Redundancy confirmation and safety intervention method and system for sudden faint of driver

Abstract

The invention belongs to the technical field of intelligent driving, and provides a redundancy confirmation and safety intervention method and system for sudden faint of a driver. According to the method, suspected faint is monitored and judged based on the physiological or visual state of a driver; then triggering a vehicle micro-intervention mode, and exploring conscious feedback of a driver by applying hidden physical stimulation; and finally, whether a security intervention plan is started or not is judged according to whether effective active feedback is monitored or not. By introducing a micro intervention-feedback confirmation man-machine interaction verification mechanism, a single signal criterion is upgraded into two-way behavior confirmation, so that false triggering can be avoided to the greatest extent while effective treatment of real faint is guaranteed, the reliability and user acceptability of the system are remarkably improved, and the user experience is improved. The problems of safety risk and poor user experience caused by direct triggering of emergency braking by misjudgment in the prior art are effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving, more particularly, to a redundancy confirmation and safety intervention method and system for sudden syncope of a driver. BACKGROUND

[0002] In the field of automotive active safety, especially for monitoring and responding to sudden health conditions of drivers (such as syncope), existing technologies have proposed various solutions. These solutions usually rely on real-time monitoring of physiological signals of drivers (such as heart rate, brain waves) or visual behaviors (such as facial orientation, eyelid opening). The basic logic is: once the sensor detects abnormal signals that meet the preset danger criteria, the system will directly trigger the highest level of safety measures, such as starting the vehicle's emergency automatic braking system (AEB), or automatically calling emergency rescue services. The original intention of this design is to intervene as soon as possible when real danger occurs, in order to protect the life safety of the driver.

[0003] However, the above-mentioned "detection-intervention" aggressive strategy has exposed serious defects and risks in practice. First, the consequences of false triggering are extremely serious. Due to temporary interference of sensors, sudden changes in light, or factors such as the driver just closing his eyes for a short rest, false judgments of the system are difficult to completely avoid. If false judgment occurs while driving at high speed, sudden emergency braking can easily cause fatal rear-end collisions, creating even greater safety hazards. Second, poor user experience leads to system being disabled. Overly sensitive and frequent false alarms can seriously interfere with normal driving experience, causing drivers to become annoyed and frustrated, leading them to manually turn off the entire safety system, rendering the advanced protection function useless. Finally, the confidence of the decision-making basis is insufficient. Making the highest risk decision based on the instantaneous abnormality of a single type of sensor signal lacks cross-validation, and the reliability of the judgment is questionable.

[0004] In summary, the core problem of existing technologies is the lack of a buffer and confirmation mechanism between preliminary warning and final intervention, which cannot effectively distinguish between real syncope and temporary abnormality. Therefore, there is an urgent need in the field for a new safety system that can significantly reduce false positives, avoid secondary accidents, and improve user trust and acceptance. SUMMARY

[0005] The present application provides a redundancy confirmation and safety intervention method and system for sudden syncope of a driver, which introduces a redundancy confirmation mechanism based on micro-intervention actions, solving the technical problems of false triggering of emergency braking, poor user experience, and insufficient decision-making confidence caused by temporary physiological abnormalities of drivers or sensor interference in existing technologies.

[0006] According to a first aspect of the present application, a method for driver sudden faintness oriented redundancy confirmation and safety intervention is provided, comprising: S1, determining that the driver is in a suspected faintness state based on monitoring of the physiological state and / or visual state of the driver; S2, triggering a vehicle micro-intervention mode based on the suspected faintness state, and monitoring active feedback of the driver to the micro-intervention mode; S3, determining whether to execute a safety intervention plan based on the monitoring result of the active feedback.

[0007] On the basis of the above technical solution, the present application can also be improved as follows.

[0008] Optionally, step S1 comprises: monitoring at least one vital sign signal of the driver through a physiological sensor, and / or monitoring the facial posture and eye state of the driver through a visual sensor; when the monitoring data exceeds a preset threshold and lasts for a certain time, determining that the driver is in the suspected faintness state.

[0009] Optionally, in step S2, the triggering of the vehicle micro-intervention mode comprises: controlling the vehicle to generate periodic steering wheel micro-vibration; and controlling the vehicle to apply a transient point brake to a single front wheel on the side opposite to the driving direction.

[0010] Optionally, the controlling of the vehicle to generate periodic steering wheel micro-vibration comprises: calculating the micro-vibration amplitude of the steering wheel based on the driving state of the vehicle using an amplitude adaptive algorithm; controlling the vehicle to generate periodic steering wheel micro-vibration based on the micro-vibration amplitude.

[0011] Optionally, the controlling of the vehicle to apply a transient point brake to a single front wheel on the side opposite to the driving direction comprises: selecting a single front wheel on the side opposite to the driving direction, the selection rule being: if the vehicle is driving straight, randomly selecting the left front wheel or the right front wheel; if the vehicle is slightly turning left, selecting the right front wheel; if the vehicle is slightly turning right, selecting the left front wheel; applying a braking force to the selected front wheel to realize the transient point brake based on the driving state of the vehicle, the braking force being determined based on the vehicle mass, the vehicle speed and the tire rolling radius of the selected front wheel.

