communication protocol (e.g., HTTP or FTP) to send the latest model to the device. The device receives this model and saves it in local storage. This allows the device to use the generative AI model in an offline environment, even if the communication infrastructure goes down.

[0963] When a user needs disaster information, they input a prompt into their device. Examples of prompts include "Where is the nearest evacuation shelter?" and "Tell me how to provide first aid." When a user inputs a prompt, the device uses a locally stored generative AI model to generate an answer to this question. For example, if the question "Where is the nearest evacuation shelter?" is input, the device will provide the location information of the nearest evacuation shelter. Also, in response to the question "Tell me how to provide first aid," it will display appropriate first aid instructions.

[0964] Consider the following scenario as a concrete example: A server collects the latest earthquake information and evacuation shelter information, trains a generative AI model based on that information, and distributes it to devices. Then, even in an offline environment, when a user asks, "Where are the evacuation shelters?", the device uses the model to provide location information for nearby evacuation shelters. This system allows users and rescue teams to quickly obtain the information they need, even if the communications infrastructure goes down during a disaster.

[0965] As described above, the present invention supports rapid and accurate decision-making during disasters by having a server collect disaster information, train a generative AI model based on that information, distribute the model to terminals, and have the terminals provide information to users in an offline environment.

[0966] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0967] Step 1:

[0968] The server collects disaster information from the Internet.

[0969] Input: API endpoint of disaster information service

[0970] Specific operation: The server sends requests to the APIs of the Japan Meteorological Agency and disaster information services to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc.

[0971] Output: JSON data of disaster information

[0972] Step 2:

[0973] The server stores the collected disaster information in a database.

[0974] Input: JSON data of disaster information

[0975] Specific operation: The server parses the received JSON data and generates and executes SQL queries to insert it into a MySQL or PostgreSQL database.

[0976] Output: Disaster information stored in a database

[0977] Step 3:

[0978] The server uses the data to train a generative AI model.

[0979] Input: Disaster information stored in the database

[0980] How it works: The server loads disaster information from a database and trains a generative AI model using machine learning frameworks such as TensorFlow and PyTorch. The training process includes data preprocessing, model building, and model training.

[0981] Output: A trained generative AI model

[0982] Step 4:

[0983] The server distributes the latest generative AI model to the device.

[0984] Input: A trained generative AI model

[0985] Specific operation: The server uses HTTP or FTP communication protocols to send the generated AI model to the terminal.

[0986] Output: The generated AI model sent to the device

[0987] Step 5:

[0988] The device stores the generated AI model locally.

[0989] Input: The generative AI model sent to the device

[0990] Specific operation: The device saves the generative AI model received from the server in a specific directory in local storage.

[0991] Output: A locally stored generative AI model

[0992] Step 6:

[0993] The user types a prompt into the terminal.

[0994] Input: The prompt text entered by the user (e.g., "Where is the nearest evacuation shelter?")

[0995] Specific operation: The user opens the application on the device, enters the information they want to know in the text box, and presses the send button.

[0996] Output: The prompt entered by the user

[0997] Step 7:

[0998] The device uses a generative AI model to generate answers to the user's questions.

[0999] Input: Locally stored generative AI model, user-entered prompt

[1000] How it works: The device application inputs a prompt sentence into the AI ​​model, which then generates an appropriate answer. This process includes interpreting the text, searching for relevant information, and generating an answer.

[1001] Output: The generated answer

[1002] Step 8:

[1003] The terminal displays the generated answer to the user.

[1004] Input: Generated answer (e.g., "The nearest evacuation shelter is the community center in XX town.")

[1005] Specific operation: The device displays the answer generated by the generative AI model on the application's user interface.

[1006] Output: The answer displayed to the user

[1007] (Application example 1)

[1008] 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."

