System

Abstract

To provide a system for accurately recognizing a situation by AI and presenting a proper countermeasure to a sudden irregular situation such as a traffic accident or an unpredictable event, especially in automatic driving.SOLUTION: An automatic driving system, comprising: means for recording a generated traffic event image; means for obtaining a driving situation in the form of a text prompt; means for generating an appropriate action instruction according to the text prompt; and means for controlling a vehicle according to the action instruction.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] Currently, autonomous driving technology demonstrates a certain level of performance in normal driving situations, but it is not able to respond adequately to sudden, irregular situations. This leaves safety and reliability issues unresolved. In particular, it is important for AI to accurately recognize situations and propose appropriate countermeasures in the event of a traffic accident or other unpredictable event. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. First, it provides a means for generating images of irregular traffic events using generative AI and recording the images in a dataset. Next, it provides a means for acquiring driving situations from sensors and describing the data in the form of text prompts. Furthermore, it introduces generative artificial intelligence means for analyzing the text prompts and generating appropriate action instructions. Finally, it provides a means for controlling the vehicle based on the generated action instructions, thereby realizing a safe and reliable autonomous driving environment.

[0006] "Generated traffic event images" are images showing traffic conditions created using generative AI, particularly those that simulate irregular situations that differ from the norm.

[0007] "Means for recording" refers to the functionality or mechanism that stores the generated traffic incident images in a dataset.

[0008] "Driving situations" refers to all environments and events that a vehicle faces while driving autonomously, such as the movement of other vehicles, the presence or absence of obstacles, and weather conditions.

[0009] The "text prompt format" describes the situation while driving in natural language and in a format that is easy for generative artificial intelligence to analyze.

[0010] "Generative artificial intelligence means" refers to an AI model that has the functionality to generate appropriate action instructions and analysis results based on input text prompts.

[0011] "Action instructions" are specific actions that an autonomous vehicle should take in a specific driving situation, such as braking or steering.

[0012] "Means for controlling a vehicle" refers to the entire system or part of the system that performs various operations on an autonomous vehicle based on the generated behavioral instructions.

[0013] "Sensor means" refers to devices and mechanisms, such as cameras and various sensors, that monitor the situation around the vehicle in real time and acquire data.

[0014] A "user interface" refers to the interface that allows a user to operate the system and check the system status, and typically consists of a screen, touch display, voice recognition system, etc.

[0015] "Analysis results" refers to the countermeasures and information generated by the generating AI by analyzing the input prompts and driving situation data. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] ---

[0038] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[0039] Server Side

[0040] ---

[0041] The server first uses the generative AI model to generate images of various traffic events. Specifically, the server loads the generative AI model and generates images by inputting the prompt "irregular traffic situation." The generated images are then saved in a dataset for use in model training. This process improves the recognition accuracy of the AI ​​model.

[0042] Terminal side

[0043] ---

[0044] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. Using data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle, the terminal analyzes the current driving situation and describes it in the form of a text prompt. For example, it generates a specific situation as text, such as "A child suddenly jumps out at an intersection." The prompt is sent to the generation AI for analysis. The resulting action instructions (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[0045] User Interface

[0046] ---

[0047] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[0048] Specific examples

[0049] ---

[0050] A specific example of system implementation is shown below.

[0051] Server-side example:

[0052] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[0053] Example of terminal operation:

[0054] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[0055] An example of how the user interface works:

[0056] The user uses the application to check the system status and input advice for specific traffic scenarios, such as entering a prompt for "advanced driving techniques in bad weather," which results in specific measures such as "slow down and increase the distance from the vehicle ahead."

[0057] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[0058] The processing flow will be explained below.

[0059] Server-side processing

[0060] ---

[0061] Step 1:

[0062] The server loads the generative AI model.

[0063] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[0064] Step 2:

[0065] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[0066] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[0067] Step 3:

[0068] The server stores the generated images in a dataset.

[0069] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[0070] Terminal-side processing

[0071] ---

[0072] Step 1:

[0073] The device acquires the current driving situation from sensors installed in the vehicle.

[0074] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[0075] Step 2:

[0076] The device analyzes the collected data and generates a text prompt.

[0077] Based on the analysis results, a specific text description of the situation, such as "A child suddenly jumps out at an intersection," is generated.

[0078] Step 3:

[0079] The terminal sends the generated text prompt to the AI.

[0080] The prompt is sent over the network to the generating AI, which begins analyzing it.

[0081] Step 4:

[0082] The terminal receives action instructions from the AI.

[0083] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[0084] Step 5:

[0085] The vehicle is controlled based on the action instructions received by the terminal.

[0086] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[0087] User Interface Handling

[0088] ---

[0089] Step 1:

[0090] The user launches an application for the mobile terminal.

[0091] Users access the system by launching the app on their smartphone or tablet.

[0092] Step 2:

[0093] The user checks the current state of the system.

[0094] The application displays irregular images generated and the latest AI analysis results.

[0095] Step 3:

[0096] The user enters any prompts.

[0097] For example, enter the prompt "Tell me how to drive on a rainy day."

[0098] Step 4:

[0099] The user checks the prompt analysis results.

[0100] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[0101] Through each of these steps, the server, terminal, and user work together seamlessly to create an autonomous driving system that can quickly respond to irregular situations.

[0102] Example 1

[0103] 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."

[0104] When autonomous vehicles encounter irregular traffic events, it is difficult for them to respond quickly and appropriately. This increases the risk of accidents in unexpected situations, potentially threatening the safety of passengers and those around them. Furthermore, conventional autonomous driving systems lack the ability to analyze data in real time and respond immediately, limiting their ability to generate optimal driving behavior. It is necessary to resolve these issues and improve the reliability and safety of autonomous driving systems.

[0105] 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.

[0106] In this invention, the server includes a means for recording the generated traffic event image, a means for acquiring the driving situation in the form of a text prompt, a generative AI model means for generating appropriate behavioral instructions based on the text prompt, a means for controlling the vehicle based on the behavioral instructions, a means for loading the generative AI model by the server, a means for inputting the traffic situation prompt and generating an image by the server, and a means for saving the generated image in a dataset, thereby enabling analysis of traffic events and generation of appropriate driving behavior in real time.

[0107] A "generative AI model" is an artificial intelligence model used to generate images or text based on specific input prompts.

[0108] A "prompt" is a text input that prompts a generative AI model to generate a specific output.

[0109] A "server" is a computer system that provides central management and processing, such as loading generative AI models, entering prompts, and performing analysis.

[0110] "Traffic incident images" are image data that visually represent traffic conditions and incidents.

[0111] A "dataset" is a collection containing a set of data used for training or analysis.

[0112] "Sensor means" means a device (e.g., camera, LiDAR, radar) installed on an autonomous vehicle for collecting data on the surrounding environment.

[0113] A "text prompt" is a prompt that expresses a traffic situation or event in text form.

[0114] A "vehicle control system" is a system for controlling the operation of a vehicle in real time.

[0115] A "user interface" is an application or device that provides a means for a user to interact with a system and input or obtain information.

[0116] "Action instructions" are operational instructions for an autonomous vehicle provided by the generative AI model based on the analysis results.

[0117] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[0118] Server Side

[0119] The server is responsible for first loading the generative AI model. The server loads a pre-trained generative AI model (e.g., the GPT-4® model) into memory and generates images by inputting traffic condition prompts. Specifically, the server inputs the prompt "irregular traffic conditions" into the generative AI model, and generates corresponding images as a result. These generated images are saved as a dataset to be used later for model training. The storage destination is an image database (e.g., Amazon S3 or Google® Cloud Storage).

[0120] A concrete example is a process in which a server generates multiple images based on the prompt "sharp obstacle at an intersection" and stores the generated images in a dataset.

[0121] Terminal side

[0122] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. The terminal includes various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation based on the data collected from these sensors. The terminal analyzes the collected data and converts it into a specific text prompt. For example, the terminal may generate a text such as "A child suddenly jumps out at an intersection" and send the prompt to the generation AI for analysis.

[0123] The behavioral instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted from the terminal to the vehicle control system and immediately executed, allowing the self-driving car to respond quickly to sudden events.

[0124] A specific example is when a self-driving car is driving in an urban area and its sensors detect an obstacle that suddenly appears on the road, the device describes the situation in text format as "An obstacle has suddenly appeared on the road" and sends a prompt to the generating AI, which then generates an action instruction such as "apply the brakes suddenly and move to the right to avoid the obstacle," which the vehicle then carries out.

[0125] User Interface

[0126] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current status of the system and input any prompts to request analysis from the generating AI.

[0127] A specific example is when a user uses an application to check the system status and enters a prompt for "advanced driving techniques in bad weather," which results in specific measures being displayed, such as "slow down and increase your distance from the vehicle ahead."

[0128] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] Server Side

[0131] Step 1: Loading the Generative AI Model

[0132] The server loads the generative AI model (e.g., the GPT-4 model) from disk into memory.

[0133] Input: Disk file of a trained generative AI model

[0134] Output: Generative AI model loaded in memory

[0135] Specific operation: Loads the model file from disk and expands the model parameters into memory.

[0136] Step 2: Entering prompts and generating images

[0137] The server inputs the prompt "irregular traffic conditions" into the generative AI model and generates an image based on the model.

[0138] Input: Text prompt "Irregular traffic conditions"

[0139] Output: Generated traffic incident images

[0140] What it does: Input a prompt into a generative AI model and get the generated image as the model's output.

[0141] Step 3: Save the generated image

[0142] The server stores the generated images in a dataset.

[0143] Input: Generated traffic incident images

[0144] Output: Images stored in a database

[0145] Specific operation: Saves image data to a specified database (e.g., Amazon S3 or Google Cloud Storage).

[0146] Terminal side

[0147] Step 1: Collect sensor data

[0148] The device collects data in real time from various sensors (cameras, LiDAR, etc.) installed in the autonomous vehicle.

[0149] Input: Sensor data such as camera and LiDAR

[0150] Output: Collected sensor data

[0151] Specific operation: Data is read from each sensor and stored in memory for real-time processing.

[0152] Step 2: Generate a text prompt

[0153] The device analyzes the sensor data collected, understands the current driving situation, and generates text prompts.

[0154] Input: Collected sensor data

[0155] Output: Text prompt (e.g. "A child suddenly jumped out at the intersection.")

[0156] Specific operation: Sensor data is run through an analysis algorithm to generate a text representation of a specific traffic event.

[0157] Step 3: AI sending and analysis of prompts

[0158] The device sends the generated text prompt to the server and requests the generative AI model to analyze it.

[0159] Input: Text prompt

[0160] Output: Analysis results (action instructions)

[0161] Specific operation: Sends a text prompt to the server and receives the analysis results from the generative AI model.

[0162] Step 4: Implementing the Action Instructions

[0163] The terminal receives action instructions from the server and transmits them to the vehicle control system.

[0164] Input: Action instructions (e.g., brake suddenly and swerve left)

[0165] Output: Vehicle control actions

[0166] Specific actions: Action instructions are input into the vehicle control system and executed in real time.

[0167] User Interface

[0168] Step 1: Check the status

[0169] The user launches the application and sees the current state of the system, the generated images, and real-time responses.

[0170] Input: Launch application

[0171] Output: System status display

[0172] Specific behavior: Retrieves system information through the application and displays it in the user interface.

[0173] Step 2: Enter the prompts

[0174] The user inputs a prompt to make an additional analysis request to the system.

[0175] Input: The prompt entered by the user (e.g., "Advanced driving tips for bad weather")

[0176] Output: Prompt sent to generative AI model

[0177] What it does: Receives user input and sends it to the server.

[0178] Step 3: Check the analysis results

[0179] The generative AI model analyzes prompts entered by the user and displays the results in real time.

[0180] Input: Analysis results of the generative AI model

[0181] Output: Display of analysis results (e.g. "Slow down and increase distance from the vehicle ahead")

[0182] Specific actions: Display the analysis results in the user interface and provide specific countermeasures.

[0183] These steps allow servers, devices, and users to work together to improve the reliability and safety of automated driving systems.

[0184] (Application example 1)

[0185] 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."

[0186] In logistics centers, automated guided vehicles and drones must deal with many abnormal events and sudden obstacles. However, current systems have difficulty responding quickly and appropriately to these irregular situations. This leads to problems such as reduced work efficiency and safety risks. In addition, there are limited means for users to monitor these abnormal events in real time and provide appropriate response methods. A comprehensive system to solve these problems is needed.

[0187] 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.

[0188] In this invention, the server includes a means for recording the generated traffic incident images, a means for acquiring driving situations in the form of text prompts, and a generating artificial intelligence means for generating appropriate action instructions based on the text prompts. This enables automated guided vehicles and drones in a logistics center to respond quickly and appropriately to irregular situations. The server also includes a means for displaying action instructions through a user interface, a means for inputting arbitrary prompts using the user interface, and a means for providing analysis results generated based on the input prompts, allowing users to monitor abnormal events in real time and instruct appropriate responses.

[0189] A "generated traffic event image" is an image captured in real time by an automated driving system that reflects a specific traffic event.

[0190] A "text prompt format" is a way of expressing a particular situation or event in a text-based input format.

[0191] "Generative AI" refers to AI that receives text prompts and generates appropriate action instructions or analytical results based on them.

[0192] "Means for controlling a vehicle" refers to equipment or systems that adjust the operation of an automated driving vehicle or automated transport vehicle in accordance with the generated behavioral instructions.

[0193] A "logistics center" is a facility where goods, materials, etc. are stored, managed, and shipped.

[0194] An "automated guided vehicle (AGV)" is an unmanned vehicle that moves automatically along a pre-set route and transports cargo.

[0195] "Drone" means an unmanned aerial vehicle that is remotely controlled or autonomously capable of aerial movement and operations.

[0196] A "user interface" is the means or method by which a system and a user interact with each other.

[0197] "Sensor means" refers to a device or equipment that senses changes in the physical environment and acquires them as data.

[0198] "Real-time" means responding or processing immediately at the moment an event occurs.

[0199] The system for implementing the present invention comprises server, terminal, and user components.

[0200] Server Side

[0201] The server first loads the generative AI model and generates images that reflect specific traffic events or abnormal events within the logistics center. For example, it generates images based on a prompt such as "there is an obstacle in the aisle." The generated images are then saved in a dataset and used for model training. This improves the recognition accuracy of the AI ​​model. The main software used by the server is TENSORFLOW (registered trademark) and Python.

[0202] Terminal side

[0203] The terminal provides a means for collecting and analyzing data in real time from drones and automated guided vehicles (AGVs) within the logistics center. The terminal uses data obtained from cameras and sensors equipped on the drones and AGVs to analyze the current situation and describe it in the form of a text prompt. For example, the situation, "Part of a shelf has fallen," is converted into text. The prompt is then sent to a generation AI for analysis. The AI ​​model generates action instructions (for example, "Stop and notify a manager") as a result of the analysis, and the terminal transmits them to the drones and AGVs for immediate execution.

[0204] User Interface

[0205] Users are provided with a smartphone application that allows them to check the system status, generated images, and the real-time response status of drones and AGVs within the logistics center. Through the application, users can monitor the current status of the system and input arbitrary prompts to request analysis from the generating AI. For example, if a user requests additional instructions on "how to respond when passing other AGVs in narrow aisles" within the logistics center, they can input the prompt into the application and receive analysis results from the AI ​​in real time.

[0206] Specific examples

[0207] Server-side example:

[0208] The server loads the generative AI model and generates multiple images based on the prompt, "There is an obstacle in the path." The generated images are saved in a dataset and used for future model training.

[0209] Example of terminal operation:

[0210] The device receives real-time images from drones and automated guided vehicles (AGVs), describes the situation in text format, such as "part of a shelf has fallen," and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "stop and notify the administrator," and the device acts in accordance with that instruction.

[0211] An example of how the user interface works:

[0212] The user checks the system status using a smartphone application and inputs a prompt for "What to do if you are passing another AGV in a narrow passage." As a result, the system displays a specific countermeasure: "Stop and wait for the other AGV to pass, then proceed."

[0213] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0214] Step 1:

[0215] The server loads the generative AI model. Specifically, it loads the generative AI model file that was previously saved using TensorFlow into the system. The input is the file path of the model. The output is the model loaded into memory.

[0216] Step 2:

[0217] The server generates images that reflect specific traffic events or abnormal events within the logistics center. Specifically, a prompt statement (e.g., "There is an obstacle in the aisle") is input into the generative AI model, and an image is generated based on that. The input is the prompt statement, and the output is the generated image.

[0218] Step 3:

[0219] The server saves the generated images as a dataset and later uses them for model training. Specifically, it saves the images in a specified directory. The generated images are used as input. The output is saved in the dataset.

[0220] Step 4:

[0221] The terminal receives image data acquired in real time from drones and automated guided vehicles (AGVs). Specifically, the terminal acquires images in real time from cameras and sensors attached to the terminal. The input is real-time data from the cameras and sensors. The output is the acquired image data.

[0222] Step 5:

[0223] The terminal analyzes the acquired image data and describes it in the form of a text prompt. Specifically, it uses an image processing algorithm to detect abnormal situations and converts them into text format. The input is the acquired image data. The output is a text prompt (e.g., "Part of the shelf has fallen").

[0224] Step 6:

[0225] The terminal sends the generated text prompt to the server. Specifically, the prompt is transferred to the server using network communication. The input is the text prompt. The output is the received data on the server side.

[0226] Step 7:

[0227] The server uses a generative AI model to generate appropriate instructions based on the received text prompt. Specifically, the server inputs the prompt into the model and generates instructions for that prompt. The input is the text prompt. The output is the instruction for that prompt (e.g., "Stop and notify an administrator").

[0228] Step 8:

[0229] The server transmits the generated action instructions to the terminal. Specifically, it transfers the action instructions to the terminal via the network. The input is the action instructions. The output is the received data on the terminal side.

[0230] Step 9:

[0231] The terminal controls drones and automated guided vehicles (AGVs) based on the received action instructions. Specifically, it analyzes the action instructions and sends corresponding control signals to the device. The action instructions are the input, and the device's control operation is the output.

[0232] Step 10:

[0233] Users use a smartphone application to check the system status and response status. Specifically, real-time data and analysis results are displayed on the app. The input is real-time data from the system. The output is user confirmation and input of additional instructions.

[0234] Step 11:

[0235] The user inputs additional prompts into the application and requests the generation AI to analyze them. Specifically, the user enters a prompt for a specific scenario into the application's input field and presses the submit button. The input is the user's prompt (e.g., "What to do if you pass another AGV in a narrow passageway"), and the output is the analysis result by the generation AI (e.g., "Stop and wait for the other AGV to pass").

[0236] The above are the specific processing steps of the system that realizes the application example.

[0237] 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.

[0238] ---

[0239] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[0240] Server Side

[0241] ---

[0242] The server first uses the generative AI model to generate images of various traffic events. This process includes the following steps: The server loads the generative AI model and generates images by entering the prompt "irregular traffic situation." The generated images are then saved in a dataset for model training. This process allows the recognition accuracy of the AI ​​model to be improved.

