Hierarchical health service switching method based on vehicle-mounted intelligent agent and related equipment

Abstract

The application provides a hierarchical health service switching method based on a vehicle-mounted intelligent agent and related equipment, the method comprising: acquiring health consultation information input by a user and collected physiological monitoring data, vehicle operating state data and navigation data; constructing an individual physiological baseline of the user and generating a long-term health portrait; generating a structured health suggestion through the vehicle-mounted intelligent agent according to the health consultation information; determining physiological index risk values, symptom risk values and driving state risk values, and obtaining a comprehensive risk score by weighted fusion; determining a risk trend sequence in a future preset period; performing risk level division according to the comprehensive risk score and the risk trend sequence to obtain a risk level; and matching a corresponding health service strategy based on the risk level and performing service switching through the vehicle-mounted intelligent agent. The method of the application makes the risk assessment result more suitable for the real physical state of the user. High-risk states can be actively warned and intervened in advance.

Description

Technical Field

[0001] This application relates to the field of intelligent cockpit technology, and in particular to a hierarchical health service transfer method and related equipment based on in-vehicle intelligent agents. Background Technology

[0002] With the rapid development of intelligent cockpits, vehicle-to-everything (V2X) technology, and large-scale modeling, more and more vehicles are beginning to possess health monitoring and health-related question-and-answer capabilities. Existing in-vehicle health systems typically use universal medical thresholds to make indiscriminate judgments on all users, ignoring individual differences. Such threshold strategies lead to high false alarm rates and even missed detections of critical risks. Furthermore, existing systems either only assess health status or only assess driving status, lacking the ability to predict the "coupled high-risk state" formed by the superposition of health risks and driving load, thus failing to intervene in advance. When health abnormalities are detected, users are often immediately alerted via voice or pop-up windows, without considering the complexity of the current driving task. During high-speed lane changes or emergency avoidance, any unnecessary voice interaction may cause cognitive overload, increasing the risk of accidents. After triggering emergency rescue, only simple call transfers or location reporting are completed, failing to form a data loop with emergency dispatch centers and target hospitals.

[0003] Therefore, there is an urgent need for a hierarchical health service transfer method and related equipment based on in-vehicle intelligent agents to solve all or part of the above problems. Summary of the Invention

[0004] In view of this, the purpose of this application is to propose a hierarchical health service transfer method and related equipment based on in-vehicle intelligent agents.

[0005] To achieve the above objectives, the first aspect of this application provides a hierarchical health service transfer method based on an in-vehicle intelligent agent, comprising: acquiring user-inputted health consultation information and collected physiological monitoring data, vehicle operating status data, and navigation data through an in-vehicle intelligent agent; constructing an individual physiological baseline for the user and generating a long-term health profile based on the physiological monitoring data and the vehicle operating status data; generating structured health suggestions through the in-vehicle intelligent agent based on the health consultation information; determining the deviation of each physiological indicator in the physiological monitoring data relative to the same scenario based on the long-term health profile, and obtaining physiological indicator risk values; determining symptom risk values ​​based on the health consultation information; determining driving status risk values ​​based on the vehicle operating status data; weightedly fusing the physiological indicator risk values, symptom risk values, and driving status risk values ​​to obtain a comprehensive risk score; determining a risk trend sequence within a future preset time period based on the physiological monitoring data, vehicle operating status data, and navigation data; classifying risk levels based on the comprehensive risk score and the risk trend sequence to obtain risk levels; and matching corresponding health service strategies based on the risk levels and transferring services through the in-vehicle intelligent agent.

[0006] Optionally, the step of constructing a user's individual physiological baseline and generating a long-term health profile based on the physiological monitoring data and the vehicle operating status data includes: acquiring the physiological monitoring data and the vehicle operating status data from historical driving cycles; in response to determining that the vehicle is in a parked state and has remained there for more than a preset time, and detecting that the driver is in the vehicle, marking the corresponding time period as a resting period, and collecting resting physiological data; classifying driving scenarios based on the vehicle operating status data to obtain driving scenario labels; grouping the physiological monitoring data according to the driving scenario labels, performing online rolling statistics on each physiological indicator under each driving scenario, and establishing a same-scenario distribution model; establishing a resting physiological baseline based on the median of the resting physiological data within a preset number of historical days; and combining the same-scenario distribution model and the resting physiological baseline to obtain the long-term health profile.

