Vehicle control method, device and electronic equipment

Abstract

The application provides a vehicle control method, device and electronic equipment, relates to the field of intelligent automobiles and active safety technology, and can determine a driving risk index reflecting the comprehensive influence degree of the driver's own factors on driving safety based on multidimensional monitoring data through multidimensional comprehensive sensing of the driver's state by a multidimensional sensor; further introducing situation data to determine the driving risk level corresponding to the driving risk index, which can reflect the vehicle driving risk level under the joint action of the driver's own state and the external environment, so as to dynamically adjust the risk judgment result according to the actual driving environment. According to the corrected driving risk level, the corresponding risk intervention strategy is determined, the controllable execution mechanism and its working parameters configured in the strategy are executed in a manner matched with the current risk level, so that the system can timely and appropriately intervene when intervention is needed, thereby ensuring driving safety while reducing unnecessary interference to normal driving of the driver.

Description

Technical Field

[0001] This invention relates to the field of intelligent vehicles and active safety technology, and in particular to a vehicle control method, device and electronic device. Background Technology

[0002] Traffic accidents can be reduced by monitoring driver condition and intervening promptly when risks occur. Existing technologies focus on monitoring and responding to the risk of sudden driver illness, while others focus on identifying and intervening in the risk of driver emotional fluctuations. However, a driver's condition during driving is not limited to a single dimension; fatigue, distraction, emotional fluctuations, and physiological abnormalities often coexist and are coupled, collectively affecting the driver's overall risk level. Furthermore, the actual risk corresponding to the same driver condition varies depending on different road sections and weather conditions.

[0003] Existing solutions typically identify a single type of risk, and their assessment results are difficult to cover the actual risks of multi-factor coupling in real driving scenarios. As a result, there is a discrepancy between the control strategies generated and the actual driving risks. Summary of the Invention

[0004] The purpose of this invention is to provide a vehicle control method, device, and electronic device that can dynamically correct driving risks by comprehensively perceiving the driver's state from multiple dimensions and using contextual data, thereby achieving graded intervention that matches the current risk level and intervening in a timely and appropriate manner when intervention is needed, thus ensuring driving safety while reducing unnecessary interference with the driver's normal driving.

[0005] In a first aspect, the present invention provides a vehicle control method, wherein the method includes: monitoring multi-dimensional monitoring data of the driver through multi-dimensional sensors inside the vehicle, and acquiring situational data of the vehicle in the current driving environment; determining the driver's driving risk index based on the multi-dimensional monitoring data; numerically correcting the driving risk index based on the situational data, and determining the latest driving risk level of the vehicle based on the corrected driving risk index; determining a risk intervention strategy for the vehicle based on the latest driving risk level; wherein the risk intervention strategy is configured with at least one controllable actuator, and operating parameters corresponding to the controllable actuator; controlling the controllable actuator to operate according to the operating parameters to intervene in the risk of the vehicle.

[0006] In an optional implementation, controlling the operation of the controllable actuator according to the operating parameters to intervene in the risk of the vehicle includes: in response to a change in the latest driving risk level compared to the previous driving risk level of the vehicle, adjusting the operating parameters of the controllable actuator according to the risk intervention strategy of the latest driving risk level, so as to intervene in the risk of the vehicle through the risk intervention strategy of the latest driving risk level.

[0007] In an optional implementation, the above-mentioned transitional adjustment of the operating parameters of the controllable actuator according to the risk intervention strategy of the latest driving risk level includes: obtaining the target controllable actuator and its target operating parameters corresponding to the risk intervention strategy of the latest driving risk level; determining the operating status adjustment target and operating parameter adjustment target of the controllable actuator corresponding to the current risk intervention strategy based on the target controllable actuator and its target operating parameters; and, within a preset transition time, adjusting the operating status and operating parameters of the controllable actuator from the current value to the operating status and operating parameters corresponding to the risk intervention strategy of the latest driving risk level according to the operating status adjustment target and operating parameter adjustment target.

