Vehicle control method, device and equipment, readable storage medium and program product

Abstract

The invention relates to a vehicle control method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring first physiological information and second physiological information of a user in a target vehicle; the first physiological information is acquired by using a non-contact device arranged in the target vehicle, and the second physiological information is acquired by using a contact device; determining the sleep state of the user according to the first physiological information and the second physiological information; and determining a vehicle-mounted equipment control strategy matched with the sleep state from a plurality of vehicle-mounted equipment control strategies in a preset strategy library, and controlling working parameters of vehicle-mounted equipment in the target vehicle based on the vehicle-mounted equipment control strategy. By adopting the method, the comfort of the vehicle-mounted environment can be improved under the scene that the user sleeps in the vehicle.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a vehicle control method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] As automotive scenarios become more diversified, users are sleeping in their cars more frequently, such as short breaks during long drives or outdoor camping, which places higher demands on the comfort of the in-vehicle environment.

[0003] Therefore, there is an urgent need for a vehicle control method that can improve the comfort of the in-vehicle environment when users are sleeping in the car. Summary of the Invention

[0004] Therefore, it is necessary to provide a vehicle control method, device, computer equipment, computer-readable storage medium, and computer program product that can improve the comfort of the in-vehicle environment in scenarios where users sleep in the vehicle, in order to address the above-mentioned technical problems.

[0005] In a first aspect, this application provides a vehicle control method, including:

[0006] Acquire the first and second physiological information of the user inside the target vehicle; the first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device.

[0007] The user's sleep state is determined based on the first and second physiological information;

[0008] From multiple vehicle device control strategies in the preset strategy library, determine the vehicle device control strategy that matches the sleep state, and control the operating parameters of the vehicle device in the target vehicle based on the vehicle device control strategy.

[0009] In one embodiment, determining a user's sleep state based on first physiological information and second physiological information includes: determining the first information quality of the first physiological information and the second information quality of the second physiological information; determining the user's target physiological characteristics based on the first information quality, the second information quality, the first physiological information, and the second physiological information; and inputting the target physiological characteristics into a sleep analysis model to obtain the user's sleep state output by the sleep analysis model.

[0010] In one embodiment, determining a user's target physiological characteristics based on first information quality, second information quality, first physiological information, and second physiological information includes: performing feature extraction processing on the first physiological information to obtain target physiological characteristics when the first information quality meets preset conditions and the second information quality does not meet preset conditions; performing feature extraction on the second physiological information to obtain target physiological characteristics when the second information quality meets preset conditions and the first information quality does not meet preset conditions; and comparing the first information quality and the second information quality when both meet preset conditions, and determining the user's target physiological characteristics based on the comparison result, the first physiological information, and the second physiological information.

[0011] In one embodiment, feature extraction processing is performed on the first physiological information to obtain target physiological features, including: determining a first feature based on the first physiological information using a time-domain feature extraction algorithm; determining a second feature based on the first physiological information using a frequency-domain feature extraction algorithm; determining a third feature based on the first physiological information using a nonlinear feature extraction algorithm; inputting the first physiological information into a convolutional neural network model to obtain a fourth feature output by the convolutional neural network model; and concatenating the first feature, second feature, third feature, and fourth feature to obtain the target physiological features.

[0012] In one embodiment, determining a user's target physiological characteristics based on a comparison result, first physiological information, and second physiological information includes: if the comparison result indicates that the quality of the first information is greater than the quality of the second information, then calibrating the second physiological information using the first physiological information, and extracting features from the first physiological information and the calibrated second physiological information to obtain the target physiological characteristics; if the comparison result indicates that the quality of the first information is less than the quality of the second information, then calibrating the first physiological information using the second physiological information, and extracting features from the second physiological information and the calibrated first physiological information to obtain the target physiological characteristics; if the comparison result indicates that the quality of the first information is equal to the quality of the second information, then extracting features from the first physiological information and the second physiological information to obtain the target physiological characteristics.

[0013] In one embodiment, calibrating the first physiological information using the second physiological information includes: determining a first target feature from the first physiological information and determining a second target feature from the second physiological information, wherein the feature types and acquisition time windows of the first and second target features are the same; inputting the first and second target features into a pre-built regression model so that the regression model calibrates the first target feature using the second target feature to obtain the calibrated first target feature; and calibrating other features in the first physiological information other than the first target feature based on the calibrated first target feature.

[0014] In one embodiment, controlling the operating parameters of the on-board equipment in the target vehicle based on the on-board equipment control strategy includes: acquiring environmental parameters of the target vehicle; and controlling the operating parameters of the on-board equipment in the target vehicle based on the environmental parameters and the on-board equipment control strategy.

[0015] In one embodiment, the method further includes: acquiring the user's sleep information after the user's sleep ends; and optimizing the control strategy of the on-board equipment of the target vehicle based on the sleep information.

[0016] Secondly, this application also provides a vehicle control device, comprising:

[0017] The acquisition module is used to acquire the first physiological information and the second physiological information of the user inside the target vehicle; the first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device.

[0018] The determination module is used to determine the user's sleep state based on the first physiological information and the second physiological information;

[0019] The execution module is used to determine the vehicle device control strategy that matches the sleep state from multiple vehicle device control strategies in the preset strategy library, and control the operating parameters of the vehicle device in the target vehicle based on the vehicle device control strategy.

[0020] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the embodiments of the first aspect above.

[0021] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the embodiments of the first aspect above.

[0022] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the embodiments of the first aspect above.

[0023] The aforementioned vehicle control method, device, computer equipment, computer-readable storage medium, and computer program product acquire first and second physiological information of a user inside a target vehicle. The first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device. Then, the user's sleep state is determined based on the first and second physiological information. Next, a vehicle device control strategy matching the sleep state is determined from multiple vehicle device control strategies in a preset strategy library. Finally, the operating parameters of the vehicle devices inside the target vehicle are controlled based on the vehicle device control strategy. The vehicle control method provided in this application acquires the user's physiological information through non-contact and contact devices installed inside the target vehicle, and then controls the vehicle devices inside the target vehicle based on the user's physiological information, effectively improving the comfort of the in-vehicle environment when the user sleeps inside the vehicle. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating a vehicle control method in one embodiment;

[0026] Figure 2 This is a flowchart illustrating a method for determining a user's sleep state in one embodiment;

[0027] Figure 3 This is a flowchart illustrating a method for determining a user's target physiological characteristics in one embodiment.

[0028] Figure 4 This is a flowchart illustrating a method for obtaining target physiological characteristics in one embodiment;

[0029] Figure 5 This is a flowchart illustrating a method for determining a user's target physiological characteristics in another embodiment;

[0030] Figure 6 This is a flowchart illustrating a method for calibrating first physiological information in one embodiment;

[0031] Figure 7 This is a flowchart illustrating a method for determining a control strategy for an on-board device in one embodiment.