[0243] Terminal side

[0244] ---

[0245] The device collects data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle in real time, analyzes the current driving situation, and describes it in the form of a text prompt. For example, it generates a specific situation such as "A child suddenly jumps out into the street at an intersection" as text. The prompt is sent to the generation AI for analysis. The action instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[0246] Emotion Engine

[0247] ---

[0248] The emotion engine is responsible for analyzing the user's emotions in real time and providing that data to the generative artificial intelligence means. The emotion engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. Based on this emotional data, the system can provide optimal driving control according to the user's state, such as whether they are stressed or relaxed.

[0249] User Interface

[0250] ---

[0251] Users are provided with a mobile application that allows them to view the system's status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[0252] Specific examples

[0253] ---

[0254] A specific example of system implementation is shown below.

[0255] Server-side example:

[0256] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[0257] Example of terminal operation:

[0258] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[0259] Example of the Emotion Engine in action:

[0260] The emotion engine detects when the user's heart rate spikes and they are feeling stressed. Based on this information, the system switches the driving mode to a safety-first mode and adjusts the music and lighting in the system to ease the user's tension.

[0261] An example of how the user interface works:

[0262] The user uses the application to check the system status and input advice for specific traffic scenarios, such as the prompt "Tell me some advanced driving tips for bad weather," which results in specific measures such as "slow down and increase your distance from the vehicle ahead."

[0263] By using these means, the automated driving system of the present invention can improve the recognition accuracy in irregular situations and realize safe and reliable automated driving. Furthermore, by combining it with an emotion engine, it becomes possible to provide an optimal driving environment according to the user's condition.

[0264] The processing flow will be explained below.

[0265] Server-side processing

[0266] ---

[0267] Step 1:

[0268] The server loads the generative AI model.

[0269] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[0270] Step 2:

[0271] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[0272] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[0273] Step 3:

[0274] The server stores the generated images in a dataset.

[0275] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[0276] Terminal-side processing

[0277] ---

[0278] Step 1:

[0279] The device acquires the current driving situation from sensors installed in the vehicle.

[0280] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[0281] Step 2:

[0282] The device analyzes the collected data and generates a text prompt.

[0283] It generates a specific textual description of the situation, such as "A child suddenly jumped out at the intersection."

[0284] Step 3:

[0285] The terminal sends the generated text prompt to the AI.

[0286] The prompt is sent over the network to the generating AI, which begins analyzing it.

[0287] Step 4:

[0288] The terminal receives action instructions from the AI.

[0289] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[0290] Step 5:

[0291] The vehicle is controlled based on the action instructions received by the terminal.

[0292] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[0293] Emotion engine processing

[0294] ---

[0295] Step 1:

[0296] The emotion engine captures the user's emotions in real time.

[0297] The emotion engine analyzes the user's facial expressions, voice, and heart rate.

[0298] Step 2:

[0299] The emotion engine analyzes the acquired emotion data and determines the user's emotional state.

[0300] Determine whether the user is stressed or relaxed.

[0301] Step 3:

[0302] The emotion engine generates response instructions according to the emotional state.

[0303] For example, if the user is feeling stressed, an instruction to switch to a safe driving mode is generated.

[0304] Step 4:

[0305] The terminal receives response instructions from the emotion engine and transmits them to the vehicle control system.

[0306] The vehicle control system adjusts the driving mode based on the corresponding instruction.

[0307] User Interface Handling

[0308] ---

[0309] Step 1:

[0310] The user launches an application for the mobile terminal.

[0311] Users access the system by launching the app on their smartphone or tablet.

[0312] Step 2:

[0313] The user checks the current state of the system.

[0314] The application displays irregular images generated and the latest AI analysis results.

[0315] Step 3:

[0316] The user enters any prompts.

[0317] For example, enter the prompt "Tell me how to drive on a rainy day."

[0318] Step 4:

[0319] The user checks the prompt analysis results.

[0320] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[0321] Example 2

[0322] 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."

[0323] Current automated driving systems have issues with responding to irregular traffic events and performing appropriate driving control based on the user's emotional state. In particular, rapid and accurate responses to unforeseen situations are required, but current systems are insufficient in this regard. In addition, the lack of a mechanism to recognize the user's emotions and adjust driving control based on them makes it difficult to achieve both safety and comfort.

[0324] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, an artificial intelligence generation means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, an emotion analysis means for analyzing the user's emotions in real time, and a means for adjusting driving control based on the emotion analysis result. This makes it possible to quickly respond to irregular traffic incidents and provide optimal driving control based on the user's emotional state.

[0325] "Means for recording generated traffic event images" refers to a device or software that has the function of saving images of traffic events generated by a generative AI model on a recording medium.

[0326] "Means for obtaining driving status in text prompt format" refers to a device or software that has the function of analyzing information obtained from sensors installed in the vehicle and generating the current driving status in text format.

[0327] "Generative artificial intelligence means" refers to a system or program that implements an artificial intelligence algorithm that generates optimal action instructions based on text prompts.

[0328] "Vehicle control means" refers to a device or software that has the function of performing actions such as accelerating, decelerating, avoiding, and stopping the autonomous vehicle based on behavioral instructions obtained from the generating artificial intelligence means.

[0329] "Emotion analysis means for analyzing a user's emotions in real time" refers to a device or software that has the function of collecting and analyzing data such as a user's facial expressions, voice, and heart rate in real time and determining the user's emotional state.

[0330] "Means for adjusting driving control" refers to a device or software that has the function of adjusting the driving mode and various comfort functions of an autonomous vehicle based on emotional state data obtained from the emotion analysis means.

[0331] "Sensor means" refers to various sensors (camera, LiDAR, microphone, pulse sensor, etc.) installed in a vehicle to detect the surrounding conditions and the internal state of the vehicle.

[0332] "User interface" refers to a device or software that provides an interface for a user to check the system status and input any prompts.

[0333] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[0334] Server Side

[0335] The server first loads the generative AI model. For example, OpenAI's DALL-E and Stable Diffusion are commonly used AI models for image generation. The server uses these models to generate images of various traffic events. The prompt uses short text, such as "A sudden obstacle at an intersection" or "A sudden obstacle appeared on the road." The images generated in this way are saved in a dataset and used for subsequent model training. This process improves the recognition accuracy of the AI ​​model, enabling more accurate analysis.

[0336] Terminal side

[0337] The device collects data in real time from various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation. For example, it combines images captured by the camera and distance data obtained from LiDAR to generate a text prompt describing a specific situation, such as "A child suddenly jumps out at an intersection." The generated prompt is sent to a generative AI model, which analyzes it and generates optimal instructions for action. In this case, the generative AI issues an instruction such as "apply sudden brakes and swerve left," and the device transmits the instruction to the vehicle control system, which immediately executes it.

[0338] Emotion Engine

[0339] The emotion engine analyzes the user's emotions in real time. This engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. For example, facial expression analysis technology is used to analyze the user's facial expressions captured by a camera, and tone of voice analysis technology is used to analyze their voice. Heart rate is also monitored using a biosensor. Based on the emotional data obtained in this way, the system adjusts driving control. Specifically, if the user is feeling stressed, the system will switch to a safety-first setting and change the in-car music and lighting to relaxing ones.

[0340] User Interface

[0341] Users are provided with a mobile application. Using this application, they can check the system status, generated images, and real-time response status. Users can also input any prompt through the application and ask the AI ​​to analyze it. For example, by inputting a prompt such as "Tell me how to drive in bad weather," they can immediately receive specific advice such as "Slow down and increase the distance from the vehicle ahead." In this way, users can always take optimal driving actions based on the latest information.

[0342] By combining these elements, the automated driving system according to the present invention can quickly respond to irregular traffic events and provide optimal driving control based on the user's emotional state.

[0343] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0344] Step 1:

[0345] The server loads the generative AI model. Specifically, it selects a generative AI model such as DALL-E or Stable Diffusion and loads the necessary libraries and model files into memory. The input for this step is the file path and configuration information of the generative AI model, and the output is an instance of the loaded generative AI model.

[0346] Step 2:

[0347] The server generates a prompt for the irregular traffic event and inputs it to the generative AI model. For example, it generates a prompt saying "a sharp obstacle at an intersection." The input of this step is the configuration information about the irregular traffic event, and the output is the generated prompt sentence.

[0348] Step 3:

[0349] The server inputs a prompt sentence into the generative AI model and generates an image. The generated image is saved in a dataset. The input of this step is the prompt sentence, and the output is a file of the generated image. For example, an image of a vehicle or pedestrian suddenly appearing at an intersection is generated.

[0350] Step 4:

[0351] The terminal collects data in real time from sensors (cameras, LiDAR, etc.) installed on the vehicle. For example, the forward-facing camera detects obstacles on the road. The input of this step is the sensor data, and the output is a set of collected raw data.

[0352] Step 5:

[0353] The device analyzes the collected sensor data and generates a textual description of the driving situation. Specifically, it uses an image analysis algorithm to analyze the sensor data and generate a specific text prompt, such as "A child suddenly jumped out at the intersection." The input of this step is the sensor data, and the output is the generated text prompt.

[0354] Step 6:

[0355] The device sends the generated text prompt to the generative AI model for analysis. The generative AI model generates optimal action instructions based on the prompt. For example, an instruction such as "apply sudden brakes and swerve left" may be generated. The input of this step is the text prompt, and the output is an action instruction.

[0356] Step 7:

[0357] The terminal transmits the action instructions obtained from the generative AI model to the vehicle control system and executes them immediately. Specifically, it transmits instructions to the vehicle's control unit, for example, to apply sudden brakes and swerve to the left. The input of this step is the action instructions, and the output is the executed vehicle behavior.

[0358] Step 8:

[0359] The emotion engine collects and analyzes the user's facial expressions, voice, heart rate, etc. in real time. For example, it uses the user's facial expressions captured by a camera and heart rate data obtained from a biosensor. The input of this step is the user's biometric data, and the output is the analyzed emotional state.

[0360] Step 9:

[0361] The emotion engine adjusts driving control based on the emotion analysis results. Specifically, if the user is feeling stressed, it takes measures such as switching to a safety-first driving mode. The input of this step is the emotion analysis results, and the output is the adjusted driving control settings.

[0362] Step 10:

[0363] The user uses a mobile application to check the system status and input a prompt, such as "Tell me how to drive in bad weather." The input for this step is the user's input prompt, and the output is the analysis result from the generative AI model.

[0364] Step 11:

[0365] The user checks the analysis results obtained through the application and puts them into action. For example, they receive advice such as "slow down and increase the distance from the vehicle ahead" and follow it. The input of this step is the analysis results, and the output is the driving behavior that the user follows.

[0366] Through these processing steps, the system can respond quickly and appropriately to irregular traffic events and provide optimal driving control based on the user's emotional state.

[0367] (Application example 2)

[0368] 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."

[0369] Conventional technologies have had difficulty in controlling autonomous vehicles to appropriately reflect irregular traffic events and the emotional state of the user. In particular, existing technologies have been inadequate in situations where driving operations are required to take into account risk avoidance due to sudden traffic events and the stress and anxiety felt by the user. For this reason, improving the safety of autonomous vehicles and the sense of security of users is an important issue.

[0370] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0371] In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, a generating artificial intelligence means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, a means for analyzing the emotional state, and a means for adjusting driving control based on the emotional state, thereby enabling a quick and appropriate response to irregular traffic incidents and enabling safe driving control that takes the user's emotional state into consideration.

[0372] The "means for recording generated traffic event images" refers to a device or mechanism for storing image data relating to traffic events generated by the server.

[0373] The "means for acquiring driving status in text prompt format" refers to a device or mechanism for converting information obtained from various sensors and cameras into text format and acquiring the information while the vehicle is in operation.

[0374] A "generative artificial intelligence means" is a device or system that utilizes artificial intelligence techniques used to generate appropriate action instructions based on text prompts.

[0375] The "means for controlling the vehicle" refers to a device or mechanism for executing driving operations of the vehicle based on the generated behavioral instructions.

[0376] The "means for analyzing emotional state" is a device or system for analyzing data such as the user's facial expression, voice, and heart rate to determine the user's emotional state.

[0377] A "means for adjusting driving controls" is a device or mechanism for adjusting the vehicle's driving operations or environmental settings based on the analyzed emotional state.

[0378] "Sensor means" refers to sensor devices such as cameras, LiDAR, and heart rate sensors for monitoring driving conditions in real time.

[0379] The "means for describing in text form" is a device or mechanism for converting data obtained from the sensor means into a text prompt format and describing it.

[0380] A "user interface" is an interface device or software that allows a user to check the system status and input any prompts.

[0381] A "prompt input means" is a mechanism or device that allows a user to input instructions in the form of text prompts relating to a particular scenario or situation.

[0382] The "means for providing analysis results" is a device or mechanism for presenting to the user action instructions and analysis results generated based on the text prompts.

[0383] This invention is an automated driving system that responds quickly and appropriately to irregular traffic events and realizes optimal driving control based on the user's emotional state. This system is composed of several main components, including a server, a terminal, a user, and an emotion engine.

[0384] Server-side configuration and operation

[0385] The server is responsible for loading the generative AI model and generating images of traffic events, following these steps:

[0386] 1. Loading a generative AI model: For example, using a deep learning library (e.g., TensorFlow), load a generative AI model that simulates traffic events.

[0387] 2. Entering a prompt: Enter the prompt "irregular traffic conditions" into the model.

[0388] 3. Image generation: The generative AI model generates an image of the traffic incident based on the prompt.

[0389] 4. Saving the dataset: The generated images are saved as a dataset on the server and used for model training.

[0390] Terminal configuration and operation

[0391] The device collects and analyzes data in real time from sensors (e.g., cameras, LiDAR) installed in the vehicle. This includes the following steps:

[0392] 1. Description of the situation in the form of a text prompt: For example, describe a specific situation in the form of a text prompt, such as, "A child suddenly jumps out at an intersection."

[0393] 2. Analysis and generation of action instructions: The prompt sentence is sent to the generative AI model, which analyzes it. As a result of the analysis, an action instruction such as "apply sudden brakes and move left to avoid the collision" is generated.

[0394] 3. Transmission to vehicle control: Control the vehicle based on the action instructions.

[0395] Emotion engine configuration and operation

[0396] The emotion engine is responsible for analyzing the user's emotional state in real time, specifically including the following steps:

[0397] 1. Emotion data collection: Data such as the user's facial expressions, voice, and heart rate are collected using sensors (e.g., smartphone built-in camera, heart rate sensor).

[0398] 2. Emotional state determination: Analyze the collected data to determine the user's emotional state (e.g., stressed or relaxed).

[0399] 3. Driving control adjustment: If the user is feeling stressed, the system will switch the driving mode to safety priority and even adjust the music and lighting within the system.

[0400] User interface configuration and operation

[0401] Through a mobile terminal application, users can check the system status, generated images, real-time response status, and input prompt statements.

[0402] Specific examples:

[0403] Situation: User is in the driver's seat, vehicle is traveling through a congested city.

[0404] Example prompt: "A sharp obstacle at the intersection."

[0405] Generated action instructions: "Suddenly brake and move to the right."

[0406] In this way, safe and optimal driving control based on the user's emotional state is possible.

[0407] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0408] Server-side processing steps

[0409] Step 1:

[0410] The server loads the generative AI model, which involves initializing it based on the prompt sentence "irregular traffic conditions," for example, using the TensorFlow library.

[0411] Input: Generative AI model, prompt "Irregular traffic conditions"

[0412] Output: Initialized model

[0413] Step 2:

[0414] The server inputs the prompt sentence into the generative AI model and generates an image of the traffic event. The generated image is a simulated traffic event.

[0415] Input: prompt statement, initialized model

[0416] Output: Generated traffic incident images

[0417] Step 3:

[0418] The server saves the generated images in a dataset for future model training, which improves the model's recognition accuracy.

[0419] Input: Generated image

[0420] Output: A saved image dataset

[0421] Terminal processing steps

[0422] Step 1:

[0423] The terminal collects data in real time from various sensors (e.g., cameras, LiDAR) installed in the vehicle.

[0424] Input: Sensor output data

[0425] Output: Real-time data collected

[0426] Step 2:

[0427] The device analyzes the collected data and describes the current driving situation in the form of a text prompt, such as "A child suddenly jumped out at the intersection."

[0428] Input: Real-time data collected

[0429] Output: Driving status in text prompt format

[0430] Step 3:

[0431] The device sends the prompt text to the generative AI model for analysis, which then generates appropriate action instructions.

[0432] Input: Driving situation in text prompt format, Generative AI model

[0433] Output: Action instructions

[0434] Step 4:

[0435] The terminal transmits the generated action instruction to the vehicle control system and executes it immediately.

[0436] Input: Action instructions

[0437] Output: Vehicle control executed

[0438] Emotion Engine Processing Steps

[0439] Step 1:

[0440] The emotion engine collects the user's facial expressions, voice, heart rate, and other data in real time.

[0441] Input: User's facial expression, voice, and heart rate data

[0442] Output: Collected emotion data

[0443] Step 2:

[0444] The emotion engine analyzes the collected data to determine the user's emotional state, for example, whether the user is stressed or relaxed.

[0445] Input: Collected emotion data

[0446] Output: User's emotional state

[0447] Step 3:

[0448] The emotion engine adjusts driving controls based on the user's emotional state. For example, if the user is stressed, it switches the driving mode to safety-first and adjusts the music and lighting in the system.

[0449] Input: User's emotional state

[0450] Output: Coordinated driving control

[0451] User Interface Processing Steps

[0452] Step 1:

[0453] Users can check the system status, generated images, and real-time response status through a mobile device application.

[0454] Input: System status, generated images, response status

[0455] Output: Visual information presented to the user

[0456] Step 2:

[0457] The user uses the application to input a prompt, for example, "I would like to know the best way to drive at a busy intersection."

[0458] Input: The prompt text entered by the user

[0459] Output: The prompt sent to the system

[0460] Step 3:

[0461] The device uses generative artificial intelligence to provide analysis results based on the input prompt. For example, in response to the prompt "Sudden obstacle at intersection," the device generates and presents the action instruction "Apply sudden braking and move right to avoid the obstacle" to the user.

[0462] Input: A prompt entered by the user, a generative AI model

[0463] Output: Analysis results and action instructions

[0464] 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.

[0465] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0466] 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.

[0467] [Second embodiment]

[0468] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0469] 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.

[0470] 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).

[0471] 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.

[0472] 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.

[0473] 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).

[0474] 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.

[0475] 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.

[0476] 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.

[0477] 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.

[0478] 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.

[0479] 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."

[0480] ---

[0481] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[0482] Server Side

[0483] ---

[0484] The server first uses the generative AI model to generate images of various traffic events. Specifically, the server loads the generative AI model and generates images by inputting the prompt "irregular traffic situation." The generated images are then saved in a dataset for use in model training. This process improves the recognition accuracy of the AI ​​model.