[1009] During disasters, communication infrastructure often goes down, making it difficult for users to obtain the disaster information they need. Therefore, there is a need to provide a system that allows users to quickly and accurately obtain the disaster information they need even when communication infrastructure goes down. In particular, within factories, it is necessary to ensure the safety of workers by providing disaster information in a timely manner. In addition, there is a need for a system that can provide information via voice even in offline environments.

[1010] 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.

[1011] In this invention, the server includes means for collecting disaster information from an information network, means for training a generative AI model based on the collected disaster information, means for periodically distributing the trained generative AI model to a terminal, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, and means for using the generative AI model installed in the robot to perform speech recognition and speech synthesis and provide appropriate disaster information by voice in response to user questions. This makes it possible to provide necessary disaster information to users in an offline environment using the locally stored model even if the communication infrastructure is down. In addition, because the robot provides information using speech recognition and speech synthesis, workers can obtain information such as safe evacuation locations and appropriate first aid methods without using their hands.

[1012] A "server" is a computing device whose role is to collect disaster information, train a generative AI model based on that information, and periodically distribute the trained model to terminals.

[1013] "Disaster information" refers to information including the latest information on natural disasters such as earthquakes, floods, and fires, the location of evacuation shelters, first aid procedures, and so on.

[1014] "Information network" means a communications infrastructure for gathering information, including the Internet and other online networks.

[1015] A "generative AI model" is an AI model trained based on collected disaster information and used to generate appropriate information in response to user questions.

[1016] "Training" is the process of using collected data to teach an AI model so that it can perform a specific task.

[1017] "Regularly" means repeated at regular time intervals, for example, once a day.

[1018] A "terminal" is a device used by a user to obtain information, and includes smartphones, tablets, PCs, etc.

[1019] "Locally stored" means that the trained generative AI model is stored in the device's internal storage and is available even in offline environments.

[1020] "Offline environment" refers to a state where there is no connection to the Internet or other external networks.

[1021] "User" refers to anyone who uses the system to refer to information, including ordinary people who need disaster information and workers in factories.

[1022] A "robot" is an autonomous mechanical device that uses voice recognition and voice synthesis to provide users with necessary information during disasters.

[1023] "Speech recognition" is a technology that converts speech into text and is used to recognize user questions.

[1024] "Speech synthesis" is a technology that converts text into speech and is used to provide audible responses to the user.

[1025] This invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure goes down during a disaster. This system is composed of a server, a terminal, and a robot.

[1026] The server collects disaster information from the Internet and stores it in a database. This disaster information includes the latest information on earthquakes and floods, the locations of evacuation shelters, first aid procedures, and so on. Based on the collected data, the server trains a generative AI model. The generative AI model is designed to provide appropriate information in response to user questions during disasters. The trained generative AI model is updated periodically (e.g., once a day) and distributed to terminals and robots.

[1027] The device locally stores the generative AI model received from the server. This allows disaster information to be provided to users in an offline environment even if the communication infrastructure is down. When a user needs disaster information, they can input a question into the device, and the locally stored generative AI model will be used to generate an answer to that question. For example, in response to the question, "Where is the nearest evacuation shelter?", it is possible to provide the location information of the nearest evacuation shelter.

[1028] The robot utilizes speech recognition and speech synthesis technologies to provide appropriate disaster information via voice in response to voice questions from users. A generative AI model is installed in the robot, allowing it to operate in offline environments. This allows users to obtain information by voice without using their hands. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot can provide specific instructions via voice. It is also designed for use in factories, ensuring the safety of workers by quickly providing information on safe evacuation locations and emergency response methods.

[1029] The software used includes Python for processing the collected disaster information, BeautifulSoup for scraping, TensorFlow for training the model, SpeechRecognition for speech recognition, and pyttsx3 for speech synthesis, as well as PostgreSQL as the database system for managing the stored data.

[1030] Examples of prompts include:

[1031] "Where is the nearest evacuation shelter in case of an earthquake?"

[1032] "Please tell me the emergency evacuation route in case of a fire."