[0007] Optionally, determining the risk trend sequence within a future preset time period based on the physiological monitoring data, the vehicle operating status data, and the navigation data includes: extracting time-series segments of a preset historical duration from the physiological monitoring data as historical time-series physiological data; extracting time-series segments of a preset historical duration from the vehicle operating status data as historical driving status sequences; dividing the future path in the navigation data into multiple time-segment road segments, encoding each segment using road type labels and expected driving load levels to obtain future path information; inputting the historical time-series physiological data, the historical driving status sequences, and the future path information into a pre-trained risk trend sequence prediction model, and outputting a risk prediction value sequence within the future preset time period as the risk trend sequence; and, in response to determining that there is a risk prediction value greater than a preset threshold within a future preset window in the risk prediction value sequence and the road type label of the corresponding time period belongs to a preset high driving load label set, marking the driving state as a high-risk state.

[0008] Optionally, after obtaining the risk level, the method further includes: generating risk-dominant source labels based on the comprehensive risk score; acquiring environmental data and combining different risk-dominant source labels to generate corresponding cabin environment intervention strategies.

[0009] Optionally, after generating the corresponding cockpit environment intervention strategy, the method further includes: calculating a driving task complexity index based on the vehicle operating status data; responding to determining that the driving task complexity index is lower than a first threshold, using full voice for active health interaction with the user; responding to determining that the driving task complexity index is greater than or equal to the first threshold and less than a second threshold, prohibiting voice interaction during active health interaction with the user, and only reminding the user through icons or brief text displayed on the dashboard; responding to determining that the driving task complexity index is greater than or equal to the second threshold, not engaging in active health interaction with the user until the driving task complexity index drops below the first threshold, and then popping up a reminder queue according to priority.

[0010] Optionally, the step of matching the corresponding health service strategy based on the risk level and transferring the service through the in-vehicle intelligent agent includes: in response to determining that the risk level is low risk, outputting the structured health advice to the user without initiating voice interaction with the user; in response to determining that the risk level is medium risk, after outputting the structured health advice and initiating voice interaction, if the user does not respond with voice and the driving task complexity index is lower than a first threshold, outputting a voice message providing health services to the user and delaying the reminder; in response to determining that the risk level is high risk and the driving state is the high-risk state, prompting the user to park safely and recommending the nearest parking area based on the navigation data, and directly providing online doctor services or emergency rescue services after the vehicle speed is reduced to a preset speed or switched to park; in response to determining that the risk level is high risk and the driving state is not the high-risk state, directly providing online doctor services or emergency rescue services.

[0011] Optionally, the physiological monitoring data includes at least one of heart rate, heart rate variability, body temperature, respiratory rate, and skin conductance; the vehicle operating status data includes at least one of vehicle speed, steering wheel angle, brake pedal depth, turn signal status, driving duration, mileage, following distance, lane departure frequency, and fatigue warning trigger frequency; and the navigation data includes at least one of road type, estimated remaining driving time, and road conditions ahead.

[0012] To achieve the above objectives, a second aspect of this application provides a hierarchical health service transfer device based on an in-vehicle intelligent agent, comprising: a data acquisition module configured to acquire user-inputted health consultation information and collected physiological monitoring data, vehicle operating status data, and navigation data through the in-vehicle intelligent agent; a first generation module configured to construct an individual physiological baseline for the user based on the physiological monitoring data and the vehicle operating status data, and generate a long-term health profile; a second generation module configured to generate structured health recommendations based on the health consultation information through the in-vehicle intelligent agent; and a risk assessment module configured to determine the risk based on the long-term health profile. The deviation of each physiological indicator from the physiological monitoring data relative to the same scenario is used to obtain the physiological indicator risk value; the symptom risk value is determined based on the health consultation information; the driving status risk value is determined based on the vehicle operation status data; the physiological indicator risk value, the symptom risk value, and the driving status risk value are weighted and fused to obtain a comprehensive risk score; the risk prediction module is configured to determine a risk trend sequence within a future preset time period based on the physiological monitoring data, the vehicle operation status data, and the navigation data; the risk level classification module is configured to classify the risk level according to the comprehensive risk score and the risk trend sequence to obtain the risk level.

[0013] The service transfer module is configured to match the corresponding health service strategy based on the risk level and transfer the service through the in-vehicle intelligent agent.

[0014] Based on the same inventive concept, a third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in the first aspect.

[0015] Based on the same inventive concept, a fourth aspect of this application provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the first aspect.

