Method and device for mobile-based AI-assisted vehicle-mounted alarm system

Abstract

This invention provides a method and apparatus for AI-assisted vehicle-mounted alarm systems based on a mobile terminal. The method includes: a mobile terminal receiving vehicle accident alarm signals and accident information data from the vehicle-mounted system; initiating an alarm call and data transmission dual-link with the police receiving personnel, configuring a preset large model, and simultaneously generating a vehicle accident identification code and local location information for transmission to the police receiving personnel; parsing the alarm call audio using the preset large model to obtain semantic and speech analysis results; combining driver and passenger health monitoring data with preset judgment conditions to determine the effective communication status of the driver and passengers; if communication is impossible, the large model takes over the call link; detecting the network quality of the data transmission link, and transmitting accident information data based on the detection results; this invention enables vehicle accident alarms with low intervention, improves the comprehensiveness and accuracy of alarm data, and provides reliable technical support for accident emergency rescue plans.

Description

Technical Field

[0001] This invention belongs to the fields of artificial intelligence and automotive safety technology, and in particular relates to a method and device for providing alarms to vehicle-mounted systems with artificial intelligence assistance based on a mobile terminal. Background Technology

[0002] With the continuous growth of motor vehicle ownership and the increasing complexity of road traffic environment, the efficiency of emergency rescue response to vehicle traffic accidents is directly related to the level of life safety protection for drivers and passengers. Currently, the traditional vehicle accident alarm method mainly relies on drivers and passengers manually triggering alarm terminals and dialing emergency rescue numbers to report accidents, which has many limitations in actual road rescue scenarios.

[0003] On the one hand, when traffic accidents cause drivers and passengers to be unconscious or physically disabled and unable to operate independently, it is difficult to manually call the alarm, which can easily lead to missing the best rescue opportunity. On the other hand, most existing vehicle alarm systems can only trigger simple accident signals, and emergency rescue agencies cannot fully obtain accident-related data when they receive the alarm, resulting in insufficient accuracy of rescue deployment. Summary of the Invention

[0004] This invention provides a method for AI-assisted vehicle-mounted alarms based on a mobile terminal. Through coordinated operation and dual-link transmission between the vehicle-mounted and mobile terminals, combined with network communication quality detection, it achieves vehicle accident alarms with low intervention, improving the comprehensiveness and accuracy of alarm data, enhancing data transmission stability, and providing reliable technical support for emergency rescue solutions. This method is applied to a mobile terminal and includes: The system receives vehicle accident alarm signals and accident information data sent from the vehicle terminal; initiates an alarm call link and data transmission link with the police based on the vehicle accident alarm signal and configures a preset large model; generates a vehicle accident identification code and local location information, and transmits them to the police through the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data. By performing semantic and speech parsing on the audio of the alarm call link using a pre-set large model, the semantic and speech parsing results of the driver and passengers and the police receiving the call are obtained. Based on semantic analysis results, voice analysis results, and health monitoring data of drivers and passengers, the effective communication status of drivers and passengers is determined by combining preset judgment conditions. When it is determined that drivers and passengers cannot communicate effectively, the alarm call link is taken over by a preset large model. The network quality of the data transmission link is checked, and based on the network quality check results, the accident information data is transmitted to the police receiving the case through the data transmission link.

[0005] This invention provides a mobile-based AI-assisted vehicle-mounted alarm device. Through coordinated operation and dual-link transmission between the vehicle-mounted and mobile terminals, combined with network communication quality detection, it achieves vehicle accident alarms with low intervention, improving the comprehensiveness and accuracy of alarm data, enhancing data transmission stability, and providing reliable technical support for emergency rescue solutions. This device is applied to a mobile terminal and includes: The alarm configuration module is used to receive vehicle accident alarm signals and accident information data sent by the vehicle terminal, initiate an alarm call link and data transmission link with the police based on the vehicle accident alarm signal, and configure a preset large model; generate a vehicle accident identification code and local location information, and transmit them to the police through the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data; The audio parsing module is used to perform semantic and speech parsing processing on the audio of the alarm call link through a preset large model, so as to obtain the semantic and speech parsing results of the driver and passengers and the police receiving the call. The status determination module is used to determine the effective communication status of the driver and passengers based on semantic parsing results, voice parsing results, and driver and passenger health monitoring data, combined with preset determination conditions. When it is determined that the driver and passengers cannot communicate effectively, the alarm call link is taken over through a preset large model. The accident information transmission module is used to detect the network quality of the data transmission link and, based on the network quality detection results, transmits the accident information data to the police receiving the case through the data transmission link.

