System

Abstract

To provide a system for quickly coping with a real-time traffic situation and a sudden obstacle which a driver faces.SOLUTION: A data collection means for collecting video data and environmental data in real time from an in-vehicle camera and a sensor, a data analysis means for analyzing the collected data using a machine learning model and extracting recognition and position information of pedestrians, bicycles, motorcycles, and the like, an analysis result notification means for notifying a driver of the analyzed information by voice and display, a voice command recognition means for recognizing a voice command of the driver and analyzing and converting the voice command into text, a route calculation means for calculating an optimal route based on traffic congestion information and road condition data, and a navigation means for guiding the driver with the calculated optimal route information by voice and display.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] Preventing traffic accidents and supporting safe and comfortable driving are extremely important issues in modern transportation systems. However, many current driver assistance systems do not fully utilize the individual driver's driving style or actual road environment information, and more effective hazard prediction and notification are required. Furthermore, it is necessary to develop systems that can quickly respond to real-time traffic conditions and unexpected obstacles that drivers face. [Means for solving the problem]

[0005] The present invention provides a system including a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. The data collection means collects video data and environmental data in real time from an on-board camera and sensors. The data analysis means analyzes the collected data using a machine learning model to identify pedestrians, bicycles, motorcycles, etc. and extract their location information. The analysis result notification means notifies the driver of the analyzed information by voice and on a display. The voice command recognition means recognizes the driver's voice commands, analyzes the voice commands, and converts them into text. The route calculation means calculates the optimal route based on congestion information and road condition data. The navigation means provides the calculated optimal route information to the driver by voice and on a display. This enables accurate hazard prediction and notification in real time based on the driver's individual driving style and actual road environment information, supporting safe and comfortable driving.

[0006] The "data collection means" is a mechanism that collects video data and environmental data in real time using on-board cameras and sensors.

[0007] The "data analysis means" is a mechanism that analyzes collected data using a machine learning model to recognize pedestrians, bicycles, motorbikes, etc. and extract their location information.

[0008] The "analysis result notification means" is a mechanism that notifies the driver of the analyzed information by voice and on a display.

[0009] The "voice command recognition means" is a mechanism that recognizes the driver's voice commands, analyzes the voice commands, and converts them into text.

[0010] The "route calculation means" is a mechanism that calculates the optimum route based on traffic congestion information and road condition data.

[0011] The "navigation means" is a mechanism that provides the driver with calculated optimum route information by voice and display.

[0012] An "in-vehicle camera" is a camera installed in a vehicle, which is a device that captures images of the front and surrounding area.

[0013] A "sensor" is a device that acquires location information and obstacle information in real time, such as LiDAR, radar, and GPS.

[0014] A "machine learning model" is an algorithm that recognizes patterns based on large amounts of data and makes future predictions and classifications.

[0015] "Traffic congestion information" is information about traffic congestion conditions obtained in real time.

[0016] "Road condition data" is information relating to road conditions, traffic restrictions, construction status, and the like.

[0017] "Driving style" refers to the characteristics of an individual driver, referring to their driving habits and tendencies. [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] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system are described below.

[0040] System Overview

[0041] The DriveGuard AI system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[0042] Data collection and analysis

[0043] Data collection methods

[0044] Terminal (in-vehicle device)

[0045] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[0046] The acquired data is sent to a server via the edge computing unit.

[0047] Data Analysis Methods

[0048] server

[0049] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[0050] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[0051] Notices and Instructions

[0052] Analysis result notification means

[0053] server

[0054] Based on the analysis results, risk information is generated and alerts and advice are created for drivers.

[0055] This information is sent to the terminal.

[0056] Terminal (in-vehicle device)

[0057] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[0058] Voice commands and route guidance

[0059] Voice command recognition means

[0060] Terminal (in-vehicle device)

[0061] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0062] Route calculation method

[0063] server

[0064] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[0065] Navigation methods

[0066] Terminal (in-vehicle device)

[0067] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[0068] Specific examples

[0069] Example 1: Danger notification

[0070] Terminal (in-vehicle device)

[0071] The onboard camera detects pedestrians ahead and sends that information to a server.

[0072] server

[0073] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[0074] Terminal (in-vehicle device)

[0075] A voice notification will be played saying "There is a pedestrian ahead. Be careful." and a warning will also be displayed on the display.

[0076] Example 2: Traffic jam avoidance route guide

[0077] server

[0078] Calculates the optimal detour route based on the latest traffic congestion information.

[0079] Terminal (in-vehicle device)

[0080] It receives new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[0081] In this way, the DriveGuard AI system predicts hazards in real time and supports safe driving by collecting, analyzing, notifying, and processing the driver's voice commands.

[0082] The processing flow will be explained below.

[0083] Step 1:

[0084] The terminal (in-vehicle device) collects video data and environmental data in real time using on-board cameras and sensors. The on-board cameras capture images of the front and surrounding area, and the sensors acquire information from LiDAR, radar, GPS, etc.

[0085] Step 2:

[0086] The terminal (in-vehicle device) sends the collected data to a server via the edge computing unit, where the data is transmitted quickly, encrypted, and securely.

[0087] Step 3:

[0088] The server receives the data sent from the terminal and pre-processes it through noise filtering and format normalization, thereby improving the quality of the data.

[0089] Step 4:

[0090] The server analyzes the data using sophisticated image processing algorithms and machine learning models to recognize objects such as pedestrians, cyclists, and motorbikes, and calculates the position, speed, and direction of each object.

[0091] Step 5:

[0092] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[0093] Step 6:

[0094] The server sends the generated alert information and safe driving advice to the terminal, so that the driver can be notified in a timely manner.

[0095] Step 7:

[0096] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it may play a voice message saying, "There is a pedestrian ahead. Please be careful."

[0097] Step 8:

[0098] The user (driver) inputs voice commands through a microphone, such as setting a destination or requesting a new route.

[0099] Step 9:

[0100] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[0101] Step 10:

[0102] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion information and road condition data, and the calculation results are sent back to the terminal.

[0103] Step 11:

[0104] The terminal (in-vehicle device) receives the optimal route information from the server and provides navigation to the driver by voice and display, for example, "Turn right at the next intersection."

[0105] Example 1

[0106] 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."

[0107] Conventional driving assistance systems have problems in that they do not adequately provide drivers with appropriate information or provide navigation based on voice commands. Furthermore, because there is no consistent processing from data collection to analysis, notification, and navigation, it is difficult to provide sufficient functionality to support safe driving. To solve this problem, there is a need for real-time data collection and analysis, appropriate hazard notification, and rapid navigation based on voice commands.

[0108] 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.

[0109] In this invention, the server includes a data preprocessing means, a data analysis means, and an analysis result notification means, which allows for quick and accurate preprocessing and analysis of data collected in real time, and provides appropriate alerts and navigation instructions to the driver.

[0110] "Data collection means" means for collecting video and environmental data in real time using digital cameras and sensing devices.

[0111] "Data pre-processing means" refers to means for performing noise filtering and format normalization on collected data.

[0112] The "data analysis means" is a means for analyzing collected and preprocessed data using a machine learning model to extract information about objects and their locations.

[0113] The "analysis result notification means" is a means for generating danger information based on the analysis results and providing alerts and advice to the driver.

[0114] The "voice command recognition means" is a means for converting a user's voice command into text and performing processing in accordance with the content of the text.

[0115] The "route calculation means" is a means for calculating the optimum route based on the latest traffic information.

[0116] "Navigation means" refers to a means for providing audio and visual navigation to the driver based on the calculated optimum route information.

[0117] The present invention relates to a system for supporting safe and comfortable driving for drivers, and this system includes a data collection means, a data preprocessing means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system will be described below.

[0118] System Overview

[0119] This system collects data using cameras and various sensors mounted on the vehicle, preprocesses and analyzes the data on a server, and provides appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[0120] Data Acquisition and Transmission

[0121] Terminal (in-vehicle device)

[0122] The device collects data using a digital camera and various detection devices (LiDAR, radar, GPS, etc.), thereby obtaining real-time image data of the front and surrounding area and environmental data. The acquired data is sent to a server via an edge computing unit.

[0123] Data preprocessing and analysis

[0124] server

[0125] The server receives video data and environmental data sent from the device. First, noise filtering and format normalization are performed by the data preprocessing means. Next, the collected and preprocessed data is analyzed by the data analysis means using a machine learning model to extract objects and their location information. For example, pedestrians, bicycles, motorbikes, etc. are recognized, and their speed and direction of movement are estimated.

[0126] Notices and Instructions

[0127] server

[0128] The server generates danger information based on the analysis results and includes an analysis result notification means for providing alerts and advice to the driver. For example, a warning message such as "There is a pedestrian ahead. Please slow down" is generated and sent to the terminal.

[0129] Terminal (in-vehicle device)

[0130] The device receives the danger information and advice sent from the server, and displays it on the display and notifies the driver by voice, for example, "There is a pedestrian ahead. Be careful."

[0131] Voice commands and route guidance

[0132] Terminal (in-vehicle device)

[0133] The device receives the user's voice commands through a microphone, analyzes them with a voice recognition engine, converts the voice commands into text, and processes them accordingly.

[0134] server

[0135] The server includes a route calculation means for calculating the optimum route based on voice commands, and calculates the optimum route based on traffic congestion information and road condition data, and transmits the calculated route to the terminal.

[0136] Terminal (in-vehicle device)

[0137] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[0138] Specific examples

[0139] Example 1: Danger notification

[0140] The terminal (on-board device) detects pedestrians ahead using the on-board camera and sends the data to the server.

[0141] The server uses data analysis means to recognize the pedestrian's location and direction of travel, and if it determines that there is danger, it generates an alert message and sends it to the terminal.

[0142] The terminal (in-vehicle device) will play a voice notification saying, "There is a pedestrian ahead. Please be careful," and will also display a warning on the display.

[0143] Example 2: Traffic jam avoidance route guide

[0144] The server calculates the optimal detour route based on the latest traffic congestion information.

[0145] The terminal (in-vehicle device) receives the new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[0146] Prompt Sentence Examples

[0147] Prompt 1: "When an in-car camera detects a pedestrian ahead, how is the data processed?"

[0148] Prompt 2: "Describe the process by which the optimal route is calculated and how it is communicated to the driver."

[0149] As described above, the present invention is a system that supports safe driving by realizing consistent processing from data acquisition, preprocessing, analysis, notification, and route guidance in response to voice commands.

[0150] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0151] Step 1:

[0152] Acquiring and Sending Data

[0153] The terminal (on-board device) uses a digital camera and sensors (LiDAR, radar, GPS, etc.) to acquire front and surrounding image data and environmental data in real time. Specifically, the on-board camera captures images of the front, and the GPS acquires current location information. This data is temporarily stored through the edge computing unit and sent to a server. The input is camera and sensor data, and the output is data sent to the server.

[0154] Step 2:

[0155] Data Preprocessing

[0156] The server receives data sent from the terminal. Data preprocessing means performs noise filtering and format normalization. Specific operations include removing outliers and missing values, and adjusting the resolution of image data. The input is raw data received from the terminal, and the output is preprocessed data.

[0157] Step 3:

[0158] Data analysis

[0159] The server analyzes the preprocessed data using data analysis means. Here, a machine learning model is used to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information, speed, direction of travel, etc. Specifically, an image recognition algorithm detects the objects and calculates their coordinate data. The input is the preprocessed data, and the output is the analysis results (object type, position, speed, etc.).

[0160] Step 4:

[0161] Hazard information generation and notification

[0162] The server generates danger information based on the analysis results. The analysis result notification means creates a message to notify the driver of the generated danger information and sends it to the terminal. Specifically, a warning message such as "There is a pedestrian ahead. Please slow down" is generated. The input is the analysis results, and the output is the notification message.

[0163] The terminal (in-vehicle device) receives danger information and advice sent from the server, displays it on the display, and notifies the driver by voice. The input is the notification message received from the server, and the output is warning information by voice and display. Specifically, the terminal displays a warning message on the in-vehicle display and plays a voice notification from the speaker.

[0164] Step 5:

[0165] Receiving and processing voice commands

[0166] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them using a voice recognition engine. The voice commands are converted into text and processing is carried out based on the content. The input is the driver's voice command and the output is the textual command content. Specifically, the voice command "Please tell me the best route" is analyzed.

[0167] Step 6:

[0168] Calculating the best route

[0169] The server receives data for calculating the optimal route based on the voice command. The route calculation means calculates the optimal route based on traffic congestion information and road condition data. The input is the voice command and the latest traffic information, and the output is the calculated optimal route information. Specifically, the system uses a traffic information API to obtain real-time traffic congestion information and calculates the optimal route using an algorithm.

[0170] Step 7:

[0171] Providing navigation

[0172] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information received from the server, and the output is navigation instructions to the driver. Specific operations include voice guidance such as "Turn left at the next intersection" and a map display.

[0173] (Application example 1)

[0174] 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."

[0175] Current autonomous vehicles are required to collect and analyze environmental data in real time and make appropriate decisions to ensure safe driving. However, existing systems lack effective notification methods for the driver and the autonomous driving system, preventing accident risk reduction and efficient route guidance. In particular, the limited use of prompts using voice command recognition and generative AI models poses a challenge for further improving driver convenience and safety.

[0176] 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.

[0177] In this invention, the server includes a data analysis means, a prompt generation means, and a notification means using a generative AI model. This allows the server to analyze environmental data in real time and provide appropriate alerts and route guidance prompts to the driver and the automated driving system using the generative AI model. This reduces the risk of traffic accidents and provides efficient driving assistance.

[0178] The "data collection means" is a means for collecting video data and environmental data in real time from on-board cameras and sensors.

[0179] The "data analysis means" is a means of analyzing collected data using a machine learning model to extract recognition and location information for pedestrians, bicycles, motorbikes, etc.

[0180] The "analysis result notification means" is a means for notifying the driver or the automated driving system of the analysis results obtained by the data analysis means.

[0181] The "voice command recognition means" is a means for recognizing a voice command from the driver and analyzing the content of the command.

[0182] The "route calculation means" is a means for calculating the optimum route based on information on the current position and destination, as well as traffic congestion information and road condition data.

[0183] "Navigation means" refers to a means for providing voice and display navigation to the driver or the automated driving system based on the calculated optimal route information.

[0184] The "prompt generating means" is a means for generating an appropriate instruction or alert prompt in response to an input voice command or a recognized situation.

[0185] The "prompt processing means" is a means for performing processing to appropriately notify the driver of the generated prompt.

[0186] "Notification means using a generative AI model" refers to a means of generating appropriate notification content from data analyzed using a generative AI model and notifying the driver or automated driving system in real time.

[0187] This invention aims to realize a safe driving support system for autonomous vehicles called "AutoNav AI." This system includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, a prompt generation means, a prompt processing means, and a notification means using a generative AI model.

[0188] System Overview

[0189] The data collection method uses on-board cameras and sensors to collect video data and environmental data in real time. Specific hardware used includes on-board cameras such as the Logitech C920, as well as LiDAR, radar, and GPS sensors. This data is sent to a server via an edge computing unit.

[0190] The data analysis means receives the video data and environmental data collected on the server and analyzes them using machine learning models (e.g., OpenCV and YOLO). The data undergoes pre-processing such as noise filtering and format normalization, and then uses analysis algorithms to recognize objects such as pedestrians, bicycles, and motorbikes and extract their location information.

[0191] The analysis result notification means generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. These notifications are sent to the terminal in real time via the edge computing unit and are displayed as voice and on a display.

[0192] The voice command recognition means receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine (e.g., Google's SpeechRecognition API). The recognized voice commands are converted into text and processed accordingly.

[0193] The route calculation means calculates the optimal route based on the current location and destination information, as well as traffic congestion and road condition data. Specifically, route calculation is performed using Google Maps API, etc. The calculated route information is sent from the server to the terminal.

[0194] The navigation means provides navigation to the driver or the automated driving system by voice and display based on the optimal route information received by the terminal from the server, for example, by providing guidance such as "Turn left at the next intersection."

[0195] The prompt generator generates appropriate instruction or alert prompts based on the input voice command and the recognized situation, using a generative AI model.

[0196] The prompt processing means performs processing to appropriately notify the driver of the generated prompt, thereby providing instructions and alerts to the driver at the appropriate time.

[0197] The notification method using the generative AI model generates appropriate notification content from data analyzed using the generative AI model and notifies the driver or autonomous driving system in real time.

[0198] Specific examples

[0199] Example 1: Traffic jam avoidance route guide

[0200] The server calculates the optimal detour route based on the latest traffic congestion information. The device receives the new route information and provides voice guidance such as, "Turn right at the next intersection to avoid traffic congestion." This supports efficient driving.

[0201] Example 2: Voice command recognition and prompt generation

[0202] The driver issues a voice command saying, "Calculate a safe and efficient route based on the latest traffic congestion information. End point: Shinjuku Station." In response, the server calculates the optimal route and generates a prompt saying, "The optimal route from your current location to Shinjuku Station is to turn right at the next intersection and go straight at the next traffic light. The route takes approximately 30 minutes. There is an accident warning on the way, so please be careful."

[0203] Thus, the present invention provides specific embodiments for significantly improving the safety and efficiency of autonomous vehicles.

[0204] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0205] Step 1:

[0206] Data collection

[0207] The onboard camera and various sensors installed in the user's vehicle collect video data and environmental data in real time. Specifically, the camera captures the surrounding image, and the LiDAR, radar, and GPS sensors acquire various environmental data. This data is sent to a server via an edge computing unit.

[0208] Input: Video data and environmental data from on-board cameras and sensors

[0209] Output: Data sent to the server through the edge computing unit

[0210] Step 2:

[0211] Data analysis

[0212] The server receives the collected video and environmental data and performs preprocessing: the data undergoes noise filtering and format normalization, and is then analyzed using machine learning models (OpenCV and YOLO, for example) to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information.

[0213] Input: Video data and environmental data collected in step 1

[0214] Output: Objects and their locations

[0215] Step 3:

[0216] Notification of analysis results

[0217] The server generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. The generated notifications are sent to the terminal via the edge computing unit and displayed by voice and on the display. For example, if there is a pedestrian ahead, the notification will say, "There is a pedestrian ahead. Please be careful."

[0218] Input: Analysis results obtained by data analysis tools

[0219] Output: Alerts and advice sent to the device

[0220] Step 4:

[0221] Voice Command Recognition

[0222] Voice commands issued by users (e.g., "End point: Shinjuku Station") are collected through microphones in the vehicle and converted into text by a speech recognition engine (e.g., Google's SpeechRecognition API). This text data is sent to a server and analyzed as a voice command.

[0223] Input: A voice command spoken by the user

[0224] Output: Text data sent to the server

[0225] Step 5:

[0226] Route calculation

[0227] The server calculates the optimal route based on the analyzed voice commands, current location information, traffic information, and road condition data. This calculation uses route calculation services such as Google Maps API. The calculated route information is sent from the server to the device.

[0228] Input: Voice commands, current location, traffic information, road condition data

[0229] Output: Calculated optimal route information

[0230] Step 6:

[0231] Navigation

[0232] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[0233] Input: Optimal route information received from the server

[0234] Output: Voice and display navigation instructions

[0235] Step 7:

[0236] Prompt Generation

[0237] The server generates appropriate prompts, such as instructions or alerts, based on the input voice command and the recognized situation. A generative AI model is used to generate these prompts. A specific example of a prompt is, "Turn right at the next intersection to avoid traffic jams."

[0238] Input: Voice commands, recognized situations

[0239] Output: Generated prompt

[0240] Step 8:

[0241] Prompt Processing

[0242] The server then processes the generated prompt to appropriately notify the driver. The notification is given via voice and display. For example, the driver may be notified, "Turn right at the next intersection to avoid traffic jams."

[0243] Input: Generated prompt

[0244] Output: Audio and visual notifications

[0245] 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.

[0246] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion engine. Specific embodiments for implementing this system are described below.

