Driver health status assessment method based on digital twinning and related equipment

Abstract

The invention discloses a digital twinning-based driver health status assessment method and related equipment, and relates to the technical field of vehicles. The digital twinning-based driver health state assessment method comprises the following steps: acquiring health data of a driver and smell information in a vehicle cabin; then, based on the health data and the smell information, the health state of the driver is analyzed and predicted through a virtual health model, the health state prediction result of the driver is obtained, and the virtual health model comprises a health prediction recognition model; the health prediction identification model is obtained by training health data and polymorphic twin odor data, and the polymorphic twin odor data is used for representing odor characteristic data of the driver in different health states; finally, under the condition that the health state prediction result shows that the health state of the driver is abnormal, early warning information is sent out and / or the driving state of the vehicle is actively intervened. According to the embodiment, the driving safety can be improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a driver health status assessment method based on digital twins and related equipment. Background Art

[0002] In the field of vehicle driving, the health of the driver is directly related to the safe operation of the vehicle, especially when faced with health problems such as drunk driving, diabetes and cardiovascular disease, which have become the main causes of serious traffic accidents. In order to reduce the occurrence of such accidents, countries have adopted preventive measures such as driving alcohol ban, safety education and training, and regular physical examinations. However, these measures have significant limitations.

[0003] The assessment of the driver's health status in related technologies often has a lag and cannot respond before the health problem affects the driver's behavior, thus causing a certain impact on driving safety. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to provide a driver health status assessment method and related equipment based on digital twins, aiming to improve the real-time performance of driver health status assessment and thereby improve driving safety.

[0005] To achieve the above purpose, one aspect of an embodiment of the present application proposes a driver health status assessment method based on digital twins, comprising the following steps:

[0006] Obtain driver health data and vehicle cabin odor information;

[0007] Based on the health data and the odor information, the health status of the driver is analyzed and predicted through a virtual health model to obtain a prediction result of the driver's health status, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained by training the health data and polymorphic twin odor data, and the polymorphic twin odor data is odor feature data used to characterize the driver in different health states;

[0008] When the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving status of the vehicle is actively intervened.

[0009] In some embodiments, after obtaining the driver's health data and the odor information in the vehicle cabin, the method further includes:

[0010] Acquiring image data including the driver's body shape, posture and facial information;

[0011] A virtual geometric model of the driver is generated based on the image data and basic personal information of the driver.

[0012] In some embodiments, the odor information includes gas information exhaled by the driver, and the driver's health status includes blood sugar information of the driver;

[0013] The analyzing and predicting the health status of the driver through a virtual health model based on the health data and the odor information to obtain a prediction result of the health status of the driver includes:

[0014] determining the driver's abnormal blood sugar history information based on the health data;

[0015] If the abnormal blood sugar history information indicates that the driver has a history of blood sugar, the driver's blood sugar is predicted according to the driver's exhaled gas information through the health prediction recognition model to obtain the driver's blood sugar information prediction result.

[0016] In some embodiments, the health prediction and recognition model includes a blood sugar prediction model, and the blood sugar of the driver is predicted according to the breath information of the driver by the health prediction and recognition model to obtain the blood sugar information prediction result of the driver, including:

[0017] Extracting a first key feature from the breath information of the driver by using a first algorithm, wherein the first algorithm includes one or more of a mean, a standard deviation, a kurtosis, a skewness, or a fast Fourier transform;

[0018] Selecting a first target feature vector through the first key feature, where the first target feature vector is used to represent a feature vector suitable for predicting blood sugar information;

[0019] The first target feature vector is input into the blood glucose prediction model to obtain the blood glucose information prediction result of the driver.

[0020] In some embodiments, the odor information includes gas information in the cockpit and gas information exhaled by the driver, and the driver's health status includes the driver's drinking information;

[0021] The analyzing and predicting the health status of the driver through a virtual health model based on the health data and the odor information to obtain a prediction result of the health status of the driver includes:

[0022] Performing alcohol content detection based on the gas information in the cockpit to obtain alcohol content indication information;

[0023] If the alcohol content indication information indicates that the gas in the cockpit contains alcohol, the health prediction and recognition model is used to identify whether the driver has been drinking alcohol based on the gas information exhaled by the driver to obtain the driver's drinking information recognition result.

[0024] In some embodiments, the health prediction and recognition model includes a drinking recognition model, and the health prediction and recognition model is used to identify whether the driver drinks alcohol according to the breath information exhaled by the driver to obtain the drinking information recognition result of the driver, including:

[0025] Extracting a second key feature from the breath information of the driver by a second algorithm, wherein the second algorithm includes one or more of a maximum value, an integral value, a wavelet transform, or a fast Fourier transform;

[0026] Selecting a second target feature vector through the second key feature, where the second target feature vector is used to represent a feature vector suitable for identifying alcohol content in gas;

[0027] The second target feature vector is input into the drinking recognition model to obtain the drinking information recognition result of the driver.

[0028] In some embodiments, the health data includes a driver's historical blood glucose level record, and the health prediction recognition model includes a blood glucose prediction model;

[0029] The blood sugar prediction model is trained and updated through the following steps:

[0030] training the blood glucose prediction model using a random forest algorithm according to the historical blood glucose level records and the polymorphic twin odor data, wherein the polymorphic twin odor data includes characteristic data of acetone concentration in the driver's breath;

[0031] The blood glucose prediction model is updated according to the latest acquired health data of the driver and the polymorphic twin odor data.

[0032] In some embodiments, the health data includes the driver's historical drinking record, and the health prediction recognition model includes a drinking recognition model;

[0033] The drinking recognition model is trained and updated through the following steps:

[0034] Training the drinking recognition model using a support vector machine algorithm based on the historical drinking records and the polymorphic twin odor data, wherein the polymorphic twin odor data includes ethanol concentration characteristic data in a vehicle cabin;

[0035] The drinking recognition model is updated according to the latest acquired health data of the driver and the polymorphic twin odor data.