[1033] This system enables users and workers to quickly obtain necessary information and respond appropriately even in the event of a disaster.

[1034] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1035] Step 1:

[1036] The server collects disaster information from the Internet.

[1037] Specifically, the server uses Python requests and the BeautifulSoup library to retrieve disaster information such as earthquakes, floods, evacuation shelter locations, first aid instructions, etc. from various websites and APIs. This collected information is then stored in a PostgreSQL database.

[1038] Input: Websites and APIs that provide disaster information

[1039] Output: Disaster information stored in a PostgreSQL database

[1040] Step 2:

[1041] The server trains a generative AI model based on the collected disaster information.

[1042] Specifically, the server uses TensorFlow to train a generative AI model using collected disaster information. This model is then able to generate appropriate answers to user questions. Training data is retrieved from a database and the model's parameters are adjusted.

[1043] Input: Disaster information stored in a PostgreSQL database

[1044] Output: A trained generative AI model

[1045] Step 3:

[1046] The server periodically distributes the trained generative AI model to terminals and robots.

[1047] Specifically, the server uses the Paramiko library to transfer the trained model to each terminal and robot via SSH, with this distribution occurring once a day.

[1048] Input: A trained generative AI model

[1049] Output: Generative AI model distributed to each device and robot

[1050] Step 4:

[1051] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[1052] When a user types a question into the device, the device uses a locally stored generative AI model to generate an answer to that question, such as "Where is the nearest evacuation shelter?", providing the location of the nearest evacuation shelter.

[1053] Input: User question (e.g., "Where is the nearest evacuation shelter?")

[1054] Output: Answer to question (e.g., location of nearest evacuation shelter)

[1055] Step 5:

[1056] The robot performs speech recognition and speech synthesis, providing appropriate disaster information in voice in response to the user's questions.

[1057] Specifically, the robot uses the SpeechRecognition library to convert the user's speech into text, and then inputs that text into a generative AI model. The generated answer is then synthesized using the pyttsx3 library and provided to the user via voice. For example, in response to a question such as "What is the emergency evacuation protocol in the event of a fire?", the robot will provide specific instructions via voice.

[1058] Input: User's spoken question (e.g., "What is the emergency evacuation protocol in case of a fire?")

[1059] Output: Answers to questions via voice (e.g., emergency evacuation protocol instructions)

[1060] 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.

[1061] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information to users that takes their emotions into consideration.

[1062] The server first collects disaster information from the internet. It connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[1063] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[1064] Furthermore, this invention incorporates an emotion engine. The emotion engine analyzes the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine recognizes this state and provides information that will reassure the user. In this way, it becomes possible to provide information that takes into consideration the user's emotions.

[1065] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[1066] With this system, even if the communications infrastructure goes down during a disaster, users can obtain the necessary disaster information offline and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[1067] The processing flow will be explained below.

[1068] Step 1:

[1069] The server collects disaster information from the internet. It connects to the APIs of governments and disaster prevention organizations to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. This information is converted into an appropriate format and stored in a database.

[1070] Step 2:

[1071] The server reads the latest disaster information data from the database, formats the data, and prepares it as a training dataset for the generative AI model.

[1072] Step 3:

[1073] The server trains the generative AI model, which learns the ability to generate appropriate answers to user questions from the collected disaster information. Once trained, the AI ​​model is saved.

[1074] Step 4:

[1075] The server periodically distributes trained generative AI models to devices. For example, the server obtains a list of each user's devices and sends it to the device to distribute the latest model once a day.

[1076] Step 5:

[1077] The device stores the generative AI model received from the server locally, allowing the device to use the generative AI model even without an internet connection.

[1078] Step 6:

[1079] The user inputs a question requiring disaster information into the terminal. For example, questions such as "Where is the nearest evacuation shelter?" or "Tell me how to provide first aid."

[1080] Step 7:

[1081] The device's emotion engine recognizes the user's emotions by analyzing the user's tone of voice and facial expressions to determine whether the user is feeling anxious or impatient.