[0247] System Overview

[0248] The "DriveGuard AI" system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate. Furthermore, it combines an emotion engine that analyzes the driver's emotional state from their voice and facial expressions to provide appropriate support according to the driver's emotional state.

[0249] Data collection and analysis

[0250] Data collection methods

[0251] Terminal (in-vehicle device)

[0252] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[0253] The acquired data is sent to a server via the edge computing unit.

[0254] Data Analysis Methods

[0255] server

[0256] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[0257] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[0258] Notices and Instructions

[0259] Analysis result notification means

[0260] server

[0261] Based on the analysis results, risk information is generated and alerts and safe driving advice are created for drivers.

[0262] This information is sent to the terminal.

[0263] Terminal (in-vehicle device)

[0264] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[0265] Voice commands and route guidance

[0266] Voice command recognition means

[0267] Terminal (in-vehicle device)

[0268] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0269] Route calculation method

[0270] server

[0271] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[0272] Navigation methods

[0273] Terminal (in-vehicle device)

[0274] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[0275] Emotion Engine

[0276] Emotion Engine

[0277] Data collection

[0278] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera.

[0279] Data analysis

[0280] The server analyzes the collected voice and facial expression data to determine the emotional state (e.g., stress, fatigue, anger, etc.).

[0281] Notices and Instructions

[0282] Based on the analysis results of the emotion engine, the server adjusts driving advice and safety suggestions and sends them to the device.

[0283] Specific examples

[0284] Example 1: Danger notification and emotional response

[0285] Terminal (in-vehicle device)

[0286] The onboard camera detects pedestrians ahead and sends that information to a server.

[0287] server

[0288] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[0289] At the same time, an emotion engine analyzes the driver's emotional state (e.g., impatience) and adjusts the tone of the alert as needed.

[0290] Terminal (in-vehicle device)

[0291] The system plays a voice notification saying "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. It also notifies in a calmer tone depending on the driver's emotional state.

[0292] Example 2: Relaxed Route Guide

[0293] server

[0294] The system calculates the optimal detour route based on the latest traffic congestion information, while the emotion engine analyzes the driver's fatigue level and suggests a less stressful route.

[0295] Terminal (in-vehicle device)

[0296] The new route information is received and a voice guides you, saying, "To avoid traffic jams, turn right at the next intersection. There is also a rest area on this route."

[0297] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[0298] The processing flow will be explained below.

[0299] Step 1:

[0300] The terminal (in-vehicle device) collects video data and environmental data in real time using an in-vehicle camera and sensors (LiDAR, radar, GPS, etc.). The in-vehicle camera captures images of the road ahead and surroundings, and the sensors obtain the necessary location information and obstacle information.

[0301] Step 2:

[0302] The terminal (in-vehicle device) performs initial processing on the collected data using an edge computing unit, compresses and encrypts it, and sends it to the server, enabling high-speed and secure data communication.

[0303] Step 3:

[0304] The server receives the data sent by the terminal and preprocesses the data by performing noise filtering and format normalization, which ensures the accuracy and consistency of the data.

[0305] Step 4:

[0306] The server analyzes the data using high-performance image processing algorithms and machine learning models, recognizing pedestrians, cyclists, motorbikes, and other objects, and calculating the position, speed, and direction of each object.

[0307] Step 5:

[0308] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[0309] Step 6:

[0310] The server sends the generated alert information and safe driving advice to the terminal, which is then ready to notify the driver in a timely manner.

[0311] Step 7:

[0312] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it conveys a message to the driver saying, "There is a pedestrian ahead. Please be careful."

[0313] Step 8:

[0314] The terminal (in-vehicle device) collects the driver's voice and facial expression data using a microphone and an in-vehicle camera, making it possible to grasp the driver's emotional state in real time.

[0315] Step 9:

[0316] The terminal (in-vehicle device) performs initial processing of the collected voice and facial expression data using an edge computing unit, compresses and encrypts it, and sends it to the server.

[0317] Step 10:

[0318] The server receives the voice and facial expression data sent from the terminal and analyzes the driver's emotional state (e.g., stress, fatigue, anger, etc.) using emotion recognition algorithms.

[0319] Step 11:

[0320] The server then adjusts driving advice and safety suggestions based on the emotion engine's analysis. For example, if the driver is feeling anxious, the server will adjust the tone of the notification to a calmer tone and generate a message encouraging more careful driving.

[0321] Step 12:

[0322] The server sends alert information and advice adjusted by the emotion engine to the terminal.

[0323] Step 13:

[0324] The terminal (in-vehicle device) receives the adjusted alert information and advice from the server and notifies the driver by voice and display, for example, a message saying, "There is a pedestrian ahead. Please proceed with caution and remain calm."

[0325] Step 14:

[0326] The user (driver) inputs voice commands through a microphone, such as setting a destination or changing a route.

[0327] Step 15:

[0328] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[0329] Step 16:

[0330] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion and road condition data, and the calculation results are then sent back to the device.

[0331] Step 17:

[0332] The terminal (in-vehicle device) receives optimal route information from the server and provides navigation to the driver by voice and display. For example, it may provide guidance such as, "Turn right at the next intersection. Also, you seem tired, so let's stop at a nearby rest area."

[0333] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[0334] Example 2

[0335] 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."

[0336] In today's automobile driving, it is important to ensure both driver safety and comfort. However, conventional driver assistance systems have difficulty adapting to changes in the driving environment and the driver's emotional state, and excessive information provision and inappropriate alerts can actually induce stress. Furthermore, there is a lack of technology to effectively combine real-time environmental data analysis and emotion analysis, resulting in situations where optimal driver assistance cannot be achieved.

[0337] 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 data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and an emotion analysis means. This makes it possible to quickly respond to changes in the driving environment and provide appropriate alerts and driving advice based on the driver's emotional state.

[0338] The "data collection means" is a component of the system that acquires video data and environmental data in real time using on-board cameras and sensors and transmits them to a server via an edge computing unit.

[0339] The "data analysis means" is a component of the system that analyzes collected data using a machine learning model after preprocessing such as noise filtering and format normalization to extract object recognition and location information.

[0340] The "analysis result notification means" is a system component that generates alerts and safe driving advice for the driver based on the analysis results obtained by the data analysis means and transmits them to the in-vehicle device.

[0341] The "voice command recognition means" is a component of the system that receives the driver's voice commands through a microphone, converts them into text using a voice recognition engine, and performs processing based on the content of the text.

[0342] The "route calculation means" is a system component that executes an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and transmits the results to the in-vehicle device.

[0343] The "navigation means" is a component of the system that provides navigation to the driver through voice and display based on the received optimum route information.

[0344] The "emotion analysis means" is a component of the system that analyzes the driver's voice and facial expression data collected through the vehicle-mounted camera and microphone, and determines the driver's emotional state.

[0345] The present invention relates to a system for supporting safe and comfortable driving of a driver, which includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion analysis means.

[0346] Data collection method configuration

[0347] The terminal (in-vehicle device) is equipped with an in-vehicle camera and various sensors (LiDAR, radar, GPS). The in-vehicle camera captures images of the front and surroundings in real time, and the sensors acquire environmental data. This data is processed by the edge computing unit and preprocessed through noise filtering. The preprocessed data is then sent to the server.

[0348] Data analysis method configuration

[0349] The server receives the video and sensor data sent from the device and performs pre-processing on the data. The pre-processed data is then analyzed using machine learning models to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel.

[0350] Configuration of analysis result notification method

[0351] The server generates risk information based on the analysis results and creates alerts and safe driving advice for the driver. These notifications are sent to the device, which displays the risk information and advice received from the server on its display and notifies the driver by voice. For example, it may notify the driver, "There is a pedestrian ahead. Please slow down."

[0352] Configuration of voice command recognition means

[0353] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0354] Configuring route calculation methods

[0355] The server runs an algorithm to calculate the optimal route based on traffic congestion and road condition data, and the calculated route information is sent to the device.

[0356] Navigation Method Configuration

[0357] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server, for example, by providing instructions such as "Turn left at the next intersection."

[0358] Sentiment analysis method configuration

[0359] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera. The server analyzes the collected voice and facial expression data to determine the driver's emotional state (e.g., stress, fatigue, anger, etc.). The server adjusts driving advice and safety suggestions based on the results of the emotion analysis and sends them to the terminal.

[0360] Specific examples

[0361] Example 1: Danger notification and emotional response

[0362] The terminal (in-vehicle device) uses an on-board camera to detect pedestrians ahead and sends that information to a server. The server uses an image analysis algorithm to identify the pedestrian's location and direction of travel, and calculates their speed and distance. If it determines that there is a danger, it generates an alert message and sends it to the terminal. Furthermore, an emotion analysis means analyzes the driver's emotional state and adjusts the tone of the alert as necessary. The terminal plays a voice notification saying, "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. The server also uses a calmer tone of voice depending on the driver's emotional state.

[0363] Example 2: Relaxed Route Guide

[0364] The server calculates the optimal detour route based on the latest traffic congestion information. At the same time, an emotion analysis tool analyzes the driver's fatigue level and suggests a less stressful route. The device receives the new route information and provides voice guidance such as, "To avoid traffic congestion, please turn right at the next intersection. There is also a rest area on this route."

[0365] Example prompts for generative AI models

[0366] Please explain in detail how "DriveGuard AI" collects sensor data, analyzes it on the server, and provides real-time alerts to the driver.

[0367] Please explain with concrete examples how the system analyzes the driver's emotional state and, based on the results, adjusts driving advice appropriately and in the right place.

[0368] This system, with the above configuration, provides a mechanism that comprehensively supports the driver in ensuring safe and comfortable driving.

[0369] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0370] Step 1:

[0371] The terminal (on-board device) acquires video data and environmental data in real time using the on-board camera and various sensors (LiDAR, radar, GPS). The acquired data is preprocessed by the edge computing unit after undergoing noise filtering. After preprocessing is complete, the data is sent from the terminal to the server. The inputs are camera video data and sensor data, and the output is the preprocessed data.

[0372] Specific examples of behavior:

[0373] The device activates the onboard camera to capture images of the road ahead, while the LiDAR and radar simultaneously acquire the position and speed of surrounding objects, and this data is then filtered by the edge computing unit.

[0374] Step 2:

[0375] The server receives video and sensor data sent from the device. As input, it receives preprocessed data. The received data is preprocessed through noise filtering and format normalization. The preprocessed data is ready to be analyzed by the machine learning model.

[0376] Specific examples of behavior:

[0377] The server receives the data stream, corrects dark areas, removes noise, and unifies the format through filtering, converting it into the appropriate data format.

[0378] Step 3:

[0379] The server uses machine learning models to analyze the noise-filtered and format-normalized data. The input is the preprocessed data. The machine learning models are used to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel. The output is the object recognition results and their location information.

[0380] Specific examples of behavior:

[0381] A machine learning algorithm scans the video data and recognizes pedestrians ahead, calculating their location and direction of travel as a result.

[0382] Step 4:

[0383] The server generates alerts and safe driving advice for the driver based on the analysis results. The input is the analysis results, and the output is an alert message or advice. The generated information is sent to the terminal.

[0384] Specific examples of behavior:

[0385] The server generates an alert message saying "There is a pedestrian ahead. Please slow down." and sends it to the terminal.

[0386] Step 5:

[0387] The terminal (in-vehicle device) displays the alerts and advice received from the server on the display and notifies the driver by voice. The input is the alert message from the server, and the output is the display and voice notification.

[0388] Specific examples of behavior:

[0389] The device will play a voice message saying, "There is a pedestrian ahead. Be careful." and will also show a warning message on the display.

[0390] Step 6:

[0391] The device receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The input is the driver's voice command, and the output is the voice command converted into text. Processing is performed based on the voice command converted into text.

[0392] Specific examples of behavior:

[0393] The driver enters the voice command "recalculate route," and a speech recognition engine converts the command into text.

[0394] Step 7:

[0395] The server receives voice commands converted into text and calculates the optimal route based on traffic congestion and road condition data. The inputs are voice commands and traffic information, and the output is the calculated optimal route information. The calculated route information is sent to the device.

[0396] Specific examples of behavior:

[0397] The server collects the latest traffic congestion information, calculates a new detour route based on that information, and sends the results to the terminal.

[0398] Step 8:

[0399] The terminal provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information from the server, and the output is navigation instructions.

[0400] Specific examples of behavior:

[0401] The device will provide voice guidance such as "Turn left at the next intersection," and will display a map and route on the screen.

[0402] Step 9:

[0403] The terminal collects the driver's voice and facial expression data through a microphone and an on-board camera. The input is the driver's voice and video data, and the output is the collected data sent to the server.

[0404] Specific examples of behavior:

[0405] The device captures the driver's facial expressions and tone of voice in real time and transmits the data to a server.

[0406] Step 10:

[0407] The server analyzes the transmitted voice and facial expression data to determine the driver's emotional state. The input is the driver's voice and facial expression data, and the output is the analysis result of the emotional state. Based on the analysis result, driving advice and safety suggestions are adjusted and sent to the terminal.

[0408] Specific examples of behavior:

[0409] The server recognizes signs of stress from the driver's tone of voice and facial expression, generates advice such as "Relax," and sends it to the terminal.

[0410] Through these processing steps, the "DriveGuard AI" system provides a mechanism to comprehensively support the driver's safety and comfort.

[0411] (Application example 2)

[0412] 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."

[0413] Conventional building security systems are limited to simple monitoring and warning using surveillance cameras and sensors, and lack the ability to detect suspicious individuals or unusual events, nor the countermeasures for their detection. Furthermore, no consideration was given to supporting security personnel in dealing with overwork and stressful environments, making it difficult to maintain efficient security. Therefore, it is necessary to reduce the mental stress of security personnel, along with early detection of suspicious individuals and appropriate countermeasures.

[0414] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. This enables the prompt detection of suspicious individuals and abnormalities, the appropriate countermeasures, and the reduction of mental stress for security personnel.

[0415] The "data collection means" is a means for collecting video data and environmental data in real time from monitoring devices and various sensors.

[0416] "Data analysis means" refers to a means of analyzing collected data using a machine learning model to detect suspicious individuals or abnormalities.

[0417] The "analysis result notification means" is a means for notifying the security officer of appropriate warnings and instructions based on the analysis results obtained by the data analysis means.

[0418] The "voice command recognition means" refers to a means for recognizing and analyzing voice commands from security personnel and performing corresponding functions in accordance with the instructions.

[0419] The "route calculation means" is a means for calculating the optimal route based on route information and situational data inside and outside the building, and predicting the route and escape route of a detected suspicious person.

[0420] The "navigation means" is a means for providing navigation to security personnel using voice and a display based on the calculated optimum route information.

[0421] The "emotion analysis engine" is a means of analyzing the voice and facial expression data of security personnel obtained through microphones and cameras to determine their mental stress and emotional state.

[0422] "Security notification means" is a means of notifying appropriate warnings and safety suggestions based on the results of detecting suspicious individuals or abnormalities and the emotional state of the person in charge.

[0423] The present invention relates to a "SecureGuard AI" system that improves building security and safety. The system includes multiple elements: a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. Specific embodiments for implementing the system are described in detail below.

[0424] Data collection and analysis

[0425] Data collection methods

[0426] Terminals (smartphones and robots) are equipped with monitoring devices and sensors that collect real-time video and environmental data from inside and outside the building. The collected data is then sent to a server via an edge computing unit. Specific hardware used includes high-resolution cameras, microphones, motion sensors, etc.

[0427] Data Analysis Methods

[0428] The server receives the collected video and sensor data. The data is preprocessed through noise filtering and format normalization, and then analyzed using analysis algorithms and machine learning models (e.g., the dlib library and open-source sentiment analysis models). This allows for the detection of suspicious individuals and unusual behavior, and the extraction of their location information and behavioral patterns.

[0429] Notices and Instructions

[0430] Analysis result notification means

[0431] The server generates risk information based on the analysis results obtained by the data analysis means, and issues alerts and safety measures to the operator, which are then sent to the terminal.

[0432] Voice command recognition means

[0433] The terminal receives voice commands from the operator through a microphone, analyzes these commands with a speech recognition engine (e.g., Google Speech-to-Text API), converts them into text, and performs the corresponding processing.

[0434] Route calculation and navigation

[0435] Route calculation method

[0436] The server runs algorithms that calculate optimal routes based on route information and situational data inside and outside the building, such as routes that bypass areas where suspicious individuals may be present, using up-to-date map data and real-time situational information.

[0437] Navigation methods

[0438] Based on the optimal route information received from the server, the terminal provides navigation to the operator by voice and display, for example, by providing instructions such as "Turn left at the next corridor."

[0439] Sentiment Analysis Engine

[0440] The terminal collects the operator's voice and facial expression data through a microphone and camera. The server analyzes the collected voice and facial expression data to determine the operator's emotional state (stress, fatigue, anger, etc.). This enables the system to take measures to reduce the operator's mental stress.

[0441] Specific examples

[0442] Example 1: Detecting and responding to suspicious individuals

[0443] The terminal transmits images captured by the surveillance cameras inside the building to the server.

[0444] The server analyzes the video and detects suspicious individuals.

[0445] The server generates a warning such as "There is a suspicious person near the entrance. Please be on guard" and sends it to the terminal.

[0446] Example 2: Controlling a surveillance camera with voice commands

[0447] The terminal receives a voice command from the operator: "Show surveillance camera 1."

[0448] A voice recognition engine converts commands into text and displays the image from the corresponding camera.

[0449] Prompt Sentence Examples

[0450] "A suspicious person has been detected near the entrance. Please implement security measures."

[0451] "Do you want to view the footage from security camera 1?"

[0452] By combining these elements, the SecureGuard AI system can significantly improve building security and provide a safe and efficient environment.

[0453] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0454] Step 1:

[0455] The terminal collects video data and environmental data from monitoring devices and sensors. The collected data is sent to a server via an edge computing unit. The input is raw data from the monitoring devices and sensors, and the output is pre-processed data. Specifically, the high-resolution camera continuously captures video, and the sensors detect temperature and movement.

[0456] Step 2:

[0457] The server receives the transmitted video data and environmental data and performs preprocessing through noise filtering and format normalization. The input is the video data and sensor data transmitted from the device, and the output is the noise-removed and normalized data. Specifically, the filtering algorithm removes noise from the video and standardizes the data format.

[0458] Step 3:

[0459] The server analyzes the preprocessed data using analysis algorithms and machine learning models (for example, dlib or open-source emotion analysis models) to detect suspicious individuals or unusual behavior. The input is noise-removed and normalized data, and the output is information about suspicious individuals or unusual behavior. Specifically, the machine learning model recognizes faces and behavioral patterns and detects suspicious behavior.

[0460] Step 4:

[0461] The server generates danger information based on the detection results, creates appropriate warnings and instructions, and sends them to the terminal. The input is information about suspicious individuals and unusual events, and the output is warning messages and instructions. Specifically, when a suspicious individual is detected, a warning message saying "There is a suspicious individual near the entrance. Please be on guard" is generated and sent to the terminal.

[0462] Step 5:

[0463] The terminal notifies the operator of warning messages and instructions received from the server by voice and on the display. The input is the warning message or instruction sent from the server, and the output is a voice notification and a display. Specifically, the speech synthesis engine reads out the warning message and displays it on the display.

[0464] Step 6:

[0465] The device receives the operator's voice commands through the microphone and analyzes them using a voice recognition engine (for example, Google Speech-to-Text API). The input is the operator's voice command, and the output is a text command. Specifically, when the operator says, "Show surveillance camera 1," the voice is converted into text.

[0466] Step 7:

[0467] The server executes the corresponding process based on the voice command. For example, it sends an instruction to the terminal to display the video from security camera 1. The input is a text-converted voice command, and the output is an instruction for the processing procedure. Specifically, in response to the command "Display security camera 1," the video from the corresponding camera is sent to the terminal.