[0485] Terminal side

[0486] ---

[0487] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. Using data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle, the terminal analyzes the current driving situation and describes it in the form of a text prompt. For example, it generates a specific situation as text, such as "A child suddenly jumps out at an intersection." The prompt is sent to the generation AI for analysis. The resulting action instructions (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[0488] User Interface

[0489] ---

[0490] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[0491] Specific examples

[0492] ---

[0493] A specific example of system implementation is shown below.

[0494] Server-side example:

[0495] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[0496] Example of terminal operation:

[0497] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[0498] An example of how the user interface works:

[0499] The user uses the application to check the system status and input advice for specific traffic scenarios, such as entering a prompt for "advanced driving techniques in bad weather," which results in specific measures such as "slow down and increase the distance from the vehicle ahead."

[0500] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[0501] The processing flow will be explained below.

[0502] Server-side processing

[0503] ---

[0504] Step 1:

[0505] The server loads the generative AI model.

[0506] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[0507] Step 2:

[0508] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[0509] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[0510] Step 3:

[0511] The server stores the generated images in a dataset.

[0512] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[0513] Terminal-side processing

[0514] ---

[0515] Step 1:

[0516] The device acquires the current driving situation from sensors installed in the vehicle.

[0517] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[0518] Step 2:

[0519] The device analyzes the collected data and generates a text prompt.

[0520] Based on the analysis results, a specific text description of the situation, such as "A child suddenly jumps out at an intersection," is generated.

[0521] Step 3:

[0522] The terminal sends the generated text prompt to the AI.

[0523] The prompt is sent over the network to the generating AI, which begins analyzing it.

[0524] Step 4:

[0525] The terminal receives action instructions from the AI.

[0526] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[0527] Step 5:

[0528] The vehicle is controlled based on the action instructions received by the terminal.

[0529] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[0530] User Interface Handling

[0531] ---

[0532] Step 1:

[0533] The user launches an application for the mobile terminal.

[0534] Users access the system by launching the app on their smartphone or tablet.

[0535] Step 2:

[0536] The user checks the current state of the system.

[0537] The application displays irregular images generated and the latest AI analysis results.

[0538] Step 3:

[0539] The user enters any prompts.

[0540] For example, enter the prompt "Tell me how to drive on a rainy day."

[0541] Step 4:

[0542] The user checks the prompt analysis results.

[0543] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[0544] Through each of these steps, the server, terminal, and user work together seamlessly to create an autonomous driving system that can quickly respond to irregular situations.

[0545] Example 1

[0546] 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."

[0547] When autonomous vehicles encounter irregular traffic events, it is difficult for them to respond quickly and appropriately. This increases the risk of accidents in unexpected situations, potentially threatening the safety of passengers and those around them. Furthermore, conventional autonomous driving systems lack the ability to analyze data in real time and respond immediately, limiting their ability to generate optimal driving behavior. It is necessary to resolve these issues and improve the reliability and safety of autonomous driving systems.

[0548] 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.

[0549] In this invention, the server includes a means for recording the generated traffic event image, a means for acquiring the driving situation in the form of a text prompt, a generative AI model means for generating appropriate behavioral instructions based on the text prompt, a means for controlling the vehicle based on the behavioral instructions, a means for loading the generative AI model by the server, a means for inputting the traffic situation prompt and generating an image by the server, and a means for saving the generated image in a dataset, thereby enabling analysis of traffic events and generation of appropriate driving behavior in real time.

[0550] A "generative AI model" is an artificial intelligence model used to generate images or text based on specific input prompts.

[0551] A "prompt" is a text input that prompts a generative AI model to generate a specific output.

[0552] A "server" is a computer system that provides central management and processing, such as loading generative AI models, entering prompts, and performing analysis.

[0553] "Traffic incident images" are image data that visually represent traffic conditions and incidents.

[0554] A "dataset" is a collection containing a set of data used for training or analysis.

[0555] "Sensor means" means a device (e.g., camera, LiDAR, radar) installed on an autonomous vehicle for collecting data on the surrounding environment.

[0556] A "text prompt" is a prompt that expresses a traffic situation or event in text form.

[0557] A "vehicle control system" is a system for controlling the operation of a vehicle in real time.

[0558] A "user interface" is an application or device that provides a means for a user to interact with a system and input or obtain information.

[0559] "Action instructions" are operational instructions for an autonomous vehicle provided by the generative AI model based on the analysis results.

[0560] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[0561] Server Side

[0562] The server is responsible for first loading the generative AI model. The server loads a pre-trained generative AI model (e.g., the GPT-4 model) into memory and generates images by inputting traffic situation prompts. Specifically, the server inputs the prompt "irregular traffic situation" into the generative AI model, and generates corresponding images as a result. These generated images are saved as a dataset to be used later for model training. The saving destination is an image database (e.g., Amazon S3 or Google Cloud Storage).

[0563] A concrete example is a process in which a server generates multiple images based on the prompt "sharp obstacle at an intersection" and stores the generated images in a dataset.

[0564] Terminal side

[0565] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. The terminal includes various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation based on the data collected from these sensors. The terminal analyzes the collected data and converts it into a specific text prompt. For example, the terminal may generate a text such as "A child suddenly jumps out at an intersection" and send the prompt to the generation AI for analysis.

[0566] The behavioral instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted from the terminal to the vehicle control system and immediately executed, allowing the self-driving car to respond quickly to sudden events.

[0567] A specific example is when a self-driving car is driving in an urban area and its sensors detect an obstacle that suddenly appears on the road, the device describes the situation in text format as "An obstacle has suddenly appeared on the road" and sends a prompt to the generating AI, which then generates an action instruction such as "apply the brakes suddenly and move to the right to avoid the obstacle," which the vehicle then carries out.

[0568] User Interface

[0569] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current status of the system and input any prompts to request analysis from the generating AI.

[0570] A specific example is when a user uses an application to check the system status and enters a prompt for "advanced driving techniques in bad weather," which results in specific measures being displayed, such as "slow down and increase your distance from the vehicle ahead."

[0571] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[0572] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0573] Server Side

[0574] Step 1: Loading the Generative AI Model

[0575] The server loads the generative AI model (e.g., the GPT-4 model) from disk into memory.

[0576] Input: Disk file of a trained generative AI model

[0577] Output: Generative AI model loaded in memory

[0578] Specific operation: Loads the model file from disk and expands the model parameters into memory.

[0579] Step 2: Entering prompts and generating images

[0580] The server inputs the prompt "irregular traffic conditions" into the generative AI model and generates an image based on the model.

[0581] Input: Text prompt "Irregular traffic conditions"

[0582] Output: Generated traffic incident images

[0583] What it does: Input a prompt into a generative AI model and get the generated image as the model's output.

[0584] Step 3: Save the generated image

[0585] The server stores the generated images in a dataset.

[0586] Input: Generated traffic incident images

[0587] Output: Images stored in a database

[0588] Specific operation: Saves image data to a specified database (e.g., Amazon S3 or Google Cloud Storage).

[0589] Terminal side

[0590] Step 1: Collect sensor data

[0591] The device collects data in real time from various sensors (cameras, LiDAR, etc.) installed in the autonomous vehicle.

[0592] Input: Sensor data such as camera and LiDAR

[0593] Output: Collected sensor data

[0594] Specific operation: Data is read from each sensor and stored in memory for real-time processing.

[0595] Step 2: Generate a text prompt

[0596] The device analyzes the sensor data collected, understands the current driving situation, and generates text prompts.

[0597] Input: Collected sensor data

[0598] Output: Text prompt (e.g. "A child suddenly jumped out at the intersection.")

[0599] Specific operation: Sensor data is run through an analysis algorithm to generate a text representation of a specific traffic event.

[0600] Step 3: AI sending and analysis of prompts

[0601] The device sends the generated text prompt to the server and requests the generative AI model to analyze it.

[0602] Input: Text prompt

[0603] Output: Analysis results (action instructions)

[0604] Specific operation: Sends a text prompt to the server and receives the analysis results from the generative AI model.

[0605] Step 4: Implementing the Action Instructions

[0606] The terminal receives action instructions from the server and transmits them to the vehicle control system.

[0607] Input: Action instructions (e.g., brake suddenly and swerve left)

[0608] Output: Vehicle control actions

[0609] Specific actions: Action instructions are input into the vehicle control system and executed in real time.

[0610] User Interface

[0611] Step 1: Check the status

[0612] The user launches the application and sees the current state of the system, the generated images, and real-time responses.

[0613] Input: Launch application

[0614] Output: System status display

[0615] Specific behavior: Retrieves system information through the application and displays it in the user interface.

[0616] Step 2: Enter the prompts

[0617] The user inputs a prompt to make an additional analysis request to the system.

[0618] Input: The prompt entered by the user (e.g., "Advanced driving tips for bad weather")

[0619] Output: Prompt sent to generative AI model

[0620] What it does: Receives user input and sends it to the server.

[0621] Step 3: Check the analysis results

[0622] The generative AI model analyzes prompts entered by the user and displays the results in real time.

[0623] Input: Analysis results of the generative AI model

[0624] Output: Display of analysis results (e.g. "Slow down and increase distance from the vehicle ahead")

[0625] Specific actions: Display the analysis results in the user interface and provide specific countermeasures.

[0626] These steps allow servers, devices, and users to work together to improve the reliability and safety of automated driving systems.

[0627] (Application example 1)

[0628] 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."

[0629] In logistics centers, automated guided vehicles and drones must deal with many abnormal events and sudden obstacles. However, current systems have difficulty responding quickly and appropriately to these irregular situations. This leads to problems such as reduced work efficiency and safety risks. In addition, there are limited means for users to monitor these abnormal events in real time and provide appropriate response methods. A comprehensive system to solve these problems is needed.

[0630] 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.

[0631] In this invention, the server includes a means for recording the generated traffic incident images, a means for acquiring driving situations in the form of text prompts, and a generating artificial intelligence means for generating appropriate action instructions based on the text prompts. This enables automated guided vehicles and drones in a logistics center to respond quickly and appropriately to irregular situations. The server also includes a means for displaying action instructions through a user interface, a means for inputting arbitrary prompts using the user interface, and a means for providing analysis results generated based on the input prompts, allowing users to monitor abnormal events in real time and instruct appropriate responses.

[0632] A "generated traffic event image" is an image captured in real time by an automated driving system that reflects a specific traffic event.

[0633] A "text prompt format" is a way of expressing a particular situation or event in a text-based input format.

[0634] "Generative AI" refers to AI that receives text prompts and generates appropriate action instructions or analytical results based on them.

[0635] "Means for controlling a vehicle" refers to equipment or systems that adjust the operation of an automated driving vehicle or automated transport vehicle in accordance with the generated behavioral instructions.

[0636] A "logistics center" is a facility where goods, materials, etc. are stored, managed, and shipped.

[0637] An "automated guided vehicle (AGV)" is an unmanned vehicle that moves automatically along a pre-set route and transports cargo.

[0638] "Drone" means an unmanned aerial vehicle that is remotely controlled or autonomously capable of aerial movement and operations.

[0639] A "user interface" is the means or method by which a system and a user interact with each other.

[0640] "Sensor means" refers to a device or equipment that senses changes in the physical environment and acquires them as data.

[0641] "Real-time" means responding or processing immediately at the moment an event occurs.

[0642] The system for implementing the present invention comprises server, terminal, and user components.

[0643] Server Side

[0644] The server first loads the generative AI model and generates images that reflect specific traffic events or abnormal events within the logistics center. For example, it generates images based on a prompt such as "there is an obstacle in the aisle." The generated images are then saved in a dataset and used for model training, improving the recognition accuracy of the AI ​​model. The main software used by the server is TensorFlow and Python.

[0645] Terminal side

[0646] The terminal provides a means for collecting and analyzing data in real time from drones and automated guided vehicles (AGVs) within the logistics center. The terminal uses data obtained from cameras and sensors equipped on the drones and AGVs to analyze the current situation and describe it in the form of a text prompt. For example, the situation, "Part of a shelf has fallen," is converted into text. The prompt is then sent to a generation AI for analysis. The AI ​​model generates action instructions (for example, "Stop and notify a manager") as a result of the analysis, and the terminal transmits them to the drones and AGVs for immediate execution.

[0647] User Interface

[0648] Users are provided with a smartphone application that allows them to check the system status, generated images, and the real-time response status of drones and AGVs within the logistics center. Through the application, users can monitor the current status of the system and input arbitrary prompts to request analysis from the generating AI. For example, if a user requests additional instructions on "how to respond when passing other AGVs in narrow aisles" within the logistics center, they can input the prompt into the application and receive analysis results from the AI ​​in real time.

[0649] Specific examples

[0650] Server-side example:

[0651] The server loads the generative AI model and generates multiple images based on the prompt, "There is an obstacle in the path." The generated images are saved in a dataset and used for future model training.

[0652] Example of terminal operation:

[0653] The device receives real-time images from drones and automated guided vehicles (AGVs), describes the situation in text format, such as "part of a shelf has fallen," and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "stop and notify the administrator," and the device acts in accordance with that instruction.

[0654] An example of how the user interface works:

[0655] The user checks the system status using a smartphone application and inputs a prompt for "What to do if you are passing another AGV in a narrow passage." As a result, the system displays a specific countermeasure: "Stop and wait for the other AGV to pass, then proceed."

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

[0657] Step 1:

[0658] The server loads the generative AI model. Specifically, it loads the generative AI model file that was previously saved using TensorFlow into the system. The input is the file path of the model. The output is the model loaded into memory.

[0659] Step 2:

[0660] The server generates images that reflect specific traffic events or abnormal events within the logistics center. Specifically, a prompt statement (e.g., "There is an obstacle in the aisle") is input into the generative AI model, and an image is generated based on that. The input is the prompt statement, and the output is the generated image.

[0661] Step 3:

[0662] The server saves the generated images as a dataset and later uses them for model training. Specifically, it saves the images in a specified directory. The generated images are used as input. The output is saved in the dataset.

[0663] Step 4:

[0664] The terminal receives image data acquired in real time from drones and automated guided vehicles (AGVs). Specifically, the terminal acquires images in real time from cameras and sensors attached to the terminal. The input is real-time data from the cameras and sensors. The output is the acquired image data.

[0665] Step 5:

[0666] The terminal analyzes the acquired image data and describes it in the form of a text prompt. Specifically, it uses an image processing algorithm to detect abnormal situations and converts them into text format. The input is the acquired image data. The output is a text prompt (e.g., "Part of the shelf has fallen").

[0667] Step 6:

[0668] The terminal sends the generated text prompt to the server. Specifically, the prompt is transferred to the server using network communication. The input is the text prompt. The output is the received data on the server side.

[0669] Step 7:

[0670] The server uses a generative AI model to generate appropriate instructions based on the received text prompt. Specifically, the server inputs the prompt into the model and generates instructions for that prompt. The input is the text prompt. The output is the instruction for that prompt (e.g., "Stop and notify an administrator").

[0671] Step 8:

[0672] The server transmits the generated action instructions to the terminal. Specifically, it transfers the action instructions to the terminal via the network. The input is the action instructions. The output is the received data on the terminal side.

[0673] Step 9:

[0674] The terminal controls drones and automated guided vehicles (AGVs) based on the received action instructions. Specifically, it analyzes the action instructions and sends corresponding control signals to the device. The action instructions are the input, and the device's control operation is the output.

[0675] Step 10:

[0676] Users use a smartphone application to check the system status and response status. Specifically, real-time data and analysis results are displayed on the app. The input is real-time data from the system. The output is user confirmation and input of additional instructions.

[0677] Step 11:

[0678] The user inputs additional prompts into the application and requests the generation AI to analyze them. Specifically, the user enters a prompt for a specific scenario into the application's input field and presses the submit button. The input is the user's prompt (e.g., "What to do if you pass another AGV in a narrow passageway"), and the output is the analysis result by the generation AI (e.g., "Stop and wait for the other AGV to pass").

[0679] The above are the specific processing steps of the system that realizes the application example.

[0680] 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.

[0681] ---

[0682] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[0683] Server Side

[0684] ---

[0685] The server first uses the generative AI model to generate images of various traffic events. This process includes the following steps: The server loads the generative AI model and generates images by entering the prompt "irregular traffic situation." The generated images are then saved in a dataset for model training. This process allows the recognition accuracy of the AI ​​model to be improved.

[0686] Terminal side

[0687] ---

[0688] The device collects data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle in real time, analyzes the current driving situation, and describes it in the form of a text prompt. For example, it generates a specific situation such as "A child suddenly jumps out into the street at an intersection" as text. The prompt is sent to the generation AI for analysis. The action instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[0689] Emotion Engine

[0690] ---

[0691] The emotion engine is responsible for analyzing the user's emotions in real time and providing that data to the generative artificial intelligence means. The emotion engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. Based on this emotional data, the system can provide optimal driving control according to the user's state, such as whether they are stressed or relaxed.

[0692] User Interface

[0693] ---

[0694] Users are provided with a mobile application that allows them to view the system's status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[0695] Specific examples

[0696] ---

[0697] A specific example of system implementation is shown below.

[0698] Server-side example:

[0699] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[0700] Example of terminal operation:

[0701] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[0702] Example of the Emotion Engine in action:

[0703] The emotion engine detects when the user's heart rate spikes and they are feeling stressed. Based on this information, the system switches the driving mode to a safety-first mode and adjusts the music and lighting in the system to ease the user's tension.

[0704] An example of how the user interface works:

[0705] The user uses the application to check the system status and input advice for specific traffic scenarios, such as the prompt "Tell me some advanced driving tips for bad weather," which results in specific measures such as "slow down and increase your distance from the vehicle ahead."

[0706] By using these means, the automated driving system of the present invention can improve the recognition accuracy in irregular situations and realize safe and reliable automated driving. Furthermore, by combining it with an emotion engine, it becomes possible to provide an optimal driving environment according to the user's condition.

[0707] The processing flow will be explained below.

[0708] Server-side processing

[0709] ---

[0710] Step 1:

[0711] The server loads the generative AI model.

[0712] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[0713] Step 2:

[0714] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[0715] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[0716] Step 3:

[0717] The server stores the generated images in a dataset.

[0718] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[0719] Terminal-side processing

[0720] ---

[0721] Step 1:

[0722] The device acquires the current driving situation from sensors installed in the vehicle.

[0723] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[0724] Step 2:

[0725] The device analyzes the collected data and generates a text prompt.

[0726] It generates a specific textual description of the situation, such as "A child suddenly jumped out at the intersection."

[0727] Step 3:

[0728] The terminal sends the generated text prompt to the AI.

[0729] The prompt is sent over the network to the generating AI, which begins analyzing it.

[0730] Step 4:

[0731] The terminal receives action instructions from the AI.

[0732] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[0733] Step 5:

[0734] The vehicle is controlled based on the action instructions received by the terminal.

[0735] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[0736] Emotion engine processing

[0737] ---

[0738] Step 1:

[0739] The emotion engine captures the user's emotions in real time.

