Vehicle-mounted health monitoring method and system
By combining in-car voice information and car driving status, and using the Deepseek model to conduct health monitoring in all scenarios, the limitations of monitoring methods in existing technologies are overcome, and the accuracy of health monitoring in all scenarios and the avoidance of misjudgments are achieved.
Patent Information
- Application Number
- CN202510799921.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-16
- Publication Date
- 2025-10-10
AI Technical Summary
The existing method of monitoring the driver's physiological indicators and behavioral status cannot be applied to all driving scenarios and has the problem of detection failure.
By acquiring, parsing and judging voice information in the car, combining it with the car's driving status, and using inference big data models such as Deepseek, the user's health status can be judged. When necessary, inquiry instructions and assisted driving can be sent to achieve health monitoring in all scenarios.
It improves the accuracy of health monitoring, is applicable to all vehicle driving conditions, avoids misjudgment, is applicable to drivers and passengers, and shortens rescue time.
Smart Images
Figure CN120756496A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of safe automobile driving, and more particularly to a vehicle-mounted health monitoring method and system. Background Art
[0002] With the advancement of technology, people's demand for cars is increasing, and their concern for driving safety is also growing. According to statistics, health risks such as fatigue driving and sudden illness account for over 20% of the causes of traffic accidents. To further reduce the probability of traffic accidents caused by health risks while driving, existing technologies have already adopted real-time monitoring of the driver's physiological indicators and behavioral status to reduce the probability of traffic accidents caused by poor physical condition. Currently, there are two main methods for monitoring the driver's physiological indicators and behavioral status. One is to install a sensor on the steering wheel to collect heart rate data and determine the driver's health status based on heart rate characteristics. The other is to use a camera to scan parameters such as the user's chest rise and fall, facial features, etc., measure the attenuation of light reflected and absorbed by blood vessels and tissues, record the pulsation state of the blood vessels, measure the pulse wave signal, apply algorithmic filtering, and perform predictive analysis on the respiratory signal, heart rate, heart rate variability, respiratory rate, and blood oxygen saturation to determine the driver's health status.
[0003] However, both of these methods for monitoring a driver's physiological indicators and behavioral status have limitations and cannot be applied to all driving scenarios. Specifically, when using sensors on the steering wheel to monitor a user's health status, the user must simultaneously touch the electrodes with both hands and maintain a clear interface between the hands and the electrodes. This detection fails in hands-free autonomous driving scenarios, or when the user is wearing gloves or a steering wheel cover. When using cameras to collect data, accurate data collection is impossible in scenarios with changing lighting, when the user is wearing makeup or glasses, or in dim lighting, which can easily lead to detection failures. Summary of the Invention
[0004] The purpose of the present invention is to overcome the limitations of the existing methods for monitoring the driver's physiological indicators and behavioral status in the prior art and the shortcomings that they cannot be applied to all driving scenarios, and to provide a vehicle-mounted health monitoring method. The vehicle-mounted health monitoring method of this solution can be applied to all driving scenarios and improve the accuracy of vehicle-mounted health monitoring.
[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A vehicle health monitoring method is provided, comprising the following steps: Step 1: Get the voice information in the car; Step 2: Analyze the voice information and determine whether the voice information contains information that affects driving. If so, proceed to step 3; otherwise, proceed to step 1; Step 3: Determine whether the car is in a safe driving state. If yes, proceed to step 1; if not, proceed to step 4; Step 4: Send a query command to the user and determine whether the user responds. If the user responds, proceed to step 5; if the user does not respond, proceed to step 6; Step 5: Analyze the voice message replied by the user and determine whether the user needs help based on the instruction information. If the user needs help, proceed to step 6; if the user does not need help, proceed to step 1; Step 6: Turn on assisted driving, drive the vehicle to a safe location and issue a vehicle warning prompt.