[0016] As can be seen from the above, the hierarchical health service transfer method and related equipment based on in-vehicle intelligent agents provided in this application effectively solve the problems of false alarms and missed alarms caused by physiological differences between individuals by establishing an individual physiological baseline for users in the same scenario instead of traditional general medical thresholds, making the risk assessment results more consistent with the user's actual physical condition. It dynamically integrates the deviation of physiological monitoring, the severity of symptoms, and driving load with weights to achieve a comprehensive quantification of user health risks. It can proactively predict high-risk states coupled with driving load, enabling early warning and intervention. Based on differentiated adjustment strategies according to the source of risk, it avoids safety hazards caused by reminders interfering with driving and increasing cognitive load. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the hierarchical health service transfer method based on in-vehicle intelligent agents according to an embodiment of this application.

[0019] Figure 2 This is a schematic diagram of the structure of the hierarchical health service transfer device based on an in-vehicle intelligent agent according to an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0022] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0023] The technical solution of this application will be described in detail below through specific embodiments.

[0024] refer to Figure 1 One embodiment of this application provides a hierarchical health service transfer method based on an in-vehicle intelligent agent, which specifically includes the following steps:

[0025] Step S101: Obtain the health consultation information input by the user, as well as the collected physiological monitoring data, vehicle operation status data, and navigation data through the in-vehicle intelligent agent.

[0026] In this step, health consultation information includes, but is not limited to, user voice Q&A content, text input content, and symptom description information. Physiological monitoring data can be obtained from the vehicle's UWB radar, seat-embedded piezoelectric sensors, or Bluetooth-connected wearable devices (such as watches), including but not limited to heart rate, heart rate variability, body temperature, respiratory rate, and skin conductance. Vehicle operating status data such as vehicle speed, steering wheel angle, brake pedal depth, longitudinal / lateral acceleration, yaw rate, turn signal status, driving duration, and mileage can be read via the CAN bus, as well as vehicle operating status data such as following distance, lane departure frequency, and fatigue warning trigger frequency obtained from the ADAS system. Navigation data includes, but is not limited to, road type, estimated remaining driving time, and road conditions ahead.

[0027] Step S102: Based on the physiological monitoring data and the vehicle operating status data, construct the user's individual physiological baseline and generate a long-term health profile.

[0028] Specifically, this step also includes the following steps:

[0029] Step S1021: Obtain the physiological monitoring data and vehicle operating status data from the historical driving cycle.

[0030] In this step, each power-on to power-off cycle is considered a driving cycle.

[0031] Step S1022: In response to determining that the vehicle is in a parked state and has stayed for more than a preset time, and detecting that the driver is in the vehicle, mark the corresponding time period as a resting period and collect resting physiological data.

[0032] In this step, when the vehicle is in Park (P) and the dwell time exceeds a preset duration (e.g., ten minutes), and the seat occupancy sensor detects that the user is inside the vehicle, the onboard intelligent agent marks this period as a resting period and collects resting physiological data such as resting heart rate, resting heart rate variability, body temperature, resting respiratory rate, and resting skin conductance. Typically, 3-5 minutes of continuous data are collected, and the median or mean of the stable period is used to avoid fluctuations at the initial seating position.

[0033] Step S1023: Based on the vehicle operating status data, classify the driving scenarios to obtain driving scenario labels.

[0034] In this step, the driving scenario tags can be divided into smooth cruise, following, lane change, intersection turning, acceleration overtaking, and braking avoidance. Specific settings can be configured according to actual conditions; no specific limitations are provided here.

[0035] Step S1024: Group the physiological monitoring data according to the driving scenario label, perform online rolling statistics on each physiological indicator under each driving scenario, and establish a distribution model for the same scenario.

[0036] In this step, for each driving scenario label, consecutive physiological data segments appearing in the corresponding scenario are extracted. Each segment lasts for at least a preset duration (e.g., 30 seconds) to ensure statistical stability. The same-scenario distribution model includes sample counts, sample mean, and sample standard deviation. The model is updated using an exponentially weighted moving average based on the same-scenario data from 30 historical driving cycles.

[0037] Step S1025: Establish a resting physiological baseline based on the median of the resting physiological data within a preset number of historical days.

[0038] Specifically, the median of the resting physiological data over the past 7 days can be used to establish resting physiological baselines: resting heart rate baseline, resting heart rate variability baseline, resting respiratory rate baseline, and resting skin conductance response baseline.

[0039] Step S1026: Combine the same scene distribution model and the resting physiological baseline to obtain the long-term health profile.

[0040] In this step, the user's health profile is stored in the vehicle's chip or in their personal cloud account after being anonymized.

[0041] By creating user health profiles and conducting personalized risk assessments, the accuracy of risk identification can be improved.

[0042] Step S103: Generate structured health recommendations based on the health consultation information through the in-vehicle intelligent agent.