[0032] Figure 8 This is a flowchart illustrating a method for handling a user after they have finished sleeping, as described in one embodiment.

[0033] Figure 9 This is a flowchart illustrating the vehicle control method in another embodiment;

[0034] Figure 10 This is a structural block diagram of a vehicle control device in one embodiment;

[0035] Figure 11 This is an internal structural diagram of a computer device in one embodiment;

[0036] Figure 12 This is a diagram of the internal structure of a computer device in another embodiment. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0038] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0039] As automotive scenarios become more diversified, users are sleeping in their cars more frequently, such as short breaks during long drives or outdoor camping, which places higher demands on the comfort of the in-vehicle environment.

[0040] Therefore, there is an urgent need for a vehicle control method that can improve the comfort of the in-vehicle environment when users are sleeping in the car.

[0041] In view of this, this application provides a vehicle control method that acquires first and second physiological information of a user inside a target vehicle. The first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device. Then, the user's sleep state is determined based on the first and second physiological information. Next, a vehicle device control strategy matching the sleep state is determined from multiple vehicle device control strategies in a preset strategy library. Finally, the operating parameters of the vehicle devices inside the target vehicle are controlled based on the vehicle device control strategy. The vehicle control method provided by this application acquires the user's physiological information through non-contact and contact devices installed inside the target vehicle, and then controls the vehicle devices inside the target vehicle based on the user's physiological information, effectively improving the comfort of the in-vehicle environment when the user sleeps in the vehicle.

[0042] The vehicle control method provided in this application can be executed by a computer device, which can be a terminal, such as the electronic control unit of the target vehicle or an in-vehicle intelligent computing platform. The computer device can also be a server, which can communicate with the target vehicle via a network, obtain the user's first and second physiological information through the network, and send the determined control strategy of the in-vehicle equipment to the target vehicle through the network.

[0043] In one exemplary embodiment, such as Figure 1 As shown, a vehicle control method is provided, which includes the following steps:

[0044] Step 101: Obtain the first and second physiological information of the user inside the target vehicle.

[0045] The first physiological information was collected using a non-contact device installed inside the target vehicle, while the second physiological information was collected using a contact device.

[0046] Optionally, a contactless device refers to a device that can collect a user's physiological information without direct physical contact with the user inside the target vehicle. Examples of contactless devices include cameras in in-vehicle driver monitoring systems, in-vehicle infrared thermal imagers, millimeter-wave radar, and in-vehicle microphones.

[0047] Furthermore, the primary physiological information can also be obtained by analyzing the user's facial video stream captured by the camera of the in-vehicle driver monitoring system, such as heart rate and respiratory rate. Specifically, the camera of the driver monitoring system can capture the user's facial video stream in real time. The computer equipment can perform face detection and alignment, region of interest (ROI) cropping, illumination equalization, and noise filtering preprocessing on the facial video stream. Then, based on the remote photoplethysmography algorithm, the temporal mean of the RGB three-channel pixel values ​​of the ROI is extracted to obtain the raw signal related to blood flow. Then, the effective physiological signal is separated by principal component analysis (PCA) or independent component analysis (ICA) algorithms to remove interference such as muscle movement and head shaking. The effective physiological signal is then subjected to frequency domain analysis to extract the main frequency and convert it into heart rate. At the same time, the computer equipment can also analyze the small fluctuations of the user's head with breathing using optical flow to obtain the respiratory rate. Finally, the heart rate and respiratory rate are determined as the primary physiological information.

[0048] Contact devices refer to devices that come into direct contact with the body of a user inside a target vehicle to collect the user's physiological information. Examples of contact devices include wearable devices, pressure sensors and heart rate sensors installed on seats inside the target vehicle, photoelectric heart rate sensors and skin conductivity sensors integrated into the steering wheel grip, blood pressure monitoring cuff assemblies installed in door armrests, and motion monitoring sensors embedded inside seat belts.

[0049] Furthermore, the second physiological information can be: real-time heart rate value collected by wearable devices; posture data collected by seat sensors, specifically the pressure distribution data of the user's buttocks and back in contact with the seat collected by seat pressure sensors, and posture-related parameters collected by seat back angle sensors and seat cushion tilt sensors. This posture data can be used to characterize whether the user's posture is upright, whether the body is leaning forward / backward, and whether the lumbar support is adequate; real-time heart rate value collected by photoelectric heart rate sensors; emotional stress index collected by skin conductivity sensors. Specifically, skin conductivity sensors can collect skin conductivity signals in real time through electrodes in contact with the user's skin, and then extract features such as fluctuation amplitude, fluctuation frequency, mean, and variance of the skin conductivity signals. Then, based on a preset feature-emotional stress index mapping list, the extracted features are quantified into an emotional stress index. For example, the index range can be 0-100, with a higher index indicating a stronger stress level; and systolic / diastolic blood pressure values ​​collected by blood pressure monitoring cuff components.

[0050] In some exemplary embodiments, a computer device may use a contactless device located within a target vehicle to obtain initial physiological information of a user within the target vehicle.

[0051] Specifically, the computer equipment can communicate with the contactless device installed in the target vehicle. The contactless device can collect the user's first physiological information in the target vehicle in real time. The computer equipment can use the communication connection to obtain the user's first physiological information collected by the contactless device installed in the target vehicle.

[0052] Furthermore, computer devices can also use contact devices to obtain users' secondary physiological information.

[0053] Specifically, the computer equipment and the contact device can communicate with each other. The contact device can collect the user's second physiological information in the target vehicle in real time, and the computer equipment can use the communication connection to obtain the user's second physiological information collected by the contact device.

[0054] In one optional embodiment of this application, the non-contact device and the contact device will only start collecting the user's first and second physiological information when preset conditions are met. For example, when the user is in sleep mode, the non-contact device and the contact device will start collecting the user's first and second physiological information. This can avoid the energy waste caused by the continuous operation of the non-contact device and the contact device, and can also realize the monitoring of the user's physiological information in the sleep scenario.

[0055] Step 102: Determine the user's sleep state based on the first physiological information and the second physiological information.

[0056] Optionally, sleep status can be used to indicate a user's sleep quality, sleep stages, abnormal sleep events, etc. For example, sleep quality can be used to indicate the degree of quality of a user's sleep. Sleep stages can include wakefulness, light sleep, deep sleep, REM sleep, and arousal. Abnormal sleep events are used to indicate abnormal physiological or behavioral manifestations that occur during sleep, such as apnea, limb twitching, or sudden changes in body temperature.

[0057] In some exemplary embodiments, after a computer device acquires a user's first physiological information using a non-contact device located in the target vehicle and acquires the user's second physiological information using a contact device, it can determine the user's sleep state based on the first and second physiological information.

[0058] Specifically, computer equipment can perform information fusion processing on the first physiological information and the second physiological information to obtain fused physiological information, and then determine the user's sleep state based on the fused physiological information.