[0008] In an optional implementation, the aforementioned multi-dimensional monitoring data includes: driver's facial feature data, physiological feature data, and voice feature data; determining the driver's driving risk index based on the multi-dimensional monitoring data includes: determining the driver's corresponding risk parameters based on the multi-dimensional monitoring data; the risk parameters include: fatigue risk value, disease risk value, distraction risk value, and emotional risk value; determining the fusion weights corresponding to each parameter in the risk parameters based on the driver's current driving environment and the driver's continuous driving time; and performing a fusion calculation on the fatigue risk value, disease risk value, distraction risk value, and emotional risk value based on the fusion weights corresponding to each parameter in the risk parameters to determine the driver's corresponding driving risk index.

[0009] In an optional implementation, the fusion weights corresponding to each of the above risk parameters are obtained in the following way: driver sample data is obtained from cloud data; wherein, the driver sample data includes historical multi-dimensional monitoring data of multiple drivers in different driving environments, historical driving risk index and corresponding actual driving status results; the fusion weights corresponding to fatigue risk value, disease risk value, distraction risk value and emotional risk value in the driving environment are determined according to the driver sample data.

[0010] In an optional implementation, the above-mentioned numerical correction of the driving risk index based on contextual data, and the determination of the latest driving risk level of the vehicle based on the corrected driving risk index, includes: determining the risk correction factor corresponding to the current driving environment based on contextual data; determining the basic risk level corresponding to the driving risk index based on the threshold range of the driving risk index among a number of preset dynamic thresholds; and correcting the level value of the basic risk level by using the risk correction factor as a level correction coefficient to obtain the latest driving risk level of the vehicle.

[0011] In an optional implementation, the aforementioned scenario data includes road type information, weather condition information, traffic condition information, and altitude information of the current driving environment. Based on the scenario data, a risk correction factor corresponding to the current driving environment is determined, including: determining the corresponding correction amount based on the risk controllability of the road type information, weather condition information, traffic condition information, and altitude information; determining the comprehensive correction amount corresponding to the scenario data based on each correction amount; and adjusting the preset base value of the correction factor based on the comprehensive correction amount to obtain the risk correction factor corresponding to the current driving environment.

[0012] In an optional implementation, the above-mentioned determination of the vehicle's risk intervention strategy based on the latest driving risk level includes: obtaining a pre-stored risk-strategy mapping table; determining the risk intervention strategy corresponding to the driving risk level based on the risk-strategy mapping table; wherein, the risk-strategy mapping table is configured with multiple driving risk levels, and the controllable actuator corresponding to each driving risk level is determined from multiple functional domains of the vehicle according to the risk level indicated by the driving risk level; the operating parameters of the controllable actuator include the output intensity parameter and / or the action time parameter of the controllable actuator; the output intensity parameter and the action time parameter are determined according to the risk level indicated by the corresponding driving risk level.

[0013] In an optional implementation, the aforementioned multiple functional domains include the cockpit domain, chassis domain, powertrain domain, autonomous driving domain, body domain, and communication domain.

[0014] In an optional implementation, the above-mentioned adjustment of the working status and working parameters of the controllable actuators within a preset transition period, according to the working status adjustment target and working parameter adjustment target, transitions the working status and working parameters of the controllable actuators from their current values ​​to the working status and working parameters corresponding to the latest driving risk level risk intervention strategy. This includes: during the first transition period, maintaining the operation of the controllable actuators corresponding to the current risk intervention strategy, and controlling the activation of the target controllable actuators required by the latest driving risk level risk intervention strategy, with their working parameters gradually increased from zero; during the second transition period after the first transition period, gradually decreasing the working parameters of the controllable actuators not needed in the current risk intervention strategy to zero, and controlling the working parameters of the target controllable actuators to continue to gradually increase to the corresponding target working parameters.

[0015] In an optional implementation, the method further includes: outputting a prompt message indicating that a risk intervention is about to be performed; and within a preset time after the prompt message is output, if a cancellation command is received from the driver, controlling the transition adjustment to stop, and controlling the operating parameters of the already activated target controllable actuator to gradually decrease to zero.

[0016] In an optional implementation, the method further includes: recording driver status data and vehicle status data before and after each risk intervention, so as to optimize the risk intervention strategy offline based on the driver status data and vehicle status data.