[0006] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for issuing alarms using an AI-assisted vehicle terminal based on a mobile terminal.

[0007] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned method for issuing alarms using an AI-assisted vehicle terminal based on a mobile terminal.

[0008] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the aforementioned method for issuing alarms using an AI-assisted vehicle terminal based on a mobile terminal.

[0009] In this embodiment of the invention, a vehicle accident alarm signal and accident information data sent by the vehicle-mounted terminal are received. Based on the vehicle accident alarm signal, an alarm call link and a data transmission link with the police are initiated, and a preset large model is configured. A vehicle accident identification code and local location information are generated and transmitted to the police via the data transmission link. The accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data. The audio of the alarm call link is semantically and verbally processed using the preset large model to obtain the semantic and verbal analysis results between the driver and passenger and the police. Based on the semantic and verbal analysis results and the driver and passenger health monitoring data, and combined with preset judgment conditions, the effective communication status of the driver and passenger is determined. When it is determined that the driver and passenger cannot communicate effectively, the alarm call link is taken over by the preset large model. The network quality of the data transmission link is detected, and based on the network quality detection results, the accident information data is transmitted to the police via the data transmission link. This embodiment of the invention achieves vehicle accident alarm in a low-intervention state through the coordinated linkage of the vehicle-mounted terminal and the mobile terminal, dual-link transmission, and network communication quality detection. This improves the comprehensiveness and accuracy of alarm data, enhances the stability of data transmission, and provides reliable technical support for accident emergency rescue solutions. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is a flowchart of a method for issuing alarms using AI-assisted vehicle-mounted systems based on mobile devices, as described in an embodiment of the present invention. Figure 2 This is a specific example diagram of semantic parsing processing in an embodiment of the present invention; Figure 3 This is a specific example diagram of speech parsing processing in an embodiment of the present invention; Figure 4 This is a specific example diagram illustrating the determination of effective communication status between drivers and passengers in an embodiment of the present invention; Figure 5 This is a specific example diagram of accident information data transmission in an embodiment of the present invention; Figure 6 This is a structural example diagram of a device for providing alarm functionality to a vehicle-mounted system based on artificial intelligence from a mobile terminal, as described in an embodiment of the present invention. Figure 7 This is a specific example diagram of the structure of the device for issuing alarms based on mobile terminal artificial intelligence-assisted vehicle terminal in an embodiment of the present invention; Figure 8 This is a structural diagram of a computer device in an embodiment of the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0012] As mentioned earlier, in the existing technology, vehicle-mounted alarms need to be transferred through the vehicle manufacturer's call center, which involves multiple communication links, low alarm efficiency, and vehicle accidents can easily cause vehicle-mounted alarms to malfunction and fail. Mobile phone-based vehicle accident alarms are prone to misjudgment, and there is a lack of reliable alarm methods when the driver is unconscious.

[0013] To address this issue, the inventors considered that the multimodal processing capabilities of mobile artificial intelligence (AI) can effectively compensate for the shortcomings of individual alarms and simple joint alarms on the vehicle-mounted and mobile phone terminals. When the driver is unconscious due to a car accident and unable to actively call for help, the alarm on the mobile phone can be triggered through the vehicle-mounted terminal, and the AI ​​on the mobile phone will take over the entire alarm process, while integrating and transmitting data. Therefore, a solution is proposed that utilizes the AI ​​capabilities of mobile devices to assist vehicle-mounted alarms. Through the collaborative cooperation between the mobile phone and the vehicle-mounted terminal, efficient and reliable alarm and rescue guidance can be achieved after an accident.

