Vehicle control method and device

Abstract

According to the vehicle control method and device, the breathing data and the body shaking data are determined through the radar data collected by the millimeter-wave radar, then the comprehensive fatigue score is determined, in this way, non-contact type fatigue driving detection can be achieved, the interference rate is low, the detection is more accurate, the comfort is better, and the driving safety is improved. The requirement for cooperation of a driver is low, popularization and use are easier, and the response speed is high. And the corresponding fatigue reminding strategy is obtained based on the comprehensive fatigue score matching, the vehicle is controlled to execute the fatigue reminding strategy, intervention reminding and timely early warning of possible fatigue driving can be performed at the initial fatigue stage of the driver, and thus the driving safety of the vehicle is further improved.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a vehicle control method and apparatus. Background Technology

[0002] Among related technologies, fatigue driving can be identified through visual recognition, control behavior analysis, and wearable physiological monitoring. Visual recognition mainly uses in-vehicle cameras to monitor behaviors such as closing the driver's eyes and yawning to determine the driver's fatigue state. However, this method is not accurate in determining driver fatigue at night, in bright light, or when the driver's face is obscured, or has a high false alarm rate. Control behavior analysis relies on vehicle control information such as steering wheel micro-movement frequency and lane departure to assess attention status, but it can only reflect lost control behavior that has already occurred, resulting in a warning lag. While wearable physiological monitoring devices can obtain high-precision physiological data such as heart rate and EEG, users need to actively wear heart rate belts or EEG caps, which is uncomfortable and difficult to widely apply in daily driving.

[0003] It is evident that fatigue driving detection technologies suffer from low accuracy, delayed warnings, poor comfort, and are not easily adopted or widely used. Summary of the Invention

[0004] This application provides a vehicle control method and apparatus to address the technical problems in related technologies, such as low accuracy, delayed warnings, poor comfort, and difficulty in widespread use of fatigue driving detection.

[0005] This application provides a vehicle control method, comprising: acquiring a driver's breathing data and body sway data, wherein the breathing data and body sway data are determined based on radar data from millimeter-wave radar; determining a comprehensive fatigue score based on the breathing data and body sway data; matching a corresponding fatigue warning strategy based on the comprehensive fatigue score; and controlling the vehicle to execute the fatigue warning strategy. Determining the breathing data and body sway data using radar data collected by millimeter-wave radar, thereby achieving a comprehensive fatigue score, enables non-contact fatigue driving detection with a low interference rate, higher accuracy, better comfort, lower requirements for driver cooperation, easier widespread adoption, and faster response speed. By matching the comprehensive fatigue score to obtain a corresponding fatigue warning strategy and controlling the vehicle to execute the fatigue warning strategy, intervention and timely warnings can be provided at the initial stage of driver fatigue, thereby further improving vehicle driving safety.

[0006] In one embodiment of this application, acquiring the driver's breathing data and body sway data includes: if the vehicle is in a running state and the driver is seated in the driver's position, activating the millimeter-wave radar and emitting electromagnetic waves through the millimeter-wave radar; receiving a first Doppler echo signal generated by the driver's chest rise and fall due to breathing, and determining the breathing data based on the first Doppler echo signal; receiving a second Doppler echo signal generated by the driver's torso displacement due to body sway, and determining the body sway data based on the second Doppler echo signal, wherein the radar data includes the first Doppler echo signal and the second Doppler echo signal. This method can filter out interference signals in the radar data, improving the accuracy of subsequent signal analysis.

[0007] In one embodiment of this application, determining the respiratory data based on the first Doppler echo signal includes: determining a first micro-motion signal matching chest rise and fall based on the first Doppler echo signal; performing Fourier transform processing on the first micro-motion signal to obtain a respiratory signal at a preset frequency at the current detection time; determining the time spectrum based on the respiratory signal to determine the current respiratory frequency; determining the number of breaths at a preset interval based on the current respiratory frequency, and using the number of breaths at the preset interval as the respiratory data. This method can filter out interference signals in radar data, improving the accuracy of subsequent signal analysis.

[0008] In one embodiment of this application, the method further includes: if the number of breaths at the preset interval is less than a preset breath count threshold, starting a timer until the number of breaths at a new preset interval is greater than or equal to the preset breath count threshold; when the timer duration is detected to be greater than a preset duration threshold, generating an early fatigue state reminder, and controlling the vehicle to display the early fatigue state reminder to alert the driver. Through this method, regardless of whether the driver's torso shows obvious signs of fatigue driving, abnormalities can still be detected promptly based on breathing patterns, and early warnings can be issued, largely avoiding the problem of delayed warnings.

[0009] In one embodiment of this application, determining the body sway data based on the second Doppler echo signal includes: determining a second micro-motion signal matching the body sway based on the second Doppler echo signal of the current sliding time window; extracting the key signal temporal features of each key sway point in the second micro-motion signal; determining a three-dimensional spatial coordinate sequence based on the key signal temporal features of a key sway point, and then determining the point acceleration of the key sway point based on the three-dimensional spatial coordinate sequence; determining the driver's overall acceleration based on the point acceleration of all key sway points; determining a body sway index based on the driver's overall acceleration and dynamic pattern confidence coefficient, and using the body sway index as the body sway data. The body sway index is obtained by matching the driver's action type, and the driver's action type is determined based on the second Doppler echo signal of the current sliding time window. This method can quantify the swaying of the torso caused by the movement of the driver's head and neck, facilitating subsequent quantitative assessment of fatigue driving levels and matching appropriate fatigue warning strategies.

[0010] In one embodiment of this application, the determination of the dynamic mode confidence coefficient includes: determining the driver's action type based on the second micro-motion signal; calculating the duration of each type of driver's action within the current sliding time window; determining the driver's action type within the current sliding time window based on a preset driver action type priority and the duration of each type of driver's action; and determining the dynamic mode confidence coefficient based on the correspondence between the driver's action type and the preset driver action type-dynamic mode confidence coefficient. Through this method, complex swaying situations can be determined based on the preset driver action type priority and the duration of each type of driver's action, identifying the main factors affecting fatigue driving detection and the corresponding dynamic mode confidence coefficient values. This makes the body sway index results more accurate and reliable, thereby making the fatigue driving detection results more accurate and reliable.