[0468] Step 8:

[0469] The terminal navigates based on the route information received from the server. The input is the route information sent from the server, and the output is navigation instructions to the operator. Specific actions include a voice prompt saying "Turn left at the next corridor" and a map being displayed on the screen.

[0470] Step 9:

[0471] The device collects the operator's voice and facial expression data through a microphone and camera, and the emotion analysis engine analyzes it. The input is the operator's voice and facial expression data, and the output is the operator's emotional state information. Specifically, if the operator is feeling stressed, that state will be displayed as the analysis result.

[0472] Step 10:

[0473] Based on the results of the emotion analysis engine, the server takes action that matches the emotional state of the operator. For example, it changes the tone of the voice notification to reduce stress. The input is the operator's emotional state information, and the output is instructions on how to respond. A specific action would be to notify the operator in a relaxed tone, saying, "It is safe, so please stay calm and take action."

[0474] The above is the flow of the main processing steps of the "SecureGuard AI" system.

[0475] 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.

[0476] 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.

[0477] 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.

[0478] [Second embodiment]

[0479] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0480] 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.

[0481] 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).

[0482] 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.

[0483] 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.

[0484] 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).

[0485] 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.

[0486] 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.

[0487] 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.

[0488] 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.

[0489] 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.

[0490] 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."

[0491] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system are described below.

[0492] System Overview

[0493] The DriveGuard AI system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[0494] Data collection and analysis

[0495] Data collection methods

[0496] Terminal (in-vehicle device)

[0497] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[0498] The acquired data is sent to a server via the edge computing unit.

[0499] Data Analysis Methods

[0500] server

[0501] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[0502] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[0503] Notices and Instructions

[0504] Analysis result notification means

[0505] server

[0506] Based on the analysis results, risk information is generated and alerts and advice are created for drivers.

[0507] This information is sent to the terminal.

[0508] Terminal (in-vehicle device)

[0509] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[0510] Voice commands and route guidance

[0511] Voice command recognition means

[0512] Terminal (in-vehicle device)

[0513] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0514] Route calculation method

[0515] server

[0516] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[0517] Navigation methods

[0518] Terminal (in-vehicle device)

[0519] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[0520] Specific examples

[0521] Example 1: Danger notification

[0522] Terminal (in-vehicle device)

[0523] The onboard camera detects pedestrians ahead and sends that information to a server.

[0524] server

[0525] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[0526] Terminal (in-vehicle device)

[0527] A voice notification will be played saying "There is a pedestrian ahead. Be careful." and a warning will also be displayed on the display.

[0528] Example 2: Traffic jam avoidance route guide

[0529] server

[0530] Calculates the optimal detour route based on the latest traffic congestion information.

[0531] Terminal (in-vehicle device)

[0532] It receives new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[0533] In this way, the DriveGuard AI system predicts hazards in real time and supports safe driving by collecting, analyzing, notifying, and processing the driver's voice commands.

[0534] The processing flow will be explained below.

[0535] Step 1:

[0536] The terminal (in-vehicle device) collects video data and environmental data in real time using on-board cameras and sensors. The on-board cameras capture images of the front and surrounding area, and the sensors acquire information from LiDAR, radar, GPS, etc.

[0537] Step 2:

[0538] The terminal (in-vehicle device) sends the collected data to a server via the edge computing unit, where the data is transmitted quickly, encrypted, and securely.

[0539] Step 3:

[0540] The server receives the data sent from the terminal and pre-processes it through noise filtering and format normalization, thereby improving the quality of the data.

[0541] Step 4:

[0542] The server analyzes the data using sophisticated image processing algorithms and machine learning models to recognize objects such as pedestrians, cyclists, and motorbikes, and calculates the position, speed, and direction of each object.

[0543] Step 5:

[0544] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[0545] Step 6:

[0546] The server sends the generated alert information and safe driving advice to the terminal, so that the driver can be notified in a timely manner.

[0547] Step 7:

[0548] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it may play a voice message saying, "There is a pedestrian ahead. Please be careful."

[0549] Step 8:

[0550] The user (driver) inputs voice commands through a microphone, such as setting a destination or requesting a new route.

[0551] Step 9:

[0552] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[0553] Step 10:

[0554] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion information and road condition data, and the calculation results are sent back to the terminal.

[0555] Step 11:

[0556] The terminal (in-vehicle device) receives the optimal route information from the server and provides navigation to the driver by voice and display, for example, "Turn right at the next intersection."

[0557] Example 1

[0558] 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."

[0559] Conventional driving assistance systems have problems in that they do not adequately provide drivers with appropriate information or provide navigation based on voice commands. Furthermore, because there is no consistent processing from data collection to analysis, notification, and navigation, it is difficult to provide sufficient functionality to support safe driving. To solve this problem, there is a need for real-time data collection and analysis, appropriate hazard notification, and rapid navigation based on voice commands.

[0560] 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.

[0561] In this invention, the server includes a data preprocessing means, a data analysis means, and an analysis result notification means, which allows for quick and accurate preprocessing and analysis of data collected in real time, and provides appropriate alerts and navigation instructions to the driver.

[0562] "Data collection means" means for collecting video and environmental data in real time using digital cameras and sensing devices.

[0563] "Data pre-processing means" refers to means for performing noise filtering and format normalization on collected data.

[0564] The "data analysis means" is a means for analyzing collected and preprocessed data using a machine learning model to extract information about objects and their locations.

[0565] The "analysis result notification means" is a means for generating danger information based on the analysis results and providing alerts and advice to the driver.

[0566] The "voice command recognition means" is a means for converting a user's voice command into text and performing processing in accordance with the content of the text.

[0567] The "route calculation means" is a means for calculating the optimum route based on the latest traffic information.

[0568] "Navigation means" refers to a means for providing audio and visual navigation to the driver based on the calculated optimum route information.

[0569] The present invention relates to a system for supporting safe and comfortable driving for drivers, and this system includes a data collection means, a data preprocessing means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system will be described below.

[0570] System Overview

[0571] This system collects data using cameras and various sensors mounted on the vehicle, preprocesses and analyzes the data on a server, and provides appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[0572] Data Acquisition and Transmission

[0573] Terminal (in-vehicle device)

[0574] The device collects data using a digital camera and various detection devices (LiDAR, radar, GPS, etc.), thereby obtaining real-time image data of the front and surrounding area and environmental data. The acquired data is sent to a server via an edge computing unit.

[0575] Data preprocessing and analysis

[0576] server

[0577] The server receives video data and environmental data sent from the device. First, noise filtering and format normalization are performed by the data preprocessing means. Next, the collected and preprocessed data is analyzed by the data analysis means using a machine learning model to extract objects and their location information. For example, pedestrians, bicycles, motorbikes, etc. are recognized, and their speed and direction of movement are estimated.

[0578] Notices and Instructions

[0579] server

[0580] The server generates danger information based on the analysis results and includes an analysis result notification means for providing alerts and advice to the driver. For example, a warning message such as "There is a pedestrian ahead. Please slow down" is generated and sent to the terminal.

[0581] Terminal (in-vehicle device)

[0582] The device receives the danger information and advice sent from the server, and displays it on the display and notifies the driver by voice, for example, "There is a pedestrian ahead. Be careful."

[0583] Voice commands and route guidance

[0584] Terminal (in-vehicle device)

[0585] The device receives the user's voice commands through a microphone, analyzes them with a voice recognition engine, converts the voice commands into text, and processes them accordingly.

[0586] server

[0587] The server includes a route calculation means for calculating the optimum route based on voice commands, and calculates the optimum route based on traffic congestion information and road condition data, and transmits the calculated route to the terminal.

[0588] Terminal (in-vehicle device)

[0589] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[0590] Specific examples

[0591] Example 1: Danger notification

[0592] The terminal (on-board device) detects pedestrians ahead using the on-board camera and sends the data to the server.

[0593] The server uses data analysis means to recognize the pedestrian's location and direction of travel, and if it determines that there is danger, it generates an alert message and sends it to the terminal.

[0594] The terminal (in-vehicle device) will play a voice notification saying, "There is a pedestrian ahead. Please be careful," and will also display a warning on the display.

[0595] Example 2: Traffic jam avoidance route guide

[0596] The server calculates the optimal detour route based on the latest traffic congestion information.

[0597] The terminal (in-vehicle device) receives the new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[0598] Prompt Sentence Examples

[0599] Prompt 1: "When an in-car camera detects a pedestrian ahead, how is the data processed?"

[0600] Prompt 2: "Describe the process by which the optimal route is calculated and how it is communicated to the driver."

[0601] As described above, the present invention is a system that supports safe driving by realizing consistent processing from data acquisition, preprocessing, analysis, notification, and route guidance in response to voice commands.

[0602] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0603] Step 1:

[0604] Acquiring and Sending Data

[0605] The terminal (on-board device) uses a digital camera and sensors (LiDAR, radar, GPS, etc.) to acquire front and surrounding image data and environmental data in real time. Specifically, the on-board camera captures images of the front, and the GPS acquires current location information. This data is temporarily stored through the edge computing unit and sent to a server. The input is camera and sensor data, and the output is data sent to the server.

[0606] Step 2:

[0607] Data Preprocessing

[0608] The server receives data sent from the terminal. Data preprocessing means performs noise filtering and format normalization. Specific operations include removing outliers and missing values, and adjusting the resolution of image data. The input is raw data received from the terminal, and the output is preprocessed data.

[0609] Step 3:

[0610] Data analysis

[0611] The server analyzes the preprocessed data using data analysis means. Here, a machine learning model is used to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information, speed, direction of travel, etc. Specifically, an image recognition algorithm detects the objects and calculates their coordinate data. The input is the preprocessed data, and the output is the analysis results (object type, position, speed, etc.).

[0612] Step 4:

[0613] Hazard information generation and notification

[0614] The server generates danger information based on the analysis results. The analysis result notification means creates a message to notify the driver of the generated danger information and sends it to the terminal. Specifically, a warning message such as "There is a pedestrian ahead. Please slow down" is generated. The input is the analysis results, and the output is the notification message.

[0615] The terminal (in-vehicle device) receives danger information and advice sent from the server, displays it on the display, and notifies the driver by voice. The input is the notification message received from the server, and the output is warning information by voice and display. Specifically, the terminal displays a warning message on the in-vehicle display and plays a voice notification from the speaker.

[0616] Step 5:

[0617] Receiving and processing voice commands

[0618] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them using a voice recognition engine. The voice commands are converted into text and processing is carried out based on the content. The input is the driver's voice command and the output is the textual command content. Specifically, the voice command "Please tell me the best route" is analyzed.

[0619] Step 6:

[0620] Calculating the best route

[0621] The server receives data for calculating the optimal route based on the voice command. The route calculation means calculates the optimal route based on traffic congestion information and road condition data. The input is the voice command and the latest traffic information, and the output is the calculated optimal route information. Specifically, the system uses a traffic information API to obtain real-time traffic congestion information and calculates the optimal route using an algorithm.

[0622] Step 7:

[0623] Providing navigation

[0624] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information received from the server, and the output is navigation instructions to the driver. Specific operations include voice guidance such as "Turn left at the next intersection" and a map display.

[0625] (Application example 1)

[0626] 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."

[0627] Current autonomous vehicles are required to collect and analyze environmental data in real time and make appropriate decisions to ensure safe driving. However, existing systems lack effective notification methods for the driver and the autonomous driving system, preventing accident risk reduction and efficient route guidance. In particular, the limited use of prompts using voice command recognition and generative AI models poses a challenge for further improving driver convenience and safety.

[0628] 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.

[0629] In this invention, the server includes a data analysis means, a prompt generation means, and a notification means using a generative AI model. This allows the server to analyze environmental data in real time and provide appropriate alerts and route guidance prompts to the driver and the automated driving system using the generative AI model. This reduces the risk of traffic accidents and provides efficient driving assistance.

[0630] The "data collection means" is a means for collecting video data and environmental data in real time from on-board cameras and sensors.

[0631] The "data analysis means" is a means of analyzing collected data using a machine learning model to extract recognition and location information for pedestrians, bicycles, motorbikes, etc.

[0632] The "analysis result notification means" is a means for notifying the driver or the automated driving system of the analysis results obtained by the data analysis means.

[0633] The "voice command recognition means" is a means for recognizing a voice command from the driver and analyzing the content of the command.

[0634] The "route calculation means" is a means for calculating the optimum route based on information on the current position and destination, as well as traffic congestion information and road condition data.

[0635] "Navigation means" refers to a means for providing voice and display navigation to the driver or the automated driving system based on the calculated optimal route information.

[0636] The "prompt generating means" is a means for generating an appropriate instruction or alert prompt in response to an input voice command or a recognized situation.

[0637] The "prompt processing means" is a means for performing processing to appropriately notify the driver of the generated prompt.

[0638] "Notification means using a generative AI model" refers to a means of generating appropriate notification content from data analyzed using a generative AI model and notifying the driver or automated driving system in real time.

[0639] This invention aims to realize a safe driving support system for autonomous vehicles called "AutoNav AI." This system includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, a prompt generation means, a prompt processing means, and a notification means using a generative AI model.

[0640] System Overview

[0641] The data collection method uses on-board cameras and sensors to collect video data and environmental data in real time. Specific hardware used includes on-board cameras such as the Logitech C920, as well as LiDAR, radar, and GPS sensors. This data is sent to a server via an edge computing unit.

[0642] The data analysis means receives the video data and environmental data collected on the server and analyzes them using machine learning models (e.g., OpenCV and YOLO). The data undergoes pre-processing such as noise filtering and format normalization, and then uses analysis algorithms to recognize objects such as pedestrians, bicycles, and motorbikes and extract their location information.

[0643] The analysis result notification means generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. These notifications are sent to the terminal in real time via the edge computing unit and are displayed as voice and on a display.

[0644] The voice command recognition means receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine (e.g., Google's SpeechRecognition API). The recognized voice commands are converted into text and processed accordingly.

[0645] The route calculation means calculates the optimal route based on the current location and destination information, as well as traffic congestion and road condition data. Specifically, route calculation is performed using Google Maps API, etc. The calculated route information is sent from the server to the terminal.

[0646] The navigation means provides navigation to the driver or the automated driving system by voice and display based on the optimal route information received by the terminal from the server, for example, by providing guidance such as "Turn left at the next intersection."

[0647] The prompt generator generates appropriate instruction or alert prompts based on the input voice command and the recognized situation, using a generative AI model.

[0648] The prompt processing means performs processing to appropriately notify the driver of the generated prompt, thereby providing instructions and alerts to the driver at the appropriate time.

[0649] The notification method using the generative AI model generates appropriate notification content from data analyzed using the generative AI model and notifies the driver or autonomous driving system in real time.

[0650] Specific examples

[0651] Example 1: Traffic jam avoidance route guide

[0652] The server calculates the optimal detour route based on the latest traffic congestion information. The device receives the new route information and provides voice guidance such as, "Turn right at the next intersection to avoid traffic congestion." This supports efficient driving.

[0653] Example 2: Voice command recognition and prompt generation

[0654] The driver issues a voice command saying, "Calculate a safe and efficient route based on the latest traffic congestion information. End point: Shinjuku Station." In response, the server calculates the optimal route and generates a prompt saying, "The optimal route from your current location to Shinjuku Station is to turn right at the next intersection and go straight at the next traffic light. The route takes approximately 30 minutes. There is an accident warning on the way, so please be careful."

[0655] Thus, the present invention provides specific embodiments for significantly improving the safety and efficiency of autonomous vehicles.

[0656] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0657] Step 1:

[0658] Data collection

[0659] The onboard camera and various sensors installed in the user's vehicle collect video data and environmental data in real time. Specifically, the camera captures the surrounding image, and the LiDAR, radar, and GPS sensors acquire various environmental data. This data is sent to a server via an edge computing unit.

[0660] Input: Video data and environmental data from on-board cameras and sensors

[0661] Output: Data sent to the server through the edge computing unit

[0662] Step 2:

[0663] Data analysis

[0664] The server receives the collected video and environmental data and performs preprocessing: the data undergoes noise filtering and format normalization, and is then analyzed using machine learning models (OpenCV and YOLO, for example) to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information.

[0665] Input: Video data and environmental data collected in step 1

[0666] Output: Objects and their locations

[0667] Step 3:

[0668] Notification of analysis results

[0669] The server generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. The generated notifications are sent to the terminal via the edge computing unit and displayed by voice and on the display. For example, if there is a pedestrian ahead, the notification will say, "There is a pedestrian ahead. Please be careful."

[0670] Input: Analysis results obtained by data analysis tools

[0671] Output: Alerts and advice sent to the device

[0672] Step 4:

[0673] Voice Command Recognition

[0674] Voice commands issued by users (e.g., "End point: Shinjuku Station") are collected through microphones in the vehicle and converted into text by a speech recognition engine (e.g., Google's SpeechRecognition API). This text data is sent to a server and analyzed as a voice command.

[0675] Input: A voice command spoken by the user

[0676] Output: Text data sent to the server

[0677] Step 5:

[0678] Route calculation

[0679] The server calculates the optimal route based on the analyzed voice commands, current location information, traffic information, and road condition data. This calculation uses route calculation services such as Google Maps API. The calculated route information is sent from the server to the device.

[0680] Input: Voice commands, current location, traffic information, road condition data

[0681] Output: Calculated optimal route information

[0682] Step 6:

[0683] Navigation

[0684] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[0685] Input: Optimal route information received from the server

[0686] Output: Voice and display navigation instructions

[0687] Step 7:

[0688] Prompt Generation

[0689] The server generates appropriate prompts, such as instructions or alerts, based on the input voice command and the recognized situation. A generative AI model is used to generate these prompts. A specific example of a prompt is, "Turn right at the next intersection to avoid traffic jams."

[0690] Input: Voice commands, recognized situations

[0691] Output: Generated prompt

[0692] Step 8:

[0693] Prompt Processing

[0694] The server then processes the generated prompt to appropriately notify the driver. The notification is given via voice and display. For example, the driver may be notified, "Turn right at the next intersection to avoid traffic jams."

[0695] Input: Generated prompt

[0696] Output: Audio and visual notifications

[0697] 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.

[0698] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion engine. Specific embodiments for implementing this system are described below.

[0699] System Overview

[0700] The "DriveGuard AI" system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate. Furthermore, it combines an emotion engine that analyzes the driver's emotional state from their voice and facial expressions to provide appropriate support according to the driver's emotional state.

[0701] Data collection and analysis

[0702] Data collection methods

[0703] Terminal (in-vehicle device)

[0704] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[0705] The acquired data is sent to a server via the edge computing unit.

[0706] Data Analysis Methods

[0707] server

[0708] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[0709] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[0710] Notices and Instructions

[0711] Analysis result notification means

[0712] server

[0713] Based on the analysis results, risk information is generated and alerts and safe driving advice are created for drivers.

[0714] This information is sent to the terminal.

[0715] Terminal (in-vehicle device)

[0716] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[0717] Voice commands and route guidance

[0718] Voice command recognition means

[0719] Terminal (in-vehicle device)

[0720] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0721] Route calculation method

[0722] server

[0723] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[0724] Navigation methods

[0725] Terminal (in-vehicle device)

[0726] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[0727] Emotion Engine

[0728] Emotion Engine

[0729] Data collection

[0730] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera.

[0731] Data analysis

[0732] The server analyzes the collected voice and facial expression data to determine the emotional state (e.g., stress, fatigue, anger, etc.).

[0733] Notices and Instructions

[0734] Based on the analysis results of the emotion engine, the server adjusts driving advice and safety suggestions and sends them to the device.

[0735] Specific examples

[0736] Example 1: Danger notification and emotional response

[0737] Terminal (in-vehicle device)

[0738] The onboard camera detects pedestrians ahead and sends that information to a server.

[0739] server

[0740] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[0741] At the same time, an emotion engine analyzes the driver's emotional state (e.g., impatience) and adjusts the tone of the alert as needed.