[0006] The in-vehicle health monitoring method of the present invention monitors the health status of the user in the car through the voice information of the user in the car. When the user in the car expresses through voice information that the user's current health status affects the safe driving of the car, it can automatically judge whether the car is in a safe driving state. When it is judged that the car is not in a safe driving state, it will further judge whether the user has the ability to drive the car safely through inquiry instructions. Whether the user can drive the car safely can be further judged based on the user's response to the inquiry instruction. When the user cannot reply to the inquiry instruction or the reply instruction information requires help, it indicates that the user is not in a healthy state at this time and needs the intervention of assisted driving to drive the car instead of the user who is in health crisis, so as to avoid traffic accidents when the user continues to drive the car. At the same time, it can also send a help message number to the outside world to help users in the car who need help to seek help, thereby shortening the user's waiting time for help.
[0007] The in-vehicle health monitoring method of the present invention uses the user's voice information and the vehicle's driving status to determine the user's health status when monitoring the user's health. It is applicable to user health monitoring in all vehicle driving conditions and further determines the user's health status based on query instructions, improving the accuracy of health monitoring and avoiding misjudgments. Furthermore, this solution can be applied to the health monitoring of all users in the vehicle, including the driver and passengers. Both drivers and passengers can use this solution for health monitoring.
[0008] Preferably, in step 1, the in-vehicle voice information includes the voice information of the user in the vehicle and the sound characteristics of the vehicle in motion. The in-vehicle voice information can be collected through an in-vehicle assistant, such as a vehicle voice assistant. The in-vehicle voice information can be obtained after the vehicle voice assistant has monitoring permissions enabled. The sound characteristics of the vehicle in motion include abnormal sounds inside and outside the vehicle and the sound of the engine during normal vehicle driving.
[0009] In step 2, the voice information can be analyzed with the help of inference big data models such as Deepseek. When the monitoring permission of the vehicle voice assistant is turned on, the big data model learns the user's daily driving habits, user emotions, and health status and forms a data set. The data set consists of 7-30 days of real driving information of the user. The features in the data set are collected, trained and extracted to obtain the user's behavior or expression characteristics that deviate from the normal health state to form a data set that affects driving information. Its characteristics include long-term hands-off driving, sudden braking, negative words in the user's language, and no response to calls to the user.
[0010] Preferably, in step 2, determining whether the voice information contains information that affects driving includes the following steps: S21: Determine whether negative words appear in the voice information. If yes, proceed to S22; if not, proceed to S24; S22: Determine whether the safety level of the negative vocabulary meets the standard of dangerous driving. If yes, proceed to S23; if not, proceed to S24; S23: Output the judgment result as yes; S24: Output the judgment result as no.
[0011] The output judgment result is no, that is, no information that affects driving appears in the voice information; the output judgment result is yes, that is, information that affects driving appears in the voice information.
[0012] Negative words include physical pain, emotional frustration, and depression. Negative words are words that express negative emotions or physical discomfort, such as "I'm so tired today," "I feel really sick in my stomach today," and "I was really angry after arguing with XX today."
[0013] When a user is identified as expressing negative language, Deepseek and other inference big data models are used to further determine whether the user's state expressed by the negative language will affect their normal driving. Specifically, whether the negative language expressed by the user meets the criteria for dangerous driving. If so, this indicates that the voice information is affecting driving. Examples of negative language that meet the dangerous driving criteria include headache, dizziness, blurred vision, and limb dysfunction. Emotional words such as quarreling and anger are specifically judged as potentially contributing to dangerous driving using Deepseek and other inference big data models. The big data models make this judgment based on the user's tone and volume.
[0014] Preferably, in step three, determining whether the vehicle is in a safe driving state includes the following steps: S31: Obtain vehicle driving road condition information, and determine whether the vehicle is speeding based on the vehicle driving status. If yes, proceed to S32; if not, proceed to S34; S32: Determine whether the user controls the vehicle normally. If yes, proceed to S33; if not, proceed to S34; S33: Output the judgment result as yes; S34: Output the judgment result as no.