[1082] Step 8:

[1083] The device uses a locally stored generative AI model to generate appropriate answers to user questions. The device then combines the information obtained from the generative AI model with the judgment results of the emotion engine to provide the user with the most appropriate information.

[1084] Step 9:

[1085] The device displays the generated answer to the user. For example, it displays specific information such as "The nearest evacuation shelter is X Park" or first aid instructions such as "First, take a deep breath. Then, stop the bleeding."

[1086] Step 10:

[1087] The system takes prompt and appropriate action based on the information provided by the user. In addition, the system provides emotional care through an emotion engine, allowing the user to respond calmly.

[1088] This series of steps allows users to obtain the necessary disaster information in an offline environment even if communication infrastructure goes down during a disaster, and provides support based on emotion recognition, enabling quick and accurate decisions.

[1089] Example 2

[1090] 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."

[1091] During disasters, communication infrastructure is prone to downtime, making it difficult to obtain necessary disaster information in a timely manner. It is also important to provide information while responding to the user's emotional state, but no systems currently exist with this functionality. Therefore, there is a need for a system that can provide disaster information offline and that can provide information that is adapted to the user's emotional state.

[1092] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting disaster information from the Internet, a means for training a generative AI model based on the collected disaster information, and a means for periodically distributing the trained generative AI model to terminals. This makes it possible to provide disaster information in an offline environment even if the communication infrastructure is down, and further to provide information adapted to the user's emotional state.

[1093] A "server" is a device that collects, stores, processes, and provides information via the Internet to other devices that require that information.

[1094] "Disaster information" refers to data such as the latest situation, countermeasures, location information of evacuation centers, and first aid methods related to natural disasters such as earthquakes, floods, and typhoons.

[1095] The "Internet" is a global communications network that interconnects computer networks around the world and enables the exchange of information.

[1096] A "generative AI model" is an algorithmic model of artificial intelligence that is trained based on accumulated data and has the ability to perform specific tasks (e.g., question answering, information generation).

[1097] "Training" is the process of using collected data to optimize the parameters of an AI model and improve its performance.

[1098] "Terminal" means a telecommunications device that allows a user to input and receive information and view the processing results.

[1099] "Stored locally" means that data or programs are stored directly on a device outside the network, making them available without the need for a remote connection.

[1100] "Offline environment" refers to a state where there is no connection to the Internet or other networks.

[1101] "Users" refer to people who interact with the system and receive particular information or services.

[1102] A "question" is a user's inquiry for information, and is input into the terminal in text or voice format.

[1103] An "emotion analysis engine" is a software or hardware system that analyzes audio and / or video input to identify a user's emotional state.

[1104] "Emotional state" is information indicating the psychological state of the user, and includes specific emotions such as anxiety, fear, and relief.

[1105] "Analysis" is a method for examining collected data in detail and understanding its meaning and structure.

[1106] The present invention provides a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. This system is composed of a server, a terminal, and an emotion analysis engine.

[1107] Server Operation

[1108] The server first connects to the APIs of government and disaster prevention agencies and collects disaster information via the internet, including the latest earthquake and flood information, evacuation shelter locations, and first aid procedures. The collected data is received in JSON format, converted into an appropriate format, and stored in a database. The server then uses this data to train a generative AI model. For training, it uses machine learning frameworks such as TensorFlow and PyTorch. Once the generative AI model is trained, it is distributed to devices periodically, for example, once a day.

[1109] Device behavior

[1110] The device locally stores the generative AI model received from the server, making it available offline even if the communications infrastructure goes down. When a user needs disaster information, they input a question into the device, such as "Where is the nearest evacuation shelter?" At this time, the device uses the locally stored generative AI model to generate and display an appropriate answer to the question. The device also uses the generative AI model to provide information on the location of evacuation shelters and first aid procedures.