[0017] In an optional implementation, the aforementioned basic risk levels include at least: normal level, mild risk level, moderate risk level, severe risk level, and emergency risk level; the multiple dynamic thresholds include the thresholds corresponding to each basic risk level.

[0018] In an optional implementation, the method further includes: obtaining the driver's record of canceling risk interventions on the vehicle during historical driving; and determining the target adjustment value of the dynamic threshold corresponding to the driving risk level based on the number of times the driving risk level was canceled in the cancellation record.

[0019] In an optional implementation, the method further includes: receiving the sensitivity level selected by the driver through the safety settings menu; determining the adjustment direction and adjustment range of multiple dynamic thresholds based on the sensitivity level; and determining the values ​​corresponding to the multiple dynamic thresholds based on the adjustment direction and adjustment range.

[0020] Secondly, embodiments of the present invention also provide a vehicle control device, comprising: a data acquisition module, configured to monitor multi-dimensional monitoring data of the driver through multi-dimensional sensors within the vehicle, and acquire contextual data of the vehicle in its current driving environment; a driving risk index determination module, configured to determine the driver's driving risk index based on the multi-dimensional monitoring data; a driving risk level determination module, configured to numerically correct the driving risk index based on the contextual data, and determine the latest driving risk level of the vehicle based on the corrected driving risk index; an execution module, configured to determine a risk intervention strategy for the vehicle based on the latest driving risk level; wherein the risk intervention strategy is configured with at least one controllable actuator and corresponding operating parameters for the controllable actuator; and a control module, configured to control the controllable actuator to operate according to the operating parameters to perform risk intervention on the vehicle.

[0021] Thirdly, embodiments of the present invention also provide an electronic device, wherein the electronic device includes a processor and a memory, the memory storing machine-executable instructions that can be executed by the processor, and the processor executing the machine-executable instructions to implement the vehicle control method of any of the above embodiments.

[0022] Fourthly, embodiments of the present invention also provide a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the vehicle control method of any of the above embodiments.

[0023] The embodiments of this invention bring the following beneficial effects: These embodiments provide a vehicle control method, device, and electronic device. Through multi-dimensional sensors, the driver's state is comprehensively perceived from multiple dimensions. Based on multi-dimensional monitoring data, a driving risk index reflecting the comprehensive impact of the driver's own factors on driving safety is determined. Furthermore, contextual data is introduced to numerically correct the driving risk index, determining the driver's risk level for the vehicle. This reflects the vehicle driving risk level under the combined influence of the driver's own state and the external environment, thereby dynamically adjusting the risk assessment result according to the actual driving environment. Based on the corrected driving risk level, a corresponding risk intervention strategy is determined. Through the controllable actuators and their operating parameters configured in the strategy, intervention is executed in a manner matching the current risk level, enabling the system to intervene promptly and appropriately when necessary, ensuring driving safety while reducing unnecessary interference with the driver's normal driving.

[0024] Other features and advantages of the invention will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0025] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0026] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0027] Figure 1 A flowchart of a vehicle control method provided in an embodiment of the present invention; Figure 2 A flowchart of another vehicle control method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a vehicle control device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0030] To facilitate understanding, a vehicle control method provided in an embodiment of the present invention will first be described, referring to... Figure 1 The method includes the following steps: Step S102: Monitor the driver's multi-dimensional monitoring data through multi-dimensional sensors inside the vehicle, and obtain the contextual data of the vehicle in the current driving environment.

[0031] The in-vehicle multidimensional sensor represents a collection of sensing devices deployed inside the vehicle cabin to simultaneously collect driver status information from multiple physical or physiological dimensions. For example, it may include optical cameras for capturing the driver's facial and eye movement features, microphones for collecting voice features, and vital sign sensing units integrated into contact locations such as the steering wheel, seat belt, or seat armrests. These units continuously acquire the driver's physiological parameters, such as heart rate, respiratory rate, blood oxygen saturation, and blood pressure. These sensors can monitor the driver's current behavioral and physiological performance from multiple dimensions to determine if the driver is in any potentially risky state that could affect safe driving.