[0740] The emotion engine analyzes the user's facial expressions, voice, and heart rate.

[0741] Step 2:

[0742] The emotion engine analyzes the acquired emotion data and determines the user's emotional state.

[0743] Determine whether the user is stressed or relaxed.

[0744] Step 3:

[0745] The emotion engine generates response instructions according to the emotional state.

[0746] For example, if the user is feeling stressed, an instruction to switch to a safe driving mode is generated.

[0747] Step 4:

[0748] The terminal receives response instructions from the emotion engine and transmits them to the vehicle control system.

[0749] The vehicle control system adjusts the driving mode based on the corresponding instruction.

[0750] User Interface Handling

[0751] ---

[0752] Step 1:

[0753] The user launches an application for the mobile terminal.

[0754] Users access the system by launching the app on their smartphone or tablet.

[0755] Step 2:

[0756] The user checks the current state of the system.

[0757] The application displays irregular images generated and the latest AI analysis results.

[0758] Step 3:

[0759] The user enters any prompts.

[0760] For example, enter the prompt "Tell me how to drive on a rainy day."

[0761] Step 4:

[0762] The user checks the prompt analysis results.

[0763] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[0764] Example 2

[0765] 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."

[0766] Current automated driving systems have issues with responding to irregular traffic events and performing appropriate driving control based on the user's emotional state. In particular, rapid and accurate responses to unforeseen situations are required, but current systems are insufficient in this regard. In addition, the lack of a mechanism to recognize the user's emotions and adjust driving control based on them makes it difficult to achieve both safety and comfort.

[0767] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, an artificial intelligence generation means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, an emotion analysis means for analyzing the user's emotions in real time, and a means for adjusting driving control based on the emotion analysis result. This makes it possible to quickly respond to irregular traffic incidents and provide optimal driving control based on the user's emotional state.

[0768] "Means for recording generated traffic event images" refers to a device or software that has the function of saving images of traffic events generated by a generative AI model on a recording medium.

[0769] "Means for obtaining driving status in text prompt format" refers to a device or software that has the function of analyzing information obtained from sensors installed in the vehicle and generating the current driving status in text format.

[0770] "Generative artificial intelligence means" refers to a system or program that implements an artificial intelligence algorithm that generates optimal action instructions based on text prompts.

[0771] "Vehicle control means" refers to a device or software that has the function of performing actions such as accelerating, decelerating, avoiding, and stopping the autonomous vehicle based on behavioral instructions obtained from the generating artificial intelligence means.

[0772] "Emotion analysis means for analyzing a user's emotions in real time" refers to a device or software that has the function of collecting and analyzing data such as a user's facial expressions, voice, and heart rate in real time and determining the user's emotional state.

[0773] "Means for adjusting driving control" refers to a device or software that has the function of adjusting the driving mode and various comfort functions of an autonomous vehicle based on emotional state data obtained from the emotion analysis means.

[0774] "Sensor means" refers to various sensors (camera, LiDAR, microphone, pulse sensor, etc.) installed in a vehicle to detect the surrounding conditions and the internal state of the vehicle.

[0775] "User interface" refers to a device or software that provides an interface for a user to check the system status and input any prompts.

[0776] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[0777] Server Side

[0778] The server first loads a generative AI model. For example, OpenAI's DALL-E and Stable Diffusion are commonly used AI models for image generation. The server uses these models to generate images of various traffic events. The prompt uses short text, such as "A sudden obstacle at an intersection" or "A sudden obstacle appeared on the road." The images generated in this way are saved in a dataset and used for subsequent model training. This process improves the recognition accuracy of the AI ​​model, enabling more accurate analysis.

[0779] Terminal side

[0780] The device collects data in real time from various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation. For example, it combines images captured by the camera and distance data obtained from LiDAR to generate a text prompt describing a specific situation, such as "A child suddenly jumps out at an intersection." The generated prompt is sent to a generative AI model, which analyzes it and generates optimal instructions for action. In this case, the generative AI issues an instruction such as "apply sudden brakes and swerve left," and the device transmits the instruction to the vehicle control system, which immediately executes it.

[0781] Emotion Engine

[0782] The emotion engine analyzes the user's emotions in real time. This engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. For example, facial expression analysis technology is used to analyze the user's facial expressions captured by a camera, and tone of voice analysis technology is used to analyze their voice. Heart rate is also monitored using a biosensor. Based on the emotional data obtained in this way, the system adjusts driving control. Specifically, if the user is feeling stressed, the system will switch to a safety-first setting and change the in-car music and lighting to relaxing ones.

[0783] User Interface

[0784] Users are provided with a mobile application. Using this application, they can check the system status, generated images, and real-time response status. Users can also input any prompt through the application and ask the AI ​​to analyze it. For example, by inputting a prompt such as "Tell me how to drive in bad weather," they can immediately receive specific advice such as "Slow down and increase the distance from the vehicle ahead." In this way, users can always take optimal driving actions based on the latest information.

[0785] By combining these elements, the automated driving system according to the present invention can quickly respond to irregular traffic events and provide optimal driving control based on the user's emotional state.

[0786] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0787] Step 1:

[0788] The server loads the generative AI model. Specifically, it selects a generative AI model such as DALL-E or Stable Diffusion and loads the necessary libraries and model files into memory. The input for this step is the file path and configuration information of the generative AI model, and the output is an instance of the loaded generative AI model.

[0789] Step 2:

[0790] The server generates a prompt for the irregular traffic event and inputs it to the generative AI model. For example, it generates a prompt saying "a sharp obstacle at an intersection." The input of this step is the configuration information about the irregular traffic event, and the output is the generated prompt sentence.

[0791] Step 3:

[0792] The server inputs a prompt sentence into the generative AI model and generates an image. The generated image is saved in a dataset. The input of this step is the prompt sentence, and the output is a file of the generated image. For example, an image of a vehicle or pedestrian suddenly appearing at an intersection is generated.

[0793] Step 4:

[0794] The terminal collects data in real time from sensors (cameras, LiDAR, etc.) installed on the vehicle. For example, the forward-facing camera detects obstacles on the road. The input of this step is the sensor data, and the output is a set of collected raw data.

[0795] Step 5:

[0796] The device analyzes the collected sensor data and generates a textual description of the driving situation. Specifically, it uses an image analysis algorithm to analyze the sensor data and generate a specific text prompt, such as "A child suddenly jumped out at the intersection." The input of this step is the sensor data, and the output is the generated text prompt.

[0797] Step 6:

[0798] The device sends the generated text prompt to the generative AI model for analysis. The generative AI model generates optimal action instructions based on the prompt. For example, an instruction such as "apply sudden brakes and swerve left" may be generated. The input of this step is the text prompt, and the output is an action instruction.

[0799] Step 7:

[0800] The terminal transmits the action instructions obtained from the generative AI model to the vehicle control system and executes them immediately. Specifically, it transmits instructions to the vehicle's control unit, for example, to apply sudden brakes and swerve to the left. The input of this step is the action instructions, and the output is the executed vehicle behavior.

[0801] Step 8:

[0802] The emotion engine collects and analyzes the user's facial expressions, voice, heart rate, etc. in real time. For example, it uses the user's facial expressions captured by a camera and heart rate data obtained from a biosensor. The input of this step is the user's biometric data, and the output is the analyzed emotional state.

[0803] Step 9:

[0804] The emotion engine adjusts driving control based on the emotion analysis results. Specifically, if the user is feeling stressed, it takes measures such as switching to a safety-first driving mode. The input of this step is the emotion analysis results, and the output is the adjusted driving control settings.

[0805] Step 10:

[0806] The user uses a mobile application to check the system status and input a prompt, such as "Tell me how to drive in bad weather." The input for this step is the user's input prompt, and the output is the analysis result from the generative AI model.

[0807] Step 11:

[0808] The user checks the analysis results obtained through the application and puts them into action. For example, they receive advice such as "slow down and increase the distance from the vehicle ahead" and follow it. The input of this step is the analysis results, and the output is the driving behavior that the user follows.

[0809] Through these processing steps, the system can respond quickly and appropriately to irregular traffic events and provide optimal driving control based on the user's emotional state.

[0810] (Application example 2)

[0811] 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."

[0812] Conventional technologies have had difficulty in controlling autonomous vehicles to appropriately reflect irregular traffic events and the emotional state of the user. In particular, existing technologies have been inadequate in situations where driving operations are required to take into account risk avoidance due to sudden traffic events and the stress and anxiety felt by the user. For this reason, improving the safety of autonomous vehicles and the sense of security of users is an important issue.

[0813] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0814] In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, a generating artificial intelligence means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, a means for analyzing the emotional state, and a means for adjusting driving control based on the emotional state, thereby enabling a quick and appropriate response to irregular traffic incidents and enabling safe driving control that takes the user's emotional state into consideration.

[0815] The "means for recording generated traffic event images" refers to a device or mechanism for storing image data relating to traffic events generated by the server.

[0816] The "means for acquiring driving status in text prompt format" refers to a device or mechanism for converting information obtained from various sensors and cameras into text format and acquiring the information while the vehicle is in operation.

[0817] A "generative artificial intelligence means" is a device or system that utilizes artificial intelligence techniques used to generate appropriate action instructions based on text prompts.

[0818] The "means for controlling the vehicle" refers to a device or mechanism for executing driving operations of the vehicle based on the generated behavioral instructions.

[0819] The "means for analyzing emotional state" is a device or system for analyzing data such as the user's facial expression, voice, and heart rate to determine the user's emotional state.

[0820] A "means for adjusting driving controls" is a device or mechanism for adjusting the vehicle's driving operations or environmental settings based on the analyzed emotional state.

[0821] "Sensor means" refers to sensor devices such as cameras, LiDAR, and heart rate sensors for monitoring driving conditions in real time.

[0822] The "means for describing in text form" is a device or mechanism for converting data obtained from the sensor means into a text prompt format and describing it.

[0823] A "user interface" is an interface device or software that allows a user to check the system status and input any prompts.

[0824] A "prompt input means" is a mechanism or device that allows a user to input instructions in the form of text prompts relating to a particular scenario or situation.

[0825] The "means for providing analysis results" is a device or mechanism for presenting to the user action instructions and analysis results generated based on the text prompts.

[0826] This invention is an automated driving system that responds quickly and appropriately to irregular traffic events and realizes optimal driving control based on the user's emotional state. This system is composed of several main components, including a server, a terminal, a user, and an emotion engine.

[0827] Server-side configuration and operation

[0828] The server is responsible for loading the generative AI model and generating images of traffic events, following these steps:

[0829] 1. Loading a generative AI model: For example, using a deep learning library (e.g., TensorFlow), load a generative AI model that simulates traffic events.

[0830] 2. Entering a prompt: Enter the prompt "irregular traffic conditions" into the model.

[0831] 3. Image generation: The generative AI model generates an image of the traffic incident based on the prompt.

[0832] 4. Saving the dataset: The generated images are saved as a dataset on the server and used for model training.

[0833] Terminal configuration and operation

[0834] The device collects and analyzes data in real time from sensors (e.g., cameras, LiDAR) installed in the vehicle. This includes the following steps:

[0835] 1. Description of the situation in the form of a text prompt: For example, describe a specific situation in the form of a text prompt, such as, "A child suddenly jumps out at an intersection."

[0836] 2. Analysis and generation of action instructions: The prompt sentence is sent to the generative AI model, which analyzes it. As a result of the analysis, an action instruction such as "apply sudden brakes and move left to avoid the collision" is generated.

[0837] 3. Transmission to vehicle control: Control the vehicle based on the action instructions.

[0838] Emotion engine configuration and operation

[0839] The emotion engine is responsible for analyzing the user's emotional state in real time, specifically including the following steps:

[0840] 1. Emotion data collection: Data such as the user's facial expressions, voice, and heart rate are collected using sensors (e.g., smartphone built-in camera, heart rate sensor).

[0841] 2. Emotional state determination: Analyze the collected data to determine the user's emotional state (e.g., stressed or relaxed).

[0842] 3. Driving control adjustment: If the user is feeling stressed, the system will switch the driving mode to safety priority and even adjust the music and lighting within the system.

[0843] User interface configuration and operation

[0844] Through a mobile terminal application, users can check the system status, generated images, real-time response status, and input prompt statements.

[0845] Specific examples:

[0846] Situation: User is in the driver's seat, vehicle is traveling through a congested city.

[0847] Example prompt: "A sharp obstacle at the intersection."

[0848] Generated action instructions: "Suddenly brake and move to the right."

[0849] In this way, safe and optimal driving control based on the user's emotional state is possible.

[0850] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0851] Server-side processing steps

[0852] Step 1:

[0853] The server loads the generative AI model, which involves initializing it based on the prompt sentence "irregular traffic conditions," for example, using the TensorFlow library.

[0854] Input: Generative AI model, prompt "Irregular traffic conditions"

[0855] Output: Initialized model

[0856] Step 2:

[0857] The server inputs the prompt sentence into the generative AI model and generates an image of the traffic event. The generated image is a simulated traffic event.

[0858] Input: prompt statement, initialized model

[0859] Output: Generated traffic incident images

[0860] Step 3:

[0861] The server saves the generated images in a dataset for future model training, which improves the model's recognition accuracy.

[0862] Input: Generated image

[0863] Output: A saved image dataset

[0864] Terminal processing steps

[0865] Step 1:

[0866] The terminal collects data in real time from various sensors (e.g., cameras, LiDAR) installed in the vehicle.

[0867] Input: Sensor output data

[0868] Output: Real-time data collected

[0869] Step 2:

[0870] The device analyzes the collected data and describes the current driving situation in the form of a text prompt, such as "A child suddenly jumped out at the intersection."

[0871] Input: Real-time data collected

[0872] Output: Driving status in text prompt format

[0873] Step 3:

[0874] The device sends the prompt text to the generative AI model for analysis, which then generates appropriate action instructions.

[0875] Input: Driving situation in text prompt format, Generative AI model

[0876] Output: Action instructions

[0877] Step 4:

[0878] The terminal transmits the generated action instruction to the vehicle control system and executes it immediately.

[0879] Input: Action instructions

[0880] Output: Vehicle control executed

[0881] Emotion Engine Processing Steps

[0882] Step 1:

[0883] The emotion engine collects the user's facial expressions, voice, heart rate, and other data in real time.

[0884] Input: User's facial expression, voice, and heart rate data

[0885] Output: Collected emotion data

[0886] Step 2:

[0887] The emotion engine analyzes the collected data to determine the user's emotional state, for example, whether the user is stressed or relaxed.

[0888] Input: Collected emotion data

[0889] Output: User's emotional state

[0890] Step 3:

[0891] The emotion engine adjusts driving controls based on the user's emotional state. For example, if the user is stressed, it switches the driving mode to safety-first and adjusts the music and lighting in the system.

[0892] Input: User's emotional state

[0893] Output: Coordinated driving control

[0894] User Interface Processing Steps

[0895] Step 1:

[0896] Users can check the system status, generated images, and real-time response status through a mobile device application.

[0897] Input: System status, generated images, response status

[0898] Output: Visual information presented to the user

[0899] Step 2:

[0900] The user uses the application to input a prompt, for example, "I would like to know the best way to drive at a busy intersection."

[0901] Input: The prompt text entered by the user

[0902] Output: The prompt sent to the system

[0903] Step 3:

[0904] The device uses generative artificial intelligence to provide analysis results based on the input prompt. For example, in response to the prompt "Sudden obstacle at intersection," the device generates and presents the action instruction "Apply sudden braking and move right to avoid the obstacle" to the user.

[0905] Input: A prompt entered by the user, a generative AI model

[0906] Output: Analysis results and action instructions

[0907] 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.

[0908] 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.

[0909] 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.

[0910] [Third embodiment]

[0911] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0912] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0913] 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).

[0914] 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.

[0915] 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.

[0916] 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).

[0917] 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.

[0918] 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.

[0919] 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.

[0920] 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.

[0921] 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.

[0922] 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."

[0923] ---

[0924] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[0925] Server Side

[0926] ---

[0927] The server first uses the generative AI model to generate images of various traffic events. Specifically, the server loads the generative AI model and generates images by inputting the prompt "irregular traffic situation." The generated images are then saved in a dataset for use in model training. This process improves the recognition accuracy of the AI ​​model.

[0928] Terminal side

[0929] ---

[0930] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. Using data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle, the terminal analyzes the current driving situation and describes it in the form of a text prompt. For example, it generates a specific situation as text, such as "A child suddenly jumps out at an intersection." The prompt is sent to the generation AI for analysis. The resulting action instructions (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[0931] User Interface

[0932] ---

[0933] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[0934] Specific examples

[0935] ---

[0936] A specific example of system implementation is shown below.

[0937] Server-side example:

[0938] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[0939] Example of terminal operation:

[0940] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[0941] An example of how the user interface works:

[0942] The user uses the application to check the system status and input advice for specific traffic scenarios, such as entering a prompt for "advanced driving techniques in bad weather," which results in specific measures such as "slow down and increase the distance from the vehicle ahead."

[0943] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[0944] The processing flow will be explained below.

[0945] Server-side processing

[0946] ---

[0947] Step 1:

[0948] The server loads the generative AI model.

[0949] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[0950] Step 2:

[0951] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[0952] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[0953] Step 3:

[0954] The server stores the generated images in a dataset.

[0955] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[0956] Terminal-side processing

[0957] ---

[0958] Step 1:

[0959] The device acquires the current driving situation from sensors installed in the vehicle.

[0960] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[0961] Step 2:

[0962] The device analyzes the collected data and generates a text prompt.

[0963] Based on the analysis results, a specific text description of the situation, such as "A child suddenly jumps out at an intersection," is generated.

[0964] Step 3:

[0965] The terminal sends the generated text prompt to the AI.

[0966] The prompt is sent over the network to the generating AI, which begins analyzing it.

[0967] Step 4:

[0968] The terminal receives action instructions from the AI.

[0969] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[0970] Step 5:

[0971] The vehicle is controlled based on the action instructions received by the terminal.

[0972] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[0973] User Interface Handling

[0974] ---

[0975] Step 1:

[0976] The user launches an application for the mobile terminal.

[0977] Users access the system by launching the app on their smartphone or tablet.

[0978] Step 2:

[0979] The user checks the current state of the system.

[0980] The application displays irregular images generated and the latest AI analysis results.

[0981] Step 3:

[0982] The user enters any prompts.

[0983] For example, enter the prompt "Tell me how to drive on a rainy day."

[0984] Step 4:

[0985] The user checks the prompt analysis results.

[0986] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[0987] Through each of these steps, the server, terminal, and user work together seamlessly to create an autonomous driving system that can quickly respond to irregular situations.

[0988] Example 1

[0989] 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."