[1111] User sentiment analysis

[1112] Furthermore, the present invention incorporates an emotion analysis engine. The emotion analysis engine uses a camera and microphone to analyze the user's voice input and facial expressions. Libraries such as OpenCV and DeepFace are used for the analysis. This analysis recognizes the user's emotional state, and if, for example, anxiety or fear is detected, the device will prioritize providing a reassuring message taking that state into consideration. For example, it will display a message such as, "First, take a deep breath. Then, do XX."

[1113] Specific examples

[1114] As a specific example, the server collects the latest information on evacuation shelters and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. If a user types "My child is injured, what should I do?" into the device, the device will use the generative AI model to provide appropriate first aid methods. It also uses an emotion analysis engine to recognize the user's sense of urgency and displays a message to calm them down. An example of a prompt sentence is "Please enter the user's question. For example, 'Where is the nearest evacuation shelter?'"

[1115] As described above, the present invention enables users to obtain necessary disaster information offline even if the communication infrastructure goes down during a disaster, and enables appropriate responses based on emotion recognition.

[1116] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1117] Step 1:

[1118] The server collects disaster information. The input is real-time disaster information received from the APIs of government and disaster prevention organizations. Specifically, the server uses the Python requests library to send HTTP requests and obtain data in JSON format. The output is the obtained disaster information in JSON format.

[1119] Step 2:

[1120] The server converts and stores the disaster information it acquires. The input is the JSON-formatted data acquired in step 1. Specifically, the server uses the Python pandas library to convert the data into an appropriate format and saves it in a database (e.g., MySQL or MongoDB). The output is the disaster information stored in the database.

[1121] Step 3:

[1122] The server trains a generative AI model based on the collected data. The input is disaster information stored in a database. Specifically, the server uses TensorFlow and PyTorch to build and train the generative AI model. The output is the fully trained generative AI model.

[1123] Step 4:

[1124] The server distributes the trained generative AI model to the terminal. The input is the generative AI model obtained in step 3. Specifically, the server uses the HTTP protocol to send the file of the generative AI model to the terminal. The output is the file of the generative AI model distributed to the terminal.

[1125] Step 5:

[1126] The device stores the generated AI model it receives locally. The input is the generated AI model file received from the server. Specifically, the device uses Python's requests library to download the generated AI model file and uses the os module to save it in a local directory. The output is the locally stored generated AI model.

[1127] Step 6:

[1128] The user inputs a question into the terminal. The input is the user's textual question (e.g., "Where is the nearest evacuation shelter?"). In concrete terms, the user inputs the question in text form through the terminal's interface. The output is the question input into the terminal.

[1129] Step 7:

[1130] The device generates an answer using a generative AI model. The input is the user's question and a locally stored generative AI model. Specifically, the device uses NLP technology to analyze the meaning of the question and generates an appropriate answer using the generative AI model. The output is the generated answer.

[1131] Step 8:

[1132] The device recognizes the user's emotional state. Input is the user's voice input and facial expression data. Specifically, the device analyzes data from the camera and microphone using an emotion analysis engine (e.g., OpenCV or DeepFace) to identify the user's emotional state. The output is the analyzed user's emotional state.

[1133] Step 9:

[1134] The device provides emotion-conscious information. The input is the answer generated in step 7 and the user's emotional state analyzed in step 8. Specifically, the device displays information by combining a message based on the user's emotional state. For example, if the device recognizes that the user is anxious, it provides an answer with a message such as "Please stay calm and take a deep breath first. Then, do XX." The output is emotion-conscious information provision.

[1135] (Application example 2)

[1136] 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."

[1137] With conventional disaster information systems, it is difficult for users to obtain the information they need when communication infrastructure goes down. Furthermore, because information is provided without consideration for users' feelings, users can become panicked. Furthermore, the provision of disaster information is limited, and information on appropriate evacuation shelters and first aid methods is often not provided promptly. This can lead to delayed responses in the event of a disaster, potentially exacerbating the damage.