[0059] The computer device can also input the first and second physiological information into a pre-trained sleep state analysis model to obtain the user's sleep state output by the sleep state analysis model.

[0060] The sleep state analysis model can be a CNN-LSTM hybrid deep learning model. For example, the sleep state analysis model uses CNN layers to extract local features of physiological information, LSTM layers to capture long-term temporal dependencies of physiological signals, and fully connected layers and a softmax classifier at the end to achieve multi-classification of sleep states. Optionally, the sleep state analysis model can also add a regression branch to output a sleep quality score.

[0061] Step 103: Determine the vehicle device control strategy that matches the sleep state from multiple vehicle device control strategies in the preset strategy library, and control the working parameters of the vehicle device in the target vehicle based on the vehicle device control strategy.

[0062] Optionally, in-vehicle equipment can be devices within the target vehicle that affect the user's sleep comfort. Examples include environmental control devices, human-machine interaction devices, and vehicle control devices.

[0063] For example, environmental control devices may include in-vehicle air conditioning systems, seat temperature control devices, air circulation devices, and fragrance release devices; human-computer interaction devices may include active noise cancellation devices, in-vehicle audio systems, ambient lighting, and roof reading lights; and body control devices may include seat adjustment systems, window and sunroof adjustment devices, sunshade drive components, seat massage modules, and headrest angle adjustment devices.

[0064] In some exemplary embodiments, after determining the user's sleep state based on first physiological information and second physiological information, the computer device can determine the control strategy of the in-vehicle equipment in the target vehicle based on the sleep state.

[0065] Specifically, the computer device can determine the vehicle device control policy that matches the sleep state from multiple vehicle device control policies in the preset policy library.

[0066] For example, when a user's sleep state is transitioning from wakefulness to light sleep, the control strategy can focus on gradually guiding the user to fall asleep. For example, the ambient lighting in the car slowly switches from warm white to warm yellow, with the brightness decreasing by 10% every 2 minutes to below 10 cd / m²; the roof reading lights automatically turn off, and non-essential screens such as the center console and instrument panel enter a screen-off or off-screen mode; the car audio system activates white noise playback, selecting low-frequency sleep-aiding white noises such as rain or ocean waves, with the volume controlled within a comfortable range of 20-30 decibels; the seat adjustment system slowly reclines to the preset 125° ergonomic sleeping angle at a low speed of 0.5° / second, while the seat massage module activates a low-frequency gentle kneading mode, acting on the muscles of the shoulders, neck, and lower back; the air conditioning system fine-tunes the temperature to a comfortable 23°C for human sleep, reduces the fan speed to level 1 silent mode, and turns off the direct airflow mode; the air circulation device switches to internal circulation mode, and the fragrance release device intermittently releases fragrance, with each release lasting 3 seconds and an interval of 10 minutes, etc.

[0067] When the user is in deep sleep, the control strategy focuses on maintaining a stable sleep environment and avoiding external interference. For example, all unnecessary light sources in the vehicle (including ambient lighting and button backlighting) are turned off to ensure a completely dark interior; the active noise cancellation system switches to deep noise cancellation mode to specifically block high-frequency external noise (such as horns and construction noise) and low-frequency vibration noise (such as road bumps); the air conditioning system maintains a constant temperature of 22-23°C, while the humidity control module keeps the humidity in the vehicle within the optimal sleep humidity range of 50%-60%; the seat massage module stops working, keeping the seat angle fixed to avoid mechanical movement from disturbing the user's sleep; the air circulation device switches to an intermittent working mode of "running for 5 minutes and then sleeping for 15 minutes" to ensure air circulation in the vehicle while avoiding noise generated by continuous operation; non-emergency alarms such as door unlocking prompts and tire pressure monitoring prompts are muted, leaving only core safety alarms such as collision warnings, and the alarm volume is set to the lowest threshold.

[0068] When a user is in REM sleep, the control strategy focuses on achieving maximum quietness and minimizing disturbances. Building upon the control strategies for deep sleep, it further enhances environmental stability. For example, the active noise cancellation system activates its sudden noise prediction and shielding function, monitoring external noise in real time via the vehicle's microphone. When a sudden high-decibel noise is detected, noise cancellation is initiated in advance. The seat adjustment system and massage module remain absolutely still, eliminating any mechanical displacement or vibration. All non-essential in-vehicle equipment, such as aromatherapy devices and air circulation systems, are turned off to minimize the subtle noise generated by the operation of in-vehicle equipment. The air conditioning system switches to a pure natural wind mode to avoid noise interference from the compressor's start-stop cycle, ensuring the continuity of the user's REM sleep.

[0069] When a user is in the wake-up phase of sleep, the control strategy can focus on simulating a natural sunrise and gently waking them up. For example, the ambient lighting gradually switches from warm yellow to warm white, with brightness increasing slowly by 15% every 3 minutes, simulating the light changes at sunrise to help the user's biological clock wake up naturally; the seat adjustment system slowly adjusts the backrest angle to a 90° sitting position at a low speed of 0.3° / second, while the seat massage module activates a high-frequency tapping mode to gently wake the user's muscles; the air conditioning system raises the temperature to the 25°C wake-up mode temperature and increases the fan speed to level 2 to promote blood circulation to the user's skin; the car audio system turns off white noise and switches to soft, light music, with the volume slowly increasing by 5 decibels every 2 minutes to 40 decibels; the air circulation system switches to external circulation mode, introducing fresh outside air to further help the user regain wakefulness; the central control screen and instrument panel automatically light up, displaying basic information such as the current time and outside temperature, providing a reference for the user's actions after waking up.

[0070] Furthermore, after determining the vehicle-mounted equipment control strategy, the computer equipment can control the operating parameters of the vehicle-mounted equipment in the target vehicle based on the vehicle-mounted equipment control strategy.

[0071] The aforementioned vehicle control method acquires first and second physiological information of the user inside the target vehicle. The first physiological information is collected using a non-contact device installed inside the target vehicle, while the second physiological information is collected using a contact device. Then, based on the first and second physiological information, the user's sleep state is determined. Next, from a preset strategy library of multiple in-vehicle device control strategies, an in-vehicle device control strategy matching the sleep state is selected. Finally, the operating parameters of the in-vehicle devices inside the target vehicle are controlled based on this strategy. The vehicle control method provided in this application acquires the user's physiological information through non-contact and contact devices installed inside the target vehicle, and then controls the in-vehicle devices based on this information, effectively improving the comfort of the in-vehicle environment when the user sleeps in the vehicle.

[0072] In one exemplary embodiment, such as Figure 2 As shown, determining a user's sleep state based on first and second physiological information includes the following steps:

[0073] Step 201: Determine the first information quality of the first physiological information and the second information quality of the second physiological information.

[0074] Optionally, the first information quality can be used to indicate the reliability, validity, accuracy, and completeness of the first physiological information. The second information quality can be used to indicate the reliability, validity, accuracy, and completeness of the second physiological information.