[0012] Optionally, in step S2, the monitoring of the active feedback of the driver to the micro-intervention mode comprises: monitoring the counteractive force moment on the steering wheel by torque sensor continuously within a preset time window of the micro-intervention mode; and / or, monitoring the vehicle acceleration or yaw rate change caused by human operation by vehicle body inertial measurement unit (IMU).

[0013] Optionally, step S3 comprises: adopting a data fusion algorithm to comprehensively judge the torque sensor monitoring signal and the vehicle body inertial measurement unit (IMU) monitoring signal to confirm whether the active feedback exists; if the effective active feedback is monitored within the preset time window, it is determined as false triggering, the system exits silently, and the event data is recorded for algorithm optimization; if the effective active feedback is not monitored within the preset time window, it is determined as confirmed syncope, and the safety intervention plan is triggered.

[0014] Optionally, the safety intervention plan comprises: determining the nearest safe parking point, generating a safe parking trajectory from the current position to the nearest safe parking point, and smoothly and safely parking along the safe parking trajectory by auxiliary driving; and, turning on the light prompt; automatically calling for rescue and sending the vehicle position.

[0015] Optionally, the method further comprises: S4, establishing a driver personalized behavior model by long-term monitoring of the driver's operation data in normal state, and regularly updating the driver personalized behavior model to adapt to the change of driver habits; dynamically adjusting the threshold for determining the suspected syncope state and / or the sensitivity for monitoring the active feedback based on the parameter change of the driver personalized behavior model.

[0016] According to the second aspect of the present application, a redundant confirmation and safety intervention system for sudden syncope of a driver is provided, comprising: a crisis determination module for determining that the driver is in a suspected syncope state based on monitoring of the physiological state and visual state of the driver; a redundant confirmation executor connected with the crisis determination module and the electric power steering system (EPS) and electronic stability program (ESP) of the vehicle, for triggering a vehicle micro-intervention mode after receiving the suspected syncope state signal, including controlling the electric power steering system (EPS) to perform a steering wheel micro-vibration action, and controlling the electronic stability program (ESP) to perform a transient point brake action on a single front wheel; The response detection module is connected with the steering wheel torque sensor and an inertial measurement unit (IMU) of the vehicle body, and is configured to monitor the operation torque of the driver on the steering wheel and the change in the motion state of the vehicle within a preset time window after the micro-intervention mode is executed, so as to determine whether the active feedback of the driver exists; The safety intervention module is connected with the response detection module, a driving assistance system of the vehicle and an emergency call system, and is configured to make a decision based on the monitoring result of the active feedback: if the active feedback is not monitored, it is determined that the syncope is confirmed, and a safety intervention plan is executed to control the driving assistance system to perform safe parking, and the emergency call system is controlled to automatically call for rescue.

[0017] According to a third aspect of the present application, an electronic device is provided, comprising a memory and a processor, the processor being configured to implement the steps of the method for redundant confirmation and safety intervention for sudden syncope of a driver when executing a computer management program stored in the memory.

[0018] According to a fourth aspect of the present application, a computer readable storage medium is provided, having a computer management program stored thereon, the computer management program being configured to implement the steps of the method for redundant confirmation and safety intervention for sudden syncope of a driver when executed by a processor.

[0019] The method, system, electronic device and storage medium for redundant confirmation and safety intervention for sudden syncope of a driver provided by the present application construct a three-level safety mechanism of “suspected early warning-redundant verification-grading decision”, and upgrade the traditional one-way criterion to a two-way confirmation logic by introducing a micro-intervention action as a human-computer interaction medium. Specifically, the system first identifies the risk through physiological / visual signals, then triggers the instinctive reaction of the driver through implicit stimuli such as steering wheel micro-vibration and single-wheel point braking, and finally determines whether to start the highest level of protection according to whether there is conscious feedback. The redundant confirmation strategy of the present application has multiple technical effects: first, the false trigger risk is transferred from the sensor signal level to the behavior feedback level, greatly reducing the false alarm rate; second, the driving continuity is ensured through non-startling interaction, improving the user trust; third, a complete criterion of “unconscious physiological abnormality + conscious behavior loss” is formed, so that the safety intervention decision has higher confidence. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 A method flow chart for redundant confirmation and safety intervention for sudden syncope of a driver provided by an embodiment of the present application; Figure 2 A method flow chart for redundant confirmation and safety intervention for sudden syncope of a driver provided by another embodiment of the present application; Figure 3A redundancy confirmation and safety intervention system for sudden fainting of a driver provided by the present application has a block diagram as shown in the figure; Figure 4 A possible hardware structure schematic diagram of an electronic device provided by the present application has a block diagram as shown in the figure; Figure 5 A possible hardware structure schematic diagram of a computer readable storage medium provided by the present application has a block diagram as shown in the figure. DETAILED DESCRIPTION

[0021] The specific embodiments of the present application are described in further detail below in combination with the accompanying drawings and examples. The following examples are used to illustrate the present application, but are not used to limit the scope of the present application.