[0011] In one embodiment of this application, determining a comprehensive fatigue score based on the breathing data and body sway data includes: determining a breathing deviation score based on the breathing data and preset breathing baseline data; determining a body sway intensity score based on the body sway data, preset maximum sway data, and preset minimum sway data; and determining the comprehensive fatigue score based on the breathing deviation score, breathing weight, sway weight, and body sway intensity score. The determination of the breathing weight and sway weight includes any one of the following: using a preset breathing weight as the breathing weight and a preset sway weight as the sway weight; adjusting the preset breathing weight and preset sway weight according to the driver's historical fatigue data to obtain the breathing weight and sway weight; obtaining the current driving scenario of the vehicle and matching the corresponding breathing weight and sway weight according to the current driving scenario. Through the above method, the degree of driver fatigue can be comprehensively evaluated from two dimensions: breathing and body sway, resulting in a comprehensive fatigue score. This makes the comprehensive fatigue score more accurate and reliable, and the weight adjustments can flexibly adapt to possible situations during driving.

[0012] In one embodiment of this application, a fatigue reminder strategy is obtained based on the comprehensive fatigue score, including at least one of the following: if the comprehensive fatigue score is greater than a preset first reminder threshold and the vehicle is in a braking state, the fatigue reminder strategy includes not reminding; if the driver is answering a phone call, the comprehensive fatigue score is reduced by a preset reduction value, and a corresponding fatigue reminder strategy is obtained based on the reduced comprehensive fatigue score; if the current driving environment of the vehicle is a preset driving environment, the comprehensive fatigue score is increased by a preset increase value, and a corresponding fatigue reminder strategy is obtained based on the increased comprehensive fatigue score; if the comprehensive fatigue score is less than or equal to the preset first reminder threshold, the fatigue reminder strategy includes not reminding; if the comprehensive fatigue score is greater than the preset first reminder threshold and the comprehensive fatigue score is less than or equal to a preset second reminder threshold, the fatigue reminder strategy includes voice prompts and / or text prompts, and the preset second reminder threshold is greater than the preset first reminder threshold; if the comprehensive fatigue score is greater than the preset second reminder threshold and the comprehensive fatigue score is less than or equal to a preset third reminder threshold, the fatigue reminder strategy includes at least one of driver's seat vibration prompts and vehicle terminal screen flashing icon prompts, and the preset... The third reminder threshold is greater than the preset second reminder threshold; if the overall fatigue score is greater than the preset third reminder threshold and the overall fatigue score is less than or equal to the preset fourth reminder threshold, the fatigue reminder strategy includes sending a fatigue reminder message to a preset terminal and sending the vehicle's current real-time location to the preset terminal, wherein the preset fourth reminder threshold is greater than the preset third reminder threshold; the overall fatigue score is matched with historical false trigger scores in the historical false trigger score set; if there is a historical false trigger score that is the same as the overall fatigue score, the historical identity corresponding to the historical false trigger score is obtained. The system uses body sway data. If the historical sway data is the same as the current body sway data, it obtains the historical micro-motion signal corresponding to the historical false trigger score. It compares the second micro-motion signal with the historical micro-motion signal. If the similarity between the second micro-motion signal and the historical micro-motion signal is greater than a preset similarity threshold, the fatigue warning strategy includes not issuing a warning. The historical false trigger score is determined based on historical body sway data and historical breathing data. The historical body sway data is determined based on the historical micro-motion signal, which is determined based on the historical second Doppler echo signal generated by the driver's torso displacement caused by body sway. Through this method, corresponding levels of warnings can be executed according to different levels of fatigue, which can minimize the possibility of driver fatigue while avoiding the risk of providing overly aggressive warnings that result in a poor user experience.

[0013] In one embodiment of this application, after controlling the vehicle to execute the fatigue reminder strategy, the method further includes: if the comprehensive fatigue score is greater than a preset fifth reminder threshold, and the number of times the fatigue reminder strategy is executed is greater than a preset number threshold, executing a preset fatigue relief strategy, the preset fatigue relief strategy including at least one of the following: controlling the vehicle to turn on the air conditioning external circulation function; controlling the vehicle to open the windows; if the vehicle is in adaptive cruise control mode, increasing the following distance. Through the above method, the risk of driver fatigue can be reduced by executable means when there is no driver response and the driver is continuously fatigued.

[0014] This application embodiment also provides a vehicle control device, the device comprising: an acquisition module for acquiring a driver's breathing data and body sway data, the breathing data and body sway data being determined based on millimeter-wave radar data; a scoring module for determining a comprehensive fatigue score based on the breathing data and body sway data; and a control module for matching a corresponding fatigue reminder strategy based on the comprehensive fatigue score and controlling the vehicle to execute the fatigue reminder strategy.

[0015] This application also provides an electronic device, including: a memory storing a computer program thereon; and a processor for executing the computer program in the memory to implement the steps of the method described in any of the above embodiments.

[0016] This invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being used to cause a computer to perform the method provided in any of the above embodiments.

[0017] The beneficial effects of this application are as follows: The vehicle control method and device proposed in this application determine breathing data and body sway data through radar data collected by millimeter-wave radar, thereby determining a comprehensive fatigue score. This enables non-contact fatigue driving detection with a low interference rate, higher accuracy, better comfort, lower requirements for driver cooperation, easier widespread adoption, and faster response speed. Based on the comprehensive fatigue score, a corresponding fatigue warning strategy is obtained, and the vehicle is controlled to execute the fatigue warning strategy. Intervention and timely warnings can be provided at the initial stage of driver fatigue, thereby further improving vehicle driving safety. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0019] In the attached diagram:

[0020] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of this application; Figure 2 A schematic flowchart of a vehicle control method provided in an embodiment of this application; Figure 3 A specific flowchart illustrating a vehicle control method provided in an embodiment of this application; Figure 4 A schematic diagram of a vehicle control device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0022] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the shape, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0023] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0024] It should be noted that the collection and processing of data such as radar data in this application must strictly comply with the requirements of relevant national laws and regulations in actual application, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0025] Please see Figure 1 , Figure 1This is a schematic diagram illustrating an application scenario provided by an embodiment of this application. For example... Figure 1 As shown, the vehicle is equipped with a millimeter-wave radar sensor (millimeter-wave radar), a signal processing unit, a logic judgment module, a warning device, a communication module, and a power management system. As an example, the millimeter-wave radar sensor is installed in the steering wheel module to collect radar data from chest rise and fall and head sway. The signal processing unit processes the collected Doppler waves (radar data), filters them, and reduces noise to obtain micro-motion signals. The logic judgment module uses a dynamic reference and coefficient of variation algorithm to determine fatigue driving, obtaining a comprehensive fatigue score and corresponding fatigue warning strategy. The warning device, for example, is integrated into the IVI (In-Vehicle Infotainment) system, triggering audible and indicator light alerts. The communication module provides GPS (Global Positioning System) positioning and 4G / 5G remote communication capabilities. The power management system can be directly powered by the vehicle's power supply, using a low-voltage 12V power supply, supporting long-term low-power operation. All of these modules can be interconnected via a CAN bus, and at least some modules support remote upgrades.