[0742] Terminal (in-vehicle device)

[0743] The system plays a voice notification saying "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. It also notifies in a calmer tone depending on the driver's emotional state.

[0744] Example 2: Relaxed Route Guide

[0745] server

[0746] The system calculates the optimal detour route based on the latest traffic congestion information, while the emotion engine analyzes the driver's fatigue level and suggests a less stressful route.

[0747] Terminal (in-vehicle device)

[0748] The new route information is received and a voice guides you, saying, "To avoid traffic jams, turn right at the next intersection. There is also a rest area on this route."

[0749] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[0750] The processing flow will be explained below.

[0751] Step 1:

[0752] The terminal (in-vehicle device) collects video data and environmental data in real time using an in-vehicle camera and sensors (LiDAR, radar, GPS, etc.). The in-vehicle camera captures images of the road ahead and surroundings, and the sensors obtain the necessary location information and obstacle information.

[0753] Step 2:

[0754] The terminal (in-vehicle device) performs initial processing on the collected data using an edge computing unit, compresses and encrypts it, and sends it to the server, enabling high-speed and secure data communication.

[0755] Step 3:

[0756] The server receives the data sent by the terminal and preprocesses the data by performing noise filtering and format normalization, which ensures the accuracy and consistency of the data.

[0757] Step 4:

[0758] The server analyzes the data using high-performance image processing algorithms and machine learning models, recognizing pedestrians, cyclists, motorbikes, and other objects, and calculating the position, speed, and direction of each object.

[0759] Step 5:

[0760] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[0761] Step 6:

[0762] The server sends the generated alert information and safe driving advice to the terminal, which is then ready to notify the driver in a timely manner.

[0763] Step 7:

[0764] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it conveys a message to the driver saying, "There is a pedestrian ahead. Please be careful."

[0765] Step 8:

[0766] The terminal (in-vehicle device) collects the driver's voice and facial expression data using a microphone and an in-vehicle camera, making it possible to grasp the driver's emotional state in real time.

[0767] Step 9:

[0768] The terminal (in-vehicle device) performs initial processing of the collected voice and facial expression data using an edge computing unit, compresses and encrypts it, and sends it to the server.

[0769] Step 10:

[0770] The server receives the voice and facial expression data sent from the terminal and analyzes the driver's emotional state (e.g., stress, fatigue, anger, etc.) using emotion recognition algorithms.

[0771] Step 11:

[0772] The server then adjusts driving advice and safety suggestions based on the emotion engine's analysis. For example, if the driver is feeling anxious, the server will adjust the tone of the notification to a calmer tone and generate a message encouraging more careful driving.

[0773] Step 12:

[0774] The server sends alert information and advice adjusted by the emotion engine to the terminal.

[0775] Step 13:

[0776] The terminal (in-vehicle device) receives the adjusted alert information and advice from the server and notifies the driver by voice and display, for example, a message saying, "There is a pedestrian ahead. Please proceed with caution and remain calm."

[0777] Step 14:

[0778] The user (driver) inputs voice commands through a microphone, such as setting a destination or changing a route.

[0779] Step 15:

[0780] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[0781] Step 16:

[0782] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion and road condition data, and the calculation results are then sent back to the device.

[0783] Step 17:

[0784] The terminal (in-vehicle device) receives optimal route information from the server and provides navigation to the driver by voice and display. For example, it may provide guidance such as, "Turn right at the next intersection. Also, you seem tired, so let's stop at a nearby rest area."

[0785] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[0786] Example 2

[0787] 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."

[0788] In today's automobile driving, it is important to ensure both driver safety and comfort. However, conventional driver assistance systems have difficulty adapting to changes in the driving environment and the driver's emotional state, and excessive information provision and inappropriate alerts can actually induce stress. Furthermore, there is a lack of technology to effectively combine real-time environmental data analysis and emotion analysis, resulting in situations where optimal driver assistance cannot be achieved.

[0789] 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 data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and an emotion analysis means. This makes it possible to quickly respond to changes in the driving environment and provide appropriate alerts and driving advice based on the driver's emotional state.

[0790] The "data collection means" is a component of the system that acquires video data and environmental data in real time using on-board cameras and sensors and transmits them to a server via an edge computing unit.

[0791] The "data analysis means" is a component of the system that analyzes collected data using a machine learning model after preprocessing such as noise filtering and format normalization to extract object recognition and location information.

[0792] The "analysis result notification means" is a system component that generates alerts and safe driving advice for the driver based on the analysis results obtained by the data analysis means and transmits them to the in-vehicle device.

[0793] The "voice command recognition means" is a component of the system that receives the driver's voice commands through a microphone, converts them into text using a voice recognition engine, and performs processing based on the content of the text.

[0794] The "route calculation means" is a system component that executes an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and transmits the results to the in-vehicle device.

[0795] The "navigation means" is a component of the system that provides navigation to the driver through voice and display based on the received optimum route information.

[0796] The "emotion analysis means" is a component of the system that analyzes the driver's voice and facial expression data collected through the vehicle-mounted camera and microphone, and determines the driver's emotional state.

[0797] The present invention relates to a system for supporting safe and comfortable driving of a driver, which includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion analysis means.

[0798] Data collection method configuration

[0799] The terminal (in-vehicle device) is equipped with an in-vehicle camera and various sensors (LiDAR, radar, GPS). The in-vehicle camera captures images of the front and surroundings in real time, and the sensors acquire environmental data. This data is processed by the edge computing unit and preprocessed through noise filtering. The preprocessed data is then sent to the server.

[0800] Data analysis method configuration

[0801] The server receives the video and sensor data sent from the device and performs pre-processing on the data. The pre-processed data is then analyzed using machine learning models to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel.

[0802] Configuration of analysis result notification method

[0803] The server generates risk information based on the analysis results and creates alerts and safe driving advice for the driver. These notifications are sent to the device, which displays the risk information and advice received from the server on its display and notifies the driver by voice. For example, it may notify the driver, "There is a pedestrian ahead. Please slow down."

[0804] Configuration of voice command recognition means

[0805] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0806] Configuring route calculation methods

[0807] The server runs an algorithm to calculate the optimal route based on traffic congestion and road condition data, and the calculated route information is sent to the device.

[0808] Navigation Method Configuration

[0809] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server, for example, by providing instructions such as "Turn left at the next intersection."

[0810] Sentiment analysis method configuration

[0811] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera. The server analyzes the collected voice and facial expression data to determine the driver's emotional state (e.g., stress, fatigue, anger, etc.). The server adjusts driving advice and safety suggestions based on the results of the emotion analysis and sends them to the terminal.

[0812] Specific examples

[0813] Example 1: Danger notification and emotional response

[0814] The terminal (in-vehicle device) uses an on-board camera to detect pedestrians ahead and sends that information to a server. The server uses an image analysis algorithm to identify the pedestrian's location and direction of travel, and calculates their speed and distance. If it determines that there is a danger, it generates an alert message and sends it to the terminal. Furthermore, an emotion analysis means analyzes the driver's emotional state and adjusts the tone of the alert as necessary. The terminal plays a voice notification saying, "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. The server also uses a calmer tone of voice depending on the driver's emotional state.

[0815] Example 2: Relaxed Route Guide

[0816] The server calculates the optimal detour route based on the latest traffic congestion information. At the same time, an emotion analysis tool analyzes the driver's fatigue level and suggests a less stressful route. The device receives the new route information and provides voice guidance such as, "To avoid traffic congestion, please turn right at the next intersection. There is also a rest area on this route."

[0817] Example prompts for generative AI models

[0818] Please explain in detail how "DriveGuard AI" collects sensor data, analyzes it on the server, and provides real-time alerts to the driver.

[0819] Please explain with concrete examples how the system analyzes the driver's emotional state and, based on the results, adjusts driving advice appropriately and in the right place.

[0820] This system, with the above configuration, provides a mechanism that comprehensively supports the driver in ensuring safe and comfortable driving.

[0821] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0822] Step 1:

[0823] The terminal (on-board device) acquires video data and environmental data in real time using the on-board camera and various sensors (LiDAR, radar, GPS). The acquired data is preprocessed by the edge computing unit after undergoing noise filtering. After preprocessing is complete, the data is sent from the terminal to the server. The inputs are camera video data and sensor data, and the output is the preprocessed data.

[0824] Specific examples of behavior:

[0825] The device activates the onboard camera to capture images of the road ahead, while the LiDAR and radar simultaneously acquire the position and speed of surrounding objects, and this data is then filtered by the edge computing unit.

[0826] Step 2:

[0827] The server receives video and sensor data sent from the device. As input, it receives preprocessed data. The received data is preprocessed through noise filtering and format normalization. The preprocessed data is ready to be analyzed by the machine learning model.

[0828] Specific examples of behavior:

[0829] The server receives the data stream, corrects dark areas, removes noise, and unifies the format through filtering, converting it into the appropriate data format.

[0830] Step 3:

[0831] The server uses machine learning models to analyze the noise-filtered and format-normalized data. The input is the preprocessed data. The machine learning models are used to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel. The output is the object recognition results and their location information.

[0832] Specific examples of behavior:

[0833] A machine learning algorithm scans the video data and recognizes pedestrians ahead, calculating their location and direction of travel as a result.

[0834] Step 4:

[0835] The server generates alerts and safe driving advice for the driver based on the analysis results. The input is the analysis results, and the output is an alert message or advice. The generated information is sent to the terminal.

[0836] Specific examples of behavior:

[0837] The server generates an alert message saying "There is a pedestrian ahead. Please slow down." and sends it to the terminal.

[0838] Step 5:

[0839] The terminal (in-vehicle device) displays the alerts and advice received from the server on the display and notifies the driver by voice. The input is the alert message from the server, and the output is the display and voice notification.

[0840] Specific examples of behavior:

[0841] The device will play a voice message saying, "There is a pedestrian ahead. Be careful." and will also show a warning message on the display.

[0842] Step 6:

[0843] The device receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The input is the driver's voice command, and the output is the voice command converted into text. Processing is performed based on the voice command converted into text.

[0844] Specific examples of behavior:

[0845] The driver enters the voice command "recalculate route," and a speech recognition engine converts the command into text.

[0846] Step 7:

[0847] The server receives voice commands converted into text and calculates the optimal route based on traffic congestion and road condition data. The inputs are voice commands and traffic information, and the output is the calculated optimal route information. The calculated route information is sent to the device.

[0848] Specific examples of behavior:

[0849] The server collects the latest traffic congestion information, calculates a new detour route based on that information, and sends the results to the terminal.

[0850] Step 8:

[0851] The terminal provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information from the server, and the output is navigation instructions.

[0852] Specific examples of behavior:

[0853] The device will provide voice guidance such as "Turn left at the next intersection," and will display a map and route on the screen.

[0854] Step 9:

[0855] The terminal collects the driver's voice and facial expression data through a microphone and an on-board camera. The input is the driver's voice and video data, and the output is the collected data sent to the server.

[0856] Specific examples of behavior:

[0857] The device captures the driver's facial expressions and tone of voice in real time and transmits the data to a server.

[0858] Step 10:

[0859] The server analyzes the transmitted voice and facial expression data to determine the driver's emotional state. The input is the driver's voice and facial expression data, and the output is the analysis result of the emotional state. Based on the analysis result, driving advice and safety suggestions are adjusted and sent to the terminal.

[0860] Specific examples of behavior:

[0861] The server recognizes signs of stress from the driver's tone of voice and facial expression, generates advice such as "Relax," and sends it to the terminal.

[0862] Through these processing steps, the "DriveGuard AI" system provides a mechanism to comprehensively support the driver's safety and comfort.

[0863] (Application example 2)

[0864] 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."

[0865] Conventional building security systems are limited to simple monitoring and warning using surveillance cameras and sensors, and lack the ability to detect suspicious individuals or unusual events, nor the countermeasures for their detection. Furthermore, no consideration was given to supporting security personnel in dealing with overwork and stressful environments, making it difficult to maintain efficient security. Therefore, it is necessary to reduce the mental stress of security personnel, along with early detection of suspicious individuals and appropriate countermeasures.

[0866] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. This enables the prompt detection of suspicious individuals and abnormalities, the appropriate countermeasures, and the reduction of mental stress for security personnel.

[0867] The "data collection means" is a means for collecting video data and environmental data in real time from monitoring devices and various sensors.

[0868] "Data analysis means" refers to a means of analyzing collected data using a machine learning model to detect suspicious individuals or abnormalities.

[0869] The "analysis result notification means" is a means for notifying the security officer of appropriate warnings and instructions based on the analysis results obtained by the data analysis means.

[0870] The "voice command recognition means" refers to a means for recognizing and analyzing voice commands from security personnel and performing corresponding functions in accordance with the instructions.

[0871] The "route calculation means" is a means for calculating the optimal route based on route information and situational data inside and outside the building, and predicting the route and escape route of a detected suspicious person.

[0872] The "navigation means" is a means for providing navigation to security personnel using voice and a display based on the calculated optimum route information.

[0873] The "emotion analysis engine" is a means of analyzing the voice and facial expression data of security personnel obtained through microphones and cameras to determine their mental stress and emotional state.

[0874] "Security notification means" is a means of notifying appropriate warnings and safety suggestions based on the results of detecting suspicious individuals or abnormalities and the emotional state of the person in charge.

[0875] The present invention relates to a "SecureGuard AI" system that improves building security and safety. The system includes multiple elements: a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. Specific embodiments for implementing the system are described in detail below.

[0876] Data collection and analysis

[0877] Data collection methods

[0878] Terminals (smartphones and robots) are equipped with monitoring devices and sensors that collect real-time video and environmental data from inside and outside the building. The collected data is then sent to a server via an edge computing unit. Specific hardware used includes high-resolution cameras, microphones, motion sensors, etc.

[0879] Data Analysis Methods

[0880] The server receives the collected video and sensor data. The data is preprocessed through noise filtering and format normalization, and then analyzed using analysis algorithms and machine learning models (e.g., the dlib library and open-source sentiment analysis models). This allows for the detection of suspicious individuals and unusual behavior, and the extraction of their location information and behavioral patterns.

[0881] Notices and Instructions

[0882] Analysis result notification means

[0883] The server generates risk information based on the analysis results obtained by the data analysis means, and issues alerts and safety measures to the operator, which are then sent to the terminal.

[0884] Voice command recognition means

[0885] The terminal receives voice commands from the operator through a microphone, analyzes these commands with a speech recognition engine (e.g., Google Speech-to-Text API), converts them into text, and performs the corresponding processing.

[0886] Route calculation and navigation

[0887] Route calculation method

[0888] The server runs algorithms that calculate optimal routes based on route information and situational data inside and outside the building, such as routes that bypass areas where suspicious individuals may be present, using up-to-date map data and real-time situational information.

[0889] Navigation methods

[0890] Based on the optimal route information received from the server, the terminal provides navigation to the operator by voice and display, for example, by providing instructions such as "Turn left at the next corridor."

[0891] Sentiment Analysis Engine

[0892] The terminal collects the operator's voice and facial expression data through a microphone and camera. The server analyzes the collected voice and facial expression data to determine the operator's emotional state (stress, fatigue, anger, etc.). This enables the system to take measures to reduce the operator's mental stress.

[0893] Specific examples

[0894] Example 1: Detecting and responding to suspicious individuals

[0895] The terminal transmits images captured by the surveillance cameras inside the building to the server.

[0896] The server analyzes the video and detects suspicious individuals.

[0897] The server generates a warning such as "There is a suspicious person near the entrance. Please be on guard" and sends it to the terminal.

[0898] Example 2: Controlling a surveillance camera with voice commands

[0899] The terminal receives a voice command from the operator: "Show surveillance camera 1."

[0900] A voice recognition engine converts commands into text and displays the image from the corresponding camera.

[0901] Prompt Sentence Examples

[0902] "A suspicious person has been detected near the entrance. Please implement security measures."

[0903] "Do you want to view the footage from security camera 1?"

[0904] By combining these elements, the SecureGuard AI system can significantly improve building security and provide a safe and efficient environment.

[0905] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0906] Step 1:

[0907] The terminal collects video data and environmental data from monitoring devices and sensors. The collected data is sent to a server via an edge computing unit. The input is raw data from the monitoring devices and sensors, and the output is pre-processed data. Specifically, the high-resolution camera continuously captures video, and the sensors detect temperature and movement.

[0908] Step 2:

[0909] The server receives the transmitted video data and environmental data and performs preprocessing through noise filtering and format normalization. The input is the video data and sensor data transmitted from the device, and the output is the noise-removed and normalized data. Specifically, the filtering algorithm removes noise from the video and standardizes the data format.

[0910] Step 3:

[0911] The server analyzes the preprocessed data using analysis algorithms and machine learning models (for example, dlib or open-source emotion analysis models) to detect suspicious individuals or unusual behavior. The input is noise-removed and normalized data, and the output is information about suspicious individuals or unusual behavior. Specifically, the machine learning model recognizes faces and behavioral patterns and detects suspicious behavior.

[0912] Step 4:

[0913] The server generates danger information based on the detection results, creates appropriate warnings and instructions, and sends them to the terminal. The input is information about suspicious individuals and unusual events, and the output is warning messages and instructions. Specifically, when a suspicious individual is detected, a warning message saying "There is a suspicious individual near the entrance. Please be on guard" is generated and sent to the terminal.

[0914] Step 5:

[0915] The terminal notifies the operator of warning messages and instructions received from the server by voice and on the display. The input is the warning message or instruction sent from the server, and the output is a voice notification and a display. Specifically, the speech synthesis engine reads out the warning message and displays it on the display.

[0916] Step 6:

[0917] The device receives the operator's voice commands through the microphone and analyzes them using a voice recognition engine (for example, Google Speech-to-Text API). The input is the operator's voice command, and the output is a text command. Specifically, when the operator says, "Show surveillance camera 1," the voice is converted into text.

[0918] Step 7:

[0919] The server executes the corresponding process based on the voice command. For example, it sends an instruction to the terminal to display the video from security camera 1. The input is a text-converted voice command, and the output is an instruction for the processing procedure. Specifically, in response to the command "Display security camera 1," the video from the corresponding camera is sent to the terminal.

[0920] Step 8:

[0921] The terminal navigates based on the route information received from the server. The input is the route information sent from the server, and the output is navigation instructions to the operator. Specific actions include a voice prompt saying "Turn left at the next corridor" and a map being displayed on the screen.

[0922] Step 9:

[0923] The device collects the operator's voice and facial expression data through a microphone and camera, and the emotion analysis engine analyzes it. The input is the operator's voice and facial expression data, and the output is the operator's emotional state information. Specifically, if the operator is feeling stressed, that state will be displayed as the analysis result.

[0924] Step 10:

[0925] Based on the results of the emotion analysis engine, the server takes action that matches the emotional state of the operator. For example, it changes the tone of the voice notification to reduce stress. The input is the operator's emotional state information, and the output is instructions on how to respond. A specific action would be to notify the operator in a relaxed tone, saying, "It is safe, so please stay calm and take action."

[0926] The above is the flow of the main processing steps of the "SecureGuard AI" system.

[0927] 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.

[0928] 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.

[0929] 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.

[0930] [Third embodiment]

[0931] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0932] 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.

[0933] 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).

[0934] 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.

[0935] 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.

[0936] 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).

[0937] 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.

[0938] 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.

[0939] 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.

[0940] 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.

[0941] 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.

[0942] 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."

[0943] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system are described below.

[0944] System Overview

[0945] The DriveGuard AI system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[0946] Data collection and analysis

[0947] Data collection methods

[0948] Terminal (in-vehicle device)

[0949] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[0950] The acquired data is sent to a server via the edge computing unit.

[0951] Data Analysis Methods

[0952] server

[0953] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[0954] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[0955] Notices and Instructions

[0956] Analysis result notification means

[0957] server

[0958] Based on the analysis results, risk information is generated and alerts and advice are created for drivers.

[0959] This information is sent to the terminal.