[0022] In the embodiments of the present application, when collecting, processing and storing the personal information (such as images, behavioral characteristics, etc.) of the user, the implementation of the technical solution strictly follows the principles of legality, legitimacy and necessity and the core rules of “notification-agreement”. Specifically, before information collection, the system explicitly notifies the user of the purpose, method, scope and use rules of information collection through the interactive interface, and needs to obtain the active authorization and agreement of the user; the entire information processing process uses data encryption, access control and other technical measures to ensure information security, and sets up a mechanism for the user to exercise his rights (such as query, correction, withdrawal of consent and deletion of information). For the exceptions stipulated by law (such as necessary for the performance of statutory duties or response to public health emergencies), its application is strictly limited within the scope and limits authorized by law, and the technical solution does not contain any content that violates the law, social morals or harms the public interest.

[0023] Figure 1 A redundancy confirmation and safety intervention method flow chart for sudden fainting of a driver provided by the present application has a flow chart as shown in the figure, Figure 1 The method comprises steps S1-S3: S1, determining that the driver is in a suspected fainting state based on monitoring of the physiological state and / or visual state of the driver; S2, triggering a vehicle micro-intervention mode based on the suspected fainting state, and monitoring the active feedback of the driver to the micro-intervention mode; S3, judging whether to execute a safety intervention plan based on the active feedback monitoring result.

[0024] It can be understood that, based on the defects in the background art, the embodiments of the present application propose a redundancy confirmation and safety intervention method for sudden syncope of a driver. The method first performs preliminary risk screening through a physiological or visual sensor, and when an abnormality is found, instead of directly taking emergency measures, a set of precisely calculated, slight physical stimuli, i.e., "micro intervention", is applied through a vehicle execution system (EPS / ESP), which is essentially an interactive signal that needs to be consciously responded by the driver; the system then monitors in real time whether the driver produces conscious control feedback (such as reverse correction of the steering wheel) to this signal, and finally uses the active feedback result as a decisive criterion: if there is active feedback, it is determined as a false alarm and the system exits; if there is no feedback, it is confirmed that the driver is in syncope and the system intervenes. The present application changes the traditional one-way "monitoring-action" mode into a man-machine cooperative closed loop of "monitoring-interaction verification-decision", and uses the instinctive behavior feedback of the driver as the highest level of confirmation evidence, thereby avoiding the risk of directly triggering aggressive braking due to sensor misjudgment.

[0025] The present application first transfers the system false alarm risk from the sensor level to the behavior feedback level by introducing the micro intervention mechanism as a buffer, effectively avoiding the secondary accidents that may be caused by directly triggering emergency braking due to misjudgment; secondly, the non-invasive micro-stimulation method (such as slight steering wheel vibration and single front wheel point braking) is used for interaction verification, which can effectively wake up the distracted driver without causing driving interference, significantly improving user acceptance and system trust; finally, through the double verification logic of "physiological signal abnormality + lack of behavior feedback", a complete decision evidence chain is constructed, so that the final safety intervention decision has higher confidence, realizing the unity of safety and reliability.

[0026] On the basis of the above technical solutions, the embodiments of the present application can also be improved as follows.

[0027] In one possible embodiment, step S1 includes: monitoring at least one vital sign signal of the driver through a physiological sensor, and / or monitoring the facial posture and eye state of the driver through a visual sensor; when the monitoring data exceeds a preset threshold and lasts for a certain time, the suspected syncope state is determined.

[0028] For example, when the system is running, the physiological sensor (such as a heart rate monitoring module integrated on the steering wheel) continuously detects that the driver's heart rate drops from 75 beats per minute to 40 beats per minute within 3 seconds, and the heart rate mutation value exceeds the preset threshold; at the same time, the visual sensor (such as the in-vehicle camera) monitors that the driver's facial posture is continuously leaning back, and the eye state is continuously closed for 3 seconds, and this behavior lasts for a certain length of time. At this time, the system fuses the abnormal signals from the two different sources, and since the monitoring data meets the two conditions of "exceeding the preset threshold" and "lasting for a certain time", the system determines that the driver is in a suspected syncope state, and triggers the subsequent micro-intervention action.

[0029] The embodiment embodies multi-modal perception and continuous time judgment, avoids false judgment of a single signal caused by normal behaviors such as the driver sneezing or briefly looking at the mobile phone, and thus improves the accuracy of preliminary determination, and provides a reliable starting point for the subsequent redundant confirmation process.

[0030] In a possible embodiment, in step S2, the vehicle micro-intervention mode is triggered, including: controlling the vehicle to generate periodic steering wheel micro-vibration; and controlling the vehicle to apply a transient point brake to a single front wheel on the side opposite to the driving direction.

[0031] In the embodiment, when it is determined that the driver is in a suspected syncope state, the system does not immediately brake or alarm, but performs a periodic steering wheel left-right micro-vibration with a short duration (for example, 1-2 seconds) and a very small amplitude (less than 3 newtons) through the electric power steering system (EPS) of the vehicle, simulating a road abnormal touch feeling similar to "pressing to a slight deceleration belt". At the same time, a very slight (for example, brake force less than 50 newtons) transient point brake is applied to a single front wheel through the electric power steering system (EPS), generating an imperceptible vehicle body jerk, and the two methods together constitute a physical stimulus sufficient to awaken a distracted driver without significantly interfering with the vehicle trajectory.