[0036] To achieve the above purpose, another aspect of the embodiment of the present application proposes a driver health status assessment system based on digital twins, and the driver health status assessment system based on digital twins includes:

[0037] A physical layer, which is used to collect and manage the driver's basic personal information, health information, and environmental information;

[0038] A twin model layer, wherein the twin model layer includes a blood sugar prediction model and a drinking recognition model, and is used to predict and recognize the health status of the driver;

[0039] A twin data layer, wherein the twin data layer is used to store and manage the driver's historical health data and current environment data;

[0040] An application service layer, the application service layer comprising a virtual visualization module, a health status assessment module and an intelligent driving control module;

[0041] The virtual visualization module is used to display the driver's health status and in-vehicle environment information on the display screen in a graphical form;

[0042] The health status assessment module is used to analyze the prediction and recognition results of the driver's health status and assess the driver's health status;

[0043] The intelligent driving control module is used to issue warning information and / or actively intervene in the driving state of the vehicle according to the evaluation result of the driver's health status;

[0044] The data interaction layer transmits and processes data through a data transmission protocol, sends the data to the twin model layer for predicting and identifying the driver's health status, and feeds back the results of the health status prediction and identification to the application service layer.

[0045] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application proposes a driver health status assessment device based on digital twin, and the driver health status assessment device based on digital twin includes:

[0046] An acquisition module, used to obtain the driver's health data and the odor information in the vehicle cabin;

[0047] An analysis and prediction module, configured to analyze and predict the health status of the driver through a virtual health model based on the health data and the odor information, and obtain a prediction result of the driver's health status, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained by training the health data and polymorphic twin odor data, and the polymorphic twin odor data is odor feature data used to characterize the driver in different health states;

[0048] The control module is used to issue a warning message and / or actively intervene in the driving state of the vehicle when the health state prediction result shows that the driver's health state is abnormal.

[0049] To achieve the above-mentioned purpose, another aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned digital twin-based driver health status assessment method when executing the computer program.

[0050] To achieve the above-mentioned purpose, another aspect of an embodiment of the present application proposes a vehicle, which includes a driver health status assessment method based on digital twin of the above-mentioned embodiment, a driver health status assessment system based on digital twin of the above-mentioned embodiment, a driver health status assessment device based on digital twin of the above-mentioned embodiment, or an electronic device of the above-mentioned embodiment.

[0051] The embodiments of the present application include at least the following beneficial effects:

[0052] The present application provides a driver health status assessment method and related equipment based on digital twins. In the embodiment of the present application, first, the driver's health data and the odor information in the vehicle cabin are obtained; then, the driver's health status is analyzed and predicted through a virtual health model based on the health data and the odor information to obtain the driver's health status prediction result, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained through training with health data and polymorphic twin odor data, and the polymorphic twin odor data is used to characterize the odor feature data of the driver in different health states; finally, when the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving state of the vehicle is actively intervened. In the embodiment of the present application, the driver's health status can be analyzed and predicted through a virtual health model based on the driver's health data and the odor information in the vehicle cabin. Therefore, the driver's health status can be monitored and predicted in real time while the driver is driving the vehicle, and the driver's health condition can be predicted before the driver's health has problems. When the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving state of the vehicle is actively intervened, thereby improving driving safety and reducing traffic accidents caused by health problems.

[0053] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0055] Figure 1 is a flowchart of a driver health status assessment method based on digital twins provided in some embodiments of the present application;

[0056] Figure 2 is a flowchart of a method for analyzing and predicting a driver's health status based on a virtual health model provided by some embodiments of the present application;

[0057] Figure 3 is a schematic block diagram of modules of a driver health status assessment system based on digital twins provided in some embodiments of the present application;

[0058] Figure 4 It is a module schematic block diagram of a driver health status assessment device based on digital twin provided in some embodiments of the present application;

[0059] Figure 5 It is a schematic diagram of the hardware structure of an electronic device provided in some embodiments of the present application. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the reference to "embodiment" in this article means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0061] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0062] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0063] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0064] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0065] In order to make the inventive concept of the present application easy to understand, before describing the embodiments of the present application in detail, the English abbreviations (terms) / related concepts involved in the embodiments of the present application are first described. The English abbreviations (terms) / related concepts involved in the embodiments of the present application are subject to the following explanations.

[0066] Digital Twin Technology: Digital Twin (DT) refers to a virtual model of a physical object, system or process created through digital technology. This model can reflect the state and behavior of its physical counterpart in real time, and can simulate, predict and optimize the physical counterpart. Through data collection and analysis, digital twins can provide real-time monitoring and early warning, which is helpful for system maintenance and improvement. Digital twins connect the physical world and the information world, providing a new direction for drivers to monitor abnormal conditions during driving.

[0067] In the field of automobile driving, the health status of the driver is directly related to the safe operation of the vehicle, especially when facing health problems such as drunk driving, diabetes and cardiovascular disease, these factors have become the main causes of serious traffic accidents. In order to reduce the occurrence of such accidents, countries have adopted preventive measures such as driving alcohol ban, safety education and training, and regular physical examinations. However, these measures have significant limitations.

[0068] First, the solutions to traffic accidents caused by drivers’ health conditions are mostly preventive and therefore lack real-time performance. This means that before health problems actually affect the driver’s behavior, it is often difficult to detect and intervene in time, resulting in a large lag in the assessment of health conditions.

[0069] Secondly, existing technical means lack the ability to collect and analyze drivers' physiological data in real time. Although traditional health monitoring devices can provide certain health information, these devices usually need to be in direct contact with the driver and are usually only used in specific medical scenarios. This not only makes real-time monitoring more difficult, but also limits the application of these devices in dynamic driving environments.

[0070] Finally, due to the limitations of health monitoring methods, current driving safety assessments are often in a passive assessment state. This assessment model can only investigate responsibilities after an accident occurs, but cannot provide active preventive measures before an accident occurs, making it difficult to effectively reduce the incidence of traffic accidents.

[0071] The assessment of the driver's health status in related technologies often has a lag and cannot respond before the health problem affects the driver's behavior, thus causing a certain impact on driving safety.

[0072] In view of this, the present application proposes a driver health status assessment method and related equipment based on digital twins. The scheme obtains the driver's health data and the odor information in the vehicle cabin; then, based on the health data and the odor information, the driver's health status is analyzed and predicted through a virtual health model to obtain the driver's health status prediction result, wherein the virtual health model includes a health prediction recognition model, which is obtained through training with health data and polymorphic twin odor data, and the polymorphic twin odor data is used to characterize the odor feature data of the driver in different health states; finally, when the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving state of the vehicle is actively intervened. In the embodiment of the present application, the driver's health status can be analyzed and predicted through a virtual health model based on the driver's health data and the odor information in the vehicle cabin. Therefore, the driver's health status can be monitored and predicted in real time while the driver is driving the vehicle, and the driver's health condition can be predicted before the driver's health has problems. When the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving state of the vehicle is actively intervened, thereby improving driving safety and reducing traffic accidents caused by health problems.