[0960] Terminal (in-vehicle device)

[0961] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[0962] Voice commands and route guidance

[0963] Voice command recognition means

[0964] Terminal (in-vehicle device)

[0965] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[0966] Route calculation method

[0967] server

[0968] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[0969] Navigation methods

[0970] Terminal (in-vehicle device)

[0971] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[0972] Specific examples

[0973] Example 1: Danger notification

[0974] Terminal (in-vehicle device)

[0975] The onboard camera detects pedestrians ahead and sends that information to a server.

[0976] server

[0977] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[0978] Terminal (in-vehicle device)

[0979] A voice notification will be played saying "There is a pedestrian ahead. Be careful." and a warning will also be displayed on the display.

[0980] Example 2: Traffic jam avoidance route guide

[0981] server

[0982] Calculates the optimal detour route based on the latest traffic congestion information.

[0983] Terminal (in-vehicle device)

[0984] It receives new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[0985] In this way, the DriveGuard AI system predicts hazards in real time and supports safe driving by collecting, analyzing, notifying, and processing the driver's voice commands.

[0986] The processing flow will be explained below.

[0987] Step 1:

[0988] The terminal (in-vehicle device) collects video data and environmental data in real time using on-board cameras and sensors. The on-board cameras capture images of the front and surrounding area, and the sensors acquire information from LiDAR, radar, GPS, etc.

[0989] Step 2:

[0990] The terminal (in-vehicle device) sends the collected data to a server via the edge computing unit, where the data is transmitted quickly, encrypted, and securely.

[0991] Step 3:

[0992] The server receives the data sent from the terminal and pre-processes it through noise filtering and format normalization, thereby improving the quality of the data.

[0993] Step 4:

[0994] The server analyzes the data using sophisticated image processing algorithms and machine learning models to recognize objects such as pedestrians, cyclists, and motorbikes, and calculates the position, speed, and direction of each object.

[0995] Step 5:

[0996] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[0997] Step 6:

[0998] The server sends the generated alert information and safe driving advice to the terminal, so that the driver can be notified in a timely manner.

[0999] Step 7:

[1000] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it may play a voice message saying, "There is a pedestrian ahead. Please be careful."

[1001] Step 8:

[1002] The user (driver) inputs voice commands through a microphone, such as setting a destination or requesting a new route.

[1003] Step 9:

[1004] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[1005] Step 10:

[1006] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion information and road condition data, and the calculation results are sent back to the terminal.

[1007] Step 11:

[1008] The terminal (in-vehicle device) receives the optimal route information from the server and provides navigation to the driver by voice and display, for example, "Turn right at the next intersection."

[1009] Example 1

[1010] 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."

[1011] Conventional driving assistance systems have problems in that they do not adequately provide drivers with appropriate information or provide navigation based on voice commands. Furthermore, because there is no consistent processing from data collection to analysis, notification, and navigation, it is difficult to provide sufficient functionality to support safe driving. To solve this problem, there is a need for real-time data collection and analysis, appropriate hazard notification, and rapid navigation based on voice commands.

[1012] 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.

[1013] In this invention, the server includes a data preprocessing means, a data analysis means, and an analysis result notification means, which allows for quick and accurate preprocessing and analysis of data collected in real time, and provides appropriate alerts and navigation instructions to the driver.

[1014] "Data collection means" means for collecting video and environmental data in real time using digital cameras and sensing devices.

[1015] "Data pre-processing means" refers to means for performing noise filtering and format normalization on collected data.

[1016] The "data analysis means" is a means for analyzing collected and preprocessed data using a machine learning model to extract information about objects and their locations.

[1017] The "analysis result notification means" is a means for generating danger information based on the analysis results and providing alerts and advice to the driver.

[1018] The "voice command recognition means" is a means for converting a user's voice command into text and performing processing in accordance with the content of the text.

[1019] The "route calculation means" is a means for calculating the optimum route based on the latest traffic information.

[1020] "Navigation means" refers to a means for providing audio and visual navigation to the driver based on the calculated optimum route information.

[1021] The present invention relates to a system for supporting safe and comfortable driving for drivers, and this system includes a data collection means, a data preprocessing means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system will be described below.

[1022] System Overview

[1023] This system collects data using cameras and various sensors mounted on the vehicle, preprocesses and analyzes the data on a server, and provides appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[1024] Data Acquisition and Transmission

[1025] Terminal (in-vehicle device)

[1026] The device collects data using a digital camera and various detection devices (LiDAR, radar, GPS, etc.), thereby obtaining real-time image data of the front and surrounding area and environmental data. The acquired data is sent to a server via an edge computing unit.

[1027] Data preprocessing and analysis

[1028] server

[1029] The server receives video data and environmental data sent from the device. First, noise filtering and format normalization are performed by the data preprocessing means. Next, the collected and preprocessed data is analyzed by the data analysis means using a machine learning model to extract objects and their location information. For example, pedestrians, bicycles, motorbikes, etc. are recognized, and their speed and direction of movement are estimated.

[1030] Notices and Instructions

[1031] server

[1032] The server generates danger information based on the analysis results and includes an analysis result notification means for providing alerts and advice to the driver. For example, a warning message such as "There is a pedestrian ahead. Please slow down" is generated and sent to the terminal.

[1033] Terminal (in-vehicle device)

[1034] The device receives the danger information and advice sent from the server, and displays it on the display and notifies the driver by voice, for example, "There is a pedestrian ahead. Be careful."

[1035] Voice commands and route guidance

[1036] Terminal (in-vehicle device)

[1037] The device receives the user's voice commands through a microphone, analyzes them with a voice recognition engine, converts the voice commands into text, and processes them accordingly.

[1038] server

[1039] The server includes a route calculation means for calculating the optimum route based on voice commands, and calculates the optimum route based on traffic congestion information and road condition data, and transmits the calculated route to the terminal.

[1040] Terminal (in-vehicle device)

[1041] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[1042] Specific examples

[1043] Example 1: Danger notification

[1044] The terminal (on-board device) detects pedestrians ahead using the on-board camera and sends the data to the server.

[1045] The server uses data analysis means to recognize the pedestrian's location and direction of travel, and if it determines that there is danger, it generates an alert message and sends it to the terminal.

[1046] The terminal (in-vehicle device) will play a voice notification saying, "There is a pedestrian ahead. Please be careful," and will also display a warning on the display.

[1047] Example 2: Traffic jam avoidance route guide

[1048] The server calculates the optimal detour route based on the latest traffic congestion information.

[1049] The terminal (in-vehicle device) receives the new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[1050] Prompt Sentence Examples

[1051] Prompt 1: "When an in-car camera detects a pedestrian ahead, how is the data processed?"

[1052] Prompt 2: "Describe the process by which the optimal route is calculated and how it is communicated to the driver."

[1053] As described above, the present invention is a system that supports safe driving by realizing consistent processing from data acquisition, preprocessing, analysis, notification, and route guidance in response to voice commands.

[1054] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1055] Step 1:

[1056] Acquiring and Sending Data

[1057] The terminal (on-board device) uses a digital camera and sensors (LiDAR, radar, GPS, etc.) to acquire front and surrounding image data and environmental data in real time. Specifically, the on-board camera captures images of the front, and the GPS acquires current location information. This data is temporarily stored through the edge computing unit and sent to a server. The input is camera and sensor data, and the output is data sent to the server.

[1058] Step 2:

[1059] Data Preprocessing

[1060] The server receives data sent from the terminal. Data preprocessing means performs noise filtering and format normalization. Specific operations include removing outliers and missing values, and adjusting the resolution of image data. The input is raw data received from the terminal, and the output is preprocessed data.

[1061] Step 3:

[1062] Data analysis

[1063] The server analyzes the preprocessed data using data analysis means. Here, a machine learning model is used to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information, speed, direction of travel, etc. Specifically, an image recognition algorithm detects the objects and calculates their coordinate data. The input is the preprocessed data, and the output is the analysis results (object type, position, speed, etc.).

[1064] Step 4:

[1065] Hazard information generation and notification

[1066] The server generates danger information based on the analysis results. The analysis result notification means creates a message to notify the driver of the generated danger information and sends it to the terminal. Specifically, a warning message such as "There is a pedestrian ahead. Please slow down" is generated. The input is the analysis results, and the output is the notification message.

[1067] The terminal (in-vehicle device) receives danger information and advice sent from the server, displays it on the display, and notifies the driver by voice. The input is the notification message received from the server, and the output is warning information by voice and display. Specifically, the terminal displays a warning message on the in-vehicle display and plays a voice notification from the speaker.

[1068] Step 5:

[1069] Receiving and processing voice commands

[1070] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them using a voice recognition engine. The voice commands are converted into text and processing is carried out based on the content. The input is the driver's voice command and the output is the textual command content. Specifically, the voice command "Please tell me the best route" is analyzed.

[1071] Step 6:

[1072] Calculating the best route

[1073] The server receives data for calculating the optimal route based on the voice command. The route calculation means calculates the optimal route based on traffic congestion information and road condition data. The input is the voice command and the latest traffic information, and the output is the calculated optimal route information. Specifically, the system uses a traffic information API to obtain real-time traffic congestion information and calculates the optimal route using an algorithm.

[1074] Step 7:

[1075] Providing navigation

[1076] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information received from the server, and the output is navigation instructions to the driver. Specific operations include voice guidance such as "Turn left at the next intersection" and a map display.

[1077] (Application example 1)

[1078] 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."

[1079] Current autonomous vehicles are required to collect and analyze environmental data in real time and make appropriate decisions to ensure safe driving. However, existing systems lack effective notification methods for the driver and the autonomous driving system, preventing accident risk reduction and efficient route guidance. In particular, the limited use of prompts using voice command recognition and generative AI models poses a challenge for further improving driver convenience and safety.

[1080] 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.

[1081] In this invention, the server includes a data analysis means, a prompt generation means, and a notification means using a generative AI model. This allows the server to analyze environmental data in real time and provide appropriate alerts and route guidance prompts to the driver and the automated driving system using the generative AI model. This reduces the risk of traffic accidents and provides efficient driving assistance.

[1082] The "data collection means" is a means for collecting video data and environmental data in real time from on-board cameras and sensors.

[1083] The "data analysis means" is a means of analyzing collected data using a machine learning model to extract recognition and location information for pedestrians, bicycles, motorbikes, etc.

[1084] The "analysis result notification means" is a means for notifying the driver or the automated driving system of the analysis results obtained by the data analysis means.

[1085] The "voice command recognition means" is a means for recognizing a voice command from the driver and analyzing the content of the command.

[1086] The "route calculation means" is a means for calculating the optimum route based on information on the current position and destination, as well as traffic congestion information and road condition data.

[1087] "Navigation means" refers to a means for providing voice and display navigation to the driver or the automated driving system based on the calculated optimal route information.

[1088] The "prompt generating means" is a means for generating an appropriate instruction or alert prompt in response to an input voice command or a recognized situation.

[1089] The "prompt processing means" is a means for performing processing to appropriately notify the driver of the generated prompt.

[1090] "Notification means using a generative AI model" refers to a means of generating appropriate notification content from data analyzed using a generative AI model and notifying the driver or automated driving system in real time.

[1091] This invention aims to realize a safe driving support system for autonomous vehicles called "AutoNav AI." This system includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, a prompt generation means, a prompt processing means, and a notification means using a generative AI model.

[1092] System Overview

[1093] The data collection method uses on-board cameras and sensors to collect video data and environmental data in real time. Specific hardware used includes on-board cameras such as the Logitech C920, as well as LiDAR, radar, and GPS sensors. This data is sent to a server via an edge computing unit.

[1094] The data analysis means receives the video data and environmental data collected on the server and analyzes them using machine learning models (e.g., OpenCV and YOLO). The data undergoes pre-processing such as noise filtering and format normalization, and then uses analysis algorithms to recognize objects such as pedestrians, bicycles, and motorbikes and extract their location information.

[1095] The analysis result notification means generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. These notifications are sent to the terminal in real time via the edge computing unit and are displayed as voice and on a display.

[1096] The voice command recognition means receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine (e.g., Google's SpeechRecognition API). The recognized voice commands are converted into text and processed accordingly.

[1097] The route calculation means calculates the optimal route based on the current location and destination information, as well as traffic congestion and road condition data. Specifically, route calculation is performed using Google Maps API, etc. The calculated route information is sent from the server to the terminal.

[1098] The navigation means provides navigation to the driver or the automated driving system by voice and display based on the optimal route information received by the terminal from the server, for example, by providing guidance such as "Turn left at the next intersection."

[1099] The prompt generator generates appropriate instruction or alert prompts based on the input voice command and the recognized situation, using a generative AI model.

[1100] The prompt processing means performs processing to appropriately notify the driver of the generated prompt, thereby providing instructions and alerts to the driver at the appropriate time.

[1101] The notification method using the generative AI model generates appropriate notification content from data analyzed using the generative AI model and notifies the driver or autonomous driving system in real time.

[1102] Specific examples

[1103] Example 1: Traffic jam avoidance route guide

[1104] The server calculates the optimal detour route based on the latest traffic congestion information. The device receives the new route information and provides voice guidance such as, "Turn right at the next intersection to avoid traffic congestion." This supports efficient driving.

[1105] Example 2: Voice command recognition and prompt generation

[1106] The driver issues a voice command saying, "Calculate a safe and efficient route based on the latest traffic congestion information. End point: Shinjuku Station." In response, the server calculates the optimal route and generates a prompt saying, "The optimal route from your current location to Shinjuku Station is to turn right at the next intersection and go straight at the next traffic light. The route takes approximately 30 minutes. There is an accident warning on the way, so please be careful."

[1107] Thus, the present invention provides specific embodiments for significantly improving the safety and efficiency of autonomous vehicles.

[1108] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1109] Step 1:

[1110] Data collection

[1111] The onboard camera and various sensors installed in the user's vehicle collect video data and environmental data in real time. Specifically, the camera captures the surrounding image, and the LiDAR, radar, and GPS sensors acquire various environmental data. This data is sent to a server via an edge computing unit.

[1112] Input: Video data and environmental data from on-board cameras and sensors

[1113] Output: Data sent to the server through the edge computing unit

[1114] Step 2:

[1115] Data analysis

[1116] The server receives the collected video and environmental data and performs preprocessing: the data undergoes noise filtering and format normalization, and is then analyzed using machine learning models (OpenCV and YOLO, for example) to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information.

[1117] Input: Video data and environmental data collected in step 1

[1118] Output: Objects and their locations

[1119] Step 3:

[1120] Notification of analysis results

[1121] The server generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. The generated notifications are sent to the terminal via the edge computing unit and displayed by voice and on the display. For example, if there is a pedestrian ahead, the notification will say, "There is a pedestrian ahead. Please be careful."

[1122] Input: Analysis results obtained by data analysis tools

[1123] Output: Alerts and advice sent to the device

[1124] Step 4:

[1125] Voice Command Recognition

[1126] Voice commands issued by users (e.g., "End point: Shinjuku Station") are collected through microphones in the vehicle and converted into text by a speech recognition engine (e.g., Google's SpeechRecognition API). This text data is sent to a server and analyzed as a voice command.

[1127] Input: A voice command spoken by the user

[1128] Output: Text data sent to the server

[1129] Step 5:

[1130] Route calculation

[1131] The server calculates the optimal route based on the analyzed voice commands, current location information, traffic information, and road condition data. This calculation uses route calculation services such as Google Maps API. The calculated route information is sent from the server to the device.

[1132] Input: Voice commands, current location, traffic information, road condition data

[1133] Output: Calculated optimal route information

[1134] Step 6:

[1135] Navigation

[1136] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[1137] Input: Optimal route information received from the server

[1138] Output: Voice and display navigation instructions

[1139] Step 7:

[1140] Prompt Generation

[1141] The server generates appropriate prompts, such as instructions or alerts, based on the input voice command and the recognized situation. A generative AI model is used to generate these prompts. A specific example of a prompt is, "Turn right at the next intersection to avoid traffic jams."

[1142] Input: Voice commands, recognized situations

[1143] Output: Generated prompt

[1144] Step 8:

[1145] Prompt Processing

[1146] The server then processes the generated prompt to appropriately notify the driver. The notification is given via voice and display. For example, the driver may be notified, "Turn right at the next intersection to avoid traffic jams."

[1147] Input: Generated prompt

[1148] Output: Audio and visual notifications

[1149] 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.

[1150] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion engine. Specific embodiments for implementing this system are described below.

[1151] System Overview

[1152] The "DriveGuard AI" system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate. Furthermore, it combines an emotion engine that analyzes the driver's emotional state from their voice and facial expressions to provide appropriate support according to the driver's emotional state.

[1153] Data collection and analysis

[1154] Data collection methods

[1155] Terminal (in-vehicle device)

[1156] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[1157] The acquired data is sent to a server via the edge computing unit.

[1158] Data Analysis Methods

[1159] server

[1160] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[1161] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[1162] Notices and Instructions

[1163] Analysis result notification means

[1164] server

[1165] Based on the analysis results, risk information is generated and alerts and safe driving advice are created for drivers.

[1166] This information is sent to the terminal.

[1167] Terminal (in-vehicle device)

[1168] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[1169] Voice commands and route guidance

[1170] Voice command recognition means

[1171] Terminal (in-vehicle device)

[1172] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[1173] Route calculation method

[1174] server

[1175] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[1176] Navigation methods

[1177] Terminal (in-vehicle device)

[1178] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[1179] Emotion Engine

[1180] Emotion Engine

[1181] Data collection

[1182] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera.

[1183] Data analysis

[1184] The server analyzes the collected voice and facial expression data to determine the emotional state (e.g., stress, fatigue, anger, etc.).

[1185] Notices and Instructions

[1186] Based on the analysis results of the emotion engine, the server adjusts driving advice and safety suggestions and sends them to the device.

[1187] Specific examples

[1188] Example 1: Danger notification and emotional response

[1189] Terminal (in-vehicle device)

[1190] The onboard camera detects pedestrians ahead and sends that information to a server.

[1191] server

[1192] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[1193] At the same time, an emotion engine analyzes the driver's emotional state (e.g., impatience) and adjusts the tone of the alert as needed.

[1194] Terminal (in-vehicle device)

[1195] The system plays a voice notification saying "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. It also notifies in a calmer tone depending on the driver's emotional state.

[1196] Example 2: Relaxed Route Guide

[1197] server

[1198] The system calculates the optimal detour route based on the latest traffic congestion information, while the emotion engine analyzes the driver's fatigue level and suggests a less stressful route.

[1199] Terminal (in-vehicle device)

[1200] The new route information is received and a voice guides you, saying, "To avoid traffic jams, turn right at the next intersection. There is also a rest area on this route."

[1201] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[1202] The processing flow will be explained below.

[1203] Step 1:

[1204] The terminal (in-vehicle device) collects video data and environmental data in real time using an in-vehicle camera and sensors (LiDAR, radar, GPS, etc.). The in-vehicle camera captures images of the road ahead and surroundings, and the sensors obtain the necessary location information and obstacle information.

[1205] Step 2:

[1206] The terminal (in-vehicle device) performs initial processing on the collected data using an edge computing unit, compresses and encrypts it, and sends it to the server, enabling high-speed and secure data communication.

[1207] Step 3:

[1208] The server receives the data sent by the terminal and preprocesses the data by performing noise filtering and format normalization, which ensures the accuracy and consistency of the data.

[1209] Step 4:

[1210] The server analyzes the data using high-performance image processing algorithms and machine learning models, recognizing pedestrians, cyclists, motorbikes, and other objects, and calculating the position, speed, and direction of each object.

[1211] Step 5:

[1212] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[1213] Step 6:

[1214] The server sends the generated alert information and safe driving advice to the terminal, which is then ready to notify the driver in a timely manner.

[1215] Step 7:

[1216] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it conveys a message to the driver saying, "There is a pedestrian ahead. Please be careful."