[0032] It can be understood that the embodiment forms a composite tactile stimulus by combining the micro-intervention (steering wheel micro-vibration and single-wheel point brake), avoids the risk that a single stimulus can be ignored, and ensures the continuity of the driving experience by minimizing the intervention intensity, to trigger the driver's instinctive reaction by implicit physical feedback. The embodiment balances safety and driving experience without significantly interfering with the vehicle trajectory, effectively awakening a distracted driver while minimizing the secondary risk of false triggering.

[0033] In a possible embodiment, the control of the vehicle to generate periodic steering wheel micro-vibration includes: (1) Based on the vehicle driving state, the amplitude adaptive algorithm is used to calculate the steering wheel micro-vibration amplitude. The amplitude adaptive algorithm is expressed as: A = A_base × (1 + k_v × v / v_max) Wherein: A is the calculated micro-vibration amplitude; A_base is the base amplitude, which is preferably 1.5 Nm in this embodiment; k_v is the vehicle speed influence coefficient, which is preferably 0.2 in this embodiment; v is the current vehicle speed; v_max is the maximum supported vehicle speed of the system.

[0034] (2) Based on the micro-vibration amplitude, the vehicle generates periodic steering wheel micro-vibration.

[0035] The steering wheel micro-vibration control algorithm is expressed as: T_vibration(t) = A × sin(2π × f × t) × w(t) Wherein: A is the micro-vibration amplitude coefficient, which is preferably in the range of 1-3 Nm in this embodiment and is adjusted adaptively according to the vehicle speed; f is the vibration frequency, which is preferably in the range of 2-5 Hz in this embodiment to avoid resonance with the inherent frequency of the vehicle; w(t) is a window function to ensure smooth start and stop of vibration.

[0036] It can be understood that the steering wheel micro-vibration control strategy of this embodiment can avoid unnecessary disturbance due to excessive intervention when the vehicle is driving at low speed, and can ensure the effectiveness of interaction by moderately enhancing the stimulation when the vehicle is driving at high speed, thereby achieving an optimal balance between safety and comfort in the full speed range.

[0037] In one possible embodiment, the control vehicle applies a transient point brake to a single front wheel on the side opposite to the driving direction, comprising: (1) Select a single front wheel on the side opposite to the driving direction of the vehicle, the judgment rule is: If the vehicle is driving straight, randomly select the left front or right front wheel; If the vehicle is slightly turning left, select the right front wheel; If the vehicle is slightly turning right, select the left front wheel; (2) Based on the vehicle driving state, a braking force is applied to the selected front wheel to realize a transient point brake, and the braking force is determined based on the vehicle mass, vehicle speed and tire rolling radius of the selected front wheel.

[0038] More specifically, the braking force calculation formula is: F brake = min(50N, 0.02 x m x v 2 / R) wherein: F brake is the braking force applied to the selected wheel; m is the mass of the vehicle; v is the current vehicle speed; R is the rolling radius of the selected wheel.

[0039] For example, in a certain implementation scenario, when the system determines that it is necessary to trigger a micro intervention, the vehicle is in a straight driving state on a highway. According to the judgment rule "if the vehicle is driving straight, then randomly select the left front or right front wheel", the system randomly selects the left front wheel as the target for this time.

[0040] Subsequently, the system calculates the braking force according to the formula F brake = min(50N, 0.02 x m x v 2 / R). Assuming the mass of the vehicle m is 1500 kg, the current vehicle speed v is 100 km / h (about 27.8 m / s), and the rolling radius of the tire R is 0.3 meters. After calculation, 0.02 x 1500 x (27.8)2 / 0.3 ≈ 77292N, which is much larger than 50N, so the system finally applies a braking force of 50N. This small force is enough to produce a momentary slight resistance on the left front wheel, causing the vehicle body to produce an imperceptible jerk, but will not significantly affect the straight driving trajectory of the vehicle.

[0041] This implementation scenario fully demonstrates the complete process from "target selection" to "force control". Through precise and controllable single-wheel braking, effective somatosensory stimulation is achieved while the safety of micro intervention is maximized.

[0042] In one possible implementation mode, in step S2, the monitoring of the driver's active feedback to the micro intervention mode includes: monitoring the artificially applied reverse correction torque on the steering wheel through the torque sensor continuously within the preset time window (e.g. 3-5 seconds) of the micro intervention mode; and / or, monitoring the vehicle acceleration or yaw rate changes caused by human operation through the vehicle body inertial measurement unit (IMU).

[0043] For example, (1) the torque response detection algorithm is represented as: if |T driver (t) - T driver (t0)|>T threshold for At>0.5s then the steering wheel feedback is detected wherein: T_driver(t) is the real-time driver torque, i.e. the torque applied by the human on the steering wheel continuously monitored by the torque sensor; T_driver(t0) is the reference torque before the micro intervention; |T_driver(t) - T_driver(t0)| is the counter-corrective torque applied by the human on the steering wheel; T_threshold is the torque threshold, the dynamic calculation formula of T_threshold is: T_threshold = T_noise x (1 + 0.1 x v / v_max) T_noise is the system noise level, which is obtained by online learning.