[0015] If the output judgment result is no, it means that the car is not in a safe driving state; if the output judgment result is yes, it means that the car is in a safe driving state.
[0016] In said S31, the vehicle driving road condition information includes the vehicle's geographical location, driving time and weather conditions.
[0017] After training, Deepseek and other inference big data models can determine whether a vehicle is speeding based on the vehicle's geographic information (i.e., road conditions, including lane information and speed limits), travel time, and weather conditions, combined with the vehicle's speed. For example, considering dark or rainy weather conditions, the speed limit for the assessed road section will be lowered. If the vehicle is speeding, it indicates that the vehicle is not operating safely. If the vehicle is not speeding, the system then determines whether the user is properly controlling the vehicle. If the user is manually controlling the vehicle, proper control indicates that the vehicle is operating safely. If the user is not properly controlling the vehicle, meaning that the user has not taken any control action, such as turning the steering wheel, for an extended period, the vehicle is not operating safely. When the vehicle is in autonomous driving mode, the system uses voice inquiries and other methods to determine whether the user is monitoring the vehicle's driving status in real time. If the user does not respond, the vehicle is not operating safely.
[0018] Preferably, in step six, during the assisted driving process, the vehicle's hazard lights are turned on; Safe locations include the roadside, service areas closest to the vehicle, and parking lots; Vehicle warning prompts include calling the emergency number or 120.
[0019] When assisted driving is enabled, Deepseek and other inference big data models use the vehicle's location and map information to determine the driving time to the nearest service area or parking lot. If the driving time exceeds the target time, the assisted driving system will stop the vehicle on the roadside. The target time is generally 15-20 minutes.
[0020] The present invention also provides a vehicle-mounted health monitoring system, comprising a collection module for acquiring in-vehicle voice information, an information processing module for determining the health status of the user in the vehicle, and an execution module for controlling the vehicle's travel; The acquisition module and the execution module are both connected to the information processing module; The information processing module can determine whether the car is in a safe driving state and whether the user needs help after sending the inquiry instruction.
[0021] The collection module sends the collected in-vehicle voice information to the information processing module, the information processing module analyzes the voice information and monitors the health status of the in-vehicle user through the in-vehicle user's voice information, when the in-vehicle user's current state expressed through the voice information affects the safe driving of the car, further determine whether the car is in a safe driving state, when it is determined that the car is not in a safe driving state, further determine whether the user has the ability to safely drive the car through the inquiry instruction, the information processing module will further determine whether the user can safely drive the car according to the user's reply to the inquiry instruction, when the user cannot reply to the inquiry instruction or the replied instruction information needs help, it indicates that the user is not in a healthy state at this time, and the intervention of auxiliary driving is needed, the information processing module controls the execution module to start working, the execution module starts the automatic driving of the car, and the car is driven to a safe position and an alarm prompt is issued.
[0022] The vehicle-mounted health monitoring system of the present application can automatically determine the health status of the in-vehicle user through the in-vehicle user's voice information and the driving state of the car during driving, and can be applied to user health monitoring in all driving states of the car, and further determine the health status of the user according to the inquiry instruction, improve the accuracy of health monitoring, and avoid misjudgment.
[0023] The present application also provides a computer device comprising a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the steps of the above-mentioned vehicle-mounted health monitoring method.
[0024] The present application also provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to realize the steps of the above-mentioned vehicle-mounted health monitoring method.