[1217] Step 8:

[1218] The terminal (in-vehicle device) collects the driver's voice and facial expression data using a microphone and an in-vehicle camera, making it possible to grasp the driver's emotional state in real time.

[1219] Step 9:

[1220] The terminal (in-vehicle device) performs initial processing of the collected voice and facial expression data using an edge computing unit, compresses and encrypts it, and sends it to the server.

[1221] Step 10:

[1222] The server receives the voice and facial expression data sent from the terminal and analyzes the driver's emotional state (e.g., stress, fatigue, anger, etc.) using emotion recognition algorithms.

[1223] Step 11:

[1224] The server then adjusts driving advice and safety suggestions based on the emotion engine's analysis. For example, if the driver is feeling anxious, the server will adjust the tone of the notification to a calmer tone and generate a message encouraging more careful driving.

[1225] Step 12:

[1226] The server sends alert information and advice adjusted by the emotion engine to the terminal.

[1227] Step 13:

[1228] The terminal (in-vehicle device) receives the adjusted alert information and advice from the server and notifies the driver by voice and display, for example, a message saying, "There is a pedestrian ahead. Please proceed with caution and remain calm."

[1229] Step 14:

[1230] The user (driver) inputs voice commands through a microphone, such as setting a destination or changing a route.

[1231] Step 15:

[1232] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[1233] Step 16:

[1234] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion and road condition data, and the calculation results are then sent back to the device.

[1235] Step 17:

[1236] The terminal (in-vehicle device) receives optimal route information from the server and provides navigation to the driver by voice and display. For example, it may provide guidance such as, "Turn right at the next intersection. Also, you seem tired, so let's stop at a nearby rest area."

[1237] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[1238] Example 2

[1239] 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."

[1240] In today's automobile driving, it is important to ensure both driver safety and comfort. However, conventional driver assistance systems have difficulty adapting to changes in the driving environment and the driver's emotional state, and excessive information provision and inappropriate alerts can actually induce stress. Furthermore, there is a lack of technology to effectively combine real-time environmental data analysis and emotion analysis, resulting in situations where optimal driver assistance cannot be achieved.

[1241] 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 data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and an emotion analysis means. This makes it possible to quickly respond to changes in the driving environment and provide appropriate alerts and driving advice based on the driver's emotional state.

[1242] The "data collection means" is a component of the system that acquires video data and environmental data in real time using on-board cameras and sensors and transmits them to a server via an edge computing unit.

[1243] The "data analysis means" is a component of the system that analyzes collected data using a machine learning model after preprocessing such as noise filtering and format normalization to extract object recognition and location information.

[1244] The "analysis result notification means" is a system component that generates alerts and safe driving advice for the driver based on the analysis results obtained by the data analysis means and transmits them to the in-vehicle device.

[1245] The "voice command recognition means" is a component of the system that receives the driver's voice commands through a microphone, converts them into text using a voice recognition engine, and performs processing based on the content of the text.

[1246] The "route calculation means" is a system component that executes an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and transmits the results to the in-vehicle device.

[1247] The "navigation means" is a component of the system that provides navigation to the driver through voice and display based on the received optimum route information.

[1248] The "emotion analysis means" is a component of the system that analyzes the driver's voice and facial expression data collected through the vehicle-mounted camera and microphone, and determines the driver's emotional state.

[1249] The present invention relates to a system for supporting safe and comfortable driving of a driver, which includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion analysis means.

[1250] Data collection method configuration

[1251] The terminal (in-vehicle device) is equipped with an in-vehicle camera and various sensors (LiDAR, radar, GPS). The in-vehicle camera captures images of the front and surroundings in real time, and the sensors acquire environmental data. This data is processed by the edge computing unit and preprocessed through noise filtering. The preprocessed data is then sent to the server.

[1252] Data analysis method configuration

[1253] The server receives the video and sensor data sent from the device and performs pre-processing on the data. The pre-processed data is then analyzed using machine learning models to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel.

[1254] Configuration of analysis result notification method

[1255] The server generates risk information based on the analysis results and creates alerts and safe driving advice for the driver. These notifications are sent to the device, which displays the risk information and advice received from the server on its display and notifies the driver by voice. For example, it may notify the driver, "There is a pedestrian ahead. Please slow down."

[1256] Configuration of voice command recognition means

[1257] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The voice commands are converted into text and processed accordingly.

[1258] Configuring route calculation methods

[1259] The server runs an algorithm to calculate the optimal route based on traffic congestion and road condition data, and the calculated route information is sent to the device.

[1260] Navigation Method Configuration

[1261] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server, for example, by providing instructions such as "Turn left at the next intersection."

[1262] Sentiment analysis method configuration

[1263] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera. The server analyzes the collected voice and facial expression data to determine the driver's emotional state (e.g., stress, fatigue, anger, etc.). The server adjusts driving advice and safety suggestions based on the results of the emotion analysis and sends them to the terminal.

[1264] Specific examples

[1265] Example 1: Danger notification and emotional response

[1266] The terminal (in-vehicle device) uses an on-board camera to detect pedestrians ahead and sends that information to a server. The server uses an image analysis algorithm to identify the pedestrian's location and direction of travel, and calculates their speed and distance. If it determines that there is a danger, it generates an alert message and sends it to the terminal. Furthermore, an emotion analysis means analyzes the driver's emotional state and adjusts the tone of the alert as necessary. The terminal plays a voice notification saying, "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. The server also uses a calmer tone of voice depending on the driver's emotional state.

[1267] Example 2: Relaxed Route Guide

[1268] The server calculates the optimal detour route based on the latest traffic congestion information. At the same time, an emotion analysis tool analyzes the driver's fatigue level and suggests a less stressful route. The device receives the new route information and provides voice guidance such as, "To avoid traffic congestion, please turn right at the next intersection. There is also a rest area on this route."

[1269] Example prompts for generative AI models

[1270] Please explain in detail how "DriveGuard AI" collects sensor data, analyzes it on the server, and provides real-time alerts to the driver.

[1271] Please explain with concrete examples how the system analyzes the driver's emotional state and, based on the results, adjusts driving advice appropriately and in the right place.

[1272] This system, with the above configuration, provides a mechanism that comprehensively supports the driver in ensuring safe and comfortable driving.

[1273] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1274] Step 1:

[1275] The terminal (on-board device) acquires video data and environmental data in real time using the on-board camera and various sensors (LiDAR, radar, GPS). The acquired data is preprocessed by the edge computing unit after undergoing noise filtering. After preprocessing is complete, the data is sent from the terminal to the server. The inputs are camera video data and sensor data, and the output is the preprocessed data.

[1276] Specific examples of behavior:

[1277] The device activates the onboard camera to capture images of the road ahead, while the LiDAR and radar simultaneously acquire the position and speed of surrounding objects, and this data is then filtered by the edge computing unit.

[1278] Step 2:

[1279] The server receives video and sensor data sent from the device. As input, it receives preprocessed data. The received data is preprocessed through noise filtering and format normalization. The preprocessed data is ready to be analyzed by the machine learning model.

[1280] Specific examples of behavior:

[1281] The server receives the data stream, corrects dark areas, removes noise, and unifies the format through filtering, converting it into the appropriate data format.

[1282] Step 3:

[1283] The server uses machine learning models to analyze the noise-filtered and format-normalized data. The input is the preprocessed data. The machine learning models are used to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel. The output is the object recognition results and their location information.

[1284] Specific examples of behavior:

[1285] A machine learning algorithm scans the video data and recognizes pedestrians ahead, calculating their location and direction of travel as a result.

[1286] Step 4:

[1287] The server generates alerts and safe driving advice for the driver based on the analysis results. The input is the analysis results, and the output is an alert message or advice. The generated information is sent to the terminal.

[1288] Specific examples of behavior:

[1289] The server generates an alert message saying "There is a pedestrian ahead. Please slow down." and sends it to the terminal.

[1290] Step 5:

[1291] The terminal (in-vehicle device) displays the alerts and advice received from the server on the display and notifies the driver by voice. The input is the alert message from the server, and the output is the display and voice notification.

[1292] Specific examples of behavior:

[1293] The device will play a voice message saying, "There is a pedestrian ahead. Be careful." and will also show a warning message on the display.

[1294] Step 6:

[1295] The device receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The input is the driver's voice command, and the output is the voice command converted into text. Processing is performed based on the voice command converted into text.

[1296] Specific examples of behavior:

[1297] The driver enters the voice command "recalculate route," and a speech recognition engine converts the command into text.

[1298] Step 7:

[1299] The server receives voice commands converted into text and calculates the optimal route based on traffic congestion and road condition data. The inputs are voice commands and traffic information, and the output is the calculated optimal route information. The calculated route information is sent to the device.

[1300] Specific examples of behavior:

[1301] The server collects the latest traffic congestion information, calculates a new detour route based on that information, and sends the results to the terminal.

[1302] Step 8:

[1303] The terminal provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information from the server, and the output is navigation instructions.

[1304] Specific examples of behavior:

[1305] The device will provide voice guidance such as "Turn left at the next intersection," and will display a map and route on the screen.

[1306] Step 9:

[1307] The terminal collects the driver's voice and facial expression data through a microphone and an on-board camera. The input is the driver's voice and video data, and the output is the collected data sent to the server.

[1308] Specific examples of behavior:

[1309] The device captures the driver's facial expressions and tone of voice in real time and transmits the data to a server.

[1310] Step 10:

[1311] The server analyzes the transmitted voice and facial expression data to determine the driver's emotional state. The input is the driver's voice and facial expression data, and the output is the analysis result of the emotional state. Based on the analysis result, driving advice and safety suggestions are adjusted and sent to the terminal.

[1312] Specific examples of behavior:

[1313] The server recognizes signs of stress from the driver's tone of voice and facial expression, generates advice such as "Relax," and sends it to the terminal.

[1314] Through these processing steps, the "DriveGuard AI" system provides a mechanism to comprehensively support the driver's safety and comfort.

[1315] (Application example 2)

[1316] 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."

[1317] Conventional building security systems are limited to simple monitoring and warning using surveillance cameras and sensors, and lack the ability to detect suspicious individuals or unusual events, nor the countermeasures for their detection. Furthermore, no consideration was given to supporting security personnel in dealing with overwork and stressful environments, making it difficult to maintain efficient security. Therefore, it is necessary to reduce the mental stress of security personnel, along with early detection of suspicious individuals and appropriate countermeasures.

[1318] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. This enables the prompt detection of suspicious individuals and abnormalities, the appropriate countermeasures, and the reduction of mental stress for security personnel.

[1319] The "data collection means" is a means for collecting video data and environmental data in real time from monitoring devices and various sensors.

[1320] "Data analysis means" refers to a means of analyzing collected data using a machine learning model to detect suspicious individuals or abnormalities.

[1321] The "analysis result notification means" is a means for notifying the security officer of appropriate warnings and instructions based on the analysis results obtained by the data analysis means.

[1322] The "voice command recognition means" refers to a means for recognizing and analyzing voice commands from security personnel and performing corresponding functions in accordance with the instructions.

[1323] The "route calculation means" is a means for calculating the optimal route based on route information and situational data inside and outside the building, and predicting the route and escape route of a detected suspicious person.

[1324] The "navigation means" is a means for providing navigation to security personnel using voice and a display based on the calculated optimum route information.

[1325] The "emotion analysis engine" is a means of analyzing the voice and facial expression data of security personnel obtained through microphones and cameras to determine their mental stress and emotional state.

[1326] "Security notification means" is a means of notifying appropriate warnings and safety suggestions based on the results of detecting suspicious individuals or abnormalities and the emotional state of the person in charge.

[1327] The present invention relates to a "SecureGuard AI" system that improves building security and safety. The system includes multiple elements: a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. Specific embodiments for implementing the system are described in detail below.

[1328] Data collection and analysis

[1329] Data collection methods

[1330] Terminals (smartphones and robots) are equipped with monitoring devices and sensors that collect real-time video and environmental data from inside and outside the building. The collected data is then sent to a server via an edge computing unit. Specific hardware used includes high-resolution cameras, microphones, motion sensors, etc.

[1331] Data Analysis Methods

[1332] The server receives the collected video and sensor data. The data is preprocessed through noise filtering and format normalization, and then analyzed using analysis algorithms and machine learning models (e.g., the dlib library and open-source sentiment analysis models). This allows for the detection of suspicious individuals and unusual behavior, and the extraction of their location information and behavioral patterns.

[1333] Notices and Instructions

[1334] Analysis result notification means

[1335] The server generates risk information based on the analysis results obtained by the data analysis means, and issues alerts and safety measures to the operator, which are then sent to the terminal.

[1336] Voice command recognition means

[1337] The terminal receives voice commands from the operator through a microphone, analyzes these commands with a speech recognition engine (e.g., Google Speech-to-Text API), converts them into text, and performs the corresponding processing.

[1338] Route calculation and navigation

[1339] Route calculation method

[1340] The server runs algorithms that calculate optimal routes based on route information and situational data inside and outside the building, such as routes that bypass areas where suspicious individuals may be present, using up-to-date map data and real-time situational information.

[1341] Navigation methods

[1342] Based on the optimal route information received from the server, the terminal provides navigation to the operator by voice and display, for example, by providing instructions such as "Turn left at the next corridor."

[1343] Sentiment Analysis Engine

[1344] The terminal collects the operator's voice and facial expression data through a microphone and camera. The server analyzes the collected voice and facial expression data to determine the operator's emotional state (stress, fatigue, anger, etc.). This enables the system to take measures to reduce the operator's mental stress.

[1345] Specific examples

[1346] Example 1: Detecting and responding to suspicious individuals

[1347] The terminal transmits images captured by the surveillance cameras inside the building to the server.

[1348] The server analyzes the video and detects suspicious individuals.

[1349] The server generates a warning such as "There is a suspicious person near the entrance. Please be on guard" and sends it to the terminal.

[1350] Example 2: Controlling a surveillance camera with voice commands

[1351] The terminal receives a voice command from the operator: "Show surveillance camera 1."

[1352] A voice recognition engine converts commands into text and displays the image from the corresponding camera.

[1353] Prompt Sentence Examples

[1354] "A suspicious person has been detected near the entrance. Please implement security measures."

[1355] "Do you want to view the footage from security camera 1?"

[1356] By combining these elements, the SecureGuard AI system can significantly improve building security and provide a safe and efficient environment.

[1357] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1358] Step 1:

[1359] The terminal collects video data and environmental data from monitoring devices and sensors. The collected data is sent to a server via an edge computing unit. The input is raw data from the monitoring devices and sensors, and the output is pre-processed data. Specifically, the high-resolution camera continuously captures video, and the sensors detect temperature and movement.

[1360] Step 2:

[1361] The server receives the transmitted video data and environmental data and performs preprocessing through noise filtering and format normalization. The input is the video data and sensor data transmitted from the device, and the output is the noise-removed and normalized data. Specifically, the filtering algorithm removes noise from the video and standardizes the data format.

[1362] Step 3:

[1363] The server analyzes the preprocessed data using analysis algorithms and machine learning models (for example, dlib or open-source emotion analysis models) to detect suspicious individuals or unusual behavior. The input is noise-removed and normalized data, and the output is information about suspicious individuals or unusual behavior. Specifically, the machine learning model recognizes faces and behavioral patterns and detects suspicious behavior.

[1364] Step 4:

[1365] The server generates danger information based on the detection results, creates appropriate warnings and instructions, and sends them to the terminal. The input is information about suspicious individuals and unusual events, and the output is warning messages and instructions. Specifically, when a suspicious individual is detected, a warning message saying "There is a suspicious individual near the entrance. Please be on guard" is generated and sent to the terminal.

[1366] Step 5:

[1367] The terminal notifies the operator of warning messages and instructions received from the server by voice and on the display. The input is the warning message or instruction sent from the server, and the output is a voice notification and a display. Specifically, the speech synthesis engine reads out the warning message and displays it on the display.

[1368] Step 6:

[1369] The device receives the operator's voice commands through the microphone and analyzes them using a voice recognition engine (for example, Google Speech-to-Text API). The input is the operator's voice command, and the output is a text command. Specifically, when the operator says, "Show surveillance camera 1," the voice is converted into text.

[1370] Step 7:

[1371] The server executes the corresponding process based on the voice command. For example, it sends an instruction to the terminal to display the video from security camera 1. The input is a text-converted voice command, and the output is an instruction for the processing procedure. Specifically, in response to the command "Display security camera 1," the video from the corresponding camera is sent to the terminal.

[1372] Step 8:

[1373] The terminal navigates based on the route information received from the server. The input is the route information sent from the server, and the output is navigation instructions to the operator. Specific actions include a voice prompt saying "Turn left at the next corridor" and a map being displayed on the screen.

[1374] Step 9:

[1375] The device collects the operator's voice and facial expression data through a microphone and camera, and the emotion analysis engine analyzes it. The input is the operator's voice and facial expression data, and the output is the operator's emotional state information. Specifically, if the operator is feeling stressed, that state will be displayed as the analysis result.

[1376] Step 10:

[1377] Based on the results of the emotion analysis engine, the server takes action that matches the emotional state of the operator. For example, it changes the tone of the voice notification to reduce stress. The input is the operator's emotional state information, and the output is instructions on how to respond. A specific action would be to notify the operator in a relaxed tone, saying, "It is safe, so please stay calm and take action."

[1378] The above is the flow of the main processing steps of the "SecureGuard AI" system.

[1379] 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.

[1380] 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.

[1381] 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.

[1382] [Fourth embodiment]

[1383] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1384] 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.

[1385] 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).

[1386] 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.

[1387] 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.

[1388] 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).

[1389] 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.

[1390] 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.

[1391] 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.

[1392] 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.

[1393] 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.

[1394] 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.

[1395] 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."

[1396] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system are described below.

[1397] System Overview

[1398] The DriveGuard AI system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[1399] Data collection and analysis

[1400] Data collection methods

[1401] Terminal (in-vehicle device)

[1402] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[1403] The acquired data is sent to a server via the edge computing unit.

[1404] Data Analysis Methods

[1405] server

[1406] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[1407] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[1408] Notices and Instructions

[1409] Analysis result notification means

[1410] server

[1411] Based on the analysis results, risk information is generated and alerts and advice are created for drivers.

[1412] This information is sent to the terminal.

[1413] Terminal (in-vehicle device)

[1414] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[1415] Voice commands and route guidance

[1416] Voice command recognition means

[1417] Terminal (in-vehicle device)

[1418] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[1419] Route calculation method

[1420] server

[1421] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[1422] Navigation methods

[1423] Terminal (in-vehicle device)

[1424] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[1425] Specific examples

[1426] Example 1: Danger notification

[1427] Terminal (in-vehicle device)

[1428] The onboard camera detects pedestrians ahead and sends that information to a server.

[1429] server

[1430] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[1431] Terminal (in-vehicle device)

[1432] A voice notification will be played saying "There is a pedestrian ahead. Be careful." and a warning will also be displayed on the display.

[1433] Example 2: Traffic jam avoidance route guide

[1434] server

[1435] Calculates the optimal detour route based on the latest traffic congestion information.

[1436] Terminal (in-vehicle device)

[1437] It receives new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[1438] In this way, the DriveGuard AI system predicts hazards in real time and supports safe driving by collecting, analyzing, notifying, and processing the driver's voice commands.

[1439] The processing flow will be explained below.

[1440] Step 1:

[1441] The terminal (in-vehicle device) collects video data and environmental data in real time using on-board cameras and sensors. The on-board cameras capture images of the front and surrounding area, and the sensors acquire information from LiDAR, radar, GPS, etc.

[1442] Step 2:

[1443] The terminal (in-vehicle device) sends the collected data to a server via the edge computing unit, where the data is transmitted quickly, encrypted, and securely.

[1444] Step 3:

[1445] The server receives the data sent from the terminal and pre-processes it through noise filtering and format normalization, thereby improving the quality of the data.

[1446] Step 4:

[1447] The server analyzes the data using sophisticated image processing algorithms and machine learning models to recognize objects such as pedestrians, cyclists, and motorbikes, and calculates the position, speed, and direction of each object.