[0044] (2) IMU behavior response detection algorithm First, separate human operation and vehicle motion through Kalman filter: This is the prediction equation This is the prediction covariance update equation Where, is the predicted state vector at time k, which contains the vehicle states of interest to the detection algorithm, such as: = [lateral acceleration, yaw rate] ; is the optimal estimated state vector at time k-1, i.e. the best estimate result combined with the prediction and measurement value at the last time; is the state transition matrix, which describes how the system state naturally evolves from time k-1 to time k according to the vehicle dynamics model (for example, based on the current speed and steering angle to predict the yaw rate at the next time; is the control input matrix; is the control input vector; in the present implementation scenario, may include known system interventions, such as single-wheel point braking or steering wheel micro-vibration instructions executed in the micro intervention mode. The filter predicts the vehicle motion that would be produced by these known interventions through this input; is the predicted estimation covariance matrix, which represents the uncertainty or error range of the state prediction value ; is the estimation covariance matrix at the last time; The process noise covariance matrix represents the inaccuracy of the prediction model, such as unmodeled minor wind forces and road friction variations.

[0045] Then, the following algorithm is used to detect abnormal changes in acceleration and yaw rate: if |a_measured - a_predicted|>a_threshold OR |ω_measured - ω_predicted|>ω_threshold Then, driving operation feedback was detected. Where a_measured is the lateral acceleration value actually measured by the IMU; a_predicted is the lateral acceleration value predicted by the Kalman filter (i.e., the state prediction value). (corresponding elements in) a_threshold is the threshold for determining lateral acceleration. This threshold is carefully calibrated and is much larger than [the threshold value is missing here]. This represents normal noise, but is less than the acceleration value typically generated when the driver actively steers; its purpose is to filter out minor disturbances.

[0046] ω_measured is the yaw rate value actually measured by the IMU (i.e., the angular velocity of the vehicle rotating around the vertical axis). ω_predicted is the yaw rate predicted by the Kalman filter (i.e., the state prediction). (corresponding elements in) ω_threshold is the threshold for determining the yaw rate.

[0047] The detection logic in this embodiment is as follows: If, within the monitoring window after micro-intervention, there is a significant deviation beyond the threshold between the vehicle motion (acceleration or yaw rate) actually measured by the IMU and the motion state predicted by the filter as "only affected by micro-intervention and normal noise", then the system considers this deviation to be caused by the driver's active operation (such as unconsciously counter-steering), and thus determines that a valid driving operation feedback has been detected.

[0048] It can be understood that the embodiment synchronously collects the steering wheel torque signal and the vehicle inertial measurement data after the micro-intervention trigger through the multi-sensor cooperative monitoring mechanism, and constructs a cross-validation system of the driver's awareness state. On the one hand, the torque sensor directly captures the instinctive correction action of the driver on the steering wheel, and on the other hand, the IMU senses the vehicle motion change caused by the driver's operation, and the two form a complementary evidence chain in space and time, thereby effectively distinguishing between true syncope (no feedback) and temporary distraction / false alarm (with feedback), significantly improving the accuracy of behavior judgment and the reliability of system decision.

[0049] In a possible implementation manner, the step S3 comprises: (1) using a data fusion algorithm to comprehensively judge the torque sensor monitoring signal and the vehicle body inertial measurement unit (IMU) monitoring signal to confirm whether the active feedback exists.

[0050] For example, the D-S evidence theory is used for decision-level fusion: (Feedback) = f(Torque Evidence), which represents the trust degree of the torque evidence, (Feedback) = g(IMU Evidence) m_combined = ⊕

[0051] Wherein, (Feedback) represents the trust degree of the torque evidence, (Feedback) represents the trust degree of the IMU evidence, and m_combined represents the combined trust degree.

[0052] Now a specific implementation scenario is used to illustrate this step.

[0053] The vehicle is driving straight at 100km / h on the highway, and the driver appears similar to syncope signs due to temporary fatigue, and the system triggers micro-intervention (steering wheel micro-vibration + right front wheel point brake).

[0054] Since the driver's hands are still holding the steering wheel, the micro-vibration itself and the slight deviation of the vehicle make his arm swing passively, and the torque sensor detects a torque change with an amplitude of 1.8 newton meters and a duration of only 0.3 seconds. After the function f(Torque Evidence) is processed, the trust degree of "existence of active feedback" is generated (Feedback) = 0.4 (because the duration is short, the trust degree is low).

[0055] At the same time, the Kalman filter predicts the lateral acceleration based on the vehicle state as 0.05g, but the IMU actually measures a change of 0.08g due to the combined effect of road unevenness and the point brake. The deviation |0.08g - 0.05g| = 0.03g is slightly lower than the threshold a_threshold (e.g. 0.05g). After processing by the function g(IMU evidence), the trust degree (Feedback) = 0.2 (more likely to be judged as noise).

[0056] The system uses the D-S evidence theory to fuse the above torque evidence and IMU evidence. Since both evidence sources do not provide strong support and there is uncertainty, the combined trust degree m_combined(Feedback) = 0.3 after fusion, which is the calculated value.

[0057] The system compares m_combined(Feedback) = 0.3 with the preset effective feedback threshold (e.g. 0.7). Since 0.3 < 0.7, the system comprehensively judges that no effective active feedback is monitored.