[0043] In this step, after receiving user input, the in-vehicle intelligent agent uses a pre-trained natural language understanding model to identify the user's intent and determine whether the current question falls under the category of health consultation. For identified health issues, it further extracts symptom keywords, descriptions of physical condition, and risk-related information; for non-health issues, it proceeds to the standard in-vehicle assistant processing flow. For identified health issues, it retrieves relevant knowledge from a pre-built health knowledge base. This knowledge is then processed by a pre-trained knowledge extraction model to generate structured health information, including possible causes, daily suggestions, and risk signals. The knowledge extraction model filters out technical disease terminology and complex medical expressions, converting medical knowledge into health guidance information that ordinary users can understand.

[0044] Step S104: Based on the long-term health profile, determine the deviation of each physiological indicator in the physiological monitoring data relative to the same scenario to obtain the physiological indicator risk value; determine the symptom risk value based on the health consultation information; determine the driving status risk value based on the vehicle operation status data; and perform weighted fusion of the physiological indicator risk value, the symptom risk value, and the driving status risk value to obtain a comprehensive risk score.

[0045] In this step, the risk value of physiological indicators is calculated as follows:

[0046] For each physiological indicator, two deviation assessments are performed. First, a mean deviation assessment is conducted. The deviation of the mean value from the mean value of the sliding window over the most recent 60 seconds relative to the mean value of the same scene distribution is calculated.

[0047]

[0048] in, Indicators of physiological indicators The arithmetic mean of the sampled values ​​within the most recent preset duration (e.g., 60 seconds) sliding window. Physiological indicators in the same scene distribution model The mean value in the corresponding driving scenario Physiological indicators in the same scene distribution model The standard deviation in the corresponding driving scenario.

[0049] Second, calculate the deviation of the mean of the physiological index within the sliding window from the resting physiological baseline:

[0050]

[0051] in, Indicates physiological indicators at resting physiological baseline The mean, Indicates physiological indicators at resting physiological baseline The standard deviation. Let the resting deviation threshold be... ,when When it is determined that a user has a long-term sensitivity or chronic deviation to a certain physiological indicator, the risk contribution weight of the corresponding physiological indicator is increased by a preset ratio. At that time, the weight remains unchanged.

[0052] Transform the mean deviation to prevent the impact of extreme deviations:

[0053]

[0054] in, Indicates the first The deviation of the mean of each physiological indicator.

[0055] Let the adjusted weights be... ,when At that time, the risk contribution weight of the corresponding physiological indicator will be increased by a preset ratio, that is... ,in, Preset increase ratio; otherwise .

[0056] Weighted risk values ​​for physiological indicators:

[0057]

[0058] in, Indicators of physiological indicators Corresponding weights Indicators of physiological indicators The corresponding risk component.

[0059] Symptom risk value The calculation method is as follows: if the symptoms are high-risk symptoms (such as severe chest pain, unilateral limb weakness, confusion, severe respiratory distress), then Common symptoms are mapped to severity levels: mild, slightly, mildly, a few times, with corresponding scores. Average, Medium, Mapping Score Obvious, severe, and persistent, mapping score Intense, severe, unbearable, intolerable, mapping score If the symptom description contains multiple degree terms, the highest score will be used; if there are no explicit degree terms, the median score of 35 will be used by default.

[0060] A time-based factor is applied based on the duration information in the symptom description or the current duration of symptom monitoring: Duration less than 5 minutes: Duration greater than or equal to 5 minutes and less than 15 minutes: The duration is greater than or equal to 15 minutes and less than 30 minutes: Duration greater than or equal to 30 minutes: If the user does not explicitly describe the duration, but the system detects that the physiological indicators related to the symptom (such as abnormal heart rate variability corresponding to dizziness) are continuously abnormal, the duration of the abnormality will be used for calculation.

[0061] Based on the triggering factors or current driving scenario described in the symptom description, determine whether it belongs to a high-risk triggering scenario: sudden occurrence while driving, plus factor. After exercise, the additive coefficient No obvious cause, additive coefficient Known medical history is relevant, and the additive coefficient is used. When multiple contributing factors exist, the highest coefficient is taken.

[0062] Basic risk score for general symptoms:

[0063]

[0064] in, The score represents the duration of symptoms.

[0065] When a user describes multiple symptoms simultaneously (such as headache + nausea + blurred vision), a non-linear summation method is used to avoid oversaturation caused by simple linear accumulation.

[0066] Let the risk of each symptom be calculated individually as follows: (Each symptom is calculated according to the above high-risk or general rules), then the combined symptom risk score is:

[0067]

[0068] in, Indicates symptom index, This represents the symptom correlation coefficient, indicating whether two symptoms belong to the same system (e.g., headache + dizziness belong to the nervous system). If the symptoms are medically linked across systems (e.g., dizziness + palpitations). If the symptoms are not clearly related to the symptoms, .