[0014] Figure 1 This is a flowchart of a method for issuing alarms using AI-assisted vehicle-mounted systems based on mobile devices, as described in this embodiment of the invention. The method is applied to mobile devices, such as... Figure 1 As shown, the method for issuing alarms using mobile-based AI-assisted vehicle-mounted systems includes: Step 101: Receive vehicle accident alarm signals and accident information data sent by the vehicle terminal; initiate an alarm call link and data transmission link with the police based on the vehicle accident alarm signal and configure a preset large model; generate a vehicle accident identification code and local location information, and transmit them to the police through the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data; Step 102: Perform semantic and speech parsing processing on the audio of the alarm call link using a preset large model to obtain the semantic and speech parsing results between the driver / passenger and the police officer receiving the call. Step 103: Based on the semantic parsing results, voice parsing results, and health monitoring data of drivers and passengers, determine the effective communication status of drivers and passengers in combination with preset judgment conditions. When it is determined that drivers and passengers cannot communicate effectively, take over the alarm call link through the preset large model. Step 104: Detect the network quality of the data transmission link. Based on the network quality detection results, transmit the accident information data to the police receiving the case through the data transmission link.

[0015] Depend on Figure 1 As shown in the process, in this embodiment of the invention, the vehicle terminal receives vehicle accident alarm signals and accident information data; based on the vehicle accident alarm signals, an alarm call link and data transmission link are initiated with the police and a preset large model is configured; a vehicle accident identification code and local location information are generated and transmitted to the police via the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data; the preset large model performs semantic and speech parsing processing on the audio of the alarm call link to obtain the semantic and speech parsing results between the driver and passenger and the police; based on the semantic and speech parsing results and the driver and passenger health monitoring data, combined with preset judgment conditions, the effective communication status of the driver and passenger is determined; when it is determined that the driver and passenger cannot communicate effectively, the preset large model takes over the alarm call link; the network quality of the data transmission link is detected, and based on the network quality detection results, the accident information data is transmitted to the police via the data transmission link.

[0016] Compared to existing technologies where vehicle-mounted alarms are inefficient and prone to failure due to multiple steps, and mobile phone-based alarms are prone to misjudgment and lack reliable alarm methods when drivers and passengers are unconscious, this new technology initiates a dual-link alarm and configures a large model after receiving accident alarm signals and data from the vehicle-mounted terminal. It first transmits the vehicle accident identification code and local location information, then uses the large model to analyze the call audio to determine the driver's and passengers' communication status and takes over the call link when effective communication is impossible. Combined with network quality detection results, it transmits accident information data, achieving AI-powered intelligent takeover and intelligent data transmission across the dual links. This enables automatic, efficient, and reliable alarms even when drivers and passengers are unconscious after a vehicle accident, ensuring the timeliness and accuracy of accident alarms. It provides comprehensive and accurate accident information support for police to quickly conduct rescue operations, improving rescue efficiency and success rate.

[0017] In step 101, the vehicle accident alarm signal and accident information data sent by the vehicle terminal are received. Based on the vehicle accident alarm signal, an alarm call link and data transmission link with the police are initiated and a preset large model is configured. The vehicle accident identification code and local positioning information are generated and transmitted to the police through the data transmission link. The accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data.

[0018] In a specific embodiment, receiving relevant signal data from the vehicle-mounted terminal and completing the alarm link establishment and initial information transmission includes: Receiving signals and data from the vehicle's infotainment system: The mobile device receives vehicle accident alarm signals sent by the vehicle's infotainment system after detecting that the vehicle has reached the accident alarm conditions.

[0019] Acquire driver and passenger health monitoring data transmitted by wearable health monitoring devices connected to mobile devices, such as static health records of drivers and passengers including blood type, allergy history, and past medical history, as well as dynamic physiological indicators such as current heart rate, body temperature, and blood oxygen. Simultaneously collect in-vehicle environment data acquired by the mobile device itself, such as panoramic view of the in-vehicle environment or driver's facial image captured by the camera, in-vehicle environment image, and abnormal noises in the in-vehicle environment collected by the microphone; It synchronously receives vehicle accident data transmitted from the vehicle's infotainment system, such as airbag deployment status, seat belt pretension status, vehicle speed at the moment of collision, impact location, and vehicle power source status, and integrates them to form accident information data.