[0058] As the opposite case, if the driver is conscious immediately after the micro intervention and subconsciously tightens the steering wheel and slightly corrects the direction, the torque sensor will detect a continuous and stable reverse torque, and the IMU will also detect a clear yaw rate change. At this time, (Feedback) and (Feedback) will be very high, and after fusion m_combined(Feedback) will be much higher than the threshold. In this case, the system comprehensively judges that effective active feedback is monitored.

[0059] (2) If effective active feedback is monitored within the preset time window, it is determined as a false trigger, the system exits silently and does not interfere with driving, and records event data for algorithm optimization.

[0060] (3) If no effective active feedback is monitored within the preset time window, it is determined as a confirmed syncope, and the safety intervention plan is triggered.

[0061] The safety intervention plan includes: 1) Determine the nearest safe parking point, generate a safe parking trajectory from the current position to the nearest safe parking point, and park smoothly and safely along the safe parking trajectory through auxiliary driving.

[0062] Specifically, the safe parking trajectory uses the following quintic polynomial for smooth trajectory generation: x(t) = a0+ a1t + a2t 2 + a3t 3 + a4t 4 + a5t5 y(t) = b0+ b1t + b2t 2 + b3t 3 + b4t 4 + b5t 5 Trajectory generation meets boundary conditions: current position -> nearest safe stopping point.

[0063] It can be understood that the above safe parking trajectory formula describes the movement law of the vehicle in the horizontal and vertical directions in a two-dimensional plane (usually the earth coordinate system). Among them, x(t) represents the position of the vehicle in the longitudinal direction (forward direction) at time t, y(t) represents the position of the vehicle at time t, a0~a5 and b0~b5 are the undetermined coefficients of the polynomial.

[0064] The system first determines the "nearest safe stopping point" through the perception module, thereby obtaining the position, speed, and acceleration constraints of the end point. The starting point constraint comes from the vehicle's own sensors. Substituting the boundary conditions (6 in the longitudinal direction and 6 in the transverse direction) into the polynomial equation set can uniquely solve all 12 coefficients (a0~a5, b0~b5). After the coefficients are determined, the system obtains the ideal position, speed, and acceleration of the vehicle at each time from the current time to the future T time. The steering, throttle, and brake control systems of the vehicle will closely follow the calculated ideal trajectory to make the vehicle smoothly and smoothly complete a series of actions such as merging, decelerating, and parking.

[0065] 2) Turn on the light prompt.

[0066] For example, turn on the double flash warning light.

[0067] 3) Automatically call for help and send the vehicle position.

[0068] It can be understood that when the torque and IMU data are comprehensively analyzed by D-S evidence theory and other multi-sensor fusion algorithms, and the final decision of "confirmation of syncope" is made, it is not simply emergency braking, but immediately triggers a highly structured safety intervention plan. The plan generates a smooth trajectory from the current position to the safe stopping point through a quintic polynomial trajectory planning algorithm, ensuring that the vehicle is smoothly and controllably parked by the auxiliary driving system, while automatically linking the light warning and rescue system, realizing the unity of humanization and systematization of safety intervention.

[0069] The embodiment can greatly reduce the possibility of misjudgment, avoid unnecessary alarm and interference, and ensure that the vehicle can execute the complete process of "stopping-warning-help" in the most stable and reliable manner when a real crisis occurs, thereby guaranteeing the safety of the driver and passengers and the safety of surrounding traffic participants, and embodying the reliability and completeness of the system in the final link.

[0070] In a possible embodiment mode, as shown in Figure 2 The method further comprises: S4, establishing a driver personalized behavior model by long-term monitoring of the operation data of the driver in a normal state, and regularly updating the driver personalized behavior model to adapt to the change of the driving habit of the driver; Based on the parameter change of the driver personalized behavior model, dynamically adjusting the threshold for determining the suspected syncope state and / or the sensitivity of the active feedback.

[0071] For example, when a driver who is used to gently operating the steering wheel is misjudged as syncope by the system many times at the initial stage of system activation because the baseline of the steering wheel torque is low during normal driving, the system dynamically lowers the torque threshold of the feedback monitoring according to the adaptive mechanism by recording the historical data of the false alarm events, and establishes a personalized behavior model of the driver through long-term learning, so that the system only triggers an alarm when the torque signal deviates significantly from the personal baseline, thereby optimizing the false alarm rate to the target interval and improving the accuracy and reliability of the system for different drivers.

[0072] The technical effect of the present application is verified by simulation test.

[0073] In the simulation test, the driver performs "intentional non-intervention" (simulated syncope) and "active intervention" tests on the simulator. The test results are shown in Table 1: Table 1

[0074] As can be seen from Table 1, by comparing the final decision and result of the system in the "simulated syncope" and "normal driving (short-term distraction)" scenarios, the effectiveness of the method of the present application is demonstrated: in the real syncope scenario, the system can achieve 100% correct and safe intervention, and in the false alarm scenario, the system can avoid serious false triggering through the redundant confirmation mechanism, thereby directly verifying the core advantages of the system.