[0075] In some exemplary embodiments, after obtaining the first physiological information and the second physiological information, the computer device can determine the first information quality of the first physiological information and the second information quality of the second physiological information.

[0076] Specifically, for the first physiological information, taking the user's facial video stream as an example, the computer device can first track the displacement and scaling rate of the face detection box between consecutive video frames, set displacement thresholds and scaling rate thresholds, and calculate the stability score of the facial region based on whether the displacement and scaling rate of the detection box exceed the thresholds. Then, the signal-to-noise ratio of the average pixel value is calculated in the extracted skin areas such as the forehead and cheeks to obtain the regional signal-to-noise ratio score. Next, the pixel grayscale distribution of the skin area is analyzed, the grayscale variance value is calculated, and the illumination uniformity score of the facial region is evaluated based on the comparison result of the grayscale variance value with the preset threshold. Finally, the stability score, regional signal-to-noise ratio score, and illumination uniformity score are weighted and summed to obtain the first information quality.

[0077] For the second physiological information, taking the PPG (Photoplethysmography) signal collected by the wearable device as an example, the computer device can first analyze the pulse waveform characteristics of the PPG signal, calculate the peak sharpness, diastolic notch significance, and baseline drift of the pulse waveform, and obtain the morphological score based on the matching degree between the quantification results of each index and the preset standard value. Then, a known resting period in the PPG signal is selected, and the ratio of PPG signal power to noise power in that period is calculated to obtain the signal-to-noise ratio score. Then, motion data collected by the accelerometer built into the wearable device is retrieved to detect the period of intense arm movement and quantify the motion intensity. The motion artifact suppression score is obtained by calculating the reciprocal of the motion intensity. Finally, the morphological score, signal-to-noise ratio score, and motion artifact suppression score are weighted and summed to obtain the quality of the second information.

[0078] In an optional embodiment of this application, before determining the first information quality of the first physiological information and the second information quality of the second physiological information, the computer device may first perform data preprocessing and data synchronization processing on the first physiological information and the second physiological information.

[0079] Specifically, the computer equipment can filter and denoise the first and second physiological information, and can also perform data synchronization processing on the first and second physiological information based on timestamps.

[0080] Step 202: Determine the user's target physiological characteristics based on the first information quality, the second information quality, the first physiological information, and the second physiological information.

[0081] Optionally, the target physiological characteristics are key features of sleep stages, such as the trend of decreasing heart rate, heart rate variability, and respiratory rate.

[0082] In some exemplary embodiments, after determining the first information quality of the first physiological information and the second information quality of the second physiological information, the computer device can determine the user's target physiological characteristics based on the first information quality, the second information quality, the first physiological information, and the second physiological information.

[0083] Specifically, the computer device can determine the target physiological information based on the first information quality and the second information quality, and then determine the user's target physiological characteristics based on the target physiological information.

[0084] Step 203: Input the target physiological characteristics into the sleep analysis model to obtain the user's sleep state output by the sleep analysis model.

[0085] Optionally, the sleep analysis model can be a long short-term memory network model.

[0086] For example, the Long Short-Term Memory (LSTM) network model refers to a recurrent neural network architecture adapted to temporal physiological data. The LSTM network model can include an input gate, a forget gate, an output gate, and cell states. The input of the LSTM network model is the target physiological feature. The input gate controls the input weight of the feature at the current moment. The forget gate filters and retains historical feature information that is strongly correlated with the sleep state. The output gate combines the cell state to output the sleep feature representation at the current moment. The sleep state is classified and determined by a fully connected layer and a softmax classifier.

[0087] The training process of the sleep analysis model includes: constructing a training dataset containing multi-dimensional temporal physiological features and corresponding sleep state labels, covering wakefulness, light sleep, deep sleep, and REM sleep; initializing model parameters, setting the number of hidden layer units, learning rate, and number of iterations, using the cross-entropy loss function and Adam optimizer for parameter iterative optimization, and introducing a Dropout layer to suppress overfitting; evaluating model performance using a validation set, and terminating training and saving model parameters when the sleep stage classification accuracy reaches a preset threshold, thus obtaining the sleep analysis model.

[0088] In some exemplary embodiments, after determining the user's target physiological characteristics based on the first information quality, the second information quality, the first physiological information, and the second physiological information, the computer device can input the target physiological characteristics into a sleep analysis model to obtain the user's sleep state output by the sleep analysis model.

[0089] In one exemplary embodiment, such as Figure 3As shown, determining a user's target physiological characteristics based on first information quality, second information quality, first physiological information, and second physiological information includes the following steps:

[0090] Step 301: If the quality of the first information meets the preset conditions and the quality of the second information does not meet the preset conditions, perform feature extraction processing on the first physiological information to obtain the target physiological features.

[0091] Optionally, the preset conditions can be pre-set by technicians according to actual needs. For example, the preset conditions could be that the quality score of the information quality indicator is greater than a preset quality score threshold, or that the signal-to-noise ratio of the information quality indicator is greater than a preset signal-to-noise ratio threshold, etc.

[0092] In some exemplary embodiments, after determining the first information quality and the second information quality, the computer device may determine whether the first information quality and the second information quality meet preset conditions, respectively.

[0093] Furthermore, if the quality of the first information meets the preset conditions and the quality of the second information does not meet the preset conditions, the computer device can perform feature extraction processing on the first physiological information to obtain the target physiological features.

[0094] Step 302: If the quality of the second information meets the preset conditions and the quality of the first information does not meet the preset conditions, feature extraction is performed on the second physiological information to obtain the target physiological features.

[0095] In some exemplary embodiments, after determining the first information quality and the second information quality, the computer device may determine whether the first information quality and the second information quality meet preset conditions, respectively.

[0096] Furthermore, if the quality of the second information meets the preset conditions and the quality of the first information does not meet the preset conditions, the computer device can extract features from the second physiological information to obtain the target physiological features.

[0097] Step 303: If both the first information quality and the second information quality meet the preset conditions, compare the first information quality and the second information quality, and determine the user's target physiological characteristics based on the comparison results, the first physiological information and the second physiological information.

[0098] In some exemplary embodiments, after determining the first information quality and the second information quality, the computer device may determine whether the first information quality and the second information quality meet preset conditions, respectively.

[0099] Furthermore, if both the first information quality and the second information quality meet the preset conditions, the computer device can compare the first information quality and the second information quality, and determine the user's target physiological characteristics based on the comparison result, the first physiological information, and the second physiological information.

[0100] Specifically, the computer device can determine the target physiological information from the first physiological information and the second physiological information based on the comparison results, and determine the user's target physiological characteristics based on the target physiological information.

[0101] In some exemplary embodiments, such as Figure 4 As shown, feature extraction processing is performed on the first physiological information to obtain the target physiological features, including the following steps:

[0102] Step 401: Use a time-domain feature extraction algorithm to determine the first feature based on the first physiological information.