[0026] It should be noted that the above scenario is only an example of an application scenario provided by the embodiments of this application. The method can also be applied to other scenarios according to the user's needs. The embodiments of this application do not limit the actual form of various devices, components, etc. included in the scenario. In the specific application of the solution, it can be set according to actual needs.

[0027] Please see Figure 2 , Figure 2 A schematic flowchart of a vehicle control method provided in one embodiment of this application is shown below. Figure 2 As shown, the method includes the following steps: Step S210: Obtain the driver's breathing data and body sway data.

[0028] Among them, the breathing data and body sway data were determined based on radar data from millimeter-wave radar.

[0029] In some embodiments, acquiring driver breathing data and body sway data includes: if the vehicle is running and the driver is seated in the driver's position, activating the millimeter-wave radar and emitting electromagnetic waves; receiving a first Doppler echo signal generated by the driver's chest rise and fall due to breathing, and determining breathing data based on the first Doppler echo signal; receiving a second Doppler echo signal generated by the driver's torso displacement due to body sway, and determining body sway data based on the second Doppler echo signal, wherein the radar data includes the first Doppler echo signal and the second Doppler echo signal. This method allows for simple and rapid acquisition of evaluation data related to driver fatigue indicators without requiring the driver to wear other detection equipment. Even if the driver's face is obscured due to a hat or other reasons, it does not affect the detection of fatigue driving, enabling non-intrusive and proactive fatigue driving detection.

[0030] As an example, this millimeter-wave radar can be installed in locations such as the steering wheel module. It emits electromagnetic waves that are reflected by the driver's chest, head, shoulders, neck, and torso, allowing the radar to receive data. For instance, after the driver enters the vehicle and starts the engine, the millimeter-wave radar begins operation, emitting electromagnetic waves to collect data on the rise and fall of the driver's chest. The radar then receives the Doppler echo signal (first Doppler echo signal) generated by the chest rise and fall due to breathing, thus completing the acquisition of the breathing signal. Simultaneously, it receives the Doppler echo signal (second Doppler echo signal) generated by the displacement of the driver's torso due to the amplitude and trajectory of the driver's head and shoulders, thus completing the acquisition of the body sway signal.

[0031] Following the above embodiments, determining respiratory data based on the first Doppler echo signal includes: determining a first micro-motion signal matching chest rise and fall based on the first Doppler echo signal; performing Fourier transform processing on the first micro-motion signal to obtain a respiratory signal at a preset frequency at the current detection time; determining the time spectrum based on the respiratory signal to determine the current respiratory frequency; and determining the number of breaths at a preset interval based on the current respiratory frequency to obtain respiratory data. This method can filter out interference signals in the radar data, improving the accuracy of subsequent signal analysis.

[0032] As an example, the method for determining the first micro-motion signal matching the chest rise and fall based on the first Doppler echo signal can be as follows: After the acquisition of the first Doppler echo signal of the time series is completed, the first Doppler echo signal is mixed according to the phase change of the radar echo to obtain I / Q signals (in-phase and quadrature signals); then the I / Q signals are bandpass filtered to retain the normal respiratory frequency band (the normal respiratory rate of an adult at rest is 0.1-0.6 Hz), for example, using a 0.1-0.6 Hz bandpass filter to remove noise, thereby extracting the first micro-motion signal matching the chest rise and fall.

[0033] As an example, the breathing signal can be determined by performing a Fourier transform (STFT) on the first micro-motion signal to calculate the driver's current breathing frequency f_curr (Hz). The specific calculation is as follows: The signal frame in the first micro-motion signal is timed (Based on the current detection time) For example, signals in the vicinity (a period of time before and after the current detection time) Using a window function to truncate the data, perform a Fourier transform: Formula (1), Among them, The raw time-series respiratory rate signal (first micro-motion signal) was collected. The array output by STFT calculation represents time. Respiratory signal at a frequency f in the vicinity (respiratory signal at a preset frequency at the current detection time); It is a complex exponential function that maps the signal to a frequency f (preset frequency), where j is the imaginary unit.

[0034] Calculate the square of the STFT mode as the time spectrum: Formula (2), in, The array output by STFT calculation represents time. The respiratory signal at a frequency f in the vicinity (the respiratory signal at the preset frequency at the current detection time). This represents the time spectrum.

[0035] Will The point where the mid-frequency f is the largest is taken as the current respiratory rate. f curr (Hz) can be understood as taking the maximum frequency value in the time spectrum as the current breathing frequency.

[0036] As an example, a preset interval of 1 minute is used. The specific interval can be set by those skilled in the art as needed. The method for determining the number of breaths during the preset interval is as follows: Formula (3), in, The preset interval for the number of breaths (breaths / minute). f curr (t) The current respiratory rate (breaths / second) is the current respiratory rate at the current detection time t.

[0037] In some embodiments, the method further includes: if the number of breaths at a preset interval is less than a preset breath count threshold, initiating a timer until the number of breaths at a new preset interval is greater than or equal to the preset breath count threshold; when the timer duration is detected to be greater than a preset duration threshold, generating an early fatigue state alert and controlling the vehicle to display the early fatigue state alert to remind the driver. The preset breath count threshold and preset duration threshold can be set by those skilled in the art as needed. This data can be set based on the average value of a population, or it can be personalized according to different drivers.

[0038] Using the above methods, even if the driver's torso does not show obvious signs of fatigue driving, abnormalities can still be detected in time by observing breathing patterns, and early warnings can be issued, avoiding the problem of delayed warnings.