[1138] 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.

[1139] In this invention, the server includes means for collecting disaster information from the Internet, means for training a generative AI model based on the disaster information, means for periodically distributing the trained generative AI model to terminals, means for the terminal to provide disaster information to users in an offline environment using the generative AI model stored locally, means for the terminal to analyze the user's emotions using an emotion recognition engine and provide emotion-conscious information, and means for the robot to move autonomously and provide disaster information and first aid methods. This allows users to obtain necessary disaster information offline even if the communication infrastructure is down during a disaster, and the provision of emotion-conscious information prevents panic and enables quick and appropriate responses.

[1140] "Disaster information" refers to data and events related to disasters, including, for example, the location of evacuation shelters, first aid procedures, and information on earthquakes and floods.

[1141] "Means of collecting from the Internet" refers to the method by which the server obtains the necessary information in real time using various data sources and APIs on the Internet.

[1142] A "generative AI model" refers to an artificial intelligence model that generates appropriate responses to user input based on training data.

[1143] "Storing locally" means storing collected AI models and data in the device's internal storage area, making them available even in offline environments.

[1144] An "emotion recognition engine" is a system that has the ability to analyze a user's voice, facial expressions, text input, etc., and recognize and identify their emotional state (anxiety, impatience, etc.).

[1145] An "autonomous mobile robot" is a robot that can reach its destination using its own sensors and algorithms, without relying on external instructions or infrastructure.

[1146] "Offline environment" means a situation where the Internet or other network connection is unavailable and functionality must be maintained in this state.

[1147] "Location of evacuation shelters" refers to information about places where people can evacuate to ensure their safety in the event of a disaster.

[1148] "First aid methods" refers to emergency treatment procedures and measures that can be applied to injured people in disasters and emergencies.

[1149] This invention is a system that allows users to obtain necessary disaster information even if the communication infrastructure is down during a disaster. In particular, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide information that takes emotions into consideration.

[1150] The server first collects disaster information from the internet. Specifically, it connects to the APIs of government and disaster prevention agencies to obtain the latest earthquake information, flood information, evacuation shelter locations, first aid procedures, etc. This information is converted into an appropriate format and stored in a database. Next, the server trains a generative AI model based on the collected disaster information. This training gives the AI ​​model the ability to generate appropriate answers to user questions. Once trained, the generative AI model is saved and distributed from the server to devices periodically (e.g., once a day).

[1151] The device stores the received generative AI model locally. This allows the device to use the generative AI model in an offline environment even if the communication infrastructure goes down. When a user needs disaster information, they input a question into the device. For example, a possible question might be, "Where is the nearest evacuation shelter?" The device uses the locally stored generative AI model to generate and display an appropriate answer to the user's question.

[1152] Furthermore, this invention incorporates an emotion engine. The emotion engine can analyze the user's voice input and facial expressions to recognize the user's emotional state. For example, if the user looks anxious, the emotion engine will recognize this state and provide information that will reassure the user. In this way, it becomes possible to provide information that takes the user's emotions into consideration.

[1153] As a concrete example, the server collects the latest evacuation shelter information and first aid methods, trains a generative AI model based on that information, and distributes this model to devices. When a user types "My child is injured, what should I do?" into the device, the device uses the generative AI model to provide appropriate first aid instructions. It also uses an emotion engine to recognize the user's sense of urgency and displays a message to calm them down. For example, it might say, "First, take a deep breath. Then, do XX."

[1154] With this system, even if communication infrastructure goes down during a disaster, users can obtain necessary disaster information in an offline environment and take appropriate action based on emotion recognition. This supports quick and accurate decision-making, improving the effectiveness of disaster response.

[1155] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1156] Step 1:

[1157] The server collects disaster information from the Internet.

[1158] Input: APIs from various data sources and government and disaster prevention agencies.