[0103] Optionally, the first feature can be used to characterize the user's autonomic nervous activity, cardiovascular status, etc. For example, the first feature can be heart rate, heart rate variability, pulse rise / fall time, respiratory rate, etc.

[0104] In some exemplary embodiments, the computer device may use a time-domain feature extraction algorithm to determine a first feature based on first physiological information.

[0105] Specifically, the computer device can detect peak values ​​from the waveform corresponding to the first physiological information and determine the first feature by calculating the interval. The computer device can also determine the first feature through respiratory rhythm by performing frequency domain analysis on the first physiological information.

[0106] For example, when the first physiological information is based on a facial video stream, the computer device can first analyze the facial video to obtain the rPPG (Remote Photoplethysmography) signal, then perform filtering, noise reduction, and baseline correction preprocessing on the original waveform of the rPPG signal. Then, the characteristic peaks in the waveform are detected by an adaptive threshold method, and the time interval between two adjacent peaks is calculated, which is the pulse cycle. If the average of 10 consecutively detected pulse cycles is 0.8 seconds, the heart rate can be determined by the heart rate calculation formula, and this heart rate is determined as the first feature. The heart rate calculation formula is: heart rate = 60 / average pulse cycle.

[0107] For example, when the first physiological information is based on a facial video stream, the computer device can first analyze the facial video to obtain the rPPG (Remote Photoplethysmography) signal, then perform filtering, noise reduction, and baseline correction preprocessing on the original waveform of the rPPG signal. Then, the computer device can analyze the slow fluctuation pattern of the baseline of the rPPG signal, detect the baseline fluctuation period, and determine the respiratory rate based on the mean of the baseline fluctuation period and the respiratory rate calculation formula, and determine the respiratory rate as the first feature. The respiratory rate calculation formula is: respiratory rate = 60 / mean of the baseline fluctuation period.

[0108] Step 402: Using a frequency domain feature extraction algorithm, determine the second feature based on the first physiological information.

[0109] Optionally, the second feature can be used to characterize the user's sympathetic and parasympathetic nervous system balance.

[0110] For example, the second feature could be the low-frequency / high-frequency power ratio, total power, etc.

[0111] In some exemplary embodiments, the computer device may use a frequency domain feature extraction algorithm to determine a second feature based on the first physiological information.

[0112] Specifically, the computer equipment can perform Fourier transform or wavelet transform on the heart rate sequence corresponding to the first physiological information, and determine the second feature by calculating the power ratio of different frequency bands and the total power value in the frequency domain.

[0113] Taking the rPPG signal obtained from facial video stream analysis as the first physiological information as an example, the computer device can first extract a continuous heart rate sequence based on the rPPG signal, and preprocess the heart rate sequence. By resampling, the sequence is unified to the same sampling frequency, and a sliding window filter is used to remove motion artifact interference. Then, a fast Fourier transform is performed on the preprocessed heart rate sequence to convert the time-domain heart rate fluctuation signal into a power spectral density distribution in the frequency domain. Finally, characteristic frequency bands strongly correlated with autonomic nervous activity are divided, and the frequency range corresponding to the low-frequency band can be used for... Reflecting the coordinated activity of the sympathetic and parasympathetic nervous systems, the frequency range corresponding to the high-frequency band can be used to reflect the activity of the parasympathetic nervous system. Then, the power values ​​of all frequency points in the low-frequency band are calculated to obtain the low-frequency power, and the power values ​​of all frequency points in the high-frequency band are calculated to obtain the high-frequency power. The power ratio of the low-frequency band to the high-frequency band is calculated. This power ratio is a characteristic of the balance between the sympathetic and parasympathetic nervous systems. The higher the ratio, the more active the sympathetic nervous system is, and the lower the ratio, the more active the parasympathetic nervous system is.

[0114] Simultaneously, the sum of the low-frequency power and the high-frequency power is calculated to obtain the total power. This total power can be used to reflect the overall activity intensity of the autonomic nervous system to determine the stability of the user's physiological state. Finally, the computer device can determine the power ratio of the low-frequency band to the high-frequency band and the total power as the second feature.

[0115] Step 403: Using a nonlinear feature extraction algorithm, determine the third feature based on the first physiological information.

[0116] Optionally, the third feature can be used to characterize the complexity and regulatory capacity of the user's physiological system.

[0117] For example, the third feature could be sample entropy, Poincaré plot scatter points, etc.

[0118] In some exemplary embodiments, the computer device may use a nonlinear feature extraction algorithm to determine a third feature based on first physiological information.

[0119] Specifically, taking the rPPG signal obtained from facial video stream analysis as the first physiological information as an example, the computer device can first set the embedding dimension and similarity tolerance, then extract a continuous heart rate sequence based on the rPPG signal, divide the heart rate sequence into several embedding vectors, calculate the similarity between any two embedding vectors, statistically analyze the proportion of similar vectors and calculate the logarithm, and then repeat the above operation by adjusting the embedding dimension. The absolute value of the difference obtained is determined as the sample entropy, and the sample entropy is determined as the third feature.

[0120] The computer device can also plot all adjacent heart rate values ​​on a two-dimensional plane using the nth heart rate value in the heart rate sequence as the x-axis and the (n+1)th heart rate value as the y-axis, forming a Poincaré plot. Then, based on the Poincaré plot, the scatter features of the Poincaré plot are obtained and determined as the third feature. The scatter features of the Poincaré plot can include the major axis (SD1), the minor axis (SD2), the SD1 / SD2 ratio, and the scatter cluster density. Among them, SD1 can be used to reflect the short-term fluctuation of heart rate, SD2 can be used to reflect the long-term fluctuation of heart rate, and the SD1 / SD2 ratio can be used to judge the stability of the physiological system.

[0121] Specifically, computer equipment can calculate the complexity and chaos of the time series corresponding to the first physiological information, and determine the third feature by solving the sample entropy value and analyzing the distribution characteristics of the Poincaré plot scatter points.

[0122] Step 404: Input the first physiological information into the convolutional neural network model to obtain the fourth feature output by the convolutional neural network model.

[0123] Optionally, the fourth feature can be used to characterize deep, complex patterns related to sleep in the first physiological information.

[0124] In some exemplary embodiments, a computer device may input preprocessed first physiological information into a convolutional neural network model to obtain a fourth feature.

[0125] Specifically, the computer device first performs filtering and noise reduction, baseline correction, and time series standardization preprocessing on the first physiological information. Then, the preprocessed signal segment is input into a lightweight convolutional neural network, and a high-dimensional abstract feature vector is extracted from the fully connected layer at the end of the network, thus obtaining the fourth feature. The fourth feature can be a high-dimensional numerical abstract feature vector with fixed dimensions. Each dimension corresponds to a deep latent pattern strongly correlated with the user's sleep state. Specifically, it can include subtle distortion features of the remote photoplethysmography (RP-PWC) signal waveform, the coordinated fluctuation features of heart rate and respiration, and high-frequency latent noise patterns of physiological signals.