[0073] The method provided in the embodiment of the present application can be applied to the electronic device provided in the embodiment of the present application, wherein the electronic device can be a terminal or a server.

[0074] The terminal may be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto.

[0075] The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks, as well as big data and artificial intelligence platforms.

[0076] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0077] The following will describe in detail the implementation steps of a driver health status assessment method based on digital twin provided by an embodiment of the present application in conjunction with the accompanying drawings.

[0078] Please refer to Figure 1 , Figure 1 A flowchart of a driver health status assessment method based on digital twins is provided for some embodiments of the present application. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in an order different from that shown here.

[0079] The method of the embodiment of the present application comprises the following steps:

[0080] Step 101: Obtaining driver's health data and odor information in the vehicle cabin;

[0081] Step 102: Analyze and predict the health status of the driver through a virtual health model based on the health data and the odor information to obtain a prediction result of the driver's health status, wherein the virtual health model includes a health prediction recognition model, which is obtained through training with the health data and polymorphic twin odor data, and the polymorphic twin odor data is odor feature data used to characterize the driver in different health states;

[0082] Step 103: When the health status prediction result shows that the driver's health status is abnormal, issue a warning message and / or actively intervene in the driving status of the vehicle.

[0083] Steps 101 to 103 shown in the embodiment of the present application can analyze and predict the driver's health status through a virtual health model based on the driver's health data and the odor information in the vehicle cabin. Therefore, the driver's health status can be monitored and predicted in real time while the driver is driving the vehicle, and the driver's health condition can be predicted before the driver's health problems occur. When the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving status of the vehicle is actively intervened, thereby improving driving safety and reducing traffic accidents caused by health problems.

[0084] The specific implementation methods of the above steps are introduced below.

[0085] In step 101, the driver's health data and the odor information in the vehicle cabin are obtained.

[0086] Among them, health data refers to various information related to personal health status, which can be used to evaluate, monitor and improve personal health status. Health data can generally include the following aspects:

[0087] Physiological parameters: including body temperature, heart rate, blood pressure, blood sugar, blood oxygen saturation, respiratory rate, etc.

[0088] Lifestyle: including living habits such as diet, exercise, sleep, smoking, and drinking.

[0089] Medical records: including medical history, diagnosis, treatment, medication use, surgical records, laboratory test results, etc.

[0090] Health risk assessment: including family medical history, genetic information, disease risk prediction, etc.

[0091] Mental health: including emotional state, psychological stress, mental health status, etc.

[0092] Health monitoring device data: includes data collected by devices such as smart watches, fitness trackers, sleep monitors, etc.

[0093] Health questionnaires and surveys: including personal health questionnaires, lifestyle surveys, etc.

[0094] Health insurance information: including insurance claims, insurance coverage, etc.

[0095] Health education and prevention: including vaccination records, health checks, disease prevention measures, etc.

[0096] The odor information in the vehicle cabin can be various odors in the air inside the vehicle, including the gas exhaled by the driver, the gas exhaled by the passengers, and the gas from the environment outside the vehicle that enters the vehicle.

[0097] The driver's health data can be obtained through conventional means. For example, after obtaining authorization, the driver's health data can be obtained through the public transportation system, the driver's health information can be obtained through the hospital database, or the driver's health information can be obtained through an insurance company, etc. This application does not impose any restrictions on this.

[0098] The odor information in the vehicle cabin can be obtained through an electronic nose (e-nose), which is an instrument that simulates the human sense of smell and uses multiple sensors to detect odors in the air. The electronic nose can provide information about the characteristics of the odor; it can also be used through spectral technology, such as infrared spectroscopy (IR) and ultraviolet-visible spectroscopy (UV-Vis), which can be used to analyze the chemicals in the air to infer the source of the odor; or it can be used through a real-time monitoring system to continuously monitor the odor and chemical composition in the car and provide timely feedback.

[0099] The embodiment provided in this application provides data support for subsequent evaluation of the driver's health status by acquiring the driver's health data and odor information in the vehicle cabin.

[0100] In step 102, the driver's health status is analyzed and predicted through a virtual health model based on the health data and odor information to obtain a prediction result of the driver's health status, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained through training with health data and polymorphic twin odor data, and the polymorphic twin odor data is used to characterize the odor feature data of the driver in different health states.

[0101] The virtual health model can analyze and predict the driver's health status, and then obtain the driver's health status prediction result.

[0102] For example, the virtual health model can use historical health record data such as the driver's physical examination report as a reference, and use the odor data obtained by the on-board electronic nose as input to conduct real-time analysis of the driver's health status and predict key health indicators that affect driving safety, such as the driver's alcohol concentration and blood sugar level.

[0103] In some embodiments, image data including the driver's body shape, posture and facial information may also be acquired, and then a virtual geometric model of the driver may be generated based on the image data and the driver's basic personal information.

[0104] For example, the geometric shape and basic attributes (such as age, gender, etc.) of the driver's body can be defined based on the video scan of the vehicle camera and the relevant basic information of the driver, and a virtual person that can reflect the driver's shape and behavior in real time can be generated in the virtual space. Among them, image data including the driver's body shape, posture and facial information can be obtained through the vehicle camera.

[0105] In some embodiments, the driver's basic personal information (such as age, gender, weight, height, etc.) and health information (such as past medical history, drug use record, etc.) can be collected in advance. Afterwards, the image data collected by the on-board camera is used to accurately analyze the driver's posture, facial features, etc. using computer vision algorithms, such as convolutional neural networks (CNN) in deep learning, so as to generate a three-dimensional virtual model (driver twin) that is highly matched with the actual driver. This model can reflect the driver's body dynamics and facial expressions in real time and naturally carry various information collected about the driver.

[0106] Please refer to Figure 2 , Figure 2 This is a flowchart of a method for analyzing and predicting a driver's health status based on a virtual health model provided in some embodiments of the present application.