[0069] If a user mentions a past medical history (e.g., "I have high blood pressure"), and the current symptoms are medically related to that history, the symptom risk score will be adjusted.

[0070]

[0071] in, This indicates a symptom risk score adjusted based on past medical history. This represents the historical additive coefficient, which is directly related to: Indirect connection: Unrelated: .

[0072] Normalize to the 0-100 range:

[0073]

[0074] By fusing vehicle speed, steering wheel turning rate, braking frequency, etc. with fuzzy logic, the driving task complexity index (DTCI) is mapped to 0-100. The deviation between the driving behavior in the current 5 minutes (such as the variance of lane keeping deviation and the stability of following distance) and the user's historical deviation in the same scenario is calculated and also converted to 0-100 as the degree of abnormality of driving behavior.

[0075] The calculation method for driving status risk value is as follows:

[0076]

[0077] in, This represents the complexity index of the driving task. Indicates the degree of abnormality in driving behavior.

[0078] Ultimately, the overall risk score is:

[0079]

[0080] in, , , This indicates dynamic weights.

[0081] It should be noted that the weighting of the comprehensive risk score is not a fixed value, but rather changes dynamically according to the scenario. For example, when the vehicle is cruising at high speed and there are no subjective health complaints, more attention is paid to the coupling between physiological and driving conditions. When users proactively report discomfort, the risk weight of the symptom increases: .

[0082] Step S105: Determine the risk trend sequence within a future preset time period based on the physiological monitoring data, the vehicle operation status data, and the navigation data.

[0083] This step also includes the following steps:

[0084] Step S1051: Extract time-series segments of a preset historical duration from the physiological monitoring data as historical time-series physiological data.

[0085] In this step, the extracted historical duration can be set to 5 minutes, or it can be set according to the actual situation; no specific limitation is made here.

[0086] Step S1052: Extract time-series segments of a preset historical duration from the vehicle operating status data as historical driving status sequences.

[0087] In this step, the extracted historical duration is consistent with the physiological monitoring data.

[0088] Step S1053: Divide the future path in the navigation data into multiple time-segment road segments. Encode each road segment using road type labels and expected driving load levels to obtain future path information.

[0089] In this step, the next 15 minutes can be divided into 30 30-second segments, each segment is labeled with tags such as ordinary straight road, curve, ramp, tunnel, congestion queue, and driving load level.

[0090] Step S1054: Input the historical time-series physiological data, the historical driving state sequence, and the future path information into the pre-trained risk trend sequence prediction model, and output the risk prediction value sequence within the future preset time period as the risk trend sequence.

[0091] In this step, the risk trend sequence prediction model uses a hybrid network of spatiotemporal graph convolution and gated recurrent units. Physiological indicators and driving status are extracted as features through two one-dimensional convolutional layers, concatenated with the scene context, and fed into two layers of gated recurrent units to output a sequence of risk prediction values ​​for each minute over the next 15 minutes.

[0092] Step S1055: In response to determining that there is a risk prediction value greater than a preset threshold in the risk prediction value sequence within a future preset window and the road type label of the corresponding time period belongs to a preset high driving load label set, the driving state is marked as a high-risk state.

[0093] In this step, the predicted risk threshold is set to 70, and the high-driving-load road segment label set includes: sharp bends, highway exits, complex intersections, and construction zones. When predicting the future... If a road segment simultaneously meets the criteria of a risk prediction value greater than or equal to 70 and the corresponding road segment label belongs to the high driving load label set, it is marked as a high-risk state.

[0094] Step S106: Divide the risk level according to the comprehensive risk score and the risk trend sequence to obtain the risk level.

[0095] In this step, a first threshold of 30 and a second threshold of 65 can be set. If the overall risk score is less than 30 and there is no risk prediction value greater than 50 within the next ten minutes, the risk level is determined to be low. If the overall risk score is greater than or equal to 30 and less than 65, or if the overall risk score is less than 30 but there is a risk prediction value greater than or equal to 50 within the next ten minutes, the risk level is determined to be medium. If the overall risk score is greater than or equal to 65, or the driving status is high-risk, or any risk prediction value is greater than or equal to 80, the risk level is determined to be high.

[0096] Step S107: Match the corresponding health service strategy based on the risk level and transfer the service through the vehicle-mounted intelligent agent.

[0097] This step also includes the following steps:

[0098] Step S1071: In response to determining that the risk level is low risk, output the structured health advice to the user without initiating voice interaction with the user.