[0126] The lightweight convolutional neural network can adopt an architecture of input layer-convolutional layer-pooling layer-batch normalization layer-flattening layer. The training process of this convolutional neural network can include: 1. Collecting preprocessed samples of first physiological information from users of different ages and physical conditions simulating various sleep scenarios in a target vehicle, labeling the samples with corresponding sleep-related tags, and dividing them into training, validation, and test sets proportionally; 2. Using the cross-entropy loss function as the basis for loss calculation, iterating parameters using the Adam optimizer, and introducing a Dropout layer to suppress overfitting; 3. Evaluating the model's feature extraction performance using the validation set after each round of training. Training is terminated when the validation set loss does not decrease for several consecutive rounds, and the optimal model parameters are saved.

[0127] Step 405: The first feature, second feature, third feature and fourth feature are spliced ​​together to obtain the target physiological features.

[0128] In some exemplary embodiments, after obtaining the first feature, the second feature, the third feature, and the fourth feature, the computer device can perform splicing processing on the first feature, the second feature, the third feature, and the fourth feature to obtain the target physiological feature.

[0129] In an optional embodiment of this application, the computer device may further perform feature extraction processing on the second physiological information based on the above method to obtain the target physiological features.

[0130] In one exemplary embodiment, such as Figure 5 As shown, determining the user's target physiological characteristics based on the comparison results, first physiological information, and second physiological information includes the following steps:

[0131] Step 501: If the comparison result indicates that the quality of the first information is greater than that of the second information, then the second physiological information is calibrated using the first physiological information, and features are extracted from the first physiological information and the calibrated second physiological information to obtain the target physiological features.

[0132] In some exemplary embodiments, when both the first information quality and the second information quality meet preset conditions, the computer device can compare the first information quality and the second information quality.

[0133] Furthermore, if the comparison result indicates that the quality of the first information is greater than that of the second information, the computer device can use the first physiological information to calibrate the second physiological information to obtain the calibrated second physiological information.

[0134] Furthermore, the computer device can extract features from the first physiological information and the calibrated second physiological information to obtain the target physiological features.

[0135] Step 502: If the comparison result indicates that the quality of the first information is less than that of the second information, then the first physiological information is calibrated using the second physiological information, and features are extracted from the second physiological information and the calibrated first physiological information to obtain the target physiological features.

[0136] In some exemplary embodiments, when both the first information quality and the second information quality meet preset conditions, the computer device can compare the first information quality and the second information quality.

[0137] Furthermore, if the comparison result indicates that the quality of the first information is less than that of the second information, the computer device can use the second physiological information to calibrate the first physiological information to obtain the calibrated first physiological information.

[0138] Furthermore, the computer device can extract features from the second physiological information and the calibrated first physiological information to obtain the target physiological features.

[0139] Step 503: If the comparison result indicates that the quality of the first information is equal to the quality of the second information, then feature extraction is performed on the first physiological information and the second physiological information to obtain the target physiological features.

[0140] In some exemplary embodiments, when both the first information quality and the second information quality meet preset conditions, the computer device can compare the first information quality and the second information quality.

[0141] Furthermore, if the comparison result indicates that the quality of the first information is equal to the quality of the second information, the computer device can extract features from the first physiological information and the second physiological information to obtain the target physiological features.

[0142] In one exemplary embodiment, such as Figure 6 As shown, calibrating the first physiological information using the second physiological information includes the following steps:

[0143] Step 601: Determine the first target feature from the first physiological information, and determine the second target feature from the second physiological information.

[0144] The first target feature and the second target feature have the same feature type and acquisition time window. For example, the first target feature can be the instantaneous heart rate, respiratory rate, etc., corresponding to the first physiological information, and the second target feature can be the instantaneous heart rate, respiratory rate, etc., corresponding to the second physiological information.

[0145] In some exemplary embodiments, a computer device can determine a first target feature and a second target feature from first physiological information and second physiological information, respectively.

[0146] Specifically, the computer equipment can first determine the feature type required for calibration, and then extract the corresponding type of features from the first physiological information and the second physiological information respectively within the same preset time window, and determine them as the first target feature and the second target feature respectively, so as to ensure that the feature type of the first target feature and the second target feature are the same as the acquisition time window, and avoid calibration deviation caused by feature mismatch.

[0147] Step 602: Input the first target feature and the second target feature into the pre-built regression model so that the regression model uses the second target feature to calibrate the first target feature to obtain the calibrated first target feature.

[0148] Optionally, the regression model can be a pre-built linear regression model or a ridge regression model.

[0149] For example, a computer device can collect multiple sets of first target feature samples and second target feature samples within a synchronous time window, wherein the second target feature is a high-precision benchmark feature; preprocess the first target feature samples and second target feature samples to remove outliers and standardize them to eliminate dimensional differences; construct a regression model with the first target feature sample as the independent variable and the second target feature as the benchmark dependent variable; solve the optimal parameters of the regression model using the least squares method, and evaluate the mapping accuracy using a validation set; when the mean square error between the model output value and the true second target feature is less than a preset threshold, the regression model is determined to be completed and the model is saved.

[0150] In some exemplary embodiments, after determining a first target feature from first physiological information and a second target feature from second physiological information, the computer device can input the first target feature and the second target feature into a pre-built regression model so that the regression model can calibrate the first target feature using the second target feature to obtain the calibrated first target feature.

[0151] Step 603: Based on the calibrated first target feature, calibrate other features in the first physiological information besides the first target feature.

[0152] In some exemplary embodiments, after obtaining the calibrated first target feature, the computer device can calibrate other features in the first physiological information, excluding the first target feature, based on the calibrated first target feature.

[0153] Specifically, since all features of the first physiological information originate from the same physiological signal source and there is a strong correlation between different features, other features in the first physiological information, except for the first target feature, can be calibrated based on the calibrated first target feature.

[0154] Computer equipment can obtain a deviation coefficient by comparing the first target feature value before and after calibration. This deviation coefficient can be used to characterize the overall correction ratio of the first target feature. For each other feature, the computer equipment can calculate its original value and the deviation coefficient to calibrate the other features. For example, if the original value of heart rate variability is 50ms and the deviation coefficient k=1.05, then the original value of heart rate variability after calibration = 50ms × 1.05 = 52.5ms.

[0155] In an optional embodiment of this application, the computer device may also calibrate the second physiological information using the first physiological information based on the above method.

[0156] In one exemplary embodiment, such as Figure 7 As shown, controlling the operating parameters of the on-board equipment in the target vehicle based on the on-board equipment control strategy includes the following steps:

[0157] Step 701: Obtain the environmental parameters of the target vehicle.

[0158] Optionally, environmental parameters can be temperature, humidity, light intensity, noise, etc.

[0159] In some exemplary embodiments, the computer device can acquire environmental parameters of the target vehicle.