[0107] Among them, the virtual health model includes a health prediction and recognition model, and may also include a medical history judgment model or an alcohol detection model.

[0108] The medical history judgment model can determine whether the driver has a history of abnormal blood sugar based on the driver's personal information without the driver being aware of it.

[0109] The alcohol detection model can detect whether there is alcohol exhaled in the car without the driver's awareness. For example, the alcohol concentration in the cockpit air can be detected through smart sensors. The alcohol detection model includes the process of extracting key features from cabin odor, feature selection and alcohol detection.

[0110] Optionally, the health prediction and recognition model may include a blood sugar prediction model (blood sugar prediction module) and a drinking recognition model (drinking recognition module). These two models may work simultaneously and perform calculations separately.

[0111] In some embodiments, the odor information includes the gas information exhaled by the driver, and the health status of the driver includes the blood sugar information of the driver. The driver's abnormal blood sugar history information can be determined based on the health data; if the abnormal blood sugar history information indicates that the driver has a history of blood sugar disease, the driver's blood sugar is predicted according to the gas information exhaled by the driver through a blood sugar prediction model to obtain the driver's blood sugar information prediction result.

[0112] For example, it is possible to determine whether the driver has a history of abnormal blood sugar based on the driver's twin without the driver being aware of it. When the driver's twin indicates that the driver has a history of blood sugar, the driver can be prompted to actively blow into the air inlet of the electronic nose, and the blood sugar level in the driver's body can be predicted through the blood sugar prediction model.

[0113] Optionally, after the driver actively blows into the air inlet of the electronic nose, the key features in the gas information exhaled by the driver can be extracted through a first algorithm, wherein the first algorithm includes one or more of mean, standard deviation, kurtosis, skewness or fast Fourier transform; then, a first target feature vector is selected through the key features, and the first target feature vector is used to characterize a feature vector suitable for predicting blood sugar information; then, the first target feature vector is input into a blood sugar prediction model to obtain the driver's blood sugar information prediction result.

[0114] Exemplarily, the data of the gas sample exhaled by the driver can be collected by electronic devices including electronic noses. These data may include the concentration, flow rate, temperature, etc. of the gas components. After that, the collected data is preprocessed, including removing noise, filling missing values, normalization, etc., to ensure the quality and consistency of the data. The average value of the gas component concentration can be calculated by calculating the mean value of the preprocessed data to reflect the average level of the gas components exhaled by the driver; the standard deviation of the gas component concentration can be calculated by calculating the standard deviation to reflect the discreteness of the data, that is, the fluctuation of the gas component concentration; the kurtosis of the gas component concentration can be calculated by calculating the kurtosis to reflect the sharpness of the data distribution, that is, whether there are extreme values; the skewness of the gas component concentration can be calculated by calculating the skewness to reflect the asymmetry of the data distribution, that is, whether the data is tilted in a certain direction. Then, the time series data of the gas component concentration is subjected to a fast Fourier transform (FFT) to convert the time domain signal into a frequency domain signal. FFT can reveal the periodicity and frequency characteristics of the gas component concentration over time. The key frequency components are extracted from the FFT results, which may reflect the specific pattern or periodic changes of the gas components exhaled by the driver. Analyze the amplitude and phase information of the FFT results, which may reveal the intensity and starting point of the change in gas component concentration.

[0115] Afterwards, the most valuable feature (first target feature vector) for judging the driver's state is selected according to information such as the importance and relevance of the feature. The first target feature vector can be selected by a particle swarm algorithm.

[0116] The selected features are used to train a machine learning model, such as K-nearest neighbor (KNN), support vector machine (SVM), or random forest (RF), to predict the driver's blood glucose information.

[0117] Optionally, the performance of the model can be evaluated through methods such as cross-validation and test set evaluation to ensure that the model can accurately and reliably extract key features.

[0118] Through the above steps, key features can be extracted from the driver's exhaled gas information, which can be used to evaluate the driver's status, predict his behavior or detect potential safety risks.

[0119] In some embodiments, a blood glucose prediction model can be trained using a random forest algorithm based on historical blood glucose level records and polymorphic twin odor data, wherein the polymorphic twin odor data includes characteristic data of acetone concentration in the driver's breath; and the blood glucose prediction model is updated based on the most recently acquired driver's health data and polymorphic twin odor data.

[0120] For example, the blood sugar prediction model can be trained using the driver's physiological data (such as acetone concentration in breath) and historical blood sugar level records, and the random forest algorithm can be used to effectively distinguish between normal and abnormal blood sugar levels. The blood sugar prediction model can self-adjust in combination with newly collected data, and is continuously updated and iterated to ensure the accuracy and real-time nature of the evaluation results.

[0121] In some embodiments, the odor information includes the gas information in the cockpit and the gas information exhaled by the driver, and the driver's health status includes the driver's drinking information; the alcohol content can be detected based on the gas information in the cockpit to obtain alcohol content indication information; if the alcohol content indication information indicates that the gas in the cockpit contains alcohol, the health prediction recognition model is used to identify whether the driver has been drinking based on the gas information exhaled by the driver to obtain the driver's drinking information recognition result.

[0122] For example, it is possible to detect whether there is breath exhaled after drinking in the car without the driver noticing (for example, the alcohol concentration in the air in the cockpit can be detected by a smart sensor). When breath exhaled after drinking is detected in the car, the driver can be prompted to actively blow into the air inlet of the electronic nose, and the drinking recognition model can be used to identify whether the driver has been drinking or is drunk.

[0123] Optionally, when exhaled gas after drinking is detected in the car, in order to ensure that it is the driver who blows into the air inlet of the electronic nose, a video scan can be performed through the on-board camera to monitor the entire exhalation process and compare the scan results with the driver's twin.

[0124] If the driver does not cooperate with the breathalyzer, the vehicle can be actively intervened. For example, if the vehicle is in motion, the vehicle can be controlled to actively slow down to a safe stop until the alcohol recognition model recognizes that the driver is not drunk, and then the vehicle control is returned to the driver; if the vehicle does not start, the vehicle can be controlled not to start until the alcohol recognition model recognizes that the driver is not drunk, and then the vehicle can be started normally.