[1159] Processing: The server retrieves data from these APIs, such as the latest earthquake information, flood information, evacuation shelter locations, first aid methods, etc. It converts this information into an appropriate format and stores it in a database.

[1160] Output: Organized disaster information data is stored in a database.

[1161] Step 2:

[1162] The server trains a generative AI model based on the collected disaster information.

[1163] Input: Disaster information data stored in a database.

[1164] Processing: The server uses disaster information data to train a generative AI model that is capable of generating appropriate answers to user questions. Machine learning algorithms are used for training.

[1165] Output: The trained generative AI model is saved on the server.

[1166] Step 3:

[1167] The server periodically distributes the trained generative AI model to the device.

[1168] Input: A trained generative AI model.

[1169] Processing: The server sends the trained generative AI model to the device once a day. The model is distributed via the network.

[1170] Output: A generative AI model stored locally on the device.

[1171] Step 4:

[1172] The user inputs a question requesting disaster information into the terminal.

[1173] Input: User question (e.g., "Where is the nearest shelter?").

[1174] Processing: The device receives the user's question and inputs it into a locally stored generative AI model.

[1175] Output: The answer to the question is generated.

[1176] Step 5:

[1177] The device uses a locally stored generative AI model to provide disaster information to users in offline environments.

[1178] Input: User question, locally stored generative AI model.

[1179] Processing: Using the generative AI model, the device generates an appropriate answer, which is then displayed on the device screen.

[1180] Output: Disaster information provided to the user.

[1181] Step 6:

[1182] The device uses an emotion recognition engine to analyze the user's emotions.

[1183] Input: User's voice input, facial expression data.

[1184] Processing: The device uses an emotion recognition engine to analyze the user's voice and facial expressions to determine their emotional state.

[1185] Output: User's emotional state data.

[1186] Step 7:

[1187] The device provides emotionally sensitive information.

[1188] Input: User emotional state data, generated disaster information.

[1189] Processing: The emotion engine appropriately modifies the information obtained from the generative AI model based on the user's emotional state (e.g., "First, take a deep breath.") and displays it to the user.

[1190] Output: Emotionally sensitive information is provided to the user.

[1191] Step 8:

[1192] The robot moves autonomously and provides disaster information and first aid methods.

[1193] Input: Generative AI model, emotional state data, locally stored database.

[1194] Processing: The robot uses its own sensors and algorithms to move autonomously and provide the user with the necessary information. It also provides the most appropriate information depending on the user's emotional state.

[1195] Output: Disaster information and first aid instructions provided to users appropriately.

[1196] 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.

[1197] 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.

[1198] 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.

[1199] 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.

[1200] FIG. 9 is a diagram illustrating 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 actions 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.

[1201] 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.

[1202] 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).

[1203] 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.

[1204] 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."

[1205] 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.

[1206] 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).

[1207] 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.

[1208] 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.

[1209] 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.

[1210] 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.

[1211] 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.

[1212] 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.

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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.

[1217] The following is further disclosed regarding the above embodiment.

[1218] (Claim 1)

[1219] A means for the server to collect disaster information from the Internet;

[1220] A means for the server to train a generative AI model based on the collected disaster information;

[1221] A means for the server to periodically distribute the trained generative AI model to the terminal;

[1222] A means for the device to provide disaster information to the user in an offline environment using the locally stored generative AI model; and

[1223] A system including:

[1224] (Claim 2)

[1225] The system according to claim 1, further comprising means for distributing the updated generative AI model to the terminal once a day.

[1226] (Claim 3)

[1227] 2. The system of claim 1, further comprising means for the terminal to use the generative AI model to provide information offline regarding the location of evacuation shelters and first aid methods.