[0020] Initiate dual links and configure large models: Based on the vehicle accident alarm signal, the mobile terminal immediately and automatically starts the emergency call procedure, initiates an alarm call link with the police, opens the mobile terminal data channel, establishes a data transmission link with the police, and activates and configures the preset large model in the background of the mobile terminal to prepare for subsequent audio parsing, call takeover and data processing. Generate and transmit initial core information: The mobile device automatically generates a unique vehicle accident alarm identification code and obtains accurate local location information through its own positioning function. At the same time as the alarm call link is established, the vehicle accident identification code and local location information are transmitted to the police receiving the call via the data transmission link.

[0021] In this embodiment, the preset large model includes a language large model and a speech large model.

[0022] Figure 2 This is a specific example diagram of semantic parsing processing in an embodiment of the present invention, such as... Figure 2 As shown, the preset large model includes a language large model; Semantic parsing of the audio in the alarm call chain using a pre-defined large model can include: Step 201: Obtain the audio stream of the conversation between the driver / passenger and the police officer in the alarm call link; Step 202: Input the call audio stream into the language big data model for semantic recognition and parsing to obtain the dialogue semantic information and interaction logic information in the audio stream, and generate the semantic parsing results of the driver and passengers and the police.

[0023] In a specific embodiment, semantic parsing processing of the audio in the alarm call link is performed using a preset large model, including: Get the call audio stream: The pre-set large model uses two-way audio stream separation technology to synchronously extract the two-way audio stream of the driver and passengers and the police responder from the alarm call link, and completely obtain the audio data of the conversation between the two parties to ensure the continuity and integrity of the audio stream.

[0024] Semantic recognition and parsing results: The extracted two-way call audio stream is input into a preset language model. The model performs speech-to-text and semantic recognition processing on the audio stream, analyzes the dialogue semantic information in the audio stream, and analyzes the question-and-answer interaction logic information between the two parties to determine whether there are logical inconsistencies or missing keywords in the conversation between the driver and passengers. The above-mentioned analysis information is integrated to generate the semantic analysis results between the driver and passengers and the police.

[0025] Figure 3 This is a specific example diagram of speech parsing processing in an embodiment of the present invention, such as... Figure 3 As shown, the preset large model includes a large voice model; The audio of the alarm call link is processed by voice parsing using a pre-set large model, which may include: Step 301: Obtain the audio stream of the conversation between the driver / passenger and the police officer in the alarm call link; Step 302: Input the call audio stream into the speech model for speech feature extraction to obtain the speech response status and speech feature information of the driver and passengers, and generate the speech analysis results of the driver and passengers and the police.

[0026] In a specific embodiment, the audio of the alarm call link is processed by voice parsing using a preset large model, including: Get the call audio stream: The pre-set large model uses two-way audio stream separation technology to synchronously and completely acquire the two-way audio stream of the driver and passengers and the police responder from the alarm call link, retaining the audio data of the entire call and providing a complete data source for voice analysis.

[0027] Speech feature extraction and generation results: The acquired two-way audio stream is input into a pre-set large voice model. The model extracts features from the driver's and passengers' voices, establishes a voice feature benchmark for the driver and passengers using voiceprint recognition technology, and monitors the driver's and passengers' voice response status in real time. It identifies situations such as response timeouts after the dispatcher asks questions, groans caused by pain, rapid breathing, and other non-verbal sounds. The voice feature information and voice response status of the driver and passengers are integrated to generate the voice analysis results between the driver and passengers and the dispatcher.

[0028] Figure 4 This is a specific example diagram illustrating the determination of effective communication status between drivers and passengers in an embodiment of the present invention, such as... Figure 4 As shown, based on semantic parsing results, voice parsing results, and driver / passenger health monitoring data, combined with preset judgment conditions, the effective communication status of the driver / passenger is determined. When it is determined that the driver / passenger cannot communicate effectively, the alarm call link is taken over by a preset large model, which may include: Step 401: Based on the semantic analysis results, voice analysis results, and health monitoring data of the driver and passengers, match preset judgment conditions; the preset judgment conditions include judgment of the driver and passengers' voice response status, judgment of the completeness of the dialogue semantic logic, and judgment of the driver and passengers' health status. Step 402: Determine whether the driver and passengers are in a valid communication state based on the matching results; Step 403: If it is determined that the driver and passengers are unable to communicate effectively, the voice big data model will take over the alarm call link.