[0099] Specifically, this can be displayed in concise text on the dashboard or HUD secondary area, such as "You may be a little tired, we suggest you rest at the service area ahead."

[0100] Step S1072: In response to determining that the risk level is medium risk, after outputting the structured health advice and initiating voice interaction, if the user does not respond with voice and the driving task complexity index is lower than the first threshold, output a voice message to the user to provide health services and delay the reminder.

[0101] In this step, after providing suggestions, if the in-vehicle intelligent agent detects that the user does not respond with voice, The system proactively asks via voice: "I've noticed some changes in your physical condition. Would you like me to schedule a consultation with a health advisor for you? Just tell me 'yes' or 'later'." It then provides three options: 1. Connect with your health manager immediately; 2. Set a delayed reminder (to check on you again in 30 minutes); 3. Ignore.

[0102] Step S1073: In response to determining that the risk level is high risk and the driving state is high-risk, prompt the user to stop safely and recommend the nearest parking area based on the navigation data. After the vehicle speed is reduced to the preset speed or switched to parking gear, provide online doctor service or emergency rescue service directly.

[0103] In this step, the in-vehicle intelligent agent prioritizes driving safety by guiding the driver to slow down and pull over through red warning icons on the HUD, steering wheel vibration, and brief voice prompts (no more than 5 words, such as "Please stop safely"); at the same time, it recommends the nearest parking area (such as emergency parking lanes or service area entrances) based on navigation data.

[0104] Step S1074: In response to determining that the risk level is high risk and the driving status is not the high-risk status, directly provide online doctor services or emergency rescue services.

[0105] In some embodiments, the following steps are included after step S106:

[0106] Step S201: Generate risk-dominant source labels based on the comprehensive risk score.

[0107] In this step, the risk-dominant source label is used to identify the main contributing dimensions of the overall risk score, guiding subsequent cabin intervention strategies. The contribution ratio of physiological indicator risk values, symptom risk values, and driving status risk values ​​to the overall risk score is calculated. Based on the largest contribution ratio, the risk is categorized as physiological-dominant, symptom-dominant, or driving status-dominant.

[0108] Step S202: Obtain environmental data and, in conjunction with the different risk-dominant source labels, generate corresponding cabin environment intervention strategies.

[0109] In this step, environmental data includes outside temperature, light intensity, and road type.

[0110] When the risk-dominant source is labeled as physiologically dominant, cabin environment intervention aims to regulate physiological stress by adjusting the balance of the autonomic nervous system through environmental stimuli. For example, when the skin conductance response is high, the volume is reduced by 30%, external information input is reduced, the central control screen interface is simplified, only the vehicle speed and navigation arrows are retained, and unnecessary information cards are hidden.

[0111] When the risk-dominant source label is symptom-dominant, cabin environment intervention focuses on symptom relief and safety guidance. It provides targeted environmental adjustments based on the user's specific symptoms. For example, when dizzy, the air conditioning temperature is lowered by 2.0°C, the airflow is directed at the upper body, and the fan speed is set to medium-high for 3 minutes; a mint fragrance is released to stimulate and relieve vasodilation in the head; the light is switched to cool white light and the brightness is reduced by 10% to avoid exacerbating dizziness; the nearest available parking area card is displayed (distance, estimated arrival time), and a message is displayed: "It is recommended to pull over and rest. A safe parking spot has been found for you."

[0112] When the primary risk source is identified as driving status, cabin environment interventions aim to reduce driver load and maintain alertness, avoiding unnecessary information interference and prioritizing driving safety. For example, during long periods of continuous driving, the HUD secondary area displays information about the nearest service area / rest area (distance, estimated arrival time) and prompts "Resting is recommended"; it activates lumbar massage to relieve fatigue from prolonged sitting; and it automatically switches to external air circulation to introduce fresh outside air and reduce cabin CO2 concentration.

[0113] In some embodiments, the method further includes the following after step S202:

[0114] Step S301: In response to determining that the driving task complexity index is below a first threshold, use full voice for proactive health interaction with the user.

[0115] Step S302: In response to determining that the driving task complexity index is greater than or equal to the first threshold and less than the second threshold, voice interaction is prohibited during active health interaction with the user, and only icons or short text reminders are displayed on the dashboard.

[0116] Step S303: In response to determining that the driving task complexity index is greater than or equal to the second threshold, do not perform proactive health interaction with the user until the driving task complexity index drops below the first threshold, and then pop up the reminder queue according to priority.