[0025] Compared with the prior art, the present application has the following advantages: The vehicle-mounted health monitoring method and system of the present application can determine the health status of the in-vehicle user through the in-vehicle user's voice information and the driving state of the car during driving, and can be applied to user health monitoring in all driving states of the car, and further determine the health status of the user according to the inquiry instruction, improve the accuracy of health monitoring, and avoid misjudgment. In addition, the present application can be applied to health monitoring of all users including the driver and the passenger in the car, whether the driver or the passenger, can use the present application for health monitoring. BRIEF DESCRIPTION OF DRAWINGS
[0026] Figure 1is a flow chart of a vehicle health monitoring method; Figure 2 A flow chart of a vehicle health monitoring method for determining whether voice information contains information that affects driving; Figure 3 A flow chart of a vehicle health monitoring method for determining whether a vehicle is in a safe driving state; Figure 4 This is a module diagram of a vehicle-mounted health monitoring system. The arrows in the figure indicate the direction of signal transmission. DETAILED DESCRIPTION
[0027] Example 1 This embodiment is a first embodiment of a vehicle health monitoring method. Figure 1 As shown, the following steps are included: Step 1: Get the voice information in the car; Step 2: Analyze the voice information and determine whether the voice information contains information that affects driving. If so, proceed to step 3; otherwise, proceed to step 1; Step 3: Determine whether the car is in a safe driving state. If yes, proceed to step 1; if not, proceed to step 4; Step 4: Send a query command to the user and determine whether the user responds. If the user responds, proceed to step 5; if the user does not respond, proceed to step 6; Step 5: Analyze the voice message replied by the user and determine whether the user needs help based on the instruction information. If the user needs help, proceed to step 6; if the user does not need help, proceed to step 1; Step 6: Turn on assisted driving, drive the vehicle to a safe location and issue a vehicle warning prompt.
[0028] The working principle or working process of this embodiment is as follows: The in-vehicle health monitoring method of this embodiment monitors the health status of the user in the car through the voice information of the user in the car. When the user in the car expresses through voice information that the user's current health status affects the safe driving of the car, it can automatically determine whether the car is in a safe driving state. When it is determined that the car is not in a safe driving state, it will further determine whether the user has the ability to drive the car safely through inquiry instructions. Whether the user can drive the car safely can be further determined based on the user's response to the inquiry instructions. When the user cannot reply to the inquiry instruction or the reply instruction information requires help, it indicates that the user is not in a healthy state at this time and needs the intervention of assisted driving to drive the car instead of the user who is in health crisis, so as to avoid traffic accidents when the user continues to drive the car. At the same time, it can also send a help message number to the outside world to help users in the car who need help to seek help, thereby shortening the user's waiting time for help.
[0029] The beneficial effects of this embodiment are as follows: The in-vehicle health monitoring method of this embodiment uses the user's voice information and the vehicle's driving status to determine the user's health status. This method is applicable to user health monitoring in all vehicle driving conditions. It further determines the user's health status based on query instructions, improving the accuracy of health monitoring and avoiding misjudgments. Furthermore, this solution can be applied to the health monitoring of all users in the vehicle, including the driver and passengers. Both drivers and passengers can use this solution for health monitoring.
[0030] Example 2 This embodiment is a second embodiment of a vehicle health monitoring method. Based on the first embodiment, this embodiment further defines steps one to six.
[0031] Specifically, in step 1, the in-vehicle voice information includes the voice information of the user in the vehicle and the sound characteristics of the vehicle in motion. The in-vehicle voice information can be collected through an in-vehicle assistant, such as a vehicle voice assistant. Once the vehicle voice assistant has monitoring permissions enabled, the in-vehicle voice information can be obtained. The sound characteristics of the vehicle in motion include unusual noises inside and outside the vehicle, as well as the sound of the engine during normal driving.
[0032] In step 2, the voice information can be analyzed with the help of inference big data models such as Deepseek. When the monitoring permission of the vehicle voice assistant is turned on, the big data model learns the user's daily driving habits, user emotions, and health status and forms a data set. The data set consists of 7-30 days of real driving information of the user. The features in the data set are collected, trained and extracted to obtain the user's behavior or expression characteristics that deviate from the normal health state to form a data set that affects driving information. Its characteristics include long-term hands-off driving, sudden braking, negative words in the user's language, and no response to calls to the user.
[0033] Specifically, in step 2, if Figure 2 As shown, determining whether the voice information contains information that affects driving includes the following steps: S21: Determine whether negative words appear in the voice information. If yes, proceed to S22; if not, proceed to S24; S22: Determine whether the safety level of the negative vocabulary meets the standard of dangerous driving. If yes, proceed to S23; if not, proceed to S24; S23: Output the judgment result as yes; S24: Output the judgment result as no.