[0160] Specifically, computer equipment can use in-vehicle temperature sensors, humidity sensors, light sensors, and noise sensors installed inside the target vehicle to obtain environmental parameters of the target vehicle.

[0161] Step 702: Control the operating parameters of the on-board equipment in the target vehicle based on environmental parameters and on-board equipment control strategies.

[0162] In some exemplary embodiments, after acquiring the environmental parameters of the target vehicle, the computer device controls the operating parameters of the on-board equipment in the target vehicle based on the environmental parameters and the on-board equipment control strategy.

[0163] Specifically, the computer equipment can first compare the environmental parameters with the preset comfort threshold range to determine whether the current environment is suitable for the basic needs corresponding to the user's sleep state. Then, it can retrieve the basic operating parameters in the vehicle equipment control strategy that matches the sleep state and dynamically adjust the basic parameters based on the deviation of the environmental parameters.

[0164] For example, if the current interior temperature is higher than the preset comfort threshold, the air conditioning cooling temperature will be lowered based on the basic air conditioning control strategy; if the interior noise exceeds the preset threshold, the active noise cancellation function will be activated and adapted to soothing white noise based on the basic audio control strategy; if the light sensor detects strong light, the sunshade will be closed and the interior ambient lighting will be adjusted to a low-brightness warm light mode; if the humidity is lower than the preset comfort range, the in-vehicle humidifier will be turned on and the mist output will be adjusted.

[0165] Furthermore, after the adjustment is completed, the computer equipment can send control commands to the corresponding vehicle equipment to control each vehicle equipment to operate according to the adjusted working parameters, so as to achieve precise adaptation of environmental parameters and sleep state and improve the user's sleep comfort.

[0166] In one exemplary embodiment, such as Figure 8 As shown, after the user's sleep ends, the method also includes the following steps:

[0167] Step 801: Obtain the user's sleep information.

[0168] For example, sleep information may include the time required to fall asleep, the duration of each sleep stage, sensitivity to environmental disturbances, and preferred final wake-up methods.

[0169] In some exemplary embodiments, the computer device may acquire the user's sleep information after the user has finished sleeping.

[0170] Specifically, computer devices can determine a user's sleep information based on first and second physiological information.

[0171] Step 802: Optimize the control strategy of the on-board equipment of the target vehicle based on sleep information.

[0172] In some exemplary embodiments, after acquiring a user's sleep information, the computer device can optimize the control strategy of the on-board equipment of the target vehicle based on the sleep information.

[0173] Specifically, for example, the sleep-inducing guidance time can be shortened or lengthened based on the time it takes to fall asleep, the noise reduction and light blocking intensity can be adjusted based on environmental sensitivity, and the wake-up type of light or vibration can be matched according to the user's preferred wake-up method, so as to optimize the control strategy and make it easy to be automatically called when the user falls asleep again, thus achieving personalized adaptation to meet the individual needs of the user.

[0174] In one exemplary embodiment, such as Figure 9 As shown, another vehicle control method is provided, including the following steps:

[0175] Step 901: Obtain the first and second physiological information of the user inside the target vehicle; the first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device.

[0176] Step 902: Determine the first information quality of the first physiological information and the second information quality of the second physiological information; if the first information quality meets the preset conditions and the second information quality does not meet the preset conditions, use a time-domain feature extraction algorithm to determine the first feature based on the first physiological information; use a frequency-domain feature extraction algorithm to determine the second feature based on the first physiological information; use a nonlinear feature extraction algorithm to determine the third feature based on the first physiological information; input the first physiological information into a convolutional neural network model to obtain the fourth feature output by the convolutional neural network model; concatenate the first feature, second feature, third feature, and fourth feature to obtain the target physiological feature;

[0177] Step 903: If the quality of the second information meets the preset conditions and the quality of the first information does not meet the preset conditions, feature extraction is performed on the second physiological information to obtain the target physiological features; if both the quality of the first information and the quality of the second information meet the preset conditions, the quality of the first information and the quality of the second information are compared.

[0178] Step 904: If the comparison result indicates that the quality of the first information is greater than that of the second information, then the second physiological information is calibrated using the first physiological information, and feature extraction is performed on the first physiological information and the calibrated second physiological information to obtain the target physiological features.

[0179] Step 905: If the comparison result indicates that the quality of the first information is less than that of the second information, then determine the first target feature from the first physiological information and the second target feature from the second physiological information. The feature types and acquisition time windows of the first and second target features are the same. Input the first and second target features into a pre-built regression model so that the regression model uses the second target feature to calibrate the first target feature to obtain the calibrated first target feature. Based on the calibrated first target feature, calibrate other features in the first physiological information except for the first target feature, and extract features from the second physiological information and the calibrated first physiological information to obtain the target physiological feature.

[0180] Step 906: If the comparison result indicates that the quality of the first information is equal to the quality of the second information, then perform feature extraction on the first physiological information and the second physiological information to obtain the target physiological features; input the target physiological features into the sleep analysis model to obtain the user's sleep state output by the sleep analysis model;

[0181] Step 907: Determine the vehicle device control strategy that matches the sleep state from multiple vehicle device control strategies in the preset strategy library, and obtain the environmental parameters of the target vehicle; control the working parameters of the vehicle device in the target vehicle based on the environmental parameters and the vehicle device control strategy; after the user's sleep ends, obtain the user's sleep information; optimize the control strategy of the vehicle device in the target vehicle based on the sleep information.

[0182] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0183] Based on the same inventive concept, this application also provides a vehicle control device for implementing the vehicle control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more vehicle control device embodiments provided below can be found in the limitations of the vehicle control method described above, and will not be repeated here.

[0184] In one exemplary embodiment, such as Figure 10 As shown, a vehicle control device 1000 is provided, including: an acquisition module 1001, a determination module 1002, and an execution module 1003, wherein:

[0185] The acquisition module 1001 is used to acquire the first physiological information and the second physiological information of the user inside the target vehicle; the first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device.

[0186] The determination module 1002 is used to determine the user's sleep state based on the first physiological information and the second physiological information;

[0187] The execution module 1003 is used to determine the vehicle device control strategy that matches the sleep state from multiple vehicle device control strategies in the preset strategy library, and control the operating parameters of the vehicle device in the target vehicle based on the vehicle device control strategy.

[0188] In one embodiment, the determining module 1002 is specifically used to determine the first information quality of the first physiological information and the second information quality of the second physiological information; determine the user's target physiological characteristics based on the first information quality, the second information quality, the first physiological information, and the second physiological information; and input the target physiological characteristics into the sleep analysis model to obtain the user's sleep state output by the sleep analysis model.