[0990] When autonomous vehicles encounter irregular traffic events, it is difficult for them to respond quickly and appropriately. This increases the risk of accidents in unexpected situations, potentially threatening the safety of passengers and those around them. Furthermore, conventional autonomous driving systems lack the ability to analyze data in real time and respond immediately, limiting their ability to generate optimal driving behavior. It is necessary to resolve these issues and improve the reliability and safety of autonomous driving systems.

[0991] 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.

[0992] In this invention, the server includes a means for recording the generated traffic event image, a means for acquiring the driving situation in the form of a text prompt, a generative AI model means for generating appropriate behavioral instructions based on the text prompt, a means for controlling the vehicle based on the behavioral instructions, a means for loading the generative AI model by the server, a means for inputting the traffic situation prompt and generating an image by the server, and a means for saving the generated image in a dataset, thereby enabling analysis of traffic events and generation of appropriate driving behavior in real time.

[0993] A "generative AI model" is an artificial intelligence model used to generate images or text based on specific input prompts.

[0994] A "prompt" is a text input that prompts a generative AI model to generate a specific output.

[0995] A "server" is a computer system that provides central management and processing, such as loading generative AI models, entering prompts, and performing analysis.

[0996] "Traffic incident images" are image data that visually represent traffic conditions and incidents.

[0997] A "dataset" is a collection containing a set of data used for training or analysis.

[0998] "Sensor means" means a device (e.g., camera, LiDAR, radar) installed on an autonomous vehicle for collecting data on the surrounding environment.

[0999] A "text prompt" is a prompt that expresses a traffic situation or event in text form.

[1000] A "vehicle control system" is a system for controlling the operation of a vehicle in real time.

[1001] A "user interface" is an application or device that provides a means for a user to interact with a system and input or obtain information.

[1002] "Action instructions" are operational instructions for an autonomous vehicle provided by the generative AI model based on the analysis results.

[1003] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[1004] Server Side

[1005] The server is responsible for first loading the generative AI model. The server loads a pre-trained generative AI model (e.g., the GPT-4 model) into memory and generates images by inputting traffic situation prompts. Specifically, the server inputs the prompt "irregular traffic situation" into the generative AI model, and generates corresponding images as a result. These generated images are saved as a dataset to be used later for model training. The saving destination is an image database (e.g., Amazon S3 or Google Cloud Storage).

[1006] A concrete example is a process in which a server generates multiple images based on the prompt "sharp obstacle at an intersection" and stores the generated images in a dataset.

[1007] Terminal side

[1008] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. The terminal includes various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation based on the data collected from these sensors. The terminal analyzes the collected data and converts it into a specific text prompt. For example, the terminal may generate a text such as "A child suddenly jumps out at an intersection" and send the prompt to the generation AI for analysis.

[1009] The behavioral instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted from the terminal to the vehicle control system and immediately executed, allowing the self-driving car to respond quickly to sudden events.

[1010] A specific example is when a self-driving car is driving in an urban area and its sensors detect an obstacle that suddenly appears on the road, the device describes the situation in text format as "An obstacle has suddenly appeared on the road" and sends a prompt to the generating AI, which then generates an action instruction such as "apply the brakes suddenly and move to the right to avoid the obstacle," which the vehicle then carries out.

[1011] User Interface

[1012] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current status of the system and input any prompts to request analysis from the generating AI.

[1013] A specific example is when a user uses an application to check the system status and enters a prompt for "advanced driving techniques in bad weather," which results in specific measures being displayed, such as "slow down and increase your distance from the vehicle ahead."

[1014] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[1015] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1016] Server Side

[1017] Step 1: Loading the Generative AI Model

[1018] The server loads the generative AI model (e.g., the GPT-4 model) from disk into memory.

[1019] Input: Disk file of a trained generative AI model

[1020] Output: Generative AI model loaded in memory

[1021] Specific operation: Loads the model file from disk and expands the model parameters into memory.

[1022] Step 2: Entering prompts and generating images

[1023] The server inputs the prompt "irregular traffic conditions" into the generative AI model and generates an image based on the model.

[1024] Input: Text prompt "Irregular traffic conditions"

[1025] Output: Generated traffic incident images

[1026] What it does: Input a prompt into a generative AI model and get the generated image as the model's output.

[1027] Step 3: Save the generated image

[1028] The server stores the generated images in a dataset.

[1029] Input: Generated traffic incident images

[1030] Output: Images stored in a database

[1031] Specific operation: Saves image data to a specified database (e.g., Amazon S3 or Google Cloud Storage).

[1032] Terminal side

[1033] Step 1: Collect sensor data

[1034] The device collects data in real time from various sensors (cameras, LiDAR, etc.) installed in the autonomous vehicle.

[1035] Input: Sensor data such as camera and LiDAR

[1036] Output: Collected sensor data

[1037] Specific operation: Data is read from each sensor and stored in memory for real-time processing.

[1038] Step 2: Generate a text prompt

[1039] The device analyzes the sensor data collected, understands the current driving situation, and generates text prompts.

[1040] Input: Collected sensor data

[1041] Output: Text prompt (e.g. "A child suddenly jumped out at the intersection.")

[1042] Specific operation: Sensor data is run through an analysis algorithm to generate a text representation of a specific traffic event.

[1043] Step 3: AI sending and analysis of prompts

[1044] The device sends the generated text prompt to the server and requests the generative AI model to analyze it.

[1045] Input: Text prompt

[1046] Output: Analysis results (action instructions)

[1047] Specific operation: Sends a text prompt to the server and receives the analysis results from the generative AI model.

[1048] Step 4: Implementing the Action Instructions

[1049] The terminal receives action instructions from the server and transmits them to the vehicle control system.

[1050] Input: Action instructions (e.g., brake suddenly and swerve left)

[1051] Output: Vehicle control actions

[1052] Specific actions: Action instructions are input into the vehicle control system and executed in real time.

[1053] User Interface

[1054] Step 1: Check the status

[1055] The user launches the application and sees the current state of the system, the generated images, and real-time responses.

[1056] Input: Launch application

[1057] Output: System status display

[1058] Specific behavior: Retrieves system information through the application and displays it in the user interface.

[1059] Step 2: Enter the prompts

[1060] The user inputs a prompt to make an additional analysis request to the system.

[1061] Input: The prompt entered by the user (e.g., "Advanced driving tips for bad weather")

[1062] Output: Prompt sent to generative AI model

[1063] What it does: Receives user input and sends it to the server.

[1064] Step 3: Check the analysis results

[1065] The generative AI model analyzes prompts entered by the user and displays the results in real time.

[1066] Input: Analysis results of the generative AI model

[1067] Output: Display of analysis results (e.g. "Slow down and increase distance from the vehicle ahead")

[1068] Specific actions: Display the analysis results in the user interface and provide specific countermeasures.

[1069] These steps allow servers, devices, and users to work together to improve the reliability and safety of automated driving systems.

[1070] (Application example 1)

[1071] 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."

[1072] In logistics centers, automated guided vehicles and drones must deal with many abnormal events and sudden obstacles. However, current systems have difficulty responding quickly and appropriately to these irregular situations. This leads to problems such as reduced work efficiency and safety risks. In addition, there are limited means for users to monitor these abnormal events in real time and provide appropriate response methods. A comprehensive system to solve these problems is needed.

[1073] 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.

[1074] In this invention, the server includes a means for recording the generated traffic incident images, a means for acquiring driving situations in the form of text prompts, and a generating artificial intelligence means for generating appropriate action instructions based on the text prompts. This enables automated guided vehicles and drones in a logistics center to respond quickly and appropriately to irregular situations. The server also includes a means for displaying action instructions through a user interface, a means for inputting arbitrary prompts using the user interface, and a means for providing analysis results generated based on the input prompts, allowing users to monitor abnormal events in real time and instruct appropriate responses.

[1075] A "generated traffic event image" is an image captured in real time by an automated driving system that reflects a specific traffic event.

[1076] A "text prompt format" is a way of expressing a particular situation or event in a text-based input format.

[1077] "Generative AI" refers to AI that receives text prompts and generates appropriate action instructions or analytical results based on them.

[1078] "Means for controlling a vehicle" refers to equipment or systems that adjust the operation of an automated driving vehicle or automated transport vehicle in accordance with the generated behavioral instructions.

[1079] A "logistics center" is a facility where goods, materials, etc. are stored, managed, and shipped.

[1080] An "automated guided vehicle (AGV)" is an unmanned vehicle that moves automatically along a pre-set route and transports cargo.

[1081] "Drone" means an unmanned aerial vehicle that is remotely controlled or autonomously capable of aerial movement and operations.

[1082] A "user interface" is the means or method by which a system and a user interact with each other.

[1083] "Sensor means" refers to a device or equipment that senses changes in the physical environment and acquires them as data.

[1084] "Real-time" means responding or processing immediately at the moment an event occurs.

[1085] The system for implementing the present invention comprises server, terminal, and user components.

[1086] Server Side

[1087] The server first loads the generative AI model and generates images that reflect specific traffic events or abnormal events within the logistics center. For example, it generates images based on a prompt such as "there is an obstacle in the aisle." The generated images are then saved in a dataset and used for model training, improving the recognition accuracy of the AI ​​model. The main software used by the server is TensorFlow and Python.

[1088] Terminal side

[1089] The terminal provides a means for collecting and analyzing data in real time from drones and automated guided vehicles (AGVs) within the logistics center. The terminal uses data obtained from cameras and sensors equipped on the drones and AGVs to analyze the current situation and describe it in the form of a text prompt. For example, the situation, "Part of a shelf has fallen," is converted into text. The prompt is then sent to a generation AI for analysis. The AI ​​model generates action instructions (for example, "Stop and notify a manager") as a result of the analysis, and the terminal transmits them to the drones and AGVs for immediate execution.

[1090] User Interface

[1091] Users are provided with a smartphone application that allows them to check the system status, generated images, and the real-time response status of drones and AGVs within the logistics center. Through the application, users can monitor the current status of the system and input arbitrary prompts to request analysis from the generating AI. For example, if a user requests additional instructions on "how to respond when passing other AGVs in narrow aisles" within the logistics center, they can input the prompt into the application and receive analysis results from the AI ​​in real time.

[1092] Specific examples

[1093] Server-side example:

[1094] The server loads the generative AI model and generates multiple images based on the prompt, "There is an obstacle in the path." The generated images are saved in a dataset and used for future model training.

[1095] Example of terminal operation:

[1096] The device receives real-time images from drones and automated guided vehicles (AGVs), describes the situation in text format, such as "part of a shelf has fallen," and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "stop and notify the administrator," and the device acts in accordance with that instruction.

[1097] An example of how the user interface works:

[1098] The user checks the system status using a smartphone application and inputs a prompt for "What to do if you are passing another AGV in a narrow passage." As a result, the system displays a specific countermeasure: "Stop and wait for the other AGV to pass, then proceed."

[1099] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1100] Step 1:

[1101] The server loads the generative AI model. Specifically, it loads the generative AI model file that was previously saved using TensorFlow into the system. The input is the file path of the model. The output is the model loaded into memory.

[1102] Step 2:

[1103] The server generates images that reflect specific traffic events or abnormal events within the logistics center. Specifically, a prompt statement (e.g., "There is an obstacle in the aisle") is input into the generative AI model, and an image is generated based on that. The input is the prompt statement, and the output is the generated image.

[1104] Step 3:

[1105] The server saves the generated images as a dataset and later uses them for model training. Specifically, it saves the images in a specified directory. The generated images are used as input. The output is saved in the dataset.

[1106] Step 4:

[1107] The terminal receives image data acquired in real time from drones and automated guided vehicles (AGVs). Specifically, the terminal acquires images in real time from cameras and sensors attached to the terminal. The input is real-time data from the cameras and sensors. The output is the acquired image data.

[1108] Step 5:

[1109] The terminal analyzes the acquired image data and describes it in the form of a text prompt. Specifically, it uses an image processing algorithm to detect abnormal situations and converts them into text format. The input is the acquired image data. The output is a text prompt (e.g., "Part of the shelf has fallen").

[1110] Step 6:

[1111] The terminal sends the generated text prompt to the server. Specifically, the prompt is transferred to the server using network communication. The input is the text prompt. The output is the received data on the server side.

[1112] Step 7:

[1113] The server uses a generative AI model to generate appropriate instructions based on the received text prompt. Specifically, the server inputs the prompt into the model and generates instructions for that prompt. The input is the text prompt. The output is the instruction for that prompt (e.g., "Stop and notify an administrator").

[1114] Step 8:

[1115] The server transmits the generated action instructions to the terminal. Specifically, it transfers the action instructions to the terminal via the network. The input is the action instructions. The output is the received data on the terminal side.

[1116] Step 9:

[1117] The terminal controls drones and automated guided vehicles (AGVs) based on the received action instructions. Specifically, it analyzes the action instructions and sends corresponding control signals to the device. The action instructions are the input, and the device's control operation is the output.

[1118] Step 10:

[1119] Users use a smartphone application to check the system status and response status. Specifically, real-time data and analysis results are displayed on the app. The input is real-time data from the system. The output is user confirmation and input of additional instructions.

[1120] Step 11:

[1121] The user inputs additional prompts into the application and requests the generation AI to analyze them. Specifically, the user enters a prompt for a specific scenario into the application's input field and presses the submit button. The input is the user's prompt (e.g., "What to do if you pass another AGV in a narrow passageway"), and the output is the analysis result by the generation AI (e.g., "Stop and wait for the other AGV to pass").

[1122] The above are the specific processing steps of the system that realizes the application example.

[1123] 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.

[1124] ---

[1125] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[1126] Server Side

[1127] ---

[1128] The server first uses the generative AI model to generate images of various traffic events. This process includes the following steps: The server loads the generative AI model and generates images by entering the prompt "irregular traffic situation." The generated images are then saved in a dataset for model training. This process allows the recognition accuracy of the AI ​​model to be improved.

[1129] Terminal side

[1130] ---

[1131] The device collects data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle in real time, analyzes the current driving situation, and describes it in the form of a text prompt. For example, it generates a specific situation such as "A child suddenly jumps out into the street at an intersection" as text. The prompt is sent to the generation AI for analysis. The action instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[1132] Emotion Engine

[1133] ---

[1134] The emotion engine is responsible for analyzing the user's emotions in real time and providing that data to the generative artificial intelligence means. The emotion engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. Based on this emotional data, the system can provide optimal driving control according to the user's state, such as whether they are stressed or relaxed.

[1135] User Interface

[1136] ---

[1137] Users are provided with a mobile application that allows them to view the system's status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[1138] Specific examples

[1139] ---

[1140] A specific example of system implementation is shown below.

[1141] Server-side example:

[1142] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[1143] Example of terminal operation:

[1144] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[1145] Example of the Emotion Engine in action:

[1146] The emotion engine detects when the user's heart rate spikes and they are feeling stressed. Based on this information, the system switches the driving mode to a safety-first mode and adjusts the music and lighting in the system to ease the user's tension.

[1147] An example of how the user interface works:

[1148] The user uses the application to check the system status and input advice for specific traffic scenarios, such as the prompt "Tell me some advanced driving tips for bad weather," which results in specific measures such as "slow down and increase your distance from the vehicle ahead."

[1149] By using these means, the automated driving system of the present invention can improve the recognition accuracy in irregular situations and realize safe and reliable automated driving. Furthermore, by combining it with an emotion engine, it becomes possible to provide an optimal driving environment according to the user's condition.

[1150] The processing flow will be explained below.

[1151] Server-side processing

[1152] ---

[1153] Step 1:

[1154] The server loads the generative AI model.

[1155] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[1156] Step 2:

[1157] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[1158] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[1159] Step 3:

[1160] The server stores the generated images in a dataset.

[1161] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[1162] Terminal-side processing

[1163] ---

[1164] Step 1:

[1165] The device acquires the current driving situation from sensors installed in the vehicle.

[1166] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[1167] Step 2:

[1168] The device analyzes the collected data and generates a text prompt.

[1169] It generates a specific textual description of the situation, such as "A child suddenly jumped out at the intersection."

[1170] Step 3:

[1171] The terminal sends the generated text prompt to the AI.

[1172] The prompt is sent over the network to the generating AI, which begins analyzing it.

[1173] Step 4:

[1174] The terminal receives action instructions from the AI.

[1175] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[1176] Step 5:

[1177] The vehicle is controlled based on the action instructions received by the terminal.

[1178] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[1179] Emotion engine processing

[1180] ---

[1181] Step 1:

[1182] The emotion engine captures the user's emotions in real time.

[1183] The emotion engine analyzes the user's facial expressions, voice, and heart rate.

[1184] Step 2:

[1185] The emotion engine analyzes the acquired emotion data and determines the user's emotional state.

[1186] Determine whether the user is stressed or relaxed.

[1187] Step 3:

[1188] The emotion engine generates response instructions according to the emotional state.

[1189] For example, if the user is feeling stressed, an instruction to switch to a safe driving mode is generated.

[1190] Step 4:

[1191] The terminal receives response instructions from the emotion engine and transmits them to the vehicle control system.

[1192] The vehicle control system adjusts the driving mode based on the corresponding instruction.

[1193] User Interface Handling

[1194] ---

[1195] Step 1:

[1196] The user launches an application for the mobile terminal.

[1197] Users access the system by launching the app on their smartphone or tablet.

[1198] Step 2:

[1199] The user checks the current state of the system.

[1200] The application displays irregular images generated and the latest AI analysis results.

[1201] Step 3:

[1202] The user enters any prompts.

[1203] For example, enter the prompt "Tell me how to drive on a rainy day."

[1204] Step 4:

[1205] The user checks the prompt analysis results.

[1206] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[1207] Example 2

[1208] 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."

[1209] Current automated driving systems have issues with responding to irregular traffic events and performing appropriate driving control based on the user's emotional state. In particular, rapid and accurate responses to unforeseen situations are required, but current systems are insufficient in this regard. In addition, the lack of a mechanism to recognize the user's emotions and adjust driving control based on them makes it difficult to achieve both safety and comfort.

[1210] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, an artificial intelligence generation means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, an emotion analysis means for analyzing the user's emotions in real time, and a means for adjusting driving control based on the emotion analysis result. This makes it possible to quickly respond to irregular traffic incidents and provide optimal driving control based on the user's emotional state.

[1211] "Means for recording generated traffic event images" refers to a device or software that has the function of saving images of traffic events generated by a generative AI model on a recording medium.

[1212] "Means for obtaining driving status in text prompt format" refers to a device or software that has the function of analyzing information obtained from sensors installed in the vehicle and generating the current driving status in text format.

[1213] "Generative artificial intelligence means" refers to a system or program that implements an artificial intelligence algorithm that generates optimal action instructions based on text prompts.

[1214] "Vehicle control means" refers to a device or software that has the function of performing actions such as accelerating, decelerating, avoiding, and stopping the autonomous vehicle based on behavioral instructions obtained from the generating artificial intelligence means.

[1215] "Emotion analysis means for analyzing a user's emotions in real time" refers to a device or software that has the function of collecting and analyzing data such as a user's facial expressions, voice, and heart rate in real time and determining the user's emotional state.