[0034] Negative words include physical pain, emotional frustration, and depression. Negative words are words that express negative emotions or physical discomfort, such as "I'm so tired today," "I feel really sick in my stomach today," and "I was really angry after arguing with XX today."
[0035] When a user is identified as expressing negative language, Deepseek and other inference big data models are used to determine whether the user's state expressed by these negative words will affect their normal driving. Specifically, whether the negative language expressed by the user meets the criteria for dangerous driving. If so, this indicates that the voice information is affecting driving. Examples of negative language that meet the criteria for dangerous driving include headache, dizziness, blurred vision, and limb dysfunction. Emotional language such as quarreling and anger are specifically judged as potentially contributing to dangerous driving using Deepseek and other inference big data models. The big data models make this judgment based on the user's tone and volume.
[0036] Specifically, in step three, if Figure 3 As shown, judging whether the car is in a safe driving state includes the following steps: S31: Obtain vehicle driving road condition information, and determine whether the vehicle is speeding based on the vehicle driving status. If yes, proceed to S32; if not, proceed to S34; S32: Determine whether the user controls the vehicle normally. If yes, proceed to S33; if not, proceed to S34; S33: Output the judgment result as yes; S34: Output the judgment result as no.
[0037] In said S31, the vehicle driving road condition information includes the vehicle's geographical location, driving time and weather conditions.
[0038] After training, Deepseek and other inference big data models can determine whether a vehicle is speeding based on its geographic location, travel time, and weather conditions, combined with the vehicle's speed. For example, considering conditions such as darkness and rain will lower the speed limit for the assessed road section. If the vehicle is speeding, it indicates that the vehicle is not operating safely. If the vehicle is not speeding, the system then determines whether the user is properly controlling the vehicle. If the user is manually controlling the vehicle, proper control indicates that the vehicle is operating safely. If the user is not properly controlling the vehicle, meaning that the user has not taken any control actions such as turning the steering wheel for an extended period, the vehicle is not operating safely. When the vehicle is in autonomous driving mode, voice inquiries and other methods are used to determine whether the user is monitoring the vehicle's driving status in real time. If the user does not respond, the vehicle is not operating safely.
[0039] Specifically, in step six, during the assisted driving process, the vehicle's hazard lights are turned on; Safe locations include the roadside, service areas closest to the vehicle, and parking lots; Vehicle warning prompts include calling the emergency number or 120.
[0040] When assisted driving is enabled, Deepseek and other inference big data models use the vehicle's location and map information to determine the driving time to the nearest service area or parking lot. If the driving time exceeds the target time, the assisted driving system will stop the vehicle on the roadside. The target time is generally 15-20 minutes.
[0041] Example 3 This embodiment is an embodiment of a vehicle-mounted health monitoring system. Figure 4 As shown, it includes a collection module for acquiring voice information in the car, an information processing module for judging the health status of the user in the car, and an execution module for controlling the vehicle's travel; The acquisition module and the execution module are both connected to the information processing module; The information processing module can determine whether the car is in a safe driving state and determine whether the user needs help after sending an inquiry instruction.
[0042] The working principle or working process of this embodiment is as follows: The acquisition module sends the collected in-car voice information to the information processing module. After analyzing the voice information, the information processing module monitors the health status of the user in the car through the voice information of the user in the car. When the voice information in the car expresses that the user's current status affects the safe driving of the car, it further determines whether the car is in a safe driving state. When it is determined that the car is not in a safe driving state, it will further determine whether the user has the ability to drive the car safely through inquiry instructions. The information processing module will also further determine whether the user can drive the car safely based on the user's response to the inquiry instruction. When the user cannot reply to the inquiry instruction or the reply instruction information requires help, it indicates that the user is not in a healthy state at this time and requires the intervention of assisted driving. The information processing module controls the execution module to start working, and the execution module starts the car's automatic driving, drives the car to a safe position and issues an alarm prompt.