[0189] In one embodiment, the determining module 1002 is specifically configured to: perform feature extraction processing on the first physiological information to obtain target physiological features when the first information quality meets preset conditions and the second information quality does not meet preset conditions; perform feature extraction on the second physiological information to obtain target physiological features when the second information quality meets preset conditions and the first information quality does not meet preset conditions; and compare the first information quality and the second information quality when both the first information quality and the second information quality meet preset conditions, and determine the user's target physiological features based on the comparison result, the first physiological information, and the second physiological information.

[0190] In one embodiment, the determining module 1002 is specifically used to determine a first feature based on first physiological information using a time-domain feature extraction algorithm; determine a second feature based on the first physiological information using a frequency-domain feature extraction algorithm; determine a third feature based on the first physiological information using a nonlinear feature extraction algorithm; input the first physiological information into a convolutional neural network model to obtain a fourth feature output by the convolutional neural network model; and concatenate the first feature, second feature, third feature, and fourth feature to obtain the target physiological feature.

[0191] In one embodiment, the determining module 1002 is specifically configured to: if the comparison result indicates that the quality of the first information is greater than the quality of the second information, then calibrate the second physiological information using the first physiological information, and extract features from the first physiological information and the calibrated second physiological information to obtain the target physiological feature; if the comparison result indicates that the quality of the first information is less than the quality of the second information, then calibrate the first physiological information using the second physiological information, and extract features from the second physiological information and the calibrated first physiological information to obtain the target physiological feature; if the comparison result indicates that the quality of the first information is equal to the quality of the second information, then extract features from the first physiological information and the second physiological information to obtain the target physiological feature.

[0192] In one embodiment, the determining module 1002 is specifically used to determine a first target feature from the first physiological information and a second target feature from the second physiological information, wherein the feature types and acquisition time windows of the first target feature and the second target feature are the same; input the first target feature and the second target feature into a pre-built regression model so that the regression model uses the second target feature to calibrate the first target feature to obtain the calibrated first target feature; and calibrate other features in the first physiological information other than the first target feature based on the calibrated first target feature.

[0193] In one embodiment, the execution module 1003 is specifically used to acquire environmental parameters of the target vehicle; and determine the control strategy of the on-board equipment in the target vehicle based on the environmental parameters and the sleep state.

[0194] In one embodiment, the execution module 1003 is further configured to acquire the user's sleep information after the user's sleep ends; and optimize the control strategy of the on-board equipment of the target vehicle based on the sleep information.

[0195] Each module in the aforementioned vehicle control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0196] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 11 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements a vehicle control method.

[0197] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 12 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a vehicle control method.

[0198] Those skilled in the art will understand that Figure 11 and Figure 12 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0199] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any embodiment.

[0200] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any embodiment.

[0201] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0202] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0203] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0204] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments 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 application.

[0205] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A vehicle control method, characterized in that, The method includes: Acquire first and second physiological information of the user inside the target vehicle; the first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device. The user's sleep state is determined based on the first physiological information and the second physiological information; From a set of vehicle device control policies in a preset policy library, a vehicle device control policy that matches the sleep state is determined, and the operating parameters of the vehicle device in the target vehicle are controlled based on the vehicle device control policy.

2. The method according to claim 1, characterized in that, Determining the user's sleep state based on the first physiological information and the second physiological information includes: Determine the first information quality of the first physiological information and the second information quality of the second physiological information; The user's target physiological characteristics are determined based on the first information quality, the second information quality, the first physiological information, and the second physiological information. The target physiological characteristics are input into the sleep analysis model to obtain the user's sleep state output by the sleep analysis model.

3. The method according to claim 2, characterized in that, The step of determining the user's target physiological characteristics based on the first information quality, the second information quality, the first physiological information, and the second physiological information includes: If the quality of the first information meets the preset conditions and the quality of the second information does not meet the preset conditions, feature extraction processing is performed on the first physiological information to obtain the target physiological features. If the quality of the second information meets the preset condition and the quality of the first information does not meet the preset condition, feature extraction is performed on the second physiological information to obtain the target physiological feature. If both the first information quality and the second information quality meet the preset conditions, the first information quality and the second information quality are compared, and the user's target physiological characteristics are determined based on the comparison result, the first physiological information, and the second physiological information.

4. The method according to claim 3, characterized in that, The step of performing feature extraction processing on the first physiological information to obtain the target physiological feature includes: Using a time-domain feature extraction algorithm, a first feature is determined based on the first physiological information; Using a frequency domain feature extraction algorithm, a second feature is determined based on the first physiological information; A nonlinear feature extraction algorithm is used to determine the third feature based on the first physiological information; The first physiological information is input into the convolutional neural network model to obtain the fourth feature output by the convolutional neural network model; The first feature, the second feature, the third feature, and the fourth feature are spliced ​​together to obtain the target physiological feature.

5. The method according to claim 3, characterized in that, Determining the user's target physiological characteristics based on the comparison results, the first physiological information, and the second physiological information includes: If the comparison result indicates that the quality of the first information is greater than that of the second information, then the second physiological information is calibrated using the first physiological information, and feature extraction is performed on the first physiological information and the calibrated second physiological information to obtain the target physiological feature; If the comparison result indicates that the quality of the first information is less than that of the second information, then the first physiological information is calibrated using the second physiological information, and feature extraction is performed on the second physiological information and the calibrated first physiological information to obtain the target physiological feature. If the comparison result indicates that the quality of the first information is equal to the quality of the second information, then feature extraction is performed on the first physiological information and the second physiological information to obtain the target physiological feature.

6. The method according to claim 5, characterized in that, The calibration of the first physiological information using the second physiological information includes: A first target feature is determined from the first physiological information, and a second target feature is determined from the second physiological information, wherein the first target feature and the second target feature have the same feature type and acquisition time window; The first target feature and the second target feature are input into a pre-built regression model so that the regression model uses the second target feature to calibrate the first target feature to obtain the calibrated first target feature; Based on the calibrated first target feature, other features in the first physiological information, excluding the first target feature, are calibrated.

7. The method according to any one of claims 1 to 6, characterized in that, The control of the operating parameters of the in-vehicle equipment in the target vehicle based on the in-vehicle equipment control strategy includes: Obtain the environmental parameters of the target vehicle; The operating parameters of the on-board equipment in the target vehicle are controlled based on the environmental parameters and the on-board equipment control strategy.

8. The method according to any one of claims 1 to 6, characterized in that, The method further includes: After the user's sleep ends, obtain the user's sleep information; The control strategy for the on-board equipment of the target vehicle is optimized based on the sleep information.

9. A vehicle control device, characterized in that, The device includes: The acquisition module is used to acquire first physiological information and second physiological information of a user inside the target vehicle; the first physiological information is collected using a non-contact device installed inside the target vehicle, and the second physiological information is collected using a contact device. A determination module is used to determine the user's sleep state based on the first physiological information and the second physiological information; The execution module is used to determine the vehicle device control strategy that matches the sleep state from multiple vehicle device control strategies in the preset strategy library, and control the operating parameters of the vehicle device in the target vehicle based on the vehicle device control strategy.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images