[0075] The false alarm rate comparison analysis of the simulation test is shown in Table 2: Table 2

[0076] The data in Table 2 shows that the redundant confirmation method adopted by the present application significantly reduces the false positive rate from 8.5% to 0.7% compared with the traditional single sensor scheme, and the average response time is controlled within 3.2 seconds, which greatly improves the system reliability without causing a significant increase in the false negative rate, achieving an optimal balance between safety and accuracy.

[0077] The performance at different vehicle speeds is shown in Table 3:

[0078] As can be seen from the data in Table 3, as the vehicle speed increases from 0-30km / h to 90-120km / h, the micro-intervention awareness rate of the system decreases from 95% to 85%, the correct decision rate decreases from 99.5% to 98.1%, but the safety intervention success rate always remains above 99.2%, proving that this technical solution can maintain high reliability throughout the entire vehicle speed range The user experience evaluation results obtained through testing are: Micro-intervention awareness rate: only 15% of drivers explicitly perceive system intervention; Acceptance score: 4.5 / 5.0 (traditional system: 2.8 / 5.0); System trust: 92% of drivers are willing to turn on the system for a long time.

[0079] Figure 3 A redundant confirmation and safety intervention system structure diagram for sudden syncope of a driver provided for an embodiment of the present application, as shown in Figure 3 A redundant confirmation and safety intervention system for sudden syncope of a driver, comprising a crisis determination module, a redundant confirmation executor, a response detection module and a safety intervention module, wherein: The crisis determination module is used to determine that the driver is in a suspected syncope state based on monitoring of the physiological state and visual state of the driver; The redundant confirmation executor is connected with the crisis determination module and the electric power steering system (EPS) and electronic stability program (ESP) of the vehicle, and is used to trigger a vehicle micro-intervention mode after receiving the suspected syncope state signal, including controlling the electric power steering system (EPS) to perform a steering wheel micro-vibration action, and controlling the electronic stability program (ESP) to perform a transient point braking action on a single front wheel; The response detection module is connected with a steering wheel torque sensor and an inertial measurement unit (IMU) of the vehicle body, and is used to monitor the operating torque of the steering wheel by the driver and the change in the motion state of the vehicle within a preset time window after the micro-intervention mode is executed, to determine whether there is active feedback from the driver; The safety intervention module is connected to the response detection module, the vehicle's driving assistance system, and the emergency call system. It is used to make decisions based on the monitoring results of the active feedback: if the active feedback is not detected, it is determined to be a confirmed fainting state, and the safety intervention plan is executed to control the driving assistance system to perform a safe stop, while controlling the emergency call system to automatically call for rescue.

[0080] It is understood that the redundant confirmation and safety intervention system for sudden driver fainting provided by the present invention corresponds to the redundant confirmation and safety intervention method for sudden driver fainting provided in the foregoing embodiments. The relevant technical features of the redundant confirmation and safety intervention system for sudden driver fainting can be referred to the relevant technical features of the redundant confirmation and safety intervention method for sudden driver fainting, and will not be repeated here.

[0081] Please see Figure 4 , Figure 4 A schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 4 As shown, this embodiment of the invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, it performs the following steps: S1, based on monitoring of the driver's physiological and / or visual state, determine that the driver is in a suspected state of unconsciousness; S2, based on the suspected fainting state, trigger the vehicle micro-intervention mode and monitor the driver's active feedback to the micro-intervention mode; S3, based on the results of proactive feedback monitoring, determines whether to implement the safety intervention plan.

[0082] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by the present invention. (See diagram below.) Figure 5 As shown, this embodiment provides a computer-readable storage medium 500 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, it performs the following steps: S1, based on monitoring of the driver's physiological and / or visual state, determine that the driver is in a suspected state of unconsciousness; S2, based on the suspected fainting state, trigger the vehicle micro-intervention mode and monitor the driver's active feedback to the micro-intervention mode; S3, based on the results of proactive feedback monitoring, determines whether to implement the safety intervention plan.

[0083] The embodiment of the present application provides a redundancy confirmation and safety intervention method and system for sudden syncope of a driver and a storage medium, and a closed-loop control logic of "suspected determination-micro intervention stimulation-feedback confirmation-hierarchical decision" is constructed. After the system preliminarily identifies the abnormal state of the driver through multiple sensors, instead of directly taking aggressive measures, a set of difficult-to-detect directional micro intervention (such as steering wheel micro vibration and single-wheel point braking) is applied through the vehicle actuator, so as to serve as a trigger signal for awakening the consciousness of the driver. Then, the system synchronously monitors whether the driver produces an instinctive control feedback, and takes the feedback as the core basis for the final decision: if the feedback exists, the system is exited silently, and if the feedback does not exist, the system confirms the syncope and starts a safety plan.

[0084] The present application has the following advantages: 1. The false alarm risk is transferred from the sensor layer to the behavior layer through the human-computer interaction verification mechanism, and the secondary accidents caused by false triggering are fundamentally avoided; 2. The non-invasive micro intervention is used to replace the traditional alarm or emergency braking, the coherence of the driving experience is maximally maintained, and the user trust is significantly improved; 3. The rapid perception of artificial intelligence and the instinctive reaction of human consciousness are combined to form a decision-making closed loop of bidirectional verification, so that the safety intervention action has timeliness and high reliability, and the organic unification of safety protection and driving experience is realized.