[1216] "Means for adjusting driving control" refers to a device or software that has the function of adjusting the driving mode and various comfort functions of an autonomous vehicle based on emotional state data obtained from the emotion analysis means.

[1217] "Sensor means" refers to various sensors (camera, LiDAR, microphone, pulse sensor, etc.) installed in a vehicle to detect the surrounding conditions and the internal state of the vehicle.

[1218] "User interface" refers to a device or software that provides an interface for a user to check the system status and input any prompts.

[1219] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[1220] Server Side

[1221] The server first loads a generative AI model. For example, OpenAI's DALL-E and Stable Diffusion are commonly used AI models for image generation. The server uses these models to generate images of various traffic events. The prompt uses short text, such as "A sudden obstacle at an intersection" or "A sudden obstacle appeared on the road." The images generated in this way are saved in a dataset and used for subsequent model training. This process improves the recognition accuracy of the AI ​​model, enabling more accurate analysis.

[1222] Terminal side

[1223] The device collects data in real time from various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation. For example, it combines images captured by the camera and distance data obtained from LiDAR to generate a text prompt describing a specific situation, such as "A child suddenly jumps out at an intersection." The generated prompt is sent to a generative AI model, which analyzes it and generates optimal instructions for action. In this case, the generative AI issues an instruction such as "apply sudden brakes and swerve left," and the device transmits the instruction to the vehicle control system, which immediately executes it.

[1224] Emotion Engine

[1225] The emotion engine analyzes the user's emotions in real time. This engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. For example, facial expression analysis technology is used to analyze the user's facial expressions captured by a camera, and tone of voice analysis technology is used to analyze their voice. Heart rate is also monitored using a biosensor. Based on the emotional data obtained in this way, the system adjusts driving control. Specifically, if the user is feeling stressed, the system will switch to a safety-first setting and change the in-car music and lighting to relaxing ones.

[1226] User Interface

[1227] Users are provided with a mobile application. Using this application, they can check the system status, generated images, and real-time response status. Users can also input any prompt through the application and ask the AI ​​to analyze it. For example, by inputting a prompt such as "Tell me how to drive in bad weather," they can immediately receive specific advice such as "Slow down and increase the distance from the vehicle ahead." In this way, users can always take optimal driving actions based on the latest information.

[1228] By combining these elements, the automated driving system according to the present invention can quickly respond to irregular traffic events and provide optimal driving control based on the user's emotional state.

[1229] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1230] Step 1:

[1231] The server loads the generative AI model. Specifically, it selects a generative AI model such as DALL-E or Stable Diffusion and loads the necessary libraries and model files into memory. The input for this step is the file path and configuration information of the generative AI model, and the output is an instance of the loaded generative AI model.

[1232] Step 2:

[1233] The server generates a prompt for the irregular traffic event and inputs it to the generative AI model. For example, it generates a prompt saying "a sharp obstacle at an intersection." The input of this step is the configuration information about the irregular traffic event, and the output is the generated prompt sentence.

[1234] Step 3:

[1235] The server inputs a prompt sentence into the generative AI model and generates an image. The generated image is saved in a dataset. The input of this step is the prompt sentence, and the output is a file of the generated image. For example, an image of a vehicle or pedestrian suddenly appearing at an intersection is generated.

[1236] Step 4:

[1237] The terminal collects data in real time from sensors (cameras, LiDAR, etc.) installed on the vehicle. For example, the forward-facing camera detects obstacles on the road. The input of this step is the sensor data, and the output is a set of collected raw data.

[1238] Step 5:

[1239] The device analyzes the collected sensor data and generates a textual description of the driving situation. Specifically, it uses an image analysis algorithm to analyze the sensor data and generate a specific text prompt, such as "A child suddenly jumped out at the intersection." The input of this step is the sensor data, and the output is the generated text prompt.

[1240] Step 6:

[1241] The device sends the generated text prompt to the generative AI model for analysis. The generative AI model generates optimal action instructions based on the prompt. For example, an instruction such as "apply sudden brakes and swerve left" may be generated. The input of this step is the text prompt, and the output is an action instruction.

[1242] Step 7:

[1243] The terminal transmits the action instructions obtained from the generative AI model to the vehicle control system and executes them immediately. Specifically, it transmits instructions to the vehicle's control unit, for example, to apply sudden brakes and swerve to the left. The input of this step is the action instructions, and the output is the executed vehicle behavior.

[1244] Step 8:

[1245] The emotion engine collects and analyzes the user's facial expressions, voice, heart rate, etc. in real time. For example, it uses the user's facial expressions captured by a camera and heart rate data obtained from a biosensor. The input of this step is the user's biometric data, and the output is the analyzed emotional state.

[1246] Step 9:

[1247] The emotion engine adjusts driving control based on the emotion analysis results. Specifically, if the user is feeling stressed, it takes measures such as switching to a safety-first driving mode. The input of this step is the emotion analysis results, and the output is the adjusted driving control settings.

[1248] Step 10:

[1249] The user uses a mobile application to check the system status and input a prompt, such as "Tell me how to drive in bad weather." The input for this step is the user's input prompt, and the output is the analysis result from the generative AI model.

[1250] Step 11:

[1251] The user checks the analysis results obtained through the application and puts them into action. For example, they receive advice such as "slow down and increase the distance from the vehicle ahead" and follow it. The input of this step is the analysis results, and the output is the driving behavior that the user follows.

[1252] Through these processing steps, the system can respond quickly and appropriately to irregular traffic events and provide optimal driving control based on the user's emotional state.

[1253] (Application example 2)

[1254] 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."

[1255] Conventional technologies have had difficulty in controlling autonomous vehicles to appropriately reflect irregular traffic events and the emotional state of the user. In particular, existing technologies have been inadequate in situations where driving operations are required to take into account risk avoidance due to sudden traffic events and the stress and anxiety felt by the user. For this reason, improving the safety of autonomous vehicles and the sense of security of users is an important issue.

[1256] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1257] In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, a generating artificial intelligence means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, a means for analyzing the emotional state, and a means for adjusting driving control based on the emotional state, thereby enabling a quick and appropriate response to irregular traffic incidents and enabling safe driving control that takes the user's emotional state into consideration.

[1258] The "means for recording generated traffic event images" refers to a device or mechanism for storing image data relating to traffic events generated by the server.

[1259] The "means for acquiring driving status in text prompt format" refers to a device or mechanism for converting information obtained from various sensors and cameras into text format and acquiring the information while the vehicle is in operation.

[1260] A "generative artificial intelligence means" is a device or system that utilizes artificial intelligence techniques used to generate appropriate action instructions based on text prompts.

[1261] The "means for controlling the vehicle" refers to a device or mechanism for executing driving operations of the vehicle based on the generated behavioral instructions.

[1262] The "means for analyzing emotional state" is a device or system for analyzing data such as the user's facial expression, voice, and heart rate to determine the user's emotional state.

[1263] A "means for adjusting driving controls" is a device or mechanism for adjusting the vehicle's driving operations or environmental settings based on the analyzed emotional state.

[1264] "Sensor means" refers to sensor devices such as cameras, LiDAR, and heart rate sensors for monitoring driving conditions in real time.

[1265] The "means for describing in text form" is a device or mechanism for converting data obtained from the sensor means into a text prompt format and describing it.

[1266] A "user interface" is an interface device or software that allows a user to check the system status and input any prompts.

[1267] A "prompt input means" is a mechanism or device that allows a user to input instructions in the form of text prompts relating to a particular scenario or situation.

[1268] The "means for providing analysis results" is a device or mechanism for presenting to the user action instructions and analysis results generated based on the text prompts.

[1269] This invention is an automated driving system that responds quickly and appropriately to irregular traffic events and realizes optimal driving control based on the user's emotional state. This system is composed of several main components, including a server, a terminal, a user, and an emotion engine.

[1270] Server-side configuration and operation

[1271] The server is responsible for loading the generative AI model and generating images of traffic events, following these steps:

[1272] 1. Loading a generative AI model: For example, using a deep learning library (e.g., TensorFlow), load a generative AI model that simulates traffic events.

[1273] 2. Entering a prompt: Enter the prompt "irregular traffic conditions" into the model.

[1274] 3. Image generation: The generative AI model generates an image of the traffic incident based on the prompt.

[1275] 4. Saving the dataset: The generated images are saved as a dataset on the server and used for model training.

[1276] Terminal configuration and operation

[1277] The device collects and analyzes data in real time from sensors (e.g., cameras, LiDAR) installed in the vehicle. This includes the following steps:

[1278] 1. Description of the situation in the form of a text prompt: For example, describe a specific situation in the form of a text prompt, such as, "A child suddenly jumps out at an intersection."

[1279] 2. Analysis and generation of action instructions: The prompt sentence is sent to the generative AI model, which analyzes it. As a result of the analysis, an action instruction such as "apply sudden brakes and move left to avoid the collision" is generated.

[1280] 3. Transmission to vehicle control: Control the vehicle based on the action instructions.

[1281] Emotion engine configuration and operation

[1282] The emotion engine is responsible for analyzing the user's emotional state in real time, specifically including the following steps:

[1283] 1. Emotion data collection: Data such as the user's facial expressions, voice, and heart rate are collected using sensors (e.g., smartphone built-in camera, heart rate sensor).

[1284] 2. Emotional state determination: Analyze the collected data to determine the user's emotional state (e.g., stressed or relaxed).

[1285] 3. Driving control adjustment: If the user is feeling stressed, the system will switch the driving mode to safety priority and even adjust the music and lighting within the system.

[1286] User interface configuration and operation

[1287] Through a mobile terminal application, users can check the system status, generated images, real-time response status, and input prompt statements.

[1288] Specific examples:

[1289] Situation: User is in the driver's seat, vehicle is traveling through a congested city.

[1290] Example prompt: "A sharp obstacle at the intersection."

[1291] Generated action instructions: "Suddenly brake and move to the right."

[1292] In this way, safe and optimal driving control based on the user's emotional state is possible.

[1293] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1294] Server-side processing steps

[1295] Step 1:

[1296] The server loads the generative AI model, which involves initializing it based on the prompt sentence "irregular traffic conditions," for example, using the TensorFlow library.

[1297] Input: Generative AI model, prompt "Irregular traffic conditions"

[1298] Output: Initialized model

[1299] Step 2:

[1300] The server inputs the prompt sentence into the generative AI model and generates an image of the traffic event. The generated image is a simulated traffic event.

[1301] Input: prompt statement, initialized model

[1302] Output: Generated traffic incident images

[1303] Step 3:

[1304] The server saves the generated images in a dataset for future model training, which improves the model's recognition accuracy.

[1305] Input: Generated image

[1306] Output: A saved image dataset

[1307] Terminal processing steps

[1308] Step 1:

[1309] The terminal collects data in real time from various sensors (e.g., cameras, LiDAR) installed in the vehicle.

[1310] Input: Sensor output data

[1311] Output: Real-time data collected

[1312] Step 2:

[1313] The device analyzes the collected data and describes the current driving situation in the form of a text prompt, such as "A child suddenly jumped out at the intersection."

[1314] Input: Real-time data collected

[1315] Output: Driving status in text prompt format

[1316] Step 3:

[1317] The device sends the prompt text to the generative AI model for analysis, which then generates appropriate action instructions.

[1318] Input: Driving situation in text prompt format, Generative AI model

[1319] Output: Action instructions

[1320] Step 4:

[1321] The terminal transmits the generated action instruction to the vehicle control system and executes it immediately.

[1322] Input: Action instructions

[1323] Output: Vehicle control executed

[1324] Emotion Engine Processing Steps

[1325] Step 1:

[1326] The emotion engine collects the user's facial expressions, voice, heart rate, and other data in real time.

[1327] Input: User's facial expression, voice, and heart rate data

[1328] Output: Collected emotion data

[1329] Step 2:

[1330] The emotion engine analyzes the collected data to determine the user's emotional state, for example, whether the user is stressed or relaxed.

[1331] Input: Collected emotion data

[1332] Output: User's emotional state

[1333] Step 3:

[1334] The emotion engine adjusts driving controls based on the user's emotional state. For example, if the user is stressed, it switches the driving mode to safety-first and adjusts the music and lighting in the system.

[1335] Input: User's emotional state

[1336] Output: Coordinated driving control

[1337] User Interface Processing Steps

[1338] Step 1:

[1339] Users can check the system status, generated images, and real-time response status through a mobile device application.

[1340] Input: System status, generated images, response status

[1341] Output: Visual information presented to the user

[1342] Step 2:

[1343] The user uses the application to input a prompt, for example, "I would like to know the best way to drive at a busy intersection."

[1344] Input: The prompt text entered by the user

[1345] Output: The prompt sent to the system

[1346] Step 3:

[1347] The device uses generative artificial intelligence to provide analysis results based on the input prompt. For example, in response to the prompt "Sudden obstacle at intersection," the device generates and presents the action instruction "Apply sudden braking and move right to avoid the obstacle" to the user.

[1348] Input: A prompt entered by the user, a generative AI model

[1349] Output: Analysis results and action instructions

[1350] 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.

[1351] 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.

[1352] 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.

[1353] [Fourth embodiment]

[1354] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1355] 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.

[1356] 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).

[1357] 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.

[1358] 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.

[1359] 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).

[1360] 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.

[1361] 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.

[1362] 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.

[1363] 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.

[1364] 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.

[1365] 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.

[1366] 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."

[1367] ---

[1368] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[1369] Server Side

[1370] ---

[1371] The server first uses the generative AI model to generate images of various traffic events. Specifically, the server loads the generative AI model and generates images by inputting the prompt "irregular traffic situation." The generated images are then saved in a dataset for use in model training. This process improves the recognition accuracy of the AI ​​model.

[1372] Terminal side

[1373] ---

[1374] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. Using data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle, the terminal analyzes the current driving situation and describes it in the form of a text prompt. For example, it generates a specific situation as text, such as "A child suddenly jumps out at an intersection." The prompt is sent to the generation AI for analysis. The resulting action instructions (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[1375] User Interface

[1376] ---

[1377] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[1378] Specific examples

[1379] ---

[1380] A specific example of system implementation is shown below.

[1381] Server-side example:

[1382] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[1383] Example of terminal operation:

[1384] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[1385] An example of how the user interface works:

[1386] The user uses the application to check the system status and input advice for specific traffic scenarios, such as entering a prompt for "advanced driving techniques in bad weather," which results in specific measures such as "slow down and increase the distance from the vehicle ahead."

[1387] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[1388] The processing flow will be explained below.

[1389] Server-side processing

[1390] ---

[1391] Step 1:

[1392] The server loads the generative AI model.

[1393] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[1394] Step 2:

[1395] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[1396] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[1397] Step 3:

[1398] The server stores the generated images in a dataset.

[1399] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[1400] Terminal-side processing

[1401] ---

[1402] Step 1:

[1403] The device acquires the current driving situation from sensors installed in the vehicle.

[1404] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[1405] Step 2:

[1406] The device analyzes the collected data and generates a text prompt.

[1407] Based on the analysis results, a specific text description of the situation, such as "A child suddenly jumps out at an intersection," is generated.

[1408] Step 3:

[1409] The terminal sends the generated text prompt to the AI.

[1410] The prompt is sent over the network to the generating AI, which begins analyzing it.

[1411] Step 4:

[1412] The terminal receives action instructions from the AI.

[1413] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[1414] Step 5:

[1415] The vehicle is controlled based on the action instructions received by the terminal.

[1416] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[1417] User Interface Handling

[1418] ---

[1419] Step 1:

[1420] The user launches an application for the mobile terminal.

[1421] Users access the system by launching the app on their smartphone or tablet.

[1422] Step 2:

[1423] The user checks the current state of the system.

[1424] The application displays irregular images generated and the latest AI analysis results.

[1425] Step 3:

[1426] The user enters any prompts.

[1427] For example, enter the prompt "Tell me how to drive on a rainy day."

[1428] Step 4:

[1429] The user checks the prompt analysis results.

[1430] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[1431] Through each of these steps, the server, terminal, and user work together seamlessly to create an autonomous driving system that can quickly respond to irregular situations.

[1432] Example 1

[1433] 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."

[1434] When autonomous vehicles encounter irregular traffic events, it is difficult for them to respond quickly and appropriately. This increases the risk of accidents in unexpected situations, potentially threatening the safety of passengers and those around them. Furthermore, conventional autonomous driving systems lack the ability to analyze data in real time and respond immediately, limiting their ability to generate optimal driving behavior. It is necessary to resolve these issues and improve the reliability and safety of autonomous driving systems.

[1435] 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.

[1436] In this invention, the server includes a means for recording the generated traffic event image, a means for acquiring the driving situation in the form of a text prompt, a generative AI model means for generating appropriate behavioral instructions based on the text prompt, a means for controlling the vehicle based on the behavioral instructions, a means for loading the generative AI model by the server, a means for inputting the traffic situation prompt and generating an image by the server, and a means for saving the generated image in a dataset, thereby enabling analysis of traffic events and generation of appropriate driving behavior in real time.

[1437] A "generative AI model" is an artificial intelligence model used to generate images or text based on specific input prompts.

[1438] A "prompt" is a text input that prompts a generative AI model to generate a specific output.

[1439] A "server" is a computer system that provides central management and processing, such as loading generative AI models, entering prompts, and performing analysis.

[1440] "Traffic incident images" are image data that visually represent traffic conditions and incidents.

[1441] A "dataset" is a collection containing a set of data used for training or analysis.

[1442] "Sensor means" means a device (e.g., camera, LiDAR, radar) installed on an autonomous vehicle for collecting data on the surrounding environment.

[1443] A "text prompt" is a prompt that expresses a traffic situation or event in text form.

[1444] A "vehicle control system" is a system for controlling the operation of a vehicle in real time.

[1445] A "user interface" is an application or device that provides a means for a user to interact with a system and input or obtain information.

[1446] "Action instructions" are operational instructions for an autonomous vehicle provided by the generative AI model based on the analysis results.

[1447] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events. The system is composed of several main components, including a server, a terminal, and a user.

[1448] Server Side

[1449] The server is responsible for first loading the generative AI model. The server loads a pre-trained generative AI model (e.g., the GPT-4 model) into memory and generates images by inputting traffic situation prompts. Specifically, the server inputs the prompt "irregular traffic situation" into the generative AI model, and generates corresponding images as a result. These generated images are saved as a dataset to be used later for model training. The saving destination is an image database (e.g., Amazon S3 or Google Cloud Storage).

[1450] A concrete example is a process in which a server generates multiple images based on the prompt "sharp obstacle at an intersection" and stores the generated images in a dataset.

[1451] Terminal side

[1452] The terminal provides a means for autonomous vehicles to collect and analyze data in real time. The terminal includes various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation based on the data collected from these sensors. The terminal analyzes the collected data and converts it into a specific text prompt. For example, the terminal may generate a text such as "A child suddenly jumps out at an intersection" and send the prompt to the generation AI for analysis.