[1448] Step 5:

[1449] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[1450] Step 6:

[1451] The server sends the generated alert information and safe driving advice to the terminal, so that the driver can be notified in a timely manner.

[1452] Step 7:

[1453] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it may play a voice message saying, "There is a pedestrian ahead. Please be careful."

[1454] Step 8:

[1455] The user (driver) inputs voice commands through a microphone, such as setting a destination or requesting a new route.

[1456] Step 9:

[1457] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[1458] Step 10:

[1459] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion information and road condition data, and the calculation results are sent back to the terminal.

[1460] Step 11:

[1461] The terminal (in-vehicle device) receives the optimal route information from the server and provides navigation to the driver by voice and display, for example, "Turn right at the next intersection."

[1462] Example 1

[1463] 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."

[1464] Conventional driving assistance systems have problems in that they do not adequately provide drivers with appropriate information or provide navigation based on voice commands. Furthermore, because there is no consistent processing from data collection to analysis, notification, and navigation, it is difficult to provide sufficient functionality to support safe driving. To solve this problem, there is a need for real-time data collection and analysis, appropriate hazard notification, and rapid navigation based on voice commands.

[1465] 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.

[1466] In this invention, the server includes a data preprocessing means, a data analysis means, and an analysis result notification means, which allows for quick and accurate preprocessing and analysis of data collected in real time, and provides appropriate alerts and navigation instructions to the driver.

[1467] "Data collection means" means for collecting video and environmental data in real time using digital cameras and sensing devices.

[1468] "Data pre-processing means" refers to means for performing noise filtering and format normalization on collected data.

[1469] The "data analysis means" is a means for analyzing collected and preprocessed data using a machine learning model to extract information about objects and their locations.

[1470] The "analysis result notification means" is a means for generating danger information based on the analysis results and providing alerts and advice to the driver.

[1471] The "voice command recognition means" is a means for converting a user's voice command into text and performing processing in accordance with the content of the text.

[1472] The "route calculation means" is a means for calculating the optimum route based on the latest traffic information.

[1473] "Navigation means" refers to a means for providing audio and visual navigation to the driver based on the calculated optimum route information.

[1474] The present invention relates to a system for supporting safe and comfortable driving for drivers, and this system includes a data collection means, a data preprocessing means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and a navigation means. Specific embodiments for implementing this system will be described below.

[1475] System Overview

[1476] This system collects data using cameras and various sensors mounted on the vehicle, preprocesses and analyzes the data on a server, and provides appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate.

[1477] Data Acquisition and Transmission

[1478] Terminal (in-vehicle device)

[1479] The device collects data using a digital camera and various detection devices (LiDAR, radar, GPS, etc.), thereby obtaining real-time image data of the front and surrounding area and environmental data. The acquired data is sent to a server via an edge computing unit.

[1480] Data preprocessing and analysis

[1481] server

[1482] The server receives video data and environmental data sent from the device. First, noise filtering and format normalization are performed by the data preprocessing means. Next, the collected and preprocessed data is analyzed by the data analysis means using a machine learning model to extract objects and their location information. For example, pedestrians, bicycles, motorbikes, etc. are recognized, and their speed and direction of movement are estimated.

[1483] Notices and Instructions

[1484] server

[1485] The server generates danger information based on the analysis results and includes an analysis result notification means for providing alerts and advice to the driver. For example, a warning message such as "There is a pedestrian ahead. Please slow down" is generated and sent to the terminal.

[1486] Terminal (in-vehicle device)

[1487] The device receives the danger information and advice sent from the server, and displays it on the display and notifies the driver by voice, for example, "There is a pedestrian ahead. Be careful."

[1488] Voice commands and route guidance

[1489] Terminal (in-vehicle device)

[1490] The device receives the user's voice commands through a microphone, analyzes them with a voice recognition engine, converts the voice commands into text, and processes them accordingly.

[1491] server

[1492] The server includes a route calculation means for calculating the optimum route based on voice commands, and calculates the optimum route based on traffic congestion information and road condition data, and transmits the calculated route to the terminal.

[1493] Terminal (in-vehicle device)

[1494] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[1495] Specific examples

[1496] Example 1: Danger notification

[1497] The terminal (on-board device) detects pedestrians ahead using the on-board camera and sends the data to the server.

[1498] The server uses data analysis means to recognize the pedestrian's location and direction of travel, and if it determines that there is danger, it generates an alert message and sends it to the terminal.

[1499] The terminal (in-vehicle device) will play a voice notification saying, "There is a pedestrian ahead. Please be careful," and will also display a warning on the display.

[1500] Example 2: Traffic jam avoidance route guide

[1501] The server calculates the optimal detour route based on the latest traffic congestion information.

[1502] The terminal (in-vehicle device) receives the new route information and provides voice guidance such as, "To avoid traffic jams, turn right at the next intersection."

[1503] Prompt Sentence Examples

[1504] Prompt 1: "When an in-car camera detects a pedestrian ahead, how is the data processed?"

[1505] Prompt 2: "Describe the process by which the optimal route is calculated and how it is communicated to the driver."

[1506] As described above, the present invention is a system that supports safe driving by realizing consistent processing from data acquisition, preprocessing, analysis, notification, and route guidance in response to voice commands.

[1507] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1508] Step 1:

[1509] Acquiring and Sending Data

[1510] The terminal (on-board device) uses a digital camera and sensors (LiDAR, radar, GPS, etc.) to acquire front and surrounding image data and environmental data in real time. Specifically, the on-board camera captures images of the front, and the GPS acquires current location information. This data is temporarily stored through the edge computing unit and sent to a server. The input is camera and sensor data, and the output is data sent to the server.

[1511] Step 2:

[1512] Data Preprocessing

[1513] The server receives data sent from the terminal. Data preprocessing means performs noise filtering and format normalization. Specific operations include removing outliers and missing values, and adjusting the resolution of image data. The input is raw data received from the terminal, and the output is preprocessed data.

[1514] Step 3:

[1515] Data analysis

[1516] The server analyzes the preprocessed data using data analysis means. Here, a machine learning model is used to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information, speed, direction of travel, etc. Specifically, an image recognition algorithm detects the objects and calculates their coordinate data. The input is the preprocessed data, and the output is the analysis results (object type, position, speed, etc.).

[1517] Step 4:

[1518] Hazard information generation and notification

[1519] The server generates danger information based on the analysis results. The analysis result notification means creates a message to notify the driver of the generated danger information and sends it to the terminal. Specifically, a warning message such as "There is a pedestrian ahead. Please slow down" is generated. The input is the analysis results, and the output is the notification message.

[1520] The terminal (in-vehicle device) receives danger information and advice sent from the server, displays it on the display, and notifies the driver by voice. The input is the notification message received from the server, and the output is warning information by voice and display. Specifically, the terminal displays a warning message on the in-vehicle display and plays a voice notification from the speaker.

[1521] Step 5:

[1522] Receiving and processing voice commands

[1523] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them using a voice recognition engine. The voice commands are converted into text and processing is carried out based on the content. The input is the driver's voice command and the output is the textual command content. Specifically, the voice command "Please tell me the best route" is analyzed.

[1524] Step 6:

[1525] Calculating the best route

[1526] The server receives data for calculating the optimal route based on the voice command. The route calculation means calculates the optimal route based on traffic congestion information and road condition data. The input is the voice command and the latest traffic information, and the output is the calculated optimal route information. Specifically, the system uses a traffic information API to obtain real-time traffic congestion information and calculates the optimal route using an algorithm.

[1527] Step 7:

[1528] Providing navigation

[1529] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information received from the server, and the output is navigation instructions to the driver. Specific operations include voice guidance such as "Turn left at the next intersection" and a map display.

[1530] (Application example 1)

[1531] 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."

[1532] Current autonomous vehicles are required to collect and analyze environmental data in real time and make appropriate decisions to ensure safe driving. However, existing systems lack effective notification methods for the driver and the autonomous driving system, preventing accident risk reduction and efficient route guidance. In particular, the limited use of prompts using voice command recognition and generative AI models poses a challenge for further improving driver convenience and safety.

[1533] 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.

[1534] In this invention, the server includes a data analysis means, a prompt generation means, and a notification means using a generative AI model. This allows the server to analyze environmental data in real time and provide appropriate alerts and route guidance prompts to the driver and the automated driving system using the generative AI model. This reduces the risk of traffic accidents and provides efficient driving assistance.

[1535] The "data collection means" is a means for collecting video data and environmental data in real time from on-board cameras and sensors.

[1536] The "data analysis means" is a means of analyzing collected data using a machine learning model to extract recognition and location information for pedestrians, bicycles, motorbikes, etc.

[1537] The "analysis result notification means" is a means for notifying the driver or the automated driving system of the analysis results obtained by the data analysis means.

[1538] The "voice command recognition means" is a means for recognizing a voice command from the driver and analyzing the content of the command.

[1539] The "route calculation means" is a means for calculating the optimum route based on information on the current position and destination, as well as traffic congestion information and road condition data.

[1540] "Navigation means" refers to a means for providing voice and display navigation to the driver or the automated driving system based on the calculated optimal route information.

[1541] The "prompt generating means" is a means for generating an appropriate instruction or alert prompt in response to an input voice command or a recognized situation.

[1542] The "prompt processing means" is a means for performing processing to appropriately notify the driver of the generated prompt.

[1543] "Notification means using a generative AI model" refers to a means of generating appropriate notification content from data analyzed using a generative AI model and notifying the driver or automated driving system in real time.

[1544] This invention aims to realize a safe driving support system for autonomous vehicles called "AutoNav AI." This system includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, a prompt generation means, a prompt processing means, and a notification means using a generative AI model.

[1545] System Overview

[1546] The data collection method uses on-board cameras and sensors to collect video data and environmental data in real time. Specific hardware used includes on-board cameras such as the Logitech C920, as well as LiDAR, radar, and GPS sensors. This data is sent to a server via an edge computing unit.

[1547] The data analysis means receives the video data and environmental data collected on the server and analyzes them using machine learning models (e.g., OpenCV and YOLO). The data undergoes pre-processing such as noise filtering and format normalization, and then uses analysis algorithms to recognize objects such as pedestrians, bicycles, and motorbikes and extract their location information.

[1548] The analysis result notification means generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. These notifications are sent to the terminal in real time via the edge computing unit and are displayed as voice and on a display.

[1549] The voice command recognition means receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine (e.g., Google's SpeechRecognition API). The recognized voice commands are converted into text and processed accordingly.

[1550] The route calculation means calculates the optimal route based on the current location and destination information, as well as traffic congestion and road condition data. Specifically, route calculation is performed using Google Maps API, etc. The calculated route information is sent from the server to the terminal.

[1551] The navigation means provides navigation to the driver or the automated driving system by voice and display based on the optimal route information received by the terminal from the server, for example, by providing guidance such as "Turn left at the next intersection."

[1552] The prompt generator generates appropriate instruction or alert prompts based on the input voice command and the recognized situation, using a generative AI model.

[1553] The prompt processing means performs processing to appropriately notify the driver of the generated prompt, thereby providing instructions and alerts to the driver at the appropriate time.

[1554] The notification method using the generative AI model generates appropriate notification content from data analyzed using the generative AI model and notifies the driver or autonomous driving system in real time.

[1555] Specific examples

[1556] Example 1: Traffic jam avoidance route guide

[1557] The server calculates the optimal detour route based on the latest traffic congestion information. The device receives the new route information and provides voice guidance such as, "Turn right at the next intersection to avoid traffic congestion." This supports efficient driving.

[1558] Example 2: Voice command recognition and prompt generation

[1559] The driver issues a voice command saying, "Calculate a safe and efficient route based on the latest traffic congestion information. End point: Shinjuku Station." In response, the server calculates the optimal route and generates a prompt saying, "The optimal route from your current location to Shinjuku Station is to turn right at the next intersection and go straight at the next traffic light. The route takes approximately 30 minutes. There is an accident warning on the way, so please be careful."

[1560] Thus, the present invention provides specific embodiments for significantly improving the safety and efficiency of autonomous vehicles.

[1561] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1562] Step 1:

[1563] Data collection

[1564] The onboard camera and various sensors installed in the user's vehicle collect video data and environmental data in real time. Specifically, the camera captures the surrounding image, and the LiDAR, radar, and GPS sensors acquire various environmental data. This data is sent to a server via an edge computing unit.

[1565] Input: Video data and environmental data from on-board cameras and sensors

[1566] Output: Data sent to the server through the edge computing unit

[1567] Step 2:

[1568] Data analysis

[1569] The server receives the collected video and environmental data and performs preprocessing: the data undergoes noise filtering and format normalization, and is then analyzed using machine learning models (OpenCV and YOLO, for example) to recognize objects such as pedestrians, bicycles, and motorbikes, and extract their location information.

[1570] Input: Video data and environmental data collected in step 1

[1571] Output: Objects and their locations

[1572] Step 3:

[1573] Notification of analysis results

[1574] The server generates appropriate alerts and advice for the driver and the automated driving system based on the analysis results obtained by the data analysis means. The generated notifications are sent to the terminal via the edge computing unit and displayed by voice and on the display. For example, if there is a pedestrian ahead, the notification will say, "There is a pedestrian ahead. Please be careful."

[1575] Input: Analysis results obtained by data analysis tools

[1576] Output: Alerts and advice sent to the device

[1577] Step 4:

[1578] Voice Command Recognition

[1579] Voice commands issued by users (e.g., "End point: Shinjuku Station") are collected through microphones in the vehicle and converted into text by a speech recognition engine (e.g., Google's SpeechRecognition API). This text data is sent to a server and analyzed as a voice command.

[1580] Input: A voice command spoken by the user

[1581] Output: Text data sent to the server

[1582] Step 5:

[1583] Route calculation

[1584] The server calculates the optimal route based on the analyzed voice commands, current location information, traffic information, and road condition data. This calculation uses route calculation services such as Google Maps API. The calculated route information is sent from the server to the device.

[1585] Input: Voice commands, current location, traffic information, road condition data

[1586] Output: Calculated optimal route information

[1587] Step 6:

[1588] Navigation

[1589] Based on the optimal route information received from the server, the device provides navigation to the driver by voice and display, for example, by providing instructions such as "Turn left at the next intersection."

[1590] Input: Optimal route information received from the server

[1591] Output: Voice and display navigation instructions

[1592] Step 7:

[1593] Prompt Generation

[1594] The server generates appropriate prompts, such as instructions or alerts, based on the input voice command and the recognized situation. A generative AI model is used to generate these prompts. A specific example of a prompt is, "Turn right at the next intersection to avoid traffic jams."

[1595] Input: Voice commands, recognized situations

[1596] Output: Generated prompt

[1597] Step 8:

[1598] Prompt Processing

[1599] The server then processes the generated prompt to appropriately notify the driver. The notification is given via voice and display. For example, the driver may be notified, "Turn right at the next intersection to avoid traffic jams."

[1600] Input: Generated prompt

[1601] Output: Audio and visual notifications

[1602] 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.

[1603] The present invention relates to a "DriveGuard AI" system that supports safe and comfortable driving for drivers, and includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion engine. Specific embodiments for implementing this system are described below.

[1604] System Overview

[1605] The "DriveGuard AI" system uses on-board cameras and various sensors to collect data, analyze it, and provide appropriate notifications to the driver to support safe driving. It also recognizes voice commands to calculate routes and navigate. Furthermore, it combines an emotion engine that analyzes the driver's emotional state from their voice and facial expressions to provide appropriate support according to the driver's emotional state.

[1606] Data collection and analysis

[1607] Data collection methods

[1608] Terminal (in-vehicle device)

[1609] The vehicle is equipped with an on-board camera and sensors that capture images of the front and surrounding area in real time, while the sensors acquire information from LiDAR, radar, GPS, etc.

[1610] The acquired data is sent to a server via the edge computing unit.

[1611] Data Analysis Methods

[1612] server

[1613] It receives video and sensor data sent from the device, and preprocesses the data through noise filtering and format normalization.

[1614] The system uses analysis algorithms and machine learning models to analyze data and recognize objects such as pedestrians, bicycles, and motorbikes. The system then calculates the object's position, speed, and direction of travel from the recognition results.

[1615] Notices and Instructions

[1616] Analysis result notification means

[1617] server

[1618] Based on the analysis results, risk information is generated and alerts and safe driving advice are created for drivers.

[1619] This information is sent to the terminal.

[1620] Terminal (in-vehicle device)

[1621] The system receives danger information and advice sent from the server, displays it on the screen, and notifies the driver by voice, for example, "There is a pedestrian ahead. Please slow down."

[1622] Voice commands and route guidance

[1623] Voice command recognition means

[1624] Terminal (in-vehicle device)

[1625] The driver's voice commands are received through a microphone and analyzed by a voice recognition engine. The voice commands are converted into text and processed accordingly.

[1626] Route calculation method

[1627] server

[1628] The system runs an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and the calculated route information is sent to the device.

[1629] Navigation methods

[1630] Terminal (in-vehicle device)

[1631] Based on the optimal route information received from the server, navigation is provided to the driver via voice and display. For example, guidance may be given such as "Turn left at the next intersection."

[1632] Emotion Engine

[1633] Emotion Engine

[1634] Data collection

[1635] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera.

[1636] Data analysis

[1637] The server analyzes the collected voice and facial expression data to determine the emotional state (e.g., stress, fatigue, anger, etc.).

[1638] Notices and Instructions

[1639] Based on the analysis results of the emotion engine, the server adjusts driving advice and safety suggestions and sends them to the device.

[1640] Specific examples

[1641] Example 1: Danger notification and emotional response

[1642] Terminal (in-vehicle device)

[1643] The onboard camera detects pedestrians ahead and sends that information to a server.

[1644] server

[1645] Image analysis algorithms identify the position and direction of pedestrians, calculate their speed and distance, and if a pedestrian is deemed dangerous, an alert message is generated and sent to the device.

[1646] At the same time, an emotion engine analyzes the driver's emotional state (e.g., impatience) and adjusts the tone of the alert as needed.

[1647] Terminal (in-vehicle device)

[1648] The system plays a voice notification saying "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. It also notifies in a calmer tone depending on the driver's emotional state.

[1649] Example 2: Relaxed Route Guide

[1650] server

[1651] The system calculates the optimal detour route based on the latest traffic congestion information, while the emotion engine analyzes the driver's fatigue level and suggests a less stressful route.

[1652] Terminal (in-vehicle device)

[1653] The new route information is received and a voice guides you, saying, "To avoid traffic jams, turn right at the next intersection. There is also a rest area on this route."

[1654] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[1655] The processing flow will be explained below.

[1656] Step 1:

[1657] The terminal (in-vehicle device) collects video data and environmental data in real time using an in-vehicle camera and sensors (LiDAR, radar, GPS, etc.). The in-vehicle camera captures images of the road ahead and surroundings, and the sensors obtain the necessary location information and obstacle information.

[1658] Step 2:

[1659] The terminal (in-vehicle device) performs initial processing on the collected data using an edge computing unit, compresses and encrypts it, and sends it to the server, enabling high-speed and secure data communication.

[1660] Step 3:

[1661] The server receives the data sent by the terminal and preprocesses the data by performing noise filtering and format normalization, which ensures the accuracy and consistency of the data.

[1662] Step 4:

[1663] The server analyzes the data using high-performance image processing algorithms and machine learning models, recognizing pedestrians, cyclists, motorbikes, and other objects, and calculating the position, speed, and direction of each object.

[1664] Step 5:

[1665] The server evaluates the risk level based on the analysis results and generates alert information and safe driving advice as necessary. For example, if there is a pedestrian ahead, it will generate an alert urging the driver to slow down.

[1666] Step 6:

[1667] The server sends the generated alert information and safe driving advice to the terminal, which is then ready to notify the driver in a timely manner.