[0125] Optionally, after the driver actively blows into the air inlet of the electronic nose, a second algorithm can be used to extract key features in the gas information exhaled by the driver, wherein the second algorithm includes one or more of maximum value, integral value, wavelet transform or fast Fourier transform; a second target feature vector is selected based on the key features, and the second target feature vector is used to characterize a feature vector suitable for identifying the alcohol content in the gas; then, the second target feature vector is input into a drinking recognition model to obtain the driver's drinking information recognition result.

[0126] Exemplarily, a sample of the driver's exhaled gas can be collected by an electronic nose or a breath alcohol detector. The alcohol concentration in the sample is converted into a digital signal for subsequent processing. Afterwards, the maximum value in the signal is calculated, which can reflect the peak value of the alcohol concentration in the driver's exhaled gas, wherein the maximum value can be used as an important indicator for judging whether the driver is drinking. The integral value of the signal is calculated by the integral value, which can reflect the overall level of the alcohol concentration in the driver's exhaled gas, and the integral value can provide information about the driver's drinking amount. The signal is subjected to a wavelet transform to decompose the signal into components of different frequencies, and the wavelet transform can help identify local features in the signal, such as sudden changes in alcohol concentration. By analyzing the results of the wavelet transform, key features related to the driver's drinking can be extracted. The signal is subjected to a fast Fourier transform to convert the signal from the time domain to the frequency domain, and FFT can help analyze the frequency components of the signal, thereby identifying specific frequencies related to the alcohol concentration, and by analyzing the results of the FFT, key features related to the driver's drinking can be extracted. According to the results of the maximum value, integral value, wavelet transform, and FFT, the features most related to the driver's drinking are selected. These features can be used to build a classification model to determine whether the driver is drinking.

[0127] Afterwards, the feature most relevant to the driver's drinking (the second target feature vector) is selected based on the importance and relevance of the feature. The second target feature vector can be selected by a particle swarm algorithm.

[0128] Use the selected features to build a classification model, such as support vector machine (SVM), decision tree, random forest or neural network. By training the model, accurate judgment of the driver's drinking status can be achieved.

[0129] Through the above steps, key features can be effectively extracted from the driver's exhaled gas information and used to determine whether the driver is drinking.

[0130] In some embodiments, a drinking recognition model can be trained using a support vector machine algorithm based on historical drinking records and polymorphic twin odor data, wherein the polymorphic twin odor data includes characteristic data of ethanol concentration in a vehicle cabin; and the drinking recognition model is updated based on the most recently acquired driver's health data and polymorphic twin odor data.

[0131] Optionally, a polymorphic twin odor library can be established to store polymorphic twin odor data. The polymorphic twin odor library is a database for recording and analyzing the odor characteristics of drivers in different health states, and may include the in-car odor characteristic data of drivers in different health states. The odor characteristic database may contain real odor data, and may also generate virtual odor data in different health states through generative adversarial network (GAN) technology to enrich the sample types and quantities for model training. When the driver's health status changes, slight changes in the odor in the car will be recorded and stored in the odor library, and the system uses this data to train and optimize the prediction model. Among them, the polymorphic twin odor library has the ability to be dynamically updated, and as the system continues to collect new odor data, the data in the library can be continuously added and updated, thereby improving the coverage and accuracy of the data. This dynamic update mechanism ensures that the evaluation model can become more accurate and reliable over time, providing more accurate health status assessment results.

[0132] For example, the drinking recognition model can be trained based on the driver's historical drinking record and the in-car odor data (such as ethanol concentration), using the support vector machine (SVM) algorithm, and improving the prediction accuracy through multiple iterations of optimization. The drinking recognition model can self-adjust in combination with newly collected data, and continuously update and iterate to ensure the accuracy and real-time nature of the evaluation results.

[0133] The embodiment of the present application analyzes and predicts the driver's health status through a virtual health model based on health data and odor information, and obtains the driver's health status prediction result, which can identify potential health risks in advance, such as drunkenness, drug effects, etc., thereby reducing the occurrence of traffic accidents; it can also ensure that the driver can better concentrate, improve efficiency, and reduce misoperations and delays caused by health problems; by understanding the prediction results of their own health status, the driver can also enhance health awareness and pay more attention to their physical condition.

[0134] In step 103, when the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving status of the vehicle is actively intervened.

[0135] For example, when it is determined that the driver's health status is abnormal, a reminder alarm can be issued, or the vehicle speed can be reduced and stopped safely. It is also possible to limit the vehicle speed, issue an alarm, and instantly notify emergency contacts through the Internet of Vehicles system to proactively prevent potential safety risks.

[0136] The embodiments provided in the present application can issue a warning message and / or actively intervene in the driving status of the vehicle when the health status prediction result shows that the driver's health status is abnormal. The warning message and active intervention can promptly remind the driver to pay attention to his or her health status and avoid driving errors caused by health problems, thereby reducing the occurrence of traffic accidents. The warning message and active intervention can also help the driver take timely measures when the health status is abnormal, such as stopping to rest, seeking medical help, etc., to protect the driver's life safety, and can also protect the safety of passengers and pedestrians, and reduce the harm to them caused by traffic accidents due to the driver's health problems.

[0137] The following introduces and illustrates the solution of the embodiment of the present invention in conjunction with specific application scenarios.

[0138] When the driver is driving the vehicle, the driver's digital twin can be established based on the driver's facial information and posture information collected by the on-board camera, as well as the driver's basic information (including health information) collected legally; and the driver's digital twin (driver twin) can be used to determine whether the driver has a history of abnormal blood sugar without the driver's awareness. When the driver's twin indicates a history of blood sugar, the driver is reminded to actively blow into the air inlet of the electronic nose to predict his blood sugar level. Simultaneously, it can be performed to detect whether there is exhaled gas after drinking in the car without the driver's awareness. When exhaled gas after drinking is detected in the car, the driver is reminded to actively blow into the air inlet of the electronic nose to determine whether the driver is drinking. When it is predicted that the driver's blood sugar level exceeds the preset value, or the driver is identified to be drinking, the vehicle speed is limited, an alarm is sounded, or the emergency contact is notified immediately through the vehicle networking system to proactively prevent potential safety risks.

[0139] The above is an introduction to an embodiment of the driver health status assessment method based on digital twin provided in this application.