[1453] The behavioral instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted from the terminal to the vehicle control system and immediately executed, allowing the self-driving car to respond quickly to sudden events.

[1454] A specific example is when a self-driving car is driving in an urban area and its sensors detect an obstacle that suddenly appears on the road, the device describes the situation in text format as "An obstacle has suddenly appeared on the road" and sends a prompt to the generating AI, which then generates an action instruction such as "apply the brakes suddenly and move to the right to avoid the obstacle," which the vehicle then carries out.

[1455] User Interface

[1456] Users are provided with a mobile application that allows them to check the system status, generated images, and real-time response status. Through the application, users can view the current status of the system and input any prompts to request analysis from the generating AI.

[1457] A specific example is when a user uses an application to check the system status and enters a prompt for "advanced driving techniques in bad weather," which results in specific measures being displayed, such as "slow down and increase your distance from the vehicle ahead."

[1458] By these means, the automated driving system of the present invention can improve recognition accuracy in irregular situations and achieve safe and reliable automated driving.

[1459] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1460] Server Side

[1461] Step 1: Loading the Generative AI Model

[1462] The server loads the generative AI model (e.g., the GPT-4 model) from disk into memory.

[1463] Input: Disk file of a trained generative AI model

[1464] Output: Generative AI model loaded in memory

[1465] Specific operation: Loads the model file from disk and expands the model parameters into memory.

[1466] Step 2: Entering prompts and generating images

[1467] The server inputs the prompt "irregular traffic conditions" into the generative AI model and generates an image based on the model.

[1468] Input: Text prompt "Irregular traffic conditions"

[1469] Output: Generated traffic incident images

[1470] What it does: Input a prompt into a generative AI model and get the generated image as the model's output.

[1471] Step 3: Save the generated image

[1472] The server stores the generated images in a dataset.

[1473] Input: Generated traffic incident images

[1474] Output: Images stored in a database

[1475] Specific operation: Saves image data to a specified database (e.g., Amazon S3 or Google Cloud Storage).

[1476] Terminal side

[1477] Step 1: Collect sensor data

[1478] The device collects data in real time from various sensors (cameras, LiDAR, etc.) installed in the autonomous vehicle.

[1479] Input: Sensor data such as camera and LiDAR

[1480] Output: Collected sensor data

[1481] Specific operation: Data is read from each sensor and stored in memory for real-time processing.

[1482] Step 2: Generate a text prompt

[1483] The device analyzes the sensor data collected, understands the current driving situation, and generates text prompts.

[1484] Input: Collected sensor data

[1485] Output: Text prompt (e.g. "A child suddenly jumped out at the intersection.")

[1486] Specific operation: Sensor data is run through an analysis algorithm to generate a text representation of a specific traffic event.

[1487] Step 3: AI sending and analysis of prompts

[1488] The device sends the generated text prompt to the server and requests the generative AI model to analyze it.

[1489] Input: Text prompt

[1490] Output: Analysis results (action instructions)

[1491] Specific operation: Sends a text prompt to the server and receives the analysis results from the generative AI model.

[1492] Step 4: Implementing the Action Instructions

[1493] The terminal receives action instructions from the server and transmits them to the vehicle control system.

[1494] Input: Action instructions (e.g., brake suddenly and swerve left)

[1495] Output: Vehicle control actions

[1496] Specific actions: Action instructions are input into the vehicle control system and executed in real time.

[1497] User Interface

[1498] Step 1: Check the status

[1499] The user launches the application and sees the current state of the system, the generated images, and real-time responses.

[1500] Input: Launch application

[1501] Output: System status display

[1502] Specific behavior: Retrieves system information through the application and displays it in the user interface.

[1503] Step 2: Enter the prompts

[1504] The user inputs a prompt to make an additional analysis request to the system.

[1505] Input: The prompt entered by the user (e.g., "Advanced driving tips for bad weather")

[1506] Output: Prompt sent to generative AI model

[1507] What it does: Receives user input and sends it to the server.

[1508] Step 3: Check the analysis results

[1509] The generative AI model analyzes prompts entered by the user and displays the results in real time.

[1510] Input: Analysis results of the generative AI model

[1511] Output: Display of analysis results (e.g. "Slow down and increase distance from the vehicle ahead")

[1512] Specific actions: Display the analysis results in the user interface and provide specific countermeasures.

[1513] These steps allow servers, devices, and users to work together to improve the reliability and safety of automated driving systems.

[1514] (Application example 1)

[1515] 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."

[1516] In logistics centers, automated guided vehicles and drones must deal with many abnormal events and sudden obstacles. However, current systems have difficulty responding quickly and appropriately to these irregular situations. This leads to problems such as reduced work efficiency and safety risks. In addition, there are limited means for users to monitor these abnormal events in real time and provide appropriate response methods. A comprehensive system to solve these problems is needed.

[1517] 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.

[1518] In this invention, the server includes a means for recording the generated traffic incident images, a means for acquiring driving situations in the form of text prompts, and a generating artificial intelligence means for generating appropriate action instructions based on the text prompts. This enables automated guided vehicles and drones in a logistics center to respond quickly and appropriately to irregular situations. The server also includes a means for displaying action instructions through a user interface, a means for inputting arbitrary prompts using the user interface, and a means for providing analysis results generated based on the input prompts, allowing users to monitor abnormal events in real time and instruct appropriate responses.

[1519] A "generated traffic event image" is an image captured in real time by an automated driving system that reflects a specific traffic event.

[1520] A "text prompt format" is a way of expressing a particular situation or event in a text-based input format.

[1521] "Generative AI" refers to AI that receives text prompts and generates appropriate action instructions or analytical results based on them.

[1522] "Means for controlling a vehicle" refers to equipment or systems that adjust the operation of an automated driving vehicle or automated transport vehicle in accordance with the generated behavioral instructions.

[1523] A "logistics center" is a facility where goods, materials, etc. are stored, managed, and shipped.

[1524] An "automated guided vehicle (AGV)" is an unmanned vehicle that moves automatically along a pre-set route and transports cargo.

[1525] "Drone" means an unmanned aerial vehicle that is remotely controlled or autonomously capable of aerial movement and operations.

[1526] A "user interface" is the means or method by which a system and a user interact with each other.

[1527] "Sensor means" refers to a device or equipment that senses changes in the physical environment and acquires them as data.

[1528] "Real-time" means responding or processing immediately at the moment an event occurs.

[1529] The system for implementing the present invention comprises server, terminal, and user components.

[1530] Server Side

[1531] The server first loads the generative AI model and generates images that reflect specific traffic events or abnormal events within the logistics center. For example, it generates images based on a prompt such as "there is an obstacle in the aisle." The generated images are then saved in a dataset and used for model training, improving the recognition accuracy of the AI ​​model. The main software used by the server is TensorFlow and Python.

[1532] Terminal side

[1533] The terminal provides a means for collecting and analyzing data in real time from drones and automated guided vehicles (AGVs) within the logistics center. The terminal uses data obtained from cameras and sensors equipped on the drones and AGVs to analyze the current situation and describe it in the form of a text prompt. For example, the situation, "Part of a shelf has fallen," is converted into text. The prompt is then sent to a generation AI for analysis. The AI ​​model generates action instructions (for example, "Stop and notify a manager") as a result of the analysis, and the terminal transmits them to the drones and AGVs for immediate execution.

[1534] User Interface

[1535] Users are provided with a smartphone application that allows them to check the system status, generated images, and the real-time response status of drones and AGVs within the logistics center. Through the application, users can monitor the current status of the system and input arbitrary prompts to request analysis from the generating AI. For example, if a user requests additional instructions on "how to respond when passing other AGVs in narrow aisles" within the logistics center, they can input the prompt into the application and receive analysis results from the AI ​​in real time.

[1536] Specific examples

[1537] Server-side example:

[1538] The server loads the generative AI model and generates multiple images based on the prompt, "There is an obstacle in the path." The generated images are saved in a dataset and used for future model training.

[1539] Example of terminal operation:

[1540] The device receives real-time images from drones and automated guided vehicles (AGVs), describes the situation in text format, such as "part of a shelf has fallen," and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "stop and notify the administrator," and the device acts in accordance with that instruction.

[1541] An example of how the user interface works:

[1542] The user checks the system status using a smartphone application and inputs a prompt for "What to do if you are passing another AGV in a narrow passage." As a result, the system displays a specific countermeasure: "Stop and wait for the other AGV to pass, then proceed."

[1543] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1544] Step 1:

[1545] The server loads the generative AI model. Specifically, it loads the generative AI model file that was previously saved using TensorFlow into the system. The input is the file path of the model. The output is the model loaded into memory.

[1546] Step 2:

[1547] The server generates images that reflect specific traffic events or abnormal events within the logistics center. Specifically, a prompt statement (e.g., "There is an obstacle in the aisle") is input into the generative AI model, and an image is generated based on that. The input is the prompt statement, and the output is the generated image.

[1548] Step 3:

[1549] The server saves the generated images as a dataset and later uses them for model training. Specifically, it saves the images in a specified directory. The generated images are used as input. The output is saved in the dataset.

[1550] Step 4:

[1551] The terminal receives image data acquired in real time from drones and automated guided vehicles (AGVs). Specifically, the terminal acquires images in real time from cameras and sensors attached to the terminal. The input is real-time data from the cameras and sensors. The output is the acquired image data.

[1552] Step 5:

[1553] The terminal analyzes the acquired image data and describes it in the form of a text prompt. Specifically, it uses an image processing algorithm to detect abnormal situations and converts them into text format. The input is the acquired image data. The output is a text prompt (e.g., "Part of the shelf has fallen").

[1554] Step 6:

[1555] The terminal sends the generated text prompt to the server. Specifically, the prompt is transferred to the server using network communication. The input is the text prompt. The output is the received data on the server side.

[1556] Step 7:

[1557] The server uses a generative AI model to generate appropriate instructions based on the received text prompt. Specifically, the server inputs the prompt into the model and generates instructions for that prompt. The input is the text prompt. The output is the instruction for that prompt (e.g., "Stop and notify an administrator").

[1558] Step 8:

[1559] The server transmits the generated action instructions to the terminal. Specifically, it transfers the action instructions to the terminal via the network. The input is the action instructions. The output is the received data on the terminal side.

[1560] Step 9:

[1561] The terminal controls drones and automated guided vehicles (AGVs) based on the received action instructions. Specifically, it analyzes the action instructions and sends corresponding control signals to the device. The action instructions are the input, and the device's control operation is the output.

[1562] Step 10:

[1563] Users use a smartphone application to check the system status and response status. Specifically, real-time data and analysis results are displayed on the app. The input is real-time data from the system. The output is user confirmation and input of additional instructions.

[1564] Step 11:

[1565] The user inputs additional prompts into the application and requests the generation AI to analyze them. Specifically, the user enters a prompt for a specific scenario into the application's input field and presses the submit button. The input is the user's prompt (e.g., "What to do if you pass another AGV in a narrow passageway"), and the output is the analysis result by the generation AI (e.g., "Stop and wait for the other AGV to pass").

[1566] The above are the specific processing steps of the system that realizes the application example.

[1567] 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.

[1568] ---

[1569] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[1570] Server Side

[1571] ---

[1572] The server first uses the generative AI model to generate images of various traffic events. This process includes the following steps: The server loads the generative AI model and generates images by entering the prompt "irregular traffic situation." The generated images are then saved in a dataset for model training. This process allows the recognition accuracy of the AI ​​model to be improved.

[1573] Terminal side

[1574] ---

[1575] The device collects data obtained from various sensors (cameras, LiDAR, etc.) equipped on the vehicle in real time, analyzes the current driving situation, and describes it in the form of a text prompt. For example, it generates a specific situation such as "A child suddenly jumps out into the street at an intersection" as text. The prompt is sent to the generation AI for analysis. The action instructions obtained as a result of the analysis (for example, apply sudden brakes and swerve to the left) are transmitted to the vehicle control system and immediately executed. This real-time data analysis and response allows autonomous vehicles to respond quickly to unexpected events.

[1576] Emotion Engine

[1577] ---

[1578] The emotion engine is responsible for analyzing the user's emotions in real time and providing that data to the generative artificial intelligence means. The emotion engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. Based on this emotional data, the system can provide optimal driving control according to the user's state, such as whether they are stressed or relaxed.

[1579] User Interface

[1580] ---

[1581] Users are provided with a mobile application that allows them to view the system's status, generated images, and real-time response status. Through the application, users can view the current system status and input arbitrary prompts to request analysis from the generating AI. For example, if a user requires additional instructions for a specific complex scenario, they can input the prompt in the application and receive analysis results from the AI ​​in real time.

[1582] Specific examples

[1583] ---

[1584] A specific example of system implementation is shown below.

[1585] Server-side example:

[1586] The server loads the generative AI model and generates multiple images based on the prompt "a sharp obstacle at an intersection." The generated images are saved in a dataset and used for future model training.

[1587] Example of terminal operation:

[1588] While an autonomous vehicle is driving in an urban area, its sensors detect an obstacle that suddenly appears on the road. The device describes the situation in text format as "An obstacle suddenly appeared on the road" and sends a prompt to the AI ​​generator. The AI ​​generates an action instruction to "apply sudden brakes and move to the right to avoid the obstacle," and the vehicle executes the instruction accordingly.

[1589] Example of the Emotion Engine in action:

[1590] The emotion engine detects when the user's heart rate spikes and they are feeling stressed. Based on this information, the system switches the driving mode to a safety-first mode and adjusts the music and lighting in the system to ease the user's tension.

[1591] An example of how the user interface works:

[1592] The user uses the application to check the system status and input advice for specific traffic scenarios, such as the prompt "Tell me some advanced driving tips for bad weather," which results in specific measures such as "slow down and increase your distance from the vehicle ahead."

[1593] By using these means, the automated driving system of the present invention can improve the recognition accuracy in irregular situations and realize safe and reliable automated driving. Furthermore, by combining it with an emotion engine, it becomes possible to provide an optimal driving environment according to the user's condition.

[1594] The processing flow will be explained below.

[1595] Server-side processing

[1596] ---

[1597] Step 1:

[1598] The server loads the generative AI model.

[1599] The server loads the generative AI model into memory and secures the computing resources necessary for generation.

[1600] Step 2:

[1601] The server inputs the prompt "irregular traffic conditions" into the generative AI model.

[1602] Based on this prompt, the generative AI model generates images of relevant irregular traffic events.

[1603] Step 3:

[1604] The server stores the generated images in a dataset.

[1605] The generated images are saved in a format that can be used as training data and are used for subsequent model training.

[1606] Terminal-side processing

[1607] ---

[1608] Step 1:

[1609] The device acquires the current driving situation from sensors installed in the vehicle.

[1610] The device collects real-time data from cameras, LiDAR sensors, GPS, and more.

[1611] Step 2:

[1612] The device analyzes the collected data and generates a text prompt.

[1613] It generates a specific textual description of the situation, such as "A child suddenly jumped out at the intersection."

[1614] Step 3:

[1615] The terminal sends the generated text prompt to the AI.

[1616] The prompt is sent over the network to the generating AI, which begins analyzing it.

[1617] Step 4:

[1618] The terminal receives action instructions from the AI.

[1619] The generation AI analyzes the prompts and generates specific action instructions such as "apply the brakes suddenly and move left to avoid the collision."

[1620] Step 5:

[1621] The vehicle is controlled based on the action instructions received by the terminal.

[1622] The control system of the autonomous vehicle performs the appropriate braking and steering operations based on the analysis results.

[1623] Emotion engine processing

[1624] ---

[1625] Step 1:

[1626] The emotion engine captures the user's emotions in real time.

[1627] The emotion engine analyzes the user's facial expressions, voice, and heart rate.

[1628] Step 2:

[1629] The emotion engine analyzes the acquired emotion data and determines the user's emotional state.

[1630] Determine whether the user is stressed or relaxed.

[1631] Step 3:

[1632] The emotion engine generates response instructions according to the emotional state.

[1633] For example, if the user is feeling stressed, an instruction to switch to a safe driving mode is generated.

[1634] Step 4:

[1635] The terminal receives response instructions from the emotion engine and transmits them to the vehicle control system.

[1636] The vehicle control system adjusts the driving mode based on the corresponding instruction.

[1637] User Interface Handling

[1638] ---

[1639] Step 1:

[1640] The user launches an application for the mobile terminal.

[1641] Users access the system by launching the app on their smartphone or tablet.

[1642] Step 2:

[1643] The user checks the current state of the system.

[1644] The application displays irregular images generated and the latest AI analysis results.

[1645] Step 3:

[1646] The user enters any prompts.

[1647] For example, enter the prompt "Tell me how to drive on a rainy day."

[1648] Step 4:

[1649] The user checks the prompt analysis results.

[1650] The AI ​​analyzes the input prompts and suggests specific measures, such as "slow down and increase the distance from the vehicle in front."

[1651] Example 2

[1652] 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."

[1653] Current automated driving systems have issues with responding to irregular traffic events and performing appropriate driving control based on the user's emotional state. In particular, rapid and accurate responses to unforeseen situations are required, but current systems are insufficient in this regard. In addition, the lack of a mechanism to recognize the user's emotions and adjust driving control based on them makes it difficult to achieve both safety and comfort.

[1654] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, an artificial intelligence generation means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, an emotion analysis means for analyzing the user's emotions in real time, and a means for adjusting driving control based on the emotion analysis result. This makes it possible to quickly respond to irregular traffic incidents and provide optimal driving control based on the user's emotional state.

[1655] "Means for recording generated traffic event images" refers to a device or software that has the function of saving images of traffic events generated by a generative AI model on a recording medium.

[1656] "Means for obtaining driving status in text prompt format" refers to a device or software that has the function of analyzing information obtained from sensors installed in the vehicle and generating the current driving status in text format.

[1657] "Generative artificial intelligence means" refers to a system or program that implements an artificial intelligence algorithm that generates optimal action instructions based on text prompts.

[1658] "Vehicle control means" refers to a device or software that has the function of performing actions such as accelerating, decelerating, avoiding, and stopping the autonomous vehicle based on behavioral instructions obtained from the generating artificial intelligence means.

[1659] "Emotion analysis means for analyzing a user's emotions in real time" refers to a device or software that has the function of collecting and analyzing data such as a user's facial expressions, voice, and heart rate in real time and determining the user's emotional state.

[1660] "Means for adjusting driving control" refers to a device or software that has the function of adjusting the driving mode and various comfort functions of an autonomous vehicle based on emotional state data obtained from the emotion analysis means.

[1661] "Sensor means" refers to various sensors (camera, LiDAR, microphone, pulse sensor, etc.) installed in a vehicle to detect the surrounding conditions and the internal state of the vehicle.

[1662] "User interface" refers to a device or software that provides an interface for a user to check the system status and input any prompts.

[1663] The automated driving system of the present invention uses a generative AI model to respond quickly and appropriately to irregular traffic events, and by combining it with an emotion engine that recognizes the user's emotions, it achieves optimal driving control based on the user's state. This system is composed of several main components, including a server, a terminal, a user, and the emotion engine.