[0039] For example, the system determines the number of breaths at preset intervals in real time. If the number of breaths at a preset interval is less than a preset threshold, a timer is triggered. If the newly determined number of breaths at a preset interval is greater than or equal to the preset threshold, the timer stops. If, during the timing process, the timer duration exceeds a preset threshold, an early fatigue alert is generated. For example, when... If the breathing rate is less than 10 breaths per minute and lasts for more than 60 seconds, the driver is considered to be breathing slowly, and intervention can be made to remind them that this may be an early stage of fatigue.

[0040] Early fatigue warnings can be displayed through sound, in-vehicle interior lights, or by displaying relevant text on a screen. The specific display method can be set by those skilled in the art as needed.

[0041] In some embodiments, determining body sway data based on a second Doppler echo signal includes: determining a second micro-motion signal matching the body sway based on the second Doppler echo signal of the current sliding time window; extracting key signal temporal features of each key sway point in the second micro-motion signal; determining a three-dimensional spatial coordinate sequence based on the key signal temporal features of a key sway point, and then determining the point acceleration of a key sway point based on the three-dimensional spatial coordinate sequence; determining the driver's overall acceleration based on the point acceleration of all key sway points; determining a body sway index based on the driver's overall acceleration and dynamic pattern confidence coefficient to obtain body sway data, wherein the body sway index is obtained by matching the driver's action type, and the driver's action type is determined based on the second Doppler echo signal of the current sliding time window.

[0042] The above method can quantify the swaying of the torso caused by the movement of the driver's head and neck, which facilitates the subsequent quantitative assessment of the degree of driver fatigue and allows for the matching of appropriate fatigue warning strategies.

[0043] As an example, for the second Doppler echo signal that records the phase time series of the swing amplitude and trajectory of the driver's head and neck, the I / Q signal is obtained by mixing, and then bandpass filtering is performed to retain the frequency band of body swaying. For example, a bandpass filter of 0.1-2HZ is used to remove noise, thereby extracting the micro-motion signal (second micro-motion signal) that matches the body swaying.

[0044] Based on the current detection time, the current sliding time window can be determined (which is consistent with the time period of the signal used to collect respiratory data in the aforementioned embodiment), and the second micro-motion signal of the current sliding time window is extracted for further analysis.

[0045] The signal features representing the key swaying points of the upper body (head) are extracted from the second micro-motion signal to obtain the trajectory of the position changing over time (key signal temporal features), and a 3D spatial coordinate sequence is formulated. An example of a 3D spatial coordinate sequence is as follows: Formula (4), in, Let M be the three-dimensional spatial coordinates of the key shaking point at time t. Let M be the x-axis coordinate of a key oscillation point at time t. Let M be the y-coordinate of a key oscillation point at time t. Let T be the z-axis coordinate of a key oscillation point M at time t, where T is the current sliding time window.

[0046] The point acceleration at a critical sway point can be calculated by numerically differentiating the three-dimensional spatial coordinates of that point to obtain its velocity, and then by differentiating again to obtain its point acceleration. Finally, the overall acceleration of the driver is obtained by averaging the point accelerations of all critical sway points.

[0047] As an example, the point acceleration at the key shaking point can be calculated as follows: Formula (5), Formula (6), in, Let M be the velocity of the key shaking point at time t. Let M be the point acceleration at time t, which is the key shaking point. Let be the three-dimensional spatial coordinates at time t. For (t) The three-dimensional spatial coordinates at time ) The time it takes for the body to sway and generate displacement.

[0048] The above calculation of acceleration for a point in the 3D spatial coordinate sequence is completed. Then, the overall acceleration of the driver's upper body (overall acceleration of the driver) is synthesized by using the multi-point averaging method.

[0049] As an example, the overall acceleration of a driver can be determined as follows: Formula (7), in, N represents the overall acceleration of the driver, and N is the number of points used to describe the body contour that produces the displacement (the number of key sway points). Let be the point acceleration of the key shaking point i at time t.

[0050] Then calculate the square of the modulus of the driver's upper body acceleration: Formula (8), in, This is the acceleration vector of the driver's upper body in three-dimensional space (the overall acceleration of the driver). , , The components of the driver's overall upper body acceleration as a function of time along three orthogonal axes.

[0051] The second micro-motion signal is obtained by analyzing the amplitude and trajectory of the driver's head and neck sway (second Doppler echo signal), and then the related acceleration change is calculated and defined as a quantitative evaluation index to obtain the Body Sway Index as body sway data.

[0052] An exemplary method for determining the body sway index is as follows: Formula (9), in, Let be the body sway index at time t, and T be the current sliding time window. This is the acceleration vector of the driver's upper body in three-dimensional space (the overall acceleration of the driver). is the confidence coefficient for the dynamic model.

[0053] As an example, the determination of the dynamic pattern confidence coefficient includes: determining the driver's action type based on the second micro-motion signal; calculating the duration of each type of driver's action within the current sliding time window; determining the driver's action type within the current sliding time window based on the preset driver action type priority and the duration of each type of driver's action; and determining the dynamic pattern confidence coefficient based on the correspondence between the driver's action type and the preset driver action type-dynamic pattern confidence coefficient. Through this method, complex swaying situations can be identified by determining the main factors affecting fatigue driving detection and the corresponding dynamic pattern confidence coefficient values ​​based on the preset driver action type priority and the duration of each type of driver's action. This makes the body sway index results more accurate and reliable, thereby making the fatigue driving detection results more accurate and reliable.

[0054] As an example, the driver action type of the current sliding time window is determined according to the preset driver action type priority and the duration of each type of driver action, including: taking the driver action type with the highest priority among the driver action types with a duration greater than the duration threshold as the driver action type of the current sliding time window.

[0055] As an example, the method for determining the driver's action type based on the second micro-motion signal is as follows: multiple preset driver action types are pre-defined, as well as a standard micro-motion signal (which can be one or more) corresponding to each preset driver action type. The second micro-motion signal is compared with each standard micro-motion signal for similarity. If the similarity between the second micro-motion signal and a standard micro-motion signal is greater than a preset similarity threshold, then the preset driver action type corresponding to the standard micro-motion signal is taken as the current driver action type.

[0056] As an example, the dynamic pattern confidence coefficient can be dynamically assigned based on the driver's action type. For example, driver action types include random head swaying, head nodding left and right, head tilting forward or backward significantly, and sudden head turning. An example of the preset driver action type-dynamic pattern confidence coefficient correspondence is as follows: Scenario 1: Random head movements ; Scenario 2: Head nodding from side to side ; Situation 3: Head tilted forward or backward significantly ; Situation 4: Emergency head lateral turn .