[0140] The implementation scheme of the driver health status assessment system based on digital twin provided in the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0141] For the driver health status assessment method based on digital twin provided in the above embodiment, the embodiment of the present application also provides a driver health status assessment system based on digital twin, which is used to implement the above method, such as Figure 3 As shown, Figure 3 It is a module schematic block diagram of a driver health status assessment system based on digital twins provided in some embodiments of the present application. The driver health status assessment system based on digital twins includes:

[0142] Physical layer: The physical layer is used to collect and manage the driver's basic personal information, health information, and environmental information.

[0143] For example, the physical entity layer is responsible for collecting and managing the driver's basic attributes (such as age, gender, weight, height, etc.), health attributes (such as past medical history, drug use records, etc.) and environmental information (such as in-vehicle temperature, humidity, air quality, etc.). Through devices such as body temperature sensors, pulse sensors, cameras, and on-board electronic noses, the data acquisition system transmits this information to the on-board computing unit via the CAN bus in real time, and stores it in the database of the physical entity layer, while synchronizing it with historical health data to support real-time analysis and updates of the twin model layer.

[0144] Among them, the electronic nose system consists of multiple sensitive gas sensors that can detect and quantify various gas components in the air in the car (such as ethanol, acetone, carbon dioxide, etc.). These sensors are connected to the central processing unit (CPU) via the SPI bus, and the analog signals output by the sensors are converted into digital signals through the analog-to-digital converter (ADC) and transmitted to the CPU for processing. The system adopts a multi-stage signal processing algorithm, including signal filtering, feature extraction and pattern recognition, and uses feature extraction algorithms such as fast Fourier transform (FFT) to analyze the spectral characteristics of gas signals and extract key features related to the driver's health status. After processing, the data is passed to the twin model layer through the data bus for real-time evaluation of the health status.

[0145] The data acquisition system (which can also be a non-contact odor detection system) monitors the driver's behavior in real time through the on-board camera, such as head movement and breathing rate, and combines the data from the odor sensor to infer the driver's health status. To ensure the accuracy of the detection, the system needs to go through a series of calibration steps when it is used for the first time, including zero point calibration and sensitivity adjustment of the odor sensor. The driver takes a deep breath under specific conditions, and the system stores these results as baseline data for subsequent health status assessment.

[0146] The twin model layer includes a blood sugar prediction model and a drinking recognition model, which are used to predict and identify the driver's health status.

[0147] Exemplarily, the twin model layer evaluates the driver's health status through two main machine learning models, namely the blood glucose prediction model and the drinking recognition model. The drinking recognition model is trained based on the driver's historical drinking records and in-car odor data (such as ethanol concentration), using the support vector machine (SVM) algorithm to improve the prediction accuracy through multiple iterative optimizations. The blood glucose prediction model is trained using the driver's physiological data (such as acetone concentration in breath) and historical blood glucose level records, using the random forest algorithm to effectively distinguish between normal and abnormal blood glucose levels. The models in the twin model layer will be continuously updated and iterated, and self-adjusted in combination with newly collected data to ensure the accuracy and real-time nature of the evaluation results.

[0148] Twin data layer,The twin data layer is used to store and manage the driver’s historical health data and current environment data.

[0149] Exemplarily, the twin data layer is responsible for storing and managing the driver's historical health data and current environmental data, relying on a high-performance database system to achieve fast data access and query.

[0150] Application service layer, the application service layer includes virtual visualization module, health status assessment module and intelligent driving control module. Among them, the virtual visualization module is used to display the driver's health status and in-vehicle environment information in a graphical form on the display screen; the health status assessment module is used to analyze the predicted recognition results of the driver's health status and assess the driver's health status; the intelligent driving control module is used to issue warning information and / or actively intervene in the vehicle's driving status based on the assessment results of the driver's health status.

[0151] Exemplarily, the virtual visualization module is responsible for displaying the driver's health status and in-vehicle environment information in a graphical form on the vehicle display screen, providing an intuitive display of health conditions, such as real-time blood sugar level curves and alcohol concentration charts. The health status assessment module performs real-time analysis of the model results provided by the twin model layer to accurately assess the driver's current health status. The intelligent driving control module automatically adjusts the vehicle's operation based on the health status assessment results, such as limiting the vehicle speed, sounding an alarm, or instantly notifying emergency contacts through the vehicle networking system when health abnormalities are detected, in order to proactively prevent potential safety risks.

[0152] Data interaction layer: The data interaction layer transmits and processes data through the data transmission protocol, sends the data to the twin model layer for predicting and identifying the driver's health status, and feeds back the results of health status prediction and identification to the application service layer.

[0153] Exemplarily, the data interaction layer transmits and processes data through a standardized data transmission protocol (such as MQTT or HTTP), sends the data to the twin model layer for analysis, and feeds back the evaluation results to the application service layer in real time.

[0154] The driver health status assessment system based on digital twins provided in the embodiment of the present application can fuse multi-source data (including odor data, image data, physiological data, etc.), and use multimodal data fusion technology to uniformly process these data within the machine learning framework. This fusion method can significantly improve the comprehensiveness and accuracy of health status assessment, especially in complex driving environments, and can more effectively guarantee the reliability of the assessment results. In order to achieve more accurate personalized health assessment, the assessment model uses advanced machine learning algorithms, such as particle swarm optimization (PSO) combined with support vector machine (SVM) for feature selection and classification. The particle swarm optimization algorithm can quickly search for the optimal feature set, while the support vector machine accurately distinguishes different health states during the classification process. In this way, the model can quickly adapt to the individual differences of different drivers, thereby providing a highly personalized health status assessment to ensure driving safety and efficient operation of the system.

[0155] The implementation scheme of the driver health status assessment device based on digital twin provided in the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0156] For the driver health status assessment method based on digital twin provided in the above embodiment, the embodiment of the present application also provides a driver health status assessment device based on digital twin, which is used to implement the above method, such as Figure 4 As shown, Figure 4 This is a module schematic block diagram of a driver health status assessment device based on digital twins according to an embodiment of the present application. The driver health status assessment device 400 based on digital twins includes:

[0157] An acquisition module 401 is used to acquire the driver's health data and the odor information in the vehicle cabin;

[0158] An analysis and prediction module 402 is used to analyze and predict the health status of the driver through a virtual health model based on the health data and the odor information, and obtain a prediction result of the health status of the driver, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained by training the health data and the polymorphic twin odor data, and the polymorphic twin odor data is used to characterize the odor feature data of the driver in different health states;

[0159] The control module 403 is used to issue a warning message and / or actively intervene in the driving state of the vehicle when the health state prediction result shows that the driver's health state is abnormal.