[1664] Server Side

[1665] The server first loads a generative AI model. For example, OpenAI's DALL-E and Stable Diffusion are commonly used AI models for image generation. The server uses these models to generate images of various traffic events. The prompt uses short text, such as "A sudden obstacle at an intersection" or "A sudden obstacle appeared on the road." The images generated in this way are saved in a dataset and used for subsequent model training. This process improves the recognition accuracy of the AI ​​model, enabling more accurate analysis.

[1666] Terminal side

[1667] The device collects data in real time from various sensors (cameras, LiDAR, etc.) equipped on the vehicle and analyzes the current driving situation. For example, it combines images captured by the camera and distance data obtained from LiDAR to generate a text prompt describing a specific situation, such as "A child suddenly jumps out at an intersection." The generated prompt is sent to a generative AI model, which analyzes it and generates optimal instructions for action. In this case, the generative AI issues an instruction such as "apply sudden brakes and swerve left," and the device transmits the instruction to the vehicle control system, which immediately executes it.

[1668] Emotion Engine

[1669] The emotion engine analyzes the user's emotions in real time. This engine monitors the user's facial expressions, voice, heart rate, etc. to determine their emotional state. For example, facial expression analysis technology is used to analyze the user's facial expressions captured by a camera, and tone of voice analysis technology is used to analyze their voice. Heart rate is also monitored using a biosensor. Based on the emotional data obtained in this way, the system adjusts driving control. Specifically, if the user is feeling stressed, the system will switch to a safety-first setting and change the in-car music and lighting to relaxing ones.

[1670] User Interface

[1671] Users are provided with a mobile application. Using this application, they can check the system status, generated images, and real-time response status. Users can also input any prompt through the application and ask the AI ​​to analyze it. For example, by inputting a prompt such as "Tell me how to drive in bad weather," they can immediately receive specific advice such as "Slow down and increase the distance from the vehicle ahead." In this way, users can always take optimal driving actions based on the latest information.

[1672] By combining these elements, the automated driving system according to the present invention can quickly respond to irregular traffic events and provide optimal driving control based on the user's emotional state.

[1673] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1674] Step 1:

[1675] The server loads the generative AI model. Specifically, it selects a generative AI model such as DALL-E or Stable Diffusion and loads the necessary libraries and model files into memory. The input for this step is the file path and configuration information of the generative AI model, and the output is an instance of the loaded generative AI model.

[1676] Step 2:

[1677] The server generates a prompt for the irregular traffic event and inputs it to the generative AI model. For example, it generates a prompt saying "a sharp obstacle at an intersection." The input of this step is the configuration information about the irregular traffic event, and the output is the generated prompt sentence.

[1678] Step 3:

[1679] The server inputs a prompt sentence into the generative AI model and generates an image. The generated image is saved in a dataset. The input of this step is the prompt sentence, and the output is a file of the generated image. For example, an image of a vehicle or pedestrian suddenly appearing at an intersection is generated.

[1680] Step 4:

[1681] The terminal collects data in real time from sensors (cameras, LiDAR, etc.) installed on the vehicle. For example, the forward-facing camera detects obstacles on the road. The input of this step is the sensor data, and the output is a set of collected raw data.

[1682] Step 5:

[1683] The device analyzes the collected sensor data and generates a textual description of the driving situation. Specifically, it uses an image analysis algorithm to analyze the sensor data and generate a specific text prompt, such as "A child suddenly jumped out at the intersection." The input of this step is the sensor data, and the output is the generated text prompt.

[1684] Step 6:

[1685] The device sends the generated text prompt to the generative AI model for analysis. The generative AI model generates optimal action instructions based on the prompt. For example, an instruction such as "apply sudden brakes and swerve left" may be generated. The input of this step is the text prompt, and the output is an action instruction.

[1686] Step 7:

[1687] The terminal transmits the action instructions obtained from the generative AI model to the vehicle control system and executes them immediately. Specifically, it transmits instructions to the vehicle's control unit, for example, to apply sudden brakes and swerve to the left. The input of this step is the action instructions, and the output is the executed vehicle behavior.

[1688] Step 8:

[1689] The emotion engine collects and analyzes the user's facial expressions, voice, heart rate, etc. in real time. For example, it uses the user's facial expressions captured by a camera and heart rate data obtained from a biosensor. The input of this step is the user's biometric data, and the output is the analyzed emotional state.

[1690] Step 9:

[1691] The emotion engine adjusts driving control based on the emotion analysis results. Specifically, if the user is feeling stressed, it takes measures such as switching to a safety-first driving mode. The input of this step is the emotion analysis results, and the output is the adjusted driving control settings.

[1692] Step 10:

[1693] The user uses a mobile application to check the system status and input a prompt, such as "Tell me how to drive in bad weather." The input for this step is the user's input prompt, and the output is the analysis result from the generative AI model.

[1694] Step 11:

[1695] The user checks the analysis results obtained through the application and puts them into action. For example, they receive advice such as "slow down and increase the distance from the vehicle ahead" and follow it. The input of this step is the analysis results, and the output is the driving behavior that the user follows.

[1696] Through these processing steps, the system can respond quickly and appropriately to irregular traffic events and provide optimal driving control based on the user's emotional state.

[1697] (Application example 2)

[1698] 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."

[1699] Conventional technologies have had difficulty in controlling autonomous vehicles to appropriately reflect irregular traffic events and the emotional state of the user. In particular, existing technologies have been inadequate in situations where driving operations are required to take into account risk avoidance due to sudden traffic events and the stress and anxiety felt by the user. For this reason, improving the safety of autonomous vehicles and the sense of security of users is an important issue.

[1700] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1701] In this invention, the server includes a means for recording the generated traffic incident image, a means for acquiring the driving situation in the form of a text prompt, a generating artificial intelligence means for generating appropriate action instructions based on the text prompt, a means for controlling the vehicle based on the action instructions, a means for analyzing the emotional state, and a means for adjusting driving control based on the emotional state, thereby enabling a quick and appropriate response to irregular traffic incidents and enabling safe driving control that takes the user's emotional state into consideration.

[1702] The "means for recording generated traffic event images" refers to a device or mechanism for storing image data relating to traffic events generated by the server.

[1703] The "means for acquiring driving status in text prompt format" refers to a device or mechanism for converting information obtained from various sensors and cameras into text format and acquiring the information while the vehicle is in operation.

[1704] A "generative artificial intelligence means" is a device or system that utilizes artificial intelligence techniques used to generate appropriate action instructions based on text prompts.

[1705] The "means for controlling the vehicle" refers to a device or mechanism for executing driving operations of the vehicle based on the generated behavioral instructions.

[1706] The "means for analyzing emotional state" is a device or system for analyzing data such as the user's facial expression, voice, and heart rate to determine the user's emotional state.

[1707] A "means for adjusting driving controls" is a device or mechanism for adjusting the vehicle's driving operations or environmental settings based on the analyzed emotional state.

[1708] "Sensor means" refers to sensor devices such as cameras, LiDAR, and heart rate sensors for monitoring driving conditions in real time.

[1709] The "means for describing in text form" is a device or mechanism for converting data obtained from the sensor means into a text prompt format and describing it.

[1710] A "user interface" is an interface device or software that allows a user to check the system status and input any prompts.

[1711] A "prompt input means" is a mechanism or device that allows a user to input instructions in the form of text prompts relating to a particular scenario or situation.

[1712] The "means for providing analysis results" is a device or mechanism for presenting to the user action instructions and analysis results generated based on the text prompts.

[1713] This invention is an automated driving system that responds quickly and appropriately to irregular traffic events and realizes optimal driving control based on the user's emotional state. This system is composed of several main components, including a server, a terminal, a user, and an emotion engine.

[1714] Server-side configuration and operation

[1715] The server is responsible for loading the generative AI model and generating images of traffic events, following these steps:

[1716] 1. Loading a generative AI model: For example, using a deep learning library (e.g., TensorFlow), load a generative AI model that simulates traffic events.

[1717] 2. Entering a prompt: Enter the prompt "irregular traffic conditions" into the model.

[1718] 3. Image generation: The generative AI model generates an image of the traffic incident based on the prompt.

[1719] 4. Saving the dataset: The generated images are saved as a dataset on the server and used for model training.

[1720] Terminal configuration and operation

[1721] The device collects and analyzes data in real time from sensors (e.g., cameras, LiDAR) installed in the vehicle. This includes the following steps:

[1722] 1. Description of the situation in the form of a text prompt: For example, describe a specific situation in the form of a text prompt, such as, "A child suddenly jumps out at an intersection."

[1723] 2. Analysis and generation of action instructions: The prompt sentence is sent to the generative AI model, which analyzes it. As a result of the analysis, an action instruction such as "apply sudden brakes and move left to avoid the collision" is generated.

[1724] 3. Transmission to vehicle control: Control the vehicle based on the action instructions.

[1725] Emotion engine configuration and operation

[1726] The emotion engine is responsible for analyzing the user's emotional state in real time, specifically including the following steps:

[1727] 1. Emotion data collection: Data such as the user's facial expressions, voice, and heart rate are collected using sensors (e.g., smartphone built-in camera, heart rate sensor).

[1728] 2. Emotional state determination: Analyze the collected data to determine the user's emotional state (e.g., stressed or relaxed).

[1729] 3. Driving control adjustment: If the user is feeling stressed, the system will switch the driving mode to safety priority and even adjust the music and lighting within the system.

[1730] User interface configuration and operation

[1731] Through a mobile terminal application, users can check the system status, generated images, real-time response status, and input prompt statements.

[1732] Specific examples:

[1733] Situation: User is in the driver's seat, vehicle is traveling through a congested city.

[1734] Example prompt: "A sharp obstacle at the intersection."

[1735] Generated action instructions: "Suddenly brake and move to the right."

[1736] In this way, safe and optimal driving control based on the user's emotional state is possible.

[1737] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1738] Server-side processing steps

[1739] Step 1:

[1740] The server loads the generative AI model, which involves initializing it based on the prompt sentence "irregular traffic conditions," for example, using the TensorFlow library.

[1741] Input: Generative AI model, prompt "Irregular traffic conditions"

[1742] Output: Initialized model

[1743] Step 2:

[1744] The server inputs the prompt sentence into the generative AI model and generates an image of the traffic event. The generated image is a simulated traffic event.

[1745] Input: prompt statement, initialized model

[1746] Output: Generated traffic incident images

[1747] Step 3:

[1748] The server saves the generated images in a dataset for future model training, which improves the model's recognition accuracy.

[1749] Input: Generated image

[1750] Output: A saved image dataset

[1751] Terminal processing steps

[1752] Step 1:

[1753] The terminal collects data in real time from various sensors (e.g., cameras, LiDAR) installed in the vehicle.

[1754] Input: Sensor output data

[1755] Output: Real-time data collected

[1756] Step 2:

[1757] The device analyzes the collected data and describes the current driving situation in the form of a text prompt, such as "A child suddenly jumped out at the intersection."

[1758] Input: Real-time data collected

[1759] Output: Driving status in text prompt format

[1760] Step 3:

[1761] The device sends the prompt text to the generative AI model for analysis, which then generates appropriate action instructions.

[1762] Input: Driving situation in text prompt format, Generative AI model

[1763] Output: Action instructions

[1764] Step 4:

[1765] The terminal transmits the generated action instruction to the vehicle control system and executes it immediately.

[1766] Input: Action instructions

[1767] Output: Vehicle control executed

[1768] Emotion Engine Processing Steps

[1769] Step 1:

[1770] The emotion engine collects the user's facial expressions, voice, heart rate, and other data in real time.

[1771] Input: User's facial expression, voice, and heart rate data

[1772] Output: Collected emotion data

[1773] Step 2:

[1774] The emotion engine analyzes the collected data to determine the user's emotional state, for example, whether the user is stressed or relaxed.

[1775] Input: Collected emotion data

[1776] Output: User's emotional state

[1777] Step 3:

[1778] The emotion engine adjusts driving controls based on the user's emotional state. For example, if the user is stressed, it switches the driving mode to safety-first and adjusts the music and lighting in the system.

[1779] Input: User's emotional state

[1780] Output: Coordinated driving control

[1781] User Interface Processing Steps

[1782] Step 1:

[1783] Users can check the system status, generated images, and real-time response status through a mobile device application.

[1784] Input: System status, generated images, response status

[1785] Output: Visual information presented to the user

[1786] Step 2:

[1787] The user uses the application to input a prompt, for example, "I would like to know the best way to drive at a busy intersection."

[1788] Input: The prompt text entered by the user

[1789] Output: The prompt sent to the system

[1790] Step 3:

[1791] The device uses generative artificial intelligence to provide analysis results based on the input prompt. For example, in response to the prompt "Sudden obstacle at intersection," the device generates and presents the action instruction "Apply sudden braking and move right to avoid the obstacle" to the user.

[1792] Input: A prompt entered by the user, a generative AI model

[1793] Output: Analysis results and action instructions

[1794] 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.

[1795] 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.

[1796] 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.

[1797] 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.

[1798] 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.

[1799] 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.

[1800] 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).

[1801] 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.

[1802] 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."

[1803] 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.

[1804] 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).

[1805] 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.

[1806] 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.

[1807] 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.

[1808] 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.

[1809] 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.

[1810] 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.

[1811] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1812] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1813] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1814] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1815] The following is further disclosed regarding the above embodiment.

[1816] (Claim 1)

[1817] means for recording the generated traffic incident images;

[1818] a means for obtaining driving status in the form of text prompts;

[1819] artificial intelligence generating means for generating appropriate action instructions based on said text prompts;

[1820] means for controlling the vehicle based on the action instructions;

[1821] Autonomous driving systems, including

[1822] (Claim 2)

[1823] a sensor means for monitoring conditions during driving;

[1824] means for analyzing data obtained from the sensor means and describing the current driving situation in text form;

[1825] The automated driving system of claim 1, further comprising:

[1826] (Claim 3)

[1827] means for displaying the generated action instructions through a user interface;

[1828] a means for inputting any prompts using a user interface;

[1829] means for providing an analysis result using artificial intelligence based on the input prompt;

[1830] The automated driving system of claim 1, further comprising:

[1831] "Example 1"

[1832] (Claim 1)

[1833] means for recording the generated traffic incident images;

[1834] a means for obtaining driving status in the form of text prompts;

[1835] a generative AI model means for generating appropriate action instructions based on the text prompt;

[1836] means for controlling the vehicle based on the action instructions;

[1837] A means for the server to load the generative AI model;

[1838] means for the server to input traffic condition prompts and generate images;

[1839] means for storing the generated images in a dataset;

[1840] A system including:

[1841] (Claim 2)

[1842] a sensor means for monitoring conditions during driving;

[1843] means for analyzing data obtained from the sensor means and describing the current driving situation in text form;

[1844] means for sending prompts to the generative AI model based on the sensor data;

[1845] a means for transmitting the behavioral instructions obtained from the generative AI model to a vehicle control system;

[1846] The system of claim 1 further comprising:

[1847] (Claim 3)

[1848] means for displaying the generated action instructions through a user interface;

[1849] a means for inputting any prompts using a user interface;

[1850] means for providing an analysis result using a generative AI model based on the input prompt;

[1851] The system of claim 1 further comprising:

[1852] "Application Example 1"

[1853] (Claim 1)

[1854] means for recording the generated traffic incident images;

[1855] a means for obtaining driving status in the form of text prompts;

[1856] artificial intelligence generating means for generating appropriate action instructions based on said text prompts;

[1857] means for controlling the vehicle based on the action instructions;

[1858] a means for applying the generated prompts and instructions to automated vehicles and drones in a logistics center;

[1859] A system including:

[1860] (Claim 2)

[1861] a sensor means for monitoring conditions during driving;

[1862] means for analyzing data obtained from the sensor means and describing the current driving situation in text form;

[1863] means for adapting said data to hardware used within the logistics center;

[1864] The system of claim 1 further comprising:

[1865] (Claim 3)

[1866] means for displaying the generated action instructions through a user interface;

[1867] a means for inputting any prompts using a user interface;

[1868] means for providing an analysis result using artificial intelligence based on the input prompt;

[1869] a means for monitoring abnormal events in the logistics center in real time via a user interface and generating additional instructions;

[1870] The system of claim 1 further comprising:

[1871] "Example 2: Combining Emotion Engines"

[1872] (Claim 1)

[1873] means for recording the generated traffic incident images;

[1874] a means for obtaining driving status in the form of text prompts;

[1875] artificial intelligence generating means for generating appropriate action instructions based on said text prompts;

[1876] means for controlling the vehicle based on the action instructions;

[1877] emotion analysis means for analyzing user emotions in real time;

[1878] a means for adjusting driving control based on the emotion analysis result;

[1879] A system including:

[1880] (Claim 2)

[1881] a sensor means for monitoring conditions during driving;

[1882] means for analyzing data obtained from the sensor means and describing the current driving situation in text form;

[1883] means for transmitting the text prompt to the generating artificial intelligence and receiving action instructions;

[1884] means for controlling the vehicle in real time based on the action instructions;

[1885] The system of claim 1 further comprising:

[1886] (Claim 3)

[1887] means for displaying the generated action instructions through a user interface;

[1888] a means for inputting any prompts using a user interface;

[1889] means for providing an analysis result using artificial intelligence based on the input prompt;

[1890] a means for adjusting operational control based on the analysis results;

[1891] The system of claim 1 further comprising:

[1892] "Application example 2 when combining emotion engines"

[1893] (Claim 1)

[1894] means for recording the generated traffic incident images;

[1895] a means for obtaining driving status in the form of text prompts;

[1896] artificial intelligence generating means for generating appropriate action instructions based on said text prompts;

[1897] means for controlling the vehicle based on the action instructions;

[1898] a means for analyzing an emotional state;

[1899] means for adjusting driving control based on said emotional state;

[1900] A system including:

[1901] (Claim 2)

[1902] a sensor means for monitoring conditions during driving;

[1903] means for analyzing data obtained from the sensor means and describing the current driving situation in text form;

[1904] The system of claim 1 further comprising:

[1905] (Claim 3)

[1906] means for displaying the generated action instructions through a user interface;

[1907] a means for inputting any prompts using a user interface;

[1908] means for providing an analysis result using artificial intelligence based on the input prompt;

[1909] The system of claim 1 further comprising: [Explanation of symbols]

[1910] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for recording the generated traffic incident images; a means for obtaining driving status in the form of text prompts; artificial intelligence generating means for generating appropriate action instructions based on said text prompts; means for controlling the vehicle based on the action instructions; Autonomous driving systems, including

2. a sensor means for monitoring conditions during driving; means for analyzing data obtained from the sensor means and describing the current driving situation in text form; The automated driving system of claim 1 further comprising:

3. means for displaying the generated action instructions through a user interface; a means for inputting any prompts using a user interface; means for providing an analysis result using artificial intelligence based on the input prompt; The automated driving system of claim 1 further comprising:

Images