[0117] It should be noted that when the risk level is high and the driving status is high-risk, even if the driving task complexity index is high, it is permissible to send high-priority, minimalist prompts to guide the driver to a safe stop.

[0118] By implementing tiered control over user interactions, we can avoid proactive interactions that could interfere with driving safety and ensure that health reminders are delivered effectively without increasing cognitive load.

[0119] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0120] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0121] Based on the same inventive concept, and corresponding to any of the above embodiments, this application also provides a hierarchical health service transfer device based on an in-vehicle intelligent agent.

[0122] refer to Figure 2 The tiered health service transfer device based on in-vehicle intelligent agents includes:

[0123] Data acquisition module 201: is configured to acquire user-inputted health consultation information, as well as collected physiological monitoring data, vehicle operating status data, and navigation data through the in-vehicle intelligent agent;

[0124] First generation module 202: configured to construct the user's individual physiological baseline and generate a long-term health profile based on the physiological monitoring data and the vehicle operating status data;

[0125] The second generation module 203 is configured to generate structured health recommendations based on the health consultation information through the in-vehicle intelligent agent.

[0126] Risk assessment module 204 is configured to, based on the long-term health profile, determine the deviation of each physiological indicator in the physiological monitoring data relative to the same scenario, and obtain the physiological indicator risk value; determine the symptom risk value based on the health consultation information; determine the driving status risk value based on the vehicle operation status data; and perform weighted fusion of the physiological indicator risk value, the symptom risk value, and the driving status risk value to obtain a comprehensive risk score.

[0127] Risk prediction module 205: configured to determine a risk trend sequence within a future preset time period based on the physiological monitoring data, the vehicle operating status data, and the navigation data;

[0128] Risk level classification module 206: configured to classify risk levels based on the comprehensive risk score and the risk trend sequence to obtain risk levels;

[0129] Service transfer module 207: is configured to match the corresponding health service strategy based on the risk level and transfer services through the vehicle-mounted intelligent agent.

[0130] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0131] The apparatus in the above embodiments is used to implement the corresponding hierarchical health service transfer method based on vehicle intelligent agents in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0132] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the hierarchical health service transfer method based on in-vehicle intelligent agents described in any of the above embodiments.

[0133] Figure 3This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0134] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0135] The memory 1020 can be implemented in the form of ROM (Read-Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0136] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0137] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0138] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0139] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0140] The electronic devices described above are used to implement the corresponding hierarchical health service transfer method based on vehicle-mounted intelligent agents in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0141] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the hierarchical health service transfer method based on in-vehicle intelligent agents as described in any of the above embodiments.

[0142] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0143] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the hierarchical health service transfer method based on vehicle intelligent agents as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0144] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0145] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0146] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0147] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A hierarchical health service transfer method based on in-vehicle intelligent agents, characterized in that, include: The vehicle-mounted intelligent agent acquires health consultation information input by the user, as well as collected physiological monitoring data, vehicle operating status data, and navigation data. Based on the physiological monitoring data and the vehicle operating status data, an individual physiological baseline for the user is constructed, and a long-term health profile is generated. Based on the health consultation information, structured health recommendations are generated through the in-vehicle intelligent agent; Based on the long-term health profile, the deviation of each physiological indicator in the physiological monitoring data relative to the same scenario is determined, and the risk value of the physiological indicator is obtained. Based on the health consultation information, symptom risk values ​​are determined; The driving status risk value is determined based on the vehicle operating status data; The risk values ​​of the physiological indicators, symptoms, and driving status are weighted and fused to obtain a comprehensive risk score. Based on the physiological monitoring data, the vehicle operating status data, and the navigation data, a risk trend sequence is determined for a future preset time period; The risk level is determined by classifying the risk level based on the comprehensive risk score and the risk trend sequence. Based on the risk level, a corresponding health service strategy is matched and the service is transferred through the in-vehicle intelligent agent.

2. The method according to claim 1, characterized in that, The process of constructing a user's individual physiological baseline and generating a long-term health profile based on the physiological monitoring data and the vehicle operating status data includes: Acquire the physiological monitoring data and vehicle operating status data from the historical driving cycle; In response to determining that the vehicle is in a parked state and has remained there for more than a preset time, or when the driver is detected inside the vehicle, the corresponding time period is marked as a resting period and resting physiological data is collected. Based on the vehicle operating status data, driving scenarios are classified to obtain driving scenario labels; The physiological monitoring data are grouped according to the driving scenario labels, and the physiological indicators under each driving scenario are statistically analyzed online to establish a distribution model for the same scenario. Establish a resting physiological baseline based on the median of the resting physiological data within a preset number of historical days; The long-term health profile is obtained by combining the same scene distribution model and the resting physiological baseline.