[0057] A higher confidence coefficient for the dynamic pattern indicates a greater likelihood of a correlation between driver fatigue and regular bodily swaying patterns.

[0058] When multiple combinations of the above four scenarios occur within a sliding time window T (e.g., 10 seconds), a strategy prioritizing high-risk actions based on duration is employed for judgment. For example, a preset driver action type priority order from highest to lowest is: sudden head turn, significant head tilt forward or backward, head nodding left or right, and random head swaying. The duration of each of the four action types within the current sliding time window T is then recorded, and actions are judged sequentially from highest to lowest risk. If the duration of a certain action type is greater than or equal to 1 second (or other duration thresholds set by those skilled in the art), then that action is immediately determined as the driver's primary action within that time window. Subsequent lower-risk scenarios are not judged, and the corresponding action is used. Calculate the body sway index within the sliding time window for the confidence coefficient. For example, first determine the duration of the emergency head turn. If it is 0.8s, then determine the duration of the large forward or backward head tilt. If it is 1s, then no further judgment is made, and the driver's action type is determined to be a large forward or backward head tilt, with a corresponding dynamic mode confidence coefficient of 1.

[0059] Step S220: Determine the comprehensive fatigue score based on respiratory data and body swaying data.

[0060] In one embodiment, determining a comprehensive fatigue score based on respiratory data and body swaying data includes: determining a respiratory deviation score based on respiratory data and preset respiratory baseline data; determining a body swaying intensity score based on body swaying data, preset maximum swaying data, and preset minimum swaying data; and determining a comprehensive fatigue score based on the respiratory deviation score, respiratory weight, swaying weight, and body swaying intensity score.

[0061] The intensity score for body swaying motion is greater than or equal to 0 and less than or equal to 1. If the body swaying data is greater than the preset maximum swaying data, the intensity score for body swaying motion is 1; if the body swaying data is less than the preset minimum swaying data, the intensity score for body swaying motion is 0.

[0062] The above method allows for a comprehensive evaluation of driver fatigue across two dimensions: breathing and body swaying, resulting in a comprehensive fatigue score that is more accurate and reliable.

[0063] As an example, the comprehensive fatigue score can be represented by the Comprehensive Fatigue Score (CFS), which is calculated by weighting the driver's breathing rate and body swaying behavior into two dimensions. For example: Formula (10), Formula (11), Formula (12), in, The respiratory deviation score (respiratory data) at time t. To preset respiratory baseline data, for example, you can set... As a benchmark for the number of breaths per minute, The preset interval for the number of breaths at time t is given, and k is a constant, K > 0, used to control the steepness of the curve (for example, k = 0.5 ~ 1.0). Let t be the body sway index (body sway data). To preset the maximum sway data, To preset the minimum sway data, For time t, the pair The intensity score of body swaying motion obtained after normalization (body swaying motion intensity score). The comprehensive fatigue score index at time t. For respiratory weight, For sway weights.

[0064] As an example, It can be used to measure the body sway index, which is obtained by slightly shaking the body in a static state. It can be used to measure the body sway index obtained from violent shaking of the body.

[0065] As an example, if formula (11) is calculated to obtain Then in formula (12) If the formula (11) yields Then in formula (12) That is, if the intensity score of the body swaying motion calculated by formula (11) is less than 0, then when substituted into formula (12), the intensity score of the body swaying motion is 0; if the intensity score of the body swaying motion calculated by formula (11) is greater than 1, then when substituted into formula (12), the intensity score of the body swaying motion is 1. As an example, the initial weight setting... , .

[0066] When BSI(t) ≤ Bmin in formula (11), This situation is possible. It is an empirical statistical value. This is not a mathematical absolute minimum (such as 0), but rather refers to the lower limit of the intensity of natural, slight tremors in a driver's body when they are awake, relaxed, and seated. However, during driving, when a driver is highly focused, the degree of body tremor is very likely to be lower than this. The situation; Similarly, when a driver is in a state of excitement and drives aggressively, the degree of body swaying is very likely to be lower than that of a human. The situation.

[0067] Overall Fatigue Score The higher the value, the more severe the driver's fatigue.

[0068] Following the above embodiments, the determination methods for breathing weights and swaying weights include any one of the following: Use the preset breathing weight as the breathing weight and the preset swaying weight as the swaying weight. See the above for the initial weight settings. Based on the driver's historical fatigue data, the preset breathing weight and preset swaying weight are adjusted respectively to obtain the breathing weight and swaying weight; Obtain the vehicle's current driving scenario, and match the corresponding breathing weight and sway weight based on the current driving scenario.

[0069] To better adapt to situations that may arise during driving, a dynamic weight adjustment mechanism will be considered. , The comprehensive fatigue score is dynamically adjusted during calculation; its value is no longer a fixed value but is dynamically adjusted according to driving conditions. For example: Based on driver behavior pattern learning, the system analyzes historical fatigue data to determine the developmental stages of the driver's fatigue process. If the analysis reveals that breathing accounts for an excessive proportion of the fatigue process, the weighting of breathing is appropriately increased gradually. The weight of the oscillation is reduced to 0.5-0.6, and the oscillation weight is reduced accordingly.

[0070] Adjustments based on different scenario weights: If the current driving scenario is urban traffic congestion, requiring frequent vehicle starts and stops, for example: , The weighting of breathing is equal to the weighting of body swaying. For example, in a long-term high-speed driving scenario: , The weighting is emphasized for body swaying, with swaying weighting greater than breathing weighting. For example, if the current driving scenario is nighttime driving: , The weighting is based on breathing, which is greater than the weighting of swaying.

[0071] Step S230: Based on the comprehensive fatigue score matching, obtain the corresponding fatigue reminder strategy, and control the vehicle to execute the fatigue reminder strategy.