[0043] The beneficial effects of this embodiment are as follows: The in-vehicle health monitoring method of this embodiment judges the health status of the in-vehicle user through the voice information of the in-vehicle user and the driving status of the car when monitoring the health of the in-vehicle user. It can be applied to user health monitoring under all driving conditions of the car, and will further determine the user's health status based on the inquiry instructions, thereby improving the accuracy of health monitoring and avoiding misjudgment.
[0044] Example 4 This embodiment is an embodiment of a computer device, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the vehicle health monitoring method described in the above embodiment 1 when executing the computer program.
[0045] Example 5 This embodiment is an embodiment of a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the vehicle health monitoring method described in the above embodiment 1 are implemented.
[0046] In the specific contents of the above-mentioned specific implementation methods, the various technical features can be combined in any non-contradictory manner. In order to make the description concise, not all possible combinations of the above-mentioned technical features are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0047] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.
Claims
1. A vehicle health monitoring method, characterized in that: The steps include: Step 1: Get the voice information in the car; Step 2: Analyze the voice information and determine whether the voice information contains information that affects driving. If so, proceed to step 3; otherwise, proceed to step 1; Step 3: Determine whether the car is in a safe driving state. If yes, proceed to step 1; if not, proceed to step 4; Step 4: Send a query command to the user and determine whether the user responds. If the user responds, proceed to step 5; if the user does not respond, proceed to step 6; Step 5: Analyze the voice message replied by the user and determine whether the user needs help based on the instruction information. If the user needs help, proceed to step 6; If the user does not need help, proceed to step 1; Step 6: Turn on assisted driving, drive the vehicle to a safe location and issue a vehicle warning prompt.
2. The vehicle-mounted health monitoring method according to claim 1, characterized in that: In the step 1, the in-car voice information includes the voice information of the user in the car and the sound characteristics of the car driving.
3. The vehicle-mounted health monitoring method according to claim 1, characterized in that: In step 2, determining whether the voice information contains information that affects driving includes the following steps: S21: Determine whether negative words appear in the voice information. If yes, proceed to S22; if not, proceed to S24; S22: Determine whether the safety level of the negative vocabulary meets the standard of dangerous driving. If yes, proceed to S23; if not, proceed to S24; S23: Output the judgment result as yes; S24: Output the judgment result as no.
4. The vehicle-mounted health monitoring method according to claim 3, characterized in that: In the S21, negative words include physical pain, emotional frustration, and depression.
5. The vehicle-mounted health monitoring method according to claim 1, characterized in that: In step three, determining whether the vehicle is in a safe driving state includes the following steps: S31: Obtain vehicle driving road condition information, and determine whether the vehicle is speeding based on the vehicle driving status. If yes, proceed to S32; if not, proceed to S34; S32: Determine whether the user controls the vehicle normally. If yes, proceed to S33; if not, proceed to S34; S33: Output the judgment result as yes; S34: Output the judgment result as no.
6. The vehicle-mounted health monitoring method according to claim 1, characterized in that: In said S31, the vehicle driving road condition information includes the vehicle's geographical location, driving time and weather conditions.
7. The vehicle-mounted health monitoring method according to claim 1, characterized in that: In step 6, during the assisted driving process, the car's hazard lights are turned on; Safe locations include the roadside, service areas closest to the vehicle, and parking lots; Vehicle warning prompts include calling the emergency number or 120.
8. A vehicle-mounted health monitoring system, characterized in that: It includes a collection module for acquiring voice information in the car, an information processing module for judging the health status of the user in the car, and an execution module for controlling the vehicle's driving; The acquisition module and the execution module are both connected to the information processing module; The information processing module can determine whether the car is in a safe driving state and determine whether the user needs help after sending an inquiry instruction.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the vehicle health monitoring method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the vehicle health monitoring method according to any one of claims 1 to 7 are implemented.