[0029] In a specific embodiment, based on semantic parsing results, voice parsing results, and driver / passenger health monitoring data, and combined with preset judgment conditions, the effective communication status of the driver / passenger is determined. When it is determined that the driver / passenger cannot communicate effectively, the alarm call link is taken over through a preset large model, including: Multi-dimensional data matching preset judgment conditions: The semantic and voice analysis results of the driver and passengers, as well as the health monitoring data of the driver and passengers transmitted by the wearable health monitoring device, are synchronously matched with preset judgment conditions. The preset judgment conditions include three judgment dimensions: judgment of the driver and passengers' voice response status, judgment of the completeness of the dialogue semantic logic, and judgment of the driver and passengers' health status, which correspond to the relevant data dimensions of voice analysis, semantic analysis, and health monitoring, respectively.

[0030] A comprehensive assessment of the effective communication status of the driver and passengers: Based on the matching results of multi-dimensional data and preset judgment conditions, a comprehensive judgment is made as to whether the driver and passengers are in an effective communication state: 1. If the response time of the driver or passenger exceeds the time limit after the dispatcher asks a question, for example, if the driver does not give any voice feedback within 5 seconds after the dispatcher asks a question.

[0031] 2. The semantic logic of the dialogue is incoherent or keywords are missing. For example, natural language processing can detect incoherent logic or missing keywords in the driver's speech, or identify non-verbal sounds such as groans and rapid breathing caused by pain.

[0032] 3. Sudden drop in heart rate or abnormal blood oxygen levels in drivers and passengers.

[0033] If any of the above matching results occur, it is determined that the driver and passengers cannot communicate effectively.

[0034] Preset large model to take over alarm call link: If the overall assessment determines that the driver and passengers are unable to communicate effectively, the pre-set large model will immediately take over the alarm call link, play a standardized identity statement to the dispatcher, and enter an interactive question-and-answer process with the dispatcher to complete the alarm call communication on behalf of the driver and passengers. For example, "I am the car owner's AI assistant. I have detected that the driver is unconscious. I will now report the accident situation."

[0035] Figure 5 This is a specific example diagram of accident information data transmission in an embodiment of the present invention, such as... Figure 5 As shown, the network quality of the data transmission link is detected, and based on the network quality detection results, the accident information data is transmitted to the police receiving unit via the data transmission link. This may include: Step 501: Detect the network quality of the data transmission link and obtain the network transmission rate; Step 502: Based on the network transmission rate and according to the preset priority of the accident information data, transmit the accident information data to the police through the data transmission link; the preset priority is arranged from high to low as driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data.

[0036] In a specific embodiment, the network quality of the data transmission link is detected, and based on the network quality detection results, the accident information data is transmitted to the police receiving personnel via the data transmission link, including: Network quality test results: The system uses a pre-set large-scale model to detect the network quality of the data transmission link established with the police in real time, and comprehensively obtains network quality detection results such as network transmission rate and signal stability, which serve as the basis for subsequent data transmission judgment.

[0037] Based on the network transmission rate and according to the preset priority of accident information data, the health monitoring data of drivers and passengers, the in-vehicle environment data, and the vehicle accident data are transmitted to the police receiving the case in sequence through the data transmission link. The next type of data will not be transmitted before the previous type of data has been transmitted.

[0038] The above-mentioned accident information data transmission steps based on network transmission rate can effectively solve the problem of interruption or loss of critical accident data transmission in remote road sections or accident scenarios with poor signal, and further improve the response speed and handling efficiency of emergency rescue.

[0039] In this embodiment, once the accident information data transmission is completed, the alarm call link and data transmission link with the police are disconnected.

[0040] In a specific embodiment, the alarm process is completed after the accident information data transmission is finished, including: After the pre-set large model confirms that all accident information data has been transmitted to the police station according to the transmission rules corresponding to the network quality test results, it first terminates the alarm communication link with the police station and simultaneously disconnects the data transmission link to complete the alarm process for this vehicle accident. After the link is disconnected, the pre-set large model immediately controls the mobile terminal to enter the emergency rescue waiting mode and performs subsequent rescue guidance related equipment status adjustment.