[0160] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0161] like Figure 5 As shown, the embodiment of the present application further provides an electronic device 500, the electronic device 500 includes a memory 501, one or more processors 502 ( Figure 5 Only one is shown) and a computer program stored in the memory 501 and executable on the processor 502. Among them: the memory 501 is used to store software programs and units, and the processor 502 executes various functional applications and data processing by running the software programs and units stored in the memory to obtain the resources corresponding to the above-mentioned preset events. Optionally, the processor 502 implements the above-mentioned driver health status assessment method based on digital twins by running the above-mentioned computer program stored in the memory.

[0162] The memory 501, as a non-transitory computer-readable medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory 501 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 501 may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor 502 via a network.

[0163] It can be understood that the contents of the above method embodiments are all applicable to the electronic device embodiment, the functions specifically implemented by the electronic device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0164] The embodiment of the present application also provides a vehicle, which includes the above-mentioned driver health status assessment device based on digital twin, the system or the electric drive assembly of the above-mentioned electronic device. Specifically, the vehicle can be a private car, such as a sedan, SUV, MPV or pickup truck. The vehicle can also be an operating vehicle, such as a van, bus, small truck or large trailer. The vehicle can be a fuel vehicle or a new energy vehicle. When the vehicle is a new energy vehicle, it can be a hybrid vehicle or a pure electric vehicle.

[0165] It can be understood that the contents of the above method embodiments are all applicable to the present vehicle embodiment, the functions specifically implemented by the present vehicle embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0166] An embodiment of the present application also provides a computer program product, which includes a computer program. When the computer program is executed by one or more processors, it can implement the steps of the above-mentioned driver health status assessment method based on digital twins.

[0167] It can be understood that the contents of the above method embodiments are all applicable to the present computer program product, the functions specifically implemented by the present computer program product embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0168] The driver health status assessment method, device, system, electronic device, vehicle and computer program product based on digital twin provided by the embodiment of the present application are designed for the specific complex environment of the car cockpit, and a multi-level driver health status assessment system based on digital twin is designed to make up for the limitation that the traditional health monitoring framework is mainly applied to large medical systems, and realize the dynamic monitoring and assessment of the driver's health status. Secondly, it breaks through the disadvantage that traditional medical equipment must be in contact with the driver's body to obtain physiological signals. A non-contact health data acquisition device based on an electronic nose is used to collect odor information in the car in real time and use it for health assessment without interfering with the driver. Finally, by constructing and optimizing a high-precision digital twin model and combining it with advanced machine learning algorithms, an accurate prediction of the driver's health status is achieved, and intelligent driving control suggestions can be provided based on the evaluation results. This technical solution not only improves the real-time and accuracy of the driver's health status assessment, but also can quickly issue an early warning when the health is abnormal, thereby effectively preventing the occurrence of traffic accidents, and has significant social and economic benefits.

[0169] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0170] Although specific embodiments are described herein, those of ordinary skill in the art will recognize that many other modifications or alternative embodiments are also within the scope of the present disclosure. For example, any of the functions and / or processing capabilities described in conjunction with a particular device or component may be performed by any other device or component. In addition, although various exemplary implementations and architectures have been described according to embodiments of the present disclosure, those of ordinary skill in the art will recognize that many other modifications to the exemplary implementations and architectures described herein are also within the scope of the present disclosure.

[0171] Some aspects of the present disclosure are described above with reference to the block diagram and flow chart of the system, method, system and / or computer program product according to the exemplary embodiments.It should be understood that the combination of one or more blocks in the block diagram and the flow chart and the blocks in the block diagram and the flow chart can be realized by executing computer executable program instructions respectively.Equally, according to some embodiments, some blocks in the block diagram and the flow chart may not need to be executed in the order shown, or may not need to be executed in full.In addition, additional components and / or operations beyond those components and / or operations shown in the blocks in the block diagram and the flow chart may be present in certain embodiments.

[0172] Therefore, the blocks in the block diagrams and flow charts support the combination of devices for performing the specified functions, the combination of elements or steps for performing the specified functions, and the program instruction devices for performing the specified functions. It should also be understood that each block in the block diagrams and flow charts and the combination of blocks in the block diagrams and flow charts can be implemented by a special-purpose hardware computer system or a combination of special-purpose hardware and computer instructions that performs a specific function, element or step.

[0173] The program modules, applications, etc. described herein may include one or more software components, including, for example, software objects, methods, data structures, etc. Each such software component may include computer-executable instructions that, in response to execution, cause at least a portion of the functionality described herein (e.g., one or more operations of the illustrative methods described herein) to be performed.

[0174] Software components can be encoded with any of various programming languages. An exemplary programming language can be a low-level programming language, such as an assembly language associated with a specific hardware architecture and / or an operating system platform. The software component comprising assembly language instructions may need to be converted to executable machine code by an assembler before being executed by hardware architecture and / or platform. Another exemplary programming language can be a higher-level programming language, which can be transplanted across multiple architectures. The software component comprising a higher-level programming language may need to be converted to an intermediate representation by an interpreter or a compiler before execution. Other examples of programming languages ​​include but are not limited to macro languages, shells or command languages, job control languages, scripting languages, database queries or search languages ​​or report writing languages. In one or more exemplary embodiments, the software component comprising the instruction of one of the above-mentioned programming language examples can be directly executed by an operating system or other software components, without first being converted into another form.

[0175] Software components may be stored as files or other data storage structures. Software components of similar type or related functions may be stored together, such as in a specific directory, folder, or library. Software components may be static (e.g., preset or fixed) or dynamic (e.g., created or modified at execution time).

[0176] The embodiments of the present application are described in detail above in conjunction with the accompanying drawings, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the relevant technical field without departing from the purpose of the present application.