[0085] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0086] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0087] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a machine that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0088] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0090] Although preferred embodiments of the application have been described herein, substitutions and alterations are possible in view of the teachings of this application. Accordingly, the appended claims are intended to encompass all such substitutions and alterations. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

[0091] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

Claims

1. A method for redundancy confirmation and safety intervention in case of sudden driver fainting, characterized in that, The method comprises: S1, determining that the driver is in a suspected syncope state based on monitoring of the physiological state and / or visual state of the driver; S2, triggering a vehicle micro-intervention mode based on the suspected syncope state, monitoring active feedback of the driver to the micro-intervention mode; S3, determining whether to execute a safety intervention plan based on the monitoring result of the active feedback.

2. The method of claim 1, wherein, Step S1 comprises: monitoring at least one vital sign signal of the driver through a physiological sensor, and / or monitoring the facial posture and eye state of the driver through a visual sensor; when the monitoring data exceeds a preset threshold and lasts for a certain time, it is determined that the suspected syncope state exists.

3. The method of claim 1, wherein, In step S2, the triggering of the vehicle micro-intervention mode comprises: controlling the vehicle to generate periodic steering wheel micro-vibration; and controlling the vehicle to apply a transient point brake to a single front wheel on the side opposite to the driving direction.

4. The method of claim 3, wherein, The control of the vehicle to generate periodic steering wheel micro-vibration comprises: based on the driving state of the vehicle, using an amplitude adaptive algorithm to calculate the micro-vibration amplitude of the steering wheel; controlling the vehicle to generate periodic steering wheel micro-vibration based on the micro-vibration amplitude.

5. The method of claim 3, wherein, The control of the vehicle to apply a transient point brake to a single front wheel on the side opposite to the driving direction comprises: selecting a single front wheel on the side opposite to the driving direction of the vehicle, the selection rule being: if the vehicle is driving straight, randomly selecting the left front wheel or the right front wheel; if the vehicle is slightly turning left, selecting the right front wheel; if the vehicle is slightly turning right, selecting the left front wheel; based on the driving state of the vehicle, applying a braking force to the selected front wheel to realize a transient point brake, the braking force being determined based on the vehicle mass, the vehicle speed and the tire rolling radius of the selected front wheel.

6. The method of claim 1, wherein, In step S2, the monitoring of the active feedback of the driver to the micro-intervention mode comprises: within a preset time window of the execution of the micro-intervention mode, continuously monitoring the artificially applied reverse correction torque on the steering wheel through a torque sensor; and / or monitoring the vehicle acceleration or yaw rate change caused by human operation through an inertial measurement unit (IMU) of the vehicle body.

7. The method of claim 6, wherein, Step S3 comprises: using a data fusion algorithm to comprehensively judge the torque sensor monitoring signal and the inertial measurement unit (IMU) monitoring signal to confirm whether the active feedback exists; if valid active feedback is monitored within a preset time window, it is determined that the triggering is false, the system is silenced and exits, and event data is recorded for algorithm optimization; if valid active feedback is not monitored within a preset time window, it is determined that the syncope is confirmed, and the safety intervention plan is triggered.

8. The method of claim 1 or 7, wherein, The safety intervention plan comprises: determining a nearest safe parking point, generating a safe parking trajectory from the current position to the nearest safe parking point, and smoothly and safely parking along the safe parking trajectory through auxiliary driving; and turning on the light prompt; automatically calling for rescue and sending the vehicle position.

9. The method of claim 1, wherein, The method further comprises: S4, establishing a driver personalized behavior model by long-term monitoring of the operation data of the driver in a normal state, and regularly updating the driver personalized behavior model to adapt to changes in the driving habits of the driver; based on the parameter changes of the driver personalized behavior model, dynamically adjusting the threshold for determining the suspected syncope state and / or the sensitivity for monitoring the active feedback.

10. A redundant validation and safety intervention system for driver sudden fainting, characterized in that, The method comprises: a crisis determination module configured to determine that the driver is in a suspected syncope state based on monitoring of the physiological state and the visual state of the driver; a redundancy confirmation executor connected with the crisis determination module and an electric power steering system (EPS) and an electronic stability program (ESP) of the vehicle, and configured to trigger a micro-intervention mode of the vehicle including controlling the electric power steering system (EPS) to perform a micro-vibration action on the steering wheel and controlling the electronic stability program (ESP) to perform a transient point braking action on a single front wheel, in response to receiving the suspected syncope state signal; a response detection module connected with a steering wheel torque sensor and an inertial measurement unit (IMU) of the vehicle, and configured to monitor the operation torque of the driver on the steering wheel and the change of the motion state of the vehicle within a preset time window after the micro-intervention mode is performed, to determine whether there is an active feedback of the driver; a safety intervention module connected with the response detection module, a driving assistance system and an emergency call system of the vehicle, and configured to make a decision based on the monitoring result of the active feedback: if the active feedback is not monitored, it is determined that the syncope is confirmed, and a safety intervention plan is executed to control the driving assistance system to perform a safe parking and control the emergency call system to automatically call for help.

Images