[0041] Specifically, the emergency rescue waiting mode includes the following four aspects: 1. Resource scheduling: Cut off all unnecessary background processes, shut down the 5G link, and switch to the 4G or 3G link to reduce power consumption, retaining only emergency location and network monitoring.

[0042] 2. Acoustic beacon positioning: Set the mobile phone to maximum volume. If the mobile phone can be set to another sound wave that is more obvious and easier for search and rescue personnel to hear, then change it to another sound wave that is more obvious.

[0043] 3. Optical Guidance: Combined with the ambient light sensor on the mobile phone; if the ambient light is below the threshold (such as at night or in a dark car), the LED flash on the mobile phone will cycle at a specific flashing frequency (such as an SOS code) and force the screen to turn on at maximum brightness to display a conspicuous rescue sign.

[0044] 4. Forced wake-up call: In this mode, all calls from rescuers will be forced to be played out at maximum volume and accompanied by optical guides to help search and rescue personnel quickly locate the rescue position.

[0045] Upon verification, the AI-assisted alarm method based on mobile terminals and vehicle terminals in this embodiment of the invention has the following beneficial effects: 1. Effectively solves the problems of multiple alarm links, low efficiency and easy failure of vehicle-mounted alarms, easy misjudgment of alarms on mobile phones, and lack of reliable alarm means when drivers and passengers are unconscious. It can achieve effective alarm when a serious car accident occurs and drivers and passengers are unconscious and unable to actively call for help.

[0046] 2. The alarm is triggered via a dual-link system on the mobile phone, with AI taking over the voice and data links to complete the alarm loop. This allows the system to communicate with the police in a timely manner, replacing the unconscious driver or passenger, ensuring the effectiveness and timeliness of the alarm.

[0047] 3. Utilize AI big data model capabilities to analyze multi-source accident information data, and combine this with intelligent network quality transmission to ensure that core information is delivered to the police with priority and reliability, providing comprehensive and accurate accident information for rescue operations.

[0048] 4. After the alarm is triggered, the AI-controlled mobile device enters the emergency rescue waiting mode. Through low power consumption settings and sound and light positioning guidance, it helps rescuers quickly locate the accident vehicle, improves rescue efficiency and success rate, and maximizes the safety of drivers and passengers.

[0049] This invention also provides a device for issuing alarms using AI-assisted vehicle-mounted systems based on mobile devices, as described in the following embodiments. Since the principle behind this device is similar to the method for issuing alarms using AI-assisted vehicle-mounted systems based on mobile devices, the implementation of this device can refer to the implementation of the method for issuing alarms using AI-assisted vehicle-mounted systems based on mobile devices; repeated details will not be elaborated further.

[0050] Figure 6 This is a structural example diagram of a device for issuing alarms using AI-assisted mobile terminal-based vehicle infotainment systems, as described in an embodiment of the present invention. Figure 6 As shown, the device for mobile-based AI-assisted vehicle-mounted alarm activation includes: The alarm configuration module 601 is used to receive vehicle accident alarm signals and accident information data sent by the vehicle terminal, initiate an alarm call link and data transmission link with the police based on the vehicle accident alarm signal, and configure a preset large model; generate a vehicle accident identification code and local location information, and transmit them to the police through the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data. The audio parsing module 602 is used to perform semantic parsing and speech parsing on the audio of the alarm call link through a preset large model, so as to obtain the semantic parsing results and speech parsing results between the driver and passengers and the police receiving the call. The status determination module 603 is used to determine the effective communication status of the driver and passenger based on the semantic parsing results, voice parsing results, and driver and passenger health monitoring data, combined with preset determination conditions. When it is determined that the driver and passenger cannot communicate effectively, the alarm call link is taken over through the preset large model. The accident information transmission module 604 is used to detect the network quality of the data transmission link and, based on the network quality detection results, transmits the accident information data to the police receiving the case through the data transmission link.