Claims

1. A driver health status assessment method based on digital twins, characterized in that: The following steps are involved: Obtain driver health data and vehicle cabin odor information; Based on the health data and the odor information, the health status of the driver is analyzed and predicted through a virtual health model to obtain a prediction result of the driver's health status, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained by training the health data and polymorphic twin odor data, and the polymorphic twin odor data is odor feature data used to characterize the driver in different health states; When the health status prediction result shows that the driver's health status is abnormal, a warning message is issued and / or the driving status of the vehicle is actively intervened.

2. The driver health status assessment method based on digital twin according to claim 1 is characterized in that: After obtaining the driver's health data and the odor information in the vehicle cabin, the method further includes: Acquiring image data including the driver's body shape, posture and facial information; A virtual geometric model of the driver is generated based on the image data and basic personal information of the driver.

3. The driver health status assessment method based on digital twin according to claim 1 is characterized in that: The odor information includes gas information exhaled by the driver, and the driver's health status includes blood sugar information of the driver; The analyzing and predicting the health status of the driver through a virtual health model based on the health data and the odor information to obtain a prediction result of the health status of the driver includes: determining the driver's abnormal blood sugar history information based on the health data; If the abnormal blood sugar history information indicates that the driver has a history of blood sugar, the driver's blood sugar is predicted according to the driver's exhaled gas information through the health prediction recognition model to obtain the driver's blood sugar information prediction result.

4. The driver health status assessment method based on digital twin according to claim 3 is characterized in that: The health prediction and recognition model includes a blood sugar prediction model, and the blood sugar of the driver is predicted according to the gas information exhaled by the driver by the health prediction and recognition model to obtain the blood sugar information prediction result of the driver, including: Extracting a first key feature from the breath information of the driver by using a first algorithm, wherein the first algorithm includes one or more of a mean, a standard deviation, a kurtosis, a skewness, or a fast Fourier transform; Selecting a first target feature vector through the first key feature, where the first target feature vector is used to represent a feature vector suitable for predicting blood sugar information; The first target feature vector is input into the blood glucose prediction model to obtain the blood glucose information prediction result of the driver.

5. The driver health status assessment method based on digital twin according to claim 1 is characterized in that: The odor information includes gas information in the cockpit and gas information exhaled by the driver, and the driver's health status includes the driver's drinking information; The analyzing and predicting the health status of the driver through a virtual health model based on the health data and the odor information to obtain a prediction result of the health status of the driver includes: Performing alcohol content detection based on the gas information in the cockpit to obtain alcohol content indication information; If the alcohol content indication information indicates that the gas in the cockpit contains alcohol, the health prediction and recognition model is used to identify whether the driver has been drinking alcohol based on the gas information exhaled by the driver to obtain the driver's drinking information recognition result.

6. The driver health status assessment method based on digital twin according to claim 5 is characterized in that: The health prediction and recognition model includes a drinking recognition model, and the health prediction and recognition model is used to recognize whether the driver drinks alcohol according to the gas information exhaled by the driver, and the drinking information recognition result of the driver is obtained, including: Extracting a second key feature from the breath information of the driver by a second algorithm, wherein the second algorithm includes one or more of a maximum value, an integral value, a wavelet transform, or a fast Fourier transform; Selecting a second target feature vector through the second key feature, where the second target feature vector is used to represent a feature vector suitable for identifying alcohol content in gas; The second target feature vector is input into the drinking recognition model to obtain the drinking information recognition result of the driver.

7. The driver health status assessment method based on digital twin according to claim 1 is characterized in that: The health data includes the driver's historical blood glucose level records, and the health prediction recognition model includes a blood glucose prediction model; The blood sugar prediction model is trained and updated through the following steps: training the blood glucose prediction model using a random forest algorithm according to the historical blood glucose level records and the polymorphic twin odor data, wherein the polymorphic twin odor data includes characteristic data of acetone concentration in the driver's breath; The blood glucose prediction model is updated according to the latest acquired health data of the driver and the polymorphic twin odor data.

8. The driver health status assessment method based on digital twin according to claim 1 is characterized in that: The health data includes the driver's historical drinking record, and the health prediction recognition model includes a drinking recognition model; The drinking recognition model is trained and updated through the following steps: Training the drinking recognition model using a support vector machine algorithm based on the historical drinking records and the polymorphic twin odor data, wherein the polymorphic twin odor data includes ethanol concentration characteristic data in a vehicle cabin; The drinking recognition model is updated according to the latest acquired health data of the driver and the polymorphic twin odor data.

9. A driver health status assessment system based on digital twins, characterized in that: The system comprises: A physical layer, which is used to collect and manage the driver's basic personal information, health information, and environmental information; A twin model layer, wherein the twin model layer includes a blood sugar prediction model and a drinking recognition model, and is used to predict and recognize the health status of the driver; A twin data layer, wherein the twin data layer is used to store and manage the driver's historical health data and current environment data; An application service layer, the application service layer comprising a virtual visualization module, a health status assessment module and an intelligent driving control module; The virtual visualization module is used to display the driver's health status and in-vehicle environment information on the display screen in a graphical form; The health status assessment module is used to analyze the prediction and recognition results of the driver's health status and assess the driver's health status; The intelligent driving control module is used to issue warning information and / or actively intervene in the driving state of the vehicle according to the evaluation result of the driver's health status; The data interaction layer transmits and processes data through a data transmission protocol, sends the data to the twin model layer for predicting and identifying the driver's health status, and feeds back the results of the health status prediction and identification to the application service layer.

10. A driver health status assessment device based on digital twin, characterized in that: The device comprises: An acquisition module, used to obtain the driver's health data and the odor information in the vehicle cabin; An analysis and prediction module, configured to analyze and predict the health status of the driver through a virtual health model based on the health data and the odor information, and obtain a prediction result of the driver's health status, wherein the virtual health model includes a health prediction recognition model, and the health prediction recognition model is obtained by training the health data and polymorphic twin odor data, and the polymorphic twin odor data is odor feature data used to characterize the driver in different health states; The control module is used to issue a warning message and / or actively intervene in the driving state of the vehicle when the health state prediction result shows that the driver's health state is abnormal.

11. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the driver health status assessment method based on digital twins as described in any one of claims 1 to 8 when executing the computer program.

12. A vehicle, characterized in that: The vehicle includes the digital twin-based driver health status assessment system as described in claim 9, the digital twin-based driver health status assessment device as described in claim 10, or the electronic device as described in claim 11.

Images