[1668] Step 7:

[1669] The terminal (in-vehicle device) receives the alert information from the server and notifies the driver of it by voice and display. For example, it conveys a message to the driver saying, "There is a pedestrian ahead. Please be careful."

[1670] Step 8:

[1671] The terminal (in-vehicle device) collects the driver's voice and facial expression data using a microphone and an in-vehicle camera, making it possible to grasp the driver's emotional state in real time.

[1672] Step 9:

[1673] The terminal (in-vehicle device) performs initial processing of the collected voice and facial expression data using an edge computing unit, compresses and encrypts it, and sends it to the server.

[1674] Step 10:

[1675] The server receives the voice and facial expression data sent from the terminal and analyzes the driver's emotional state (e.g., stress, fatigue, anger, etc.) using emotion recognition algorithms.

[1676] Step 11:

[1677] The server then adjusts driving advice and safety suggestions based on the emotion engine's analysis. For example, if the driver is feeling anxious, the server will adjust the tone of the notification to a calmer tone and generate a message encouraging more careful driving.

[1678] Step 12:

[1679] The server sends alert information and advice adjusted by the emotion engine to the terminal.

[1680] Step 13:

[1681] The terminal (in-vehicle device) receives the adjusted alert information and advice from the server and notifies the driver by voice and display, for example, a message saying, "There is a pedestrian ahead. Please proceed with caution and remain calm."

[1682] Step 14:

[1683] The user (driver) inputs voice commands through a microphone, such as setting a destination or changing a route.

[1684] Step 15:

[1685] The terminal (in-vehicle device) analyzes the input voice command using a voice recognition engine, converts it into text, and then sends the analysis results to the server.

[1686] Step 16:

[1687] The server receives the driver's voice commands and calculates the optimal route based on traffic congestion and road condition data, and the calculation results are then sent back to the device.

[1688] Step 17:

[1689] The terminal (in-vehicle device) receives optimal route information from the server and provides navigation to the driver by voice and display. For example, it may provide guidance such as, "Turn right at the next intersection. Also, you seem tired, so let's stop at a nearby rest area."

[1690] In this way, the DriveGuard AI system assists in safe and comfortable driving by collecting, analyzing, and notifying data, as well as processing the driver's voice commands and emotional state in real time.

[1691] Example 2

[1692] 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."

[1693] In today's automobile driving, it is important to ensure both driver safety and comfort. However, conventional driver assistance systems have difficulty adapting to changes in the driving environment and the driver's emotional state, and excessive information provision and inappropriate alerts can actually induce stress. Furthermore, there is a lack of technology to effectively combine real-time environmental data analysis and emotion analysis, resulting in situations where optimal driver assistance cannot be achieved.

[1694] 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 data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, and an emotion analysis means. This makes it possible to quickly respond to changes in the driving environment and provide appropriate alerts and driving advice based on the driver's emotional state.

[1695] The "data collection means" is a component of the system that acquires video data and environmental data in real time using on-board cameras and sensors and transmits them to a server via an edge computing unit.

[1696] The "data analysis means" is a component of the system that analyzes collected data using a machine learning model after preprocessing such as noise filtering and format normalization to extract object recognition and location information.

[1697] The "analysis result notification means" is a system component that generates alerts and safe driving advice for the driver based on the analysis results obtained by the data analysis means and transmits them to the in-vehicle device.

[1698] The "voice command recognition means" is a component of the system that receives the driver's voice commands through a microphone, converts them into text using a voice recognition engine, and performs processing based on the content of the text.

[1699] The "route calculation means" is a system component that executes an algorithm to calculate the optimal route based on traffic congestion information and road condition data, and transmits the results to the in-vehicle device.

[1700] The "navigation means" is a component of the system that provides navigation to the driver through voice and display based on the received optimum route information.

[1701] The "emotion analysis means" is a component of the system that analyzes the driver's voice and facial expression data collected through the vehicle-mounted camera and microphone, and determines the driver's emotional state.

[1702] The present invention relates to a system for supporting safe and comfortable driving of a driver, which includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, and an emotion analysis means.

[1703] Data collection method configuration

[1704] The terminal (in-vehicle device) is equipped with an in-vehicle camera and various sensors (LiDAR, radar, GPS). The in-vehicle camera captures images of the front and surroundings in real time, and the sensors acquire environmental data. This data is processed by the edge computing unit and preprocessed through noise filtering. The preprocessed data is then sent to the server.

[1705] Data analysis method configuration

[1706] The server receives the video and sensor data sent from the device and performs pre-processing on the data. The pre-processed data is then analyzed using machine learning models to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel.

[1707] Configuration of analysis result notification method

[1708] The server generates risk information based on the analysis results and creates alerts and safe driving advice for the driver. These notifications are sent to the device, which displays the risk information and advice received from the server on its display and notifies the driver by voice. For example, it may notify the driver, "There is a pedestrian ahead. Please slow down."

[1709] Configuration of voice command recognition means

[1710] The terminal (in-vehicle device) receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The voice commands are converted into text and processed accordingly.

[1711] Configuring route calculation methods

[1712] The server runs an algorithm to calculate the optimal route based on traffic congestion and road condition data, and the calculated route information is sent to the device.

[1713] Navigation Method Configuration

[1714] The terminal (in-vehicle device) provides navigation to the driver by voice and display based on the optimal route information received from the server, for example, by providing instructions such as "Turn left at the next intersection."

[1715] Sentiment analysis method configuration

[1716] The terminal (in-vehicle device) collects the driver's voice and facial expression data through a microphone and an in-vehicle camera. The server analyzes the collected voice and facial expression data to determine the driver's emotional state (e.g., stress, fatigue, anger, etc.). The server adjusts driving advice and safety suggestions based on the results of the emotion analysis and sends them to the terminal.

[1717] Specific examples

[1718] Example 1: Danger notification and emotional response

[1719] The terminal (in-vehicle device) uses an on-board camera to detect pedestrians ahead and sends that information to a server. The server uses an image analysis algorithm to identify the pedestrian's location and direction of travel, and calculates their speed and distance. If it determines that there is a danger, it generates an alert message and sends it to the terminal. Furthermore, an emotion analysis means analyzes the driver's emotional state and adjusts the tone of the alert as necessary. The terminal plays a voice notification saying, "There is a pedestrian ahead. Please be careful," and also displays a warning on the display. The server also uses a calmer tone of voice depending on the driver's emotional state.

[1720] Example 2: Relaxed Route Guide

[1721] The server calculates the optimal detour route based on the latest traffic congestion information. At the same time, an emotion analysis tool analyzes the driver's fatigue level and suggests a less stressful route. The device receives the new route information and provides voice guidance such as, "To avoid traffic congestion, please turn right at the next intersection. There is also a rest area on this route."

[1722] Example prompts for generative AI models

[1723] Please explain in detail how "DriveGuard AI" collects sensor data, analyzes it on the server, and provides real-time alerts to the driver.

[1724] Please explain with concrete examples how the system analyzes the driver's emotional state and, based on the results, adjusts driving advice appropriately and in the right place.

[1725] This system, with the above configuration, provides a mechanism that comprehensively supports the driver in ensuring safe and comfortable driving.

[1726] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1727] Step 1:

[1728] The terminal (on-board device) acquires video data and environmental data in real time using the on-board camera and various sensors (LiDAR, radar, GPS). The acquired data is preprocessed by the edge computing unit after undergoing noise filtering. After preprocessing is complete, the data is sent from the terminal to the server. The inputs are camera video data and sensor data, and the output is the preprocessed data.

[1729] Specific examples of behavior:

[1730] The device activates the onboard camera to capture images of the road ahead, while the LiDAR and radar simultaneously acquire the position and speed of surrounding objects, and this data is then filtered by the edge computing unit.

[1731] Step 2:

[1732] The server receives video and sensor data sent from the device. As input, it receives preprocessed data. The received data is preprocessed through noise filtering and format normalization. The preprocessed data is ready to be analyzed by the machine learning model.

[1733] Specific examples of behavior:

[1734] The server receives the data stream, corrects dark areas, removes noise, and unifies the format through filtering, converting it into the appropriate data format.

[1735] Step 3:

[1736] The server uses machine learning models to analyze the noise-filtered and format-normalized data. The input is the preprocessed data. The machine learning models are used to recognize objects such as pedestrians, bicycles, and motorbikes, and calculate their location, speed, and direction of travel. The output is the object recognition results and their location information.

[1737] Specific examples of behavior:

[1738] A machine learning algorithm scans the video data and recognizes pedestrians ahead, calculating their location and direction of travel as a result.

[1739] Step 4:

[1740] The server generates alerts and safe driving advice for the driver based on the analysis results. The input is the analysis results, and the output is an alert message or advice. The generated information is sent to the terminal.

[1741] Specific examples of behavior:

[1742] The server generates an alert message saying "There is a pedestrian ahead. Please slow down." and sends it to the terminal.

[1743] Step 5:

[1744] The terminal (in-vehicle device) displays the alerts and advice received from the server on the display and notifies the driver by voice. The input is the alert message from the server, and the output is the display and voice notification.

[1745] Specific examples of behavior:

[1746] The device will play a voice message saying, "There is a pedestrian ahead. Be careful." and will also show a warning message on the display.

[1747] Step 6:

[1748] The device receives the driver's voice commands through a microphone and analyzes them with a voice recognition engine. The input is the driver's voice command, and the output is the voice command converted into text. Processing is performed based on the voice command converted into text.

[1749] Specific examples of behavior:

[1750] The driver enters the voice command "recalculate route," and a speech recognition engine converts the command into text.

[1751] Step 7:

[1752] The server receives voice commands converted into text and calculates the optimal route based on traffic congestion and road condition data. The inputs are voice commands and traffic information, and the output is the calculated optimal route information. The calculated route information is sent to the device.

[1753] Specific examples of behavior:

[1754] The server collects the latest traffic congestion information, calculates a new detour route based on that information, and sends the results to the terminal.

[1755] Step 8:

[1756] The terminal provides navigation to the driver by voice and display based on the optimal route information received from the server. The input is the route information from the server, and the output is navigation instructions.

[1757] Specific examples of behavior:

[1758] The device will provide voice guidance such as "Turn left at the next intersection," and will display a map and route on the screen.

[1759] Step 9:

[1760] The terminal collects the driver's voice and facial expression data through a microphone and an on-board camera. The input is the driver's voice and video data, and the output is the collected data sent to the server.

[1761] Specific examples of behavior:

[1762] The device captures the driver's facial expressions and tone of voice in real time and transmits the data to a server.

[1763] Step 10:

[1764] The server analyzes the transmitted voice and facial expression data to determine the driver's emotional state. The input is the driver's voice and facial expression data, and the output is the analysis result of the emotional state. Based on the analysis result, driving advice and safety suggestions are adjusted and sent to the terminal.

[1765] Specific examples of behavior:

[1766] The server recognizes signs of stress from the driver's tone of voice and facial expression, generates advice such as "Relax," and sends it to the terminal.

[1767] Through these processing steps, the "DriveGuard AI" system provides a mechanism to comprehensively support the driver's safety and comfort.

[1768] (Application example 2)

[1769] 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."

[1770] Conventional building security systems are limited to simple monitoring and warning using surveillance cameras and sensors, and lack the ability to detect suspicious individuals or unusual events, nor the countermeasures for their detection. Furthermore, no consideration was given to supporting security personnel in dealing with overwork and stressful environments, making it difficult to maintain efficient security. Therefore, it is necessary to reduce the mental stress of security personnel, along with early detection of suspicious individuals and appropriate countermeasures.

[1771] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. This enables the prompt detection of suspicious individuals and abnormalities, the appropriate countermeasures, and the reduction of mental stress for security personnel.

[1772] The "data collection means" is a means for collecting video data and environmental data in real time from monitoring devices and various sensors.

[1773] "Data analysis means" refers to a means of analyzing collected data using a machine learning model to detect suspicious individuals or abnormalities.

[1774] The "analysis result notification means" is a means for notifying the security officer of appropriate warnings and instructions based on the analysis results obtained by the data analysis means.

[1775] The "voice command recognition means" refers to a means for recognizing and analyzing voice commands from security personnel and performing corresponding functions in accordance with the instructions.

[1776] The "route calculation means" is a means for calculating the optimal route based on route information and situational data inside and outside the building, and predicting the route and escape route of a detected suspicious person.

[1777] The "navigation means" is a means for providing navigation to security personnel using voice and a display based on the calculated optimum route information.

[1778] The "emotion analysis engine" is a means of analyzing the voice and facial expression data of security personnel obtained through microphones and cameras to determine their mental stress and emotional state.

[1779] "Security notification means" is a means of notifying appropriate warnings and safety suggestions based on the results of detecting suspicious individuals or abnormalities and the emotional state of the person in charge.

[1780] The present invention relates to a "SecureGuard AI" system that improves building security and safety. The system includes multiple elements: a data collection means, a data analysis means, an analysis result notification means, a voice command recognition means, a route calculation means, a navigation means, an emotion analysis engine, and a security notification means. Specific embodiments for implementing the system are described in detail below.

[1781] Data collection and analysis

[1782] Data collection methods

[1783] Terminals (smartphones and robots) are equipped with monitoring devices and sensors that collect real-time video and environmental data from inside and outside the building. The collected data is then sent to a server via an edge computing unit. Specific hardware used includes high-resolution cameras, microphones, motion sensors, etc.

[1784] Data Analysis Methods

[1785] The server receives the collected video and sensor data. The data is preprocessed through noise filtering and format normalization, and then analyzed using analysis algorithms and machine learning models (e.g., the dlib library and open-source sentiment analysis models). This allows for the detection of suspicious individuals and unusual behavior, and the extraction of their location information and behavioral patterns.

[1786] Notices and Instructions

[1787] Analysis result notification means

[1788] The server generates risk information based on the analysis results obtained by the data analysis means, and issues alerts and safety measures to the operator, which are then sent to the terminal.

[1789] Voice command recognition means

[1790] The terminal receives voice commands from the operator through a microphone, analyzes these commands with a speech recognition engine (e.g., Google Speech-to-Text API), converts them into text, and performs the corresponding processing.

[1791] Route calculation and navigation

[1792] Route calculation method

[1793] The server runs algorithms that calculate optimal routes based on route information and situational data inside and outside the building, such as routes that bypass areas where suspicious individuals may be present, using up-to-date map data and real-time situational information.

[1794] Navigation methods

[1795] Based on the optimal route information received from the server, the terminal provides navigation to the operator by voice and display, for example, by providing instructions such as "Turn left at the next corridor."

[1796] Sentiment Analysis Engine

[1797] The terminal collects the operator's voice and facial expression data through a microphone and camera. The server analyzes the collected voice and facial expression data to determine the operator's emotional state (stress, fatigue, anger, etc.). This enables the system to take measures to reduce the operator's mental stress.

[1798] Specific examples

[1799] Example 1: Detecting and responding to suspicious individuals

[1800] The terminal transmits images captured by the surveillance cameras inside the building to the server.

[1801] The server analyzes the video and detects suspicious individuals.

[1802] The server generates a warning such as "There is a suspicious person near the entrance. Please be on guard" and sends it to the terminal.

[1803] Example 2: Controlling a surveillance camera with voice commands

[1804] The terminal receives a voice command from the operator: "Show surveillance camera 1."

[1805] A voice recognition engine converts commands into text and displays the image from the corresponding camera.

[1806] Prompt Sentence Examples

[1807] "A suspicious person has been detected near the entrance. Please implement security measures."

[1808] "Do you want to view the footage from security camera 1?"

[1809] By combining these elements, the SecureGuard AI system can significantly improve building security and provide a safe and efficient environment.

[1810] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1811] Step 1:

[1812] The terminal collects video data and environmental data from monitoring devices and sensors. The collected data is sent to a server via an edge computing unit. The input is raw data from the monitoring devices and sensors, and the output is pre-processed data. Specifically, the high-resolution camera continuously captures video, and the sensors detect temperature and movement.

[1813] Step 2:

[1814] The server receives the transmitted video data and environmental data and performs preprocessing through noise filtering and format normalization. The input is the video data and sensor data transmitted from the device, and the output is the noise-removed and normalized data. Specifically, the filtering algorithm removes noise from the video and standardizes the data format.

[1815] Step 3:

[1816] The server analyzes the preprocessed data using analysis algorithms and machine learning models (for example, dlib or open-source emotion analysis models) to detect suspicious individuals or unusual behavior. The input is noise-removed and normalized data, and the output is information about suspicious individuals or unusual behavior. Specifically, the machine learning model recognizes faces and behavioral patterns and detects suspicious behavior.

[1817] Step 4:

[1818] The server generates danger information based on the detection results, creates appropriate warnings and instructions, and sends them to the terminal. The input is information about suspicious individuals and unusual events, and the output is warning messages and instructions. Specifically, when a suspicious individual is detected, a warning message saying "There is a suspicious individual near the entrance. Please be on guard" is generated and sent to the terminal.

[1819] Step 5:

[1820] The terminal notifies the operator of warning messages and instructions received from the server by voice and on the display. The input is the warning message or instruction sent from the server, and the output is a voice notification and a display. Specifically, the speech synthesis engine reads out the warning message and displays it on the display.

[1821] Step 6:

[1822] The device receives the operator's voice commands through the microphone and analyzes them using a voice recognition engine (for example, Google Speech-to-Text API). The input is the operator's voice command, and the output is a text command. Specifically, when the operator says, "Show surveillance camera 1," the voice is converted into text.

[1823] Step 7:

[1824] The server executes the corresponding process based on the voice command. For example, it sends an instruction to the terminal to display the video from security camera 1. The input is a text-converted voice command, and the output is an instruction for the processing procedure. Specifically, in response to the command "Display security camera 1," the video from the corresponding camera is sent to the terminal.

[1825] Step 8:

[1826] The terminal navigates based on the route information received from the server. The input is the route information sent from the server, and the output is navigation instructions to the operator. Specific actions include a voice prompt saying "Turn left at the next corridor" and a map being displayed on the screen.

[1827] Step 9:

[1828] The device collects the operator's voice and facial expression data through a microphone and camera, and the emotion analysis engine analyzes it. The input is the operator's voice and facial expression data, and the output is the operator's emotional state information. Specifically, if the operator is feeling stressed, that state will be displayed as the analysis result.

[1829] Step 10:

[1830] Based on the results of the emotion analysis engine, the server takes action that matches the emotional state of the operator. For example, it changes the tone of the voice notification to reduce stress. The input is the operator's emotional state information, and the output is instructions on how to respond. A specific action would be to notify the operator in a relaxed tone, saying, "It is safe, so please stay calm and take action."

[1831] The above is the flow of the main processing steps of the "SecureGuard AI" system.

[1832] 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.

[1833] 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.

[1834] 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.

[1835] 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.

[1836] 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.

[1837] 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.

[1838] 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).

[1839] 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.

[1840] 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."

[1841] 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.

[1842] 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).

[1843] 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.

[1844] 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.

[1845] 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.

[1846] 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.

[1847] 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.

[1848] 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.

[1849] 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, th...

Claims

1. data collection means; data analysis means; analysis result notification means; a voice command recognition means; A route calculation means; A system including a navigation means.

2. 2. The system according to claim 1, wherein the data collection means is means for collecting video data and environmental data in real time from an on-board camera and a sensor.

3. The system according to claim 1 , wherein the data analysis means analyzes the collected data using a machine learning model and extracts recognition and location information of pedestrians, bicycles, motorbikes, etc.

4. 2. The system according to claim 1, wherein the analysis result notification means notifies the driver of the analyzed information by voice and on a display.

5. 2. The system according to claim 1, wherein the voice command recognition means recognizes a voice command of a driver, analyzes the voice command, and converts the voice command into text.

6. 2. The system according to claim 1, wherein said route calculation means calculates an optimum route based on congestion information and road condition data.

7. 2. The system according to claim 1, wherein said navigation means is a means for providing the driver with calculated optimum route information by voice and on a display.

Images