3. The method according to claim 1, characterized in that, The step of determining the risk trend sequence within a preset future time period based on the physiological monitoring data, the vehicle operating status data, and the navigation data includes: Extract time-series segments of a preset historical duration from the physiological monitoring data as historical time-series physiological data; Extract time-series segments of preset historical duration from the vehicle operating status data as historical driving status sequences; The future path in the navigation data is divided into multiple time-segment road segments, and each segment is encoded using road type labels and expected driving load levels to obtain future path information; The historical time-series physiological data, the historical driving state sequence, and the future path information are input into a pre-trained risk trend sequence prediction model, which outputs a risk prediction value sequence within a preset future time period as the risk trend sequence. In response to determining that there is a risk prediction value greater than a preset threshold in the risk prediction value sequence within a future preset window and the road type label for the corresponding time period belongs to a preset high driving load label set, the driving state is marked as a high-risk state.

4. The method according to claim 3, characterized in that, After obtaining the risk level, the following also includes: Risk-dominant source labels are generated based on the comprehensive risk score; By acquiring environmental data and combining it with the labels of different risk-dominant sources, corresponding cabin environment intervention strategies are generated.

5. The method according to claim 4, characterized in that, After generating the corresponding cockpit environment intervention strategy, the following is also included: Calculate the driving task complexity index based on the vehicle operating status data; In response to determining that the driving task complexity index is below a first threshold, the active health interaction with the user uses full voice. When it is determined that the driving task complexity index is greater than or equal to a first threshold and less than a second threshold, voice interaction is prohibited during active health interaction with the user, and the user is only reminded by displaying icons or short text on the dashboard. In response to determining that the driving task complexity index is greater than or equal to the second threshold, no proactive health interaction will be performed with the user until the driving task complexity index drops below the first threshold, at which point a reminder queue will be popped up according to priority.

6. The method according to claim 5, characterized in that, The process of matching a corresponding health service strategy based on the risk level and transferring services through the in-vehicle intelligent agent includes: In response to determining that the risk level is low, the structured health advice is output to the user without initiating a voice interaction with the user; In response to determining the risk level as medium risk, after outputting the structured health advice and initiating voice interaction, if the user does not respond with voice and the driving task complexity index is lower than the first threshold, the system outputs a voice message to the user to provide health services and delays the reminder. In response to determining that the risk level is high risk and the driving status is high-risk, the system prompts the user to stop safely and recommends the nearest parking area based on the navigation data. Once the vehicle speed is reduced to a preset speed or the vehicle is switched to park, the system directly provides online medical services or emergency rescue services. In response to determining that the risk level is high risk and the driving condition is not the high-risk condition, online medical services or emergency rescue services are provided directly.

7. The method according to claim 1, characterized in that, The physiological monitoring data includes at least one of heart rate, heart rate variability, body temperature, respiratory rate, and skin conductance; the vehicle operating status data includes at least one of vehicle speed, steering wheel angle, brake pedal depth, turn signal status, driving time, mileage, following distance, lane departure frequency, and fatigue warning trigger frequency; the navigation data includes at least one of road type, estimated remaining driving time, and road conditions ahead.

8. A hierarchical health service transfer device based on an in-vehicle intelligent agent, characterized in that, include: The data acquisition module is configured to acquire user-inputted health consultation information, as well as collected physiological monitoring data, vehicle operating status data, and navigation data through the in-vehicle intelligent agent; The first generation module is configured to construct the user's individual physiological baseline and generate a long-term health profile based on the physiological monitoring data and the vehicle operating status data. The second generation module is configured to generate structured health recommendations based on the health consultation information through the in-vehicle intelligent agent; The risk assessment module is configured to determine the deviation of each physiological indicator in the physiological monitoring data relative to the same scenario based on the long-term health profile, and obtain the risk value of the physiological indicator. Based on the health consultation information, symptom risk values ​​are determined; The driving status risk value is determined based on the vehicle operating status data; The risk values ​​of the physiological indicators, symptoms, and driving status are weighted and fused to obtain a comprehensive risk score. The risk prediction module is configured to determine a risk trend sequence within a preset future time period based on the physiological monitoring data, the vehicle operating status data, and the navigation data. The risk level classification module is configured to classify risk levels based on the comprehensive risk score and the risk trend sequence to obtain risk levels. The service transfer module is configured to match the corresponding health service strategy based on the risk level and transfer the service through the in-vehicle intelligent agent.

9. An electronic 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 program, it implements the method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method described in any one of claims 1 to 7.

Images