[0072] In one embodiment, a corresponding fatigue reminder strategy is obtained based on a comprehensive fatigue score, including at least one of the following: If the overall fatigue score is greater than the preset first warning threshold and the vehicle is in a braking state, the fatigue warning strategy includes not issuing a warning; If the driver is on the phone, the overall fatigue score will be lowered by a preset reduction value, and a corresponding fatigue reminder strategy will be matched based on the reduced overall fatigue score. If the vehicle's current driving environment is a preset driving environment, the overall fatigue score will be increased by a preset increase value, and a corresponding fatigue reminder strategy will be matched based on the increased overall fatigue score. If the overall fatigue score is less than or equal to the preset first reminder threshold, the fatigue reminder strategy includes not reminding; If the overall fatigue score is greater than the preset first reminder threshold and the overall fatigue score is less than or equal to the preset second reminder threshold, the fatigue reminder strategy includes voice prompts and / or text prompts (such as displaying corresponding text through the screen, head-up display, etc.), and the preset second reminder threshold is greater than the preset first reminder threshold. If the overall fatigue score is greater than the preset second reminder threshold and the overall fatigue score is less than or equal to the preset third reminder threshold, the fatigue reminder strategy includes at least one of the driver's seat vibration reminder and the vehicle terminal screen flashing icon reminder, and the preset third reminder threshold is greater than the preset second reminder threshold. If the overall fatigue score is greater than the preset third reminder threshold and less than or equal to the preset fourth reminder threshold, the fatigue reminder strategy includes sending a fatigue reminder message to a preset terminal and sending the vehicle's current real-time location to the preset terminal. The preset fourth reminder threshold is greater than the preset third reminder threshold. The preset terminal can be the terminal of a pre-defined legal emergency contact, etc., and can be used to issue warnings via emergency SMS, emergency push system messages, etc. It can also be used to share the real-time location so that the emergency contact knows the vehicle's location. As an example, with the user's consent and in accordance with relevant regulations, the current in-vehicle images can be collected and sent to a preset terminal to understand the in-vehicle conditions. The comprehensive fatigue score is matched with the historical false trigger scores in the historical false trigger score set. If a historical false trigger score is the same as the comprehensive fatigue score, the historical body swaying data corresponding to the historical false trigger score is obtained. If the historical swaying data is the same as the body swaying data, the historical micro-motion signal corresponding to the historical false trigger score is obtained. The second micro-motion signal is compared with the historical micro-motion signal. If the similarity between the second micro-motion signal and the historical micro-motion signal is greater than a preset similarity threshold, the fatigue reminder strategy includes not reminding. The historical false trigger score is determined based on the historical body swaying data and historical breathing data. The historical body swaying data is determined based on the historical micro-motion signal. The historical micro-motion signal is determined based on the historical second Doppler echo signal generated by the displacement of the driver's torso caused by the driver's body swaying.

[0073] A tiered early warning and alert mechanism can be established based on the comprehensive fatigue score. For example, the calculated comprehensive fatigue score can be used to... The system is divided into four levels, with the preset first reminder threshold set at 0.3, the preset second reminder threshold at 0.5, the preset third reminder threshold at 0.7, and the preset fourth reminder threshold at 1. Please refer to Table 1 for details. Table 1

[0074] The above method allows for the implementation of appropriate reminders based on different levels of fatigue, which can both mitigate the possibility of driver fatigue and avoid the risk of providing overly aggressive reminders that would result in a poor user experience.

[0075] As an example, when a driver manually turns off a fatigue driving alert after it is triggered, it indicates that the driver is not fatigued. This triggering data is recorded, and subsequent data with the same characteristics will not trigger a fatigue driving alert. Due to individual differences, false alarms may occur. Therefore, optimization can be based on driver feedback. When a fatigue driving alert has been triggered (the fatigue driving alert strategy has been implemented), and the driver manually turns off the alert and explicitly selects "I am not fatigued," the saved data is marked as a false trigger. Subsequent occurrences of the same signal characteristics will not trigger a fatigue driving alert. The comprehensive fatigue score for this instance is recorded in the historical false trigger score set, and the corresponding body sway data constituting this comprehensive fatigue score is stored as historical body sway data. The second micro-motion signal corresponding to this historical body sway data is calculated as the historical micro-motion signal. If the same score is obtained again, first compare the historical swaying data with the body swaying data. If they are different, then proceed with the normal process. If they are the same, then compare the second micro-motion signal with the historical micro-motion signal. If they are similar, then it means that this was also a false trigger and the user was not driving while fatigued. In this case, regardless of which fatigue level the comprehensive fatigue score falls into, the fatigue reminder strategy will be set to no reminder.

[0076] In some embodiments, to fundamentally avoid the system falsely triggering fatigue alerts and causing driver distraction, the fatigue judgment results are corrected based on vehicle driving history information and environmental factors, and a whitelist is introduced to prevent false alarms. For example: When a vehicle is waiting at a traffic light, if the vehicle is stationary and the brake pedal is continuously depressed (in a braking state), even if the body swaying and breathing rate are abnormal (the overall fatigue score is greater than the preset first warning threshold), the fatigue warning will not be triggered. When a driver is on the phone, the overall fatigue score should be appropriately lowered to prevent fatigue alerts from being triggered directly. When driving in preset driving environments such as nighttime, steep slopes, mountain roads, and rainy conditions, the overall fatigue score should be appropriately increased to improve alarm sensitivity.

[0077] The preset decrease value and preset increase value can be set by those skilled in the art as needed. The preset increase value can be a fixed value or different values ​​can be set according to different preset driving environments.

[0078] In one embodiment, after controlling the vehicle to execute the fatigue reminder strategy, the method further includes: if the overall fatigue score is greater than a preset fifth reminder threshold, and the number of times the fatigue reminder strategy is executed is greater than a preset number threshold, executing a preset fatigue relief strategy, wherein the preset fatigue relief strategy includes at least one of the following: Control the vehicle to activate the air conditioning's external air circulation function; Control the vehicle to open the windows; If the vehicle is in adaptive cruise control mode, increase the following distance.

[0079] The preset fifth reminder threshold is higher than the preset first reminder threshold. The preset fifth reminder threshold can be set as needed, such as setting a corresponding value based on different preset fatigue mitigation strategies. This method allows for the effective reduction of the risk of driver fatigue even when there is no driver response and the driver remains fatigued. For a better user experience, the user can be reminded of the upcoming preset fatigue mitigation strategy via voice, text, or flashing icons before the strategy is executed.

[0080] As an example, a mechanism can be set up to handle situations after a fatigue warning is triggered. When a driver is determined to be severely fatigued and multiple warnings are ineffective, the corresponding vehicle function will be triggered. The vehicle automatically switches to external air circulation to reduce the carbon dioxide concentration inside the vehicle. The vehicle automatically opens the driver's side window to improve air circulation inside the vehicle and stimulate the driver's alertness. If a fatigue warning is triggered during adaptive cruise control, the following distance will be automatically increased.