[0051] In one embodiment, the preset large model includes a language large model; Audio parsing module 602 is specifically used for: Acquire the audio stream of the conversation between the driver / passengers and the police officer in the alarm call link; The audio stream of the call is input into a large language model for semantic recognition and parsing to obtain the dialogue semantic information and interaction logic information in the audio stream, and generate semantic parsing results for the driver and passengers and the police.

[0052] In one embodiment, the preset large model includes a large speech model; Audio parsing module 602 is specifically used for: Acquire the audio stream of the conversation between the driver / passengers and the police officer in the alarm call link; The audio stream of the call is input into a large voice model for voice feature extraction, which yields the voice response status and voice feature information of the driver and passengers, and generates the voice analysis results of the driver and passengers and the police.

[0053] In one embodiment, the state determination module 603 is specifically used for: Based on the semantic analysis results of the driver and passengers and the police, the voice analysis results, and the health monitoring data of the driver and passengers, preset judgment conditions are matched; the preset judgment conditions include the judgment of the driver and passengers' voice response status, the judgment of the completeness of the dialogue semantic logic, and the judgment of the driver and passengers' health status. The matching results determine whether the driver and passengers are in a state of effective communication. If it is determined that the driver and passengers are unable to communicate effectively, the voice big data model will take over the alarm call link.

[0054] In one embodiment, the accident information transmission module 604 is specifically used for: Detect the network quality of the data transmission link and obtain the network transmission rate; Based on the network transmission rate and according to the preset priority of the accident information data, the accident information data is transmitted to the police through the data transmission link; the preset priority is arranged from high to low as driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data.

[0055] Figure 7 This is a specific example diagram of the structure of the device for issuing alarms based on mobile terminal artificial intelligence-assisted vehicle terminal in an embodiment of the present invention, as shown below. Figure 7 As shown in one embodiment, Figure 6 The device for alarming the vehicle-mounted system based on artificial intelligence from a mobile terminal, as shown in the embodiment of the present invention, may further include: a link disconnection module 701.

[0056] In one embodiment, the link disconnection module 701 is specifically used for: Once the accident information data transmission is complete, disconnect the alarm call link and data transmission link with the police.

[0057] Based on the aforementioned inventive concept, such as Figure 8 As shown, the present invention also proposes a computer device 800, including a memory 810, a processor 820, and a computer program 830 stored in the memory 810 and executable on the processor 820. When the processor 820 executes the computer program 830, it implements the aforementioned method for issuing alarms on a mobile terminal-based artificial intelligence-assisted vehicle terminal.

[0058] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned method for issuing alarms using an AI-assisted vehicle terminal based on a mobile terminal.

[0059] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the aforementioned method for issuing alarms using an AI-assisted vehicle terminal based on a mobile terminal.

[0060] In this embodiment of the invention, a vehicle accident alarm signal and accident information data sent by the vehicle-mounted terminal are received. Based on the vehicle accident alarm signal, an alarm call link and a data transmission link with the police are initiated, and a preset large model is configured. A vehicle accident identification code and local location information are generated and transmitted to the police via the data transmission link. The accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data. The audio of the alarm call link is semantically and verbally processed using the preset large model to obtain the semantic and verbal analysis results between the driver and passenger and the police. Based on the semantic and verbal analysis results and the driver and passenger health monitoring data, and combined with preset judgment conditions, the effective communication status of the driver and passenger is determined. When it is determined that the driver and passenger cannot communicate effectively, the alarm call link is taken over by the preset large model. The network quality of the data transmission link is detected, and based on the network quality detection results, the accident information data is transmitted to the police via the data transmission link. This embodiment of the invention achieves vehicle accident alarm in a low-intervention state through the coordinated linkage of the vehicle-mounted terminal and the mobile terminal, dual-link transmission, and network communication quality detection. This improves the comprehensiveness and accuracy of alarm data, enhances the stability of data transmission, and provides reliable technical support for accident emergency rescue solutions.