[1228] "Example 1"

[1229] (Claim 1)

[1230] A means for the server to collect disaster information from the Internet;

[1231] a means for the server to store the collected disaster information in a database;

[1232] a means by which the server uses the data to train a generative AI model; and

[1233] A means for the server to periodically distribute the trained generative AI model to the terminal;

[1234] A means for storing the generated AI model distributed to the device in a local environment;

[1235] A means for a user to input a prompt sentence into a terminal to obtain disaster information;

[1236] A means for the device to provide disaster information to the user in an offline environment using the locally stored generative AI model; and

[1237] A system including:

[1238] (Claim 2)

[1239] The system of claim 1, further comprising means for distributing the updated generative AI model to the terminal once a day.

[1240] (Claim 3)

[1241] 2. The system of claim 1, further comprising means for the terminal to use the generative AI model to provide information offline regarding the location of evacuation shelters and first aid methods.

[1242] "Application Example 1"

[1243] (Claim 1)

[1244] A means for the server to collect disaster information from an information network;

[1245] A means for the server to train a generative AI model based on the collected disaster information;

[1246] A means for the server to periodically distribute the trained generative AI model to the terminal;

[1247] A means for the device to provide disaster information to the user in an offline environment using the locally stored generative AI model; and

[1248] A means to use a generative AI model installed in the robot to perform speech recognition and speech synthesis, and provide appropriate disaster information in voice in response to user questions;

[1249] A system including:

[1250] (Claim 2)

[1251] The system according to claim 1, further comprising means for distributing the updated generative AI model to the terminal once a day.

[1252] (Claim 3)

[1253] 2. The system of claim 1, further comprising means for the terminal to use the generative AI model to provide information offline regarding the location of evacuation shelters and first aid methods.

[1254] "Example 2: Combining Emotion Engines"

[1255] (Claim 1)

[1256] A means for the server to collect disaster information from the Internet;

[1257] A means for the server to train a generative AI model based on the collected disaster information;

[1258] A means for the server to periodically distribute the trained generative AI model to the terminal;

[1259] A means for the device to provide disaster information to the user in an offline environment using the locally stored generative AI model; and

[1260] a means for the device to receive a user's question and generate an answer to the question using a generative AI model;

[1261] a means for the terminal to analyze the emotional state of the user using an emotion analysis engine and provide information based on the analysis result;

[1262] A system including:

[1263] (Claim 2)

[1264] The system of claim 1, further comprising means for the server to periodically distribute updated generative AI models to the terminals.

[1265] (Claim 3)

[1266] 2. The system of claim 1, further comprising means for the terminal to use the generative AI model to provide information offline regarding the location of evacuation shelters and first aid methods.

[1267] "Application example 2 when combining emotion engines"

[1268] The original patent claims were rewritten based on the newly conceived invention.

[1269] (Claim 1)

[1270] A means for the server to collect disaster information from the Internet;

[1271] A means for the server to train a generative AI model based on the collected disaster information;

[1272] A means for the server to periodically distribute the trained generative AI model to the terminal;

[1273] A means for the device to provide disaster information to users in offline environments using the locally stored generative AI model; and

[1274] A means for the device to analyze the user's emotions using an emotion recognition engine and provide emotion-sensitive information;

[1275] A means for the robot to move autonomously and provide disaster information and first aid methods;

[1276] A system including:

[1277] (Claim 2)

[1278] The system according to claim 1, further comprising means for distributing the updated generative AI model to the terminal once a day.

[1279] (Claim 3)

[1280] 2. The system of claim 1, further comprising means for the terminal to use the generative AI model to provide information offline regarding the location of evacuation shelters and first aid methods. [Explanation of symbols]

[1281] 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 for the server to collect disaster information from the Internet; A means for the server to train a generative AI model based on the collected disaster information; A means for the server to periodically distribute the trained generative AI model to the terminal; A means for the device to provide disaster information to the user in an offline environment using the locally stored generative AI model; and A system including:

2. The system according to claim 1, further comprising means for distributing the updated generative AI model to the terminal once a day.

3. The system of claim 1, further comprising means for the terminal to use the generative AI model to provide information offline regarding the location of evacuation shelters and first aid methods.

Images