[0081] Please see Figure 3 , Figure 3 A specific flowchart illustrating a vehicle control method provided in an embodiment of this application is shown below. Figure 3As shown, the method includes the following steps: After starting the detection, respiratory and body sway data are collected (radar data acquisition). If radar data is collected, the respiratory rate is extracted, the number of breaths RR(t) at a preset interval is calculated, the body sway index BSI(t) is calculated, and then the comprehensive fatigue score (Comprehensive Fatigue Score) CFS is calculated. If CFS is greater than 0.3, a fatigue alert is triggered. If CFS ∈ [0.3, 0.5), it is considered mild fatigue; if CFS ∈ [0.5, 0.7), it is considered moderate fatigue; and if CFS ∈ [0.7, 1], it is considered severe fatigue. Different fatigue alert strategies are set according to different fatigue levels. Otherwise, if CFS is less than or equal to 0.3, the respiratory and body sway data acquisition steps are repeated.

[0082] The vehicle control method provided in the above embodiments determines breathing and body sway data using radar data collected by millimeter-wave radar, thereby determining a comprehensive fatigue score. This enables non-contact fatigue driving detection with a lower interference rate, greater accuracy, better comfort, lower requirements for driver cooperation, easier widespread adoption, and faster response. Based on the comprehensive fatigue score, a corresponding fatigue warning strategy is matched, and the vehicle is controlled to execute the fatigue warning strategy. Intervention and timely warnings can be provided at the initial stage of driver fatigue, further improving vehicle driving safety.

[0083] Utilizing millimeter-wave radar for detection achieves non-contact monitoring, eliminating the need for drivers to wear detection equipment and monitoring without affecting their driving. Secondly, the monitoring effect is significant; millimeter-wave radar can continuously monitor even in complex environments such as nighttime driving, strong sunlight, and drivers wearing masks, demonstrating high applicability. In practical vehicle deployment, it can be deployed on existing terminals without additional wiring design, resulting in high adoption rates. Compared to visual monitoring technologies like DMS (Driver Monitoring System), it offers longer early warning times. A multi-level warning mechanism is introduced to reduce traffic accidents caused by driver fatigue. By using driver breathing rate and body sway as influencing factors and incorporating a weighted formula, the degree of driver fatigue is determined. Dynamic weights are introduced during the judgment process to support personalized learning, reducing variability caused by individual driver differences. Using Doppler signals collected by millimeter-wave radar to monitor the driver's chest rise and fall to reflect breathing rate and head sway for fatigue warning, it features non-contact monitoring, low interference rate, and fast response speed. By extracting breathing frequency and head movement characteristics, the system can intervene and provide alerts in the early stages of driver fatigue, along with remote warnings, thereby improving vehicle driving safety.

[0084] In one embodiment, a vehicle control device is provided for executing the vehicle control method provided in any of the above embodiments. See also... Figure 4 , Figure 4 A schematic diagram of a vehicle control device provided in an embodiment of this application is shown below. Figure 4 As shown, the vehicle control device 500 includes: an acquisition module 410 for acquiring the driver's breathing data and body sway data, wherein the breathing data and body sway data are determined based on radar data from millimeter-wave radar; a scoring module 420 for determining a comprehensive fatigue score based on the breathing data and body sway data; and a control module 430 for matching a corresponding fatigue reminder strategy based on the comprehensive fatigue score and controlling the vehicle to execute the fatigue reminder strategy.

[0085] For specific limitations regarding the vehicle control device, please refer to the limitations on the vehicle control method above, which will not be repeated here. 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 hardware or independently of the processor in the electronic device, or stored in software in the memory of the electronic device, so that the processor can call and execute the operations corresponding to each module.

[0086] In this embodiment, the vehicle control device is essentially equipped with multiple modules to execute the vehicle control method in any of the above embodiments. The specific functions and technical effects can be referred to in the above embodiments, and will not be repeated here.

[0087] See Figure 5 , Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown below. Figure 5 As shown, this embodiment of the invention also provides an electronic device 500, including a processor 501, a memory 502, and a communication bus 503; the communication bus 503 is used to connect the processor 501 and the memory 502; the processor 501 is used to execute a computer program stored in the memory 502 to implement the method described in any of the above embodiments. As an example, the electronic device may be a vehicle infotainment system or similar device.

[0088] This invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being used to cause a computer to perform the method provided in any of the above embodiments.

[0089] This application also provides a non-volatile readable storage medium storing one or more modules (programs) that, when applied to a device, enable the device to execute the instructions included in the steps provided in this application.

[0090] This application also provides a computer program product, including a computer program that, when executed by a processor, can implement the steps and corresponding content of the aforementioned method embodiments.

[0091] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0092] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0093] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0094] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0095] It should be understood that the terms "first," "second," etc., used in this application are used to distinguish similar objects and do not necessarily indicate a specific order or sequence. The technical features to which these terms are used can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in a sequence other than that shown in the figures or text.

[0096] It should be understood that although the flowcharts provided in the embodiments of this application indicate the various steps with arrows, the order indicated by the arrows does not necessarily limit the implementation order of these steps. Those skilled in the art can perform these steps in other orders according to different implementation scenarios and requirements.

[0097] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A vehicle control method, characterized in that, The method includes: Acquire the driver's breathing data and body sway data, which are determined based on millimeter-wave radar data; A comprehensive fatigue score is determined based on the respiratory data and body swaying data. Based on the comprehensive fatigue score, a corresponding fatigue reminder strategy is obtained, and the vehicle is controlled to execute the fatigue reminder strategy.

2. The vehicle control method as described in claim 1, characterized in that, Acquire the driver's breathing and body sway data, including: If the vehicle is in the starting state and the driver is in the driver's seat, the millimeter-wave radar is activated and electromagnetic waves are emitted through the millimeter-wave radar. Receive the first Doppler echo signal generated by the driver's chest rise and fall due to breathing, and determine the breathing data based on the first Doppler echo signal; The radar receives a second Doppler echo signal generated by the displacement of the driver's torso due to the driver's body swaying, and determines the body swaying data based on the second Doppler echo signal. The radar data includes the first Doppler echo signal and the second Doppler echo signal.