[0061] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0062] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0063] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0064] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0065] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for issuing alarms using mobile-based artificial intelligence-assisted vehicle-mounted systems, characterized in that, The method is applied to mobile devices and includes: The system receives vehicle accident alarm signals and accident information data sent from the vehicle terminal; initiates an alarm call link and data transmission link with the police based on the vehicle accident alarm signal and configures a preset large model; generates a vehicle accident identification code and local location information, and transmits them to the police through the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data. By performing semantic and speech parsing on the audio of the alarm call link using a pre-set large model, the semantic and speech parsing results of the driver and passengers and the police receiving the call are obtained. Based on semantic analysis results, voice analysis results, and health monitoring data of drivers and passengers, the effective communication status of drivers and passengers is determined by combining preset judgment conditions. When it is determined that drivers and passengers cannot communicate effectively, the alarm call link is taken over by a preset large model. The network quality of the data transmission link is checked, and based on the network quality check results, the accident information data is transmitted to the police receiving the case through the data transmission link.

2. The method as described in claim 1, characterized in that, The preset large model includes a language large model; The audio of the alarm call link is semantically parsed using a pre-set large model, including: Acquire the audio stream of the conversation between the driver / passengers and the police officer in the alarm call link; The audio stream of the call is input into a large language model for semantic recognition and parsing to obtain the dialogue semantic information and interaction logic information in the audio stream, and generate semantic parsing results for the driver and passengers and the police.

3. The method as described in claim 2, characterized in that, The preset large model includes a large voice model; The audio of the alarm call link is processed through voice parsing using a pre-set large model, including: Acquire the audio stream of the conversation between the driver / passengers and the police officer in the alarm call link; The audio stream of the call is input into a large voice model for voice feature extraction, which yields the voice response status and voice feature information of the driver and passengers, and generates the voice analysis results of the driver and passengers and the police.

4. The method as described in claim 3, characterized in that, Based on semantic analysis results, voice analysis results, and driver / passenger health monitoring data, and combined with preset judgment conditions, the effective communication status of the driver / passengers is determined. When it is determined that the driver / passengers cannot communicate effectively, the alarm call link is taken over through a preset large model, including: Based on the semantic analysis results of the driver and passengers and the police, the voice analysis results, and the health monitoring data of the driver and passengers, preset judgment conditions are matched; the preset judgment conditions include the judgment of the driver and passengers' voice response status, the judgment of the completeness of the dialogue semantic logic, and the judgment of the driver and passengers' health status. The matching results determine whether the driver and passengers are in a state of effective communication. If it is determined that the driver and passengers are unable to communicate effectively, the voice big data model will take over the alarm call link.

5. The method as described in claim 1, characterized in that, The network quality of the data transmission link is checked, and based on the network quality check results, the accident information data is transmitted to the police receiving the case through the data transmission link, including: Detect the network quality of the data transmission link and obtain the network transmission rate; Based on the network transmission rate and according to the preset priority of the accident information data, the accident information data is transmitted to the police through the data transmission link; the preset priority is arranged from high to low as driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data.

6. The method as described in claim 1, characterized in that, Also includes: Once the accident information data transmission is complete, disconnect the alarm call link and data transmission link with the police.

7. A device for issuing alarms using mobile-based artificial intelligence-assisted vehicle-mounted systems, characterized in that, The device is applied to a mobile terminal and includes: The alarm configuration module is used to receive vehicle accident alarm signals and accident information data sent by the vehicle terminal, initiate an alarm call link and data transmission link with the police based on the vehicle accident alarm signal, and configure a preset large model; generate a vehicle accident identification code and local location information, and transmit them to the police through the data transmission link; the accident information data includes driver and passenger health monitoring data, in-vehicle environment data, and vehicle accident data; The audio parsing module is used to perform semantic and speech parsing processing on the audio of the alarm call link through a preset large model, so as to obtain the semantic and speech parsing results of the driver and passengers and the police receiving the call. The status determination module is used to determine the effective communication status of the driver and passengers based on semantic parsing results, voice parsing results, and driver and passenger health monitoring data, combined with preset determination conditions. When it is determined that the driver and passengers cannot communicate effectively, the alarm call link is taken over through a preset large model. The accident information transmission module is used to detect the network quality of the data transmission link and, based on the network quality detection results, transmits the accident information data to the police receiving the case through the data transmission link.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1-6.

Images