3. The vehicle control method as described in claim 2, characterized in that, The respiratory data is determined based on the first Doppler echo signal, including: The first micro-motion signal matching the chest movement is determined based on the first Doppler echo signal; The first micro-motion signal is subjected to Fourier transform processing to obtain the respiratory signal at a preset frequency at the current detection time; The respiratory signal is used to determine the time spectrum, and then the current respiratory rate is determined. The number of breaths at a preset interval is determined based on the current respiratory rate, and the number of breaths at the preset interval is used as the respiratory data.

4. The vehicle control method as described in claim 3, characterized in that, The method further includes: If the number of breaths at the preset interval is less than the preset number of breaths threshold, start timing until the number of breaths at the new preset interval is greater than or equal to the preset number of breaths threshold. When the timed duration is detected to exceed a preset time threshold, an early fatigue state reminder is generated, and the vehicle is controlled to display the early fatigue state reminder to alert the driver.

5. The vehicle control method as described in claim 2, characterized in that, Determining the body sway data based on the second Doppler echo signal includes: The second micro-motion signal matching the body sway is determined based on the second Doppler echo signal within the current sliding time window; Extract the key signal timing features of each key shaking point in the second micro-motion signal; A three-dimensional spatial coordinate sequence is determined based on the key signal timing characteristics of a key shaking point, and then the point acceleration of the key shaking point is determined based on the three-dimensional spatial coordinate sequence. The driver's overall acceleration is determined based on the point accelerations at all key points of swaying. The body sway index is determined based on the driver's overall acceleration and dynamic pattern confidence coefficient. The body sway index is used as the body sway data. The body sway index is obtained by matching the driver's action type. The driver's action type is determined based on the second Doppler echo signal of the current sliding time window.

6. The vehicle control method as described in claim 5, characterized in that, The methods for determining the confidence coefficient of the dynamic pattern include: The driver's action type is determined based on the second micro-motion signal; Calculate the duration of each type of driver action within the current sliding time window; The driver action type of the current sliding time window is determined based on the preset driver action type priority and the duration of each type of driver action. The dynamic mode confidence coefficient is determined based on the driver action type and the correspondence between the preset driver action type and the dynamic mode confidence coefficient.

7. The vehicle control method according to any one of claims 1-6, characterized in that, A comprehensive fatigue score is determined based on the respiratory data and body sway data, including: A respiratory deviation score is determined based on the respiratory data and preset respiratory baseline data. The intensity score of the body swaying motion is determined based on the body swaying data, the preset maximum swaying data, and the preset minimum swaying data; The comprehensive fatigue score is determined based on the breathing deviation score, breathing weight, swaying weight, and body swaying intensity score. The determination methods for the breathing weight and swaying weight include any one of the following: The preset breathing weight is used as the breathing weight, and the preset swaying weight is used as the swaying weight; The preset breathing weight and preset swaying weight are adjusted according to the driver's historical fatigue data to obtain the breathing weight and swaying weight respectively; The current driving scenario of the vehicle is obtained, and the corresponding breathing weight and sway weight are matched according to the current driving scenario.

8. The vehicle control method according to any one of claims 1-6, characterized in that, Based on the comprehensive fatigue score, a corresponding fatigue reminder strategy is obtained, including at least one of the following: If the overall fatigue score is greater than a preset first warning threshold and the vehicle is in a braking state, the fatigue warning strategy includes not issuing a warning; If the driver is on the phone, the overall fatigue score is reduced by a preset reduction value, and a corresponding fatigue reminder strategy is obtained based on the reduced overall fatigue score. If the current driving environment of the vehicle is a preset driving environment, the comprehensive fatigue score is increased by a preset increase value, and a corresponding fatigue reminder strategy is obtained based on the increased comprehensive fatigue score. If the overall fatigue score is less than or equal to a preset first reminder threshold, the fatigue reminder strategy includes not issuing a reminder; If the overall fatigue score is greater than the preset first reminder threshold and the overall fatigue score is less than or equal to the preset second reminder threshold, the fatigue reminder strategy includes voice prompts and / or text prompts, and the preset second reminder threshold is greater than the preset first reminder threshold; If the overall fatigue score is greater than the preset second reminder threshold, and the overall fatigue score is less than or equal to the preset third reminder threshold, the fatigue reminder strategy includes at least one of driver's seat vibration prompt and vehicle terminal screen flashing icon prompt, and the preset third reminder threshold is greater than the preset second reminder threshold; If the overall fatigue score is greater than the preset third reminder threshold and the overall fatigue score is less than or equal to the preset fourth reminder threshold, the fatigue reminder strategy includes sending a fatigue reminder message to a preset terminal and sending the vehicle's current real-time location to the preset terminal, wherein the preset fourth reminder threshold is greater than the preset third reminder threshold; The comprehensive fatigue score is matched with historical false trigger scores in the historical false trigger score set. If a historical false trigger score is the same as the comprehensive fatigue score, the historical body swaying data corresponding to the historical false trigger score is obtained. If the historical swaying data is the same as the body swaying data, the historical micro-motion signal corresponding to the historical false trigger score is obtained. The second micro-motion signal is compared with the historical micro-motion signal. If the similarity between the second micro-motion signal and the historical micro-motion signal is greater than a preset similarity threshold, the fatigue reminder strategy includes not reminding. The historical false trigger score is determined based on historical body swaying data and historical breathing data. The historical body swaying data is determined based on historical micro-motion signals. The historical micro-motion signals are determined based on historical second Doppler echo signals generated by the displacement of the driver's torso due to the driver's body swaying.

9. The vehicle control method according to any one of claims 1-6, characterized in that, After controlling the vehicle to execute the fatigue warning strategy, the method further includes: If the overall fatigue score is greater than a preset fifth reminder threshold, and the number of times the fatigue reminder strategy is executed is greater than a preset number threshold, a preset fatigue relief strategy is executed. The preset fatigue relief strategy includes at least one of the following: Control the vehicle to activate the air conditioning external circulation function; Control the vehicle to open its windows; If the vehicle is in adaptive cruise control mode, increase the following distance.

10. A vehicle control device, characterized in that, The device includes: The acquisition module is used to acquire the driver's breathing data and body sway data, which are determined based on the radar data of the millimeter-wave radar; The scoring module is used to determine a comprehensive fatigue score based on the breathing data and body swaying data; The control module is used to obtain a corresponding fatigue reminder strategy based on the comprehensive fatigue score and control the vehicle to execute the fatigue reminder strategy.

Images