[0032] Multi-dimensional monitoring data characterizes various parameters reflecting the driver's physiological and behavioral states collected by the aforementioned multi-dimensional sensors during the driver's driving process. For example, data reflecting behavioral states may include eyelid closure frequency, gaze direction, head posture, and distracting behaviors such as using a mobile phone, collected by the optical camera, which can be used to determine whether the driver exhibits risk signs such as fatigue, distraction, or emotional abnormalities. Data reflecting physiological states may include vital sign parameters such as heart rate, respiratory rate, blood oxygen, blood pressure, electrocardiogram, and heart rate variability continuously collected by the vital sign sensing unit, which can be used to identify whether the driver is experiencing physiological abnormalities such as sudden illness. By fusing the above behavioral data with physiological feature data, the driver's state can be expanded from single visual observation to multimodal physiological perception, enabling a more comprehensive capture of various potential driving risks such as fatigue, distraction, emotional abnormalities, and sudden illnesses.

[0033] The current driving environment of a vehicle represents the set of external objective conditions it faces during operation, constituting external factors that influence the difficulty of the driving task and the magnitude of potential safety risks. In actual engineering deployments, the driving load borne by the same driver varies under different driving environments. For example, in a sunny, straight, and smooth highway environment with good traffic, the driver's control burden is relatively small; while in rainy or foggy weather, on winding roads with heavy traffic, the driver's control burden increases significantly. By understanding the current driving environment, the actual driving task load borne by the driver can be determined.

[0034] Contextual data represents information about road network attributes and dynamically changing environmental parameters of a vehicle in its current driving environment. Under specific driving conditions, this contextual data may include information about the road type, weather conditions, traffic conditions, and altitude of the current road.

[0035] Among them, road type information reflects the road geometry and design speed range of the current driving segment, such as highways, urban arterial roads or mountain roads; weather condition information reflects parameters that affect the vehicle's handling boundaries, such as visibility and road surface adhesion conditions, such as rain, snowfall or fog and haze; traffic condition information reflects the density and distribution of traffic participants on the current road, such as smooth traffic, slow traffic or congestion; altitude information can be used to help determine the potential impact of changes in oxygen content on the driver's physiological state in high-altitude scenarios, as well as the adaptability of the vehicle's power system to the operating conditions in this environment.

[0036] The aforementioned contextual data can be obtained through vehicle-mounted perception sensors, high-precision maps, cloud-based weather services, or vehicle-to-everything (V2X) communication. By acquiring contextual data, it is possible to understand the actual environmental risks currently faced by the driver, thereby enabling a comprehensive assessment of the driver's condition in conjunction with the workload of the environment.

[0037] Step S104: Determine the driver's driving risk index based on multi-dimensional monitoring data.

[0038] The driving risk index reflects the degree to which a driver's own factors affect driving safety in a given situation. In practice, by acquiring multi-dimensional monitoring data, it is possible to obtain the driving safety risks posed by the driver under the superposition of multiple risk dimensions, thereby determining the driving risk index that reflects the degree of impact.

[0039] For example, when the eyelid closure frequency and head posture data collected by the optical camera show signs of driver fatigue, and the vital signs parameters such as heart rate and blood pressure collected by the vital signs sensor unit show abnormal fluctuations, it indicates that the driver is performing driving tasks under the dual risk dimensions of fatigue and disease, and the driving risk index is at a high level. When the multi-dimensional monitoring data only reflects that the driver has a slight risk in a single dimension, such as the eyelid closure frequency being slightly higher than the baseline level, while the monitoring data of other dimensions are all within the normal range, the driving risk index is at a low level.

[0040] Step S106: Adjust the driving risk index numerically based on the scenario data, and determine the latest driving risk level of the vehicle based on the adjusted driving risk index.

[0041] The aforementioned driving risk index reflects the overall risk level of a driver's condition. However, the actual driving risk level corresponding to the same driving risk index varies under different driving environments. For example, if a driver is slightly fatigued and is driving on a highway with clear weather and smooth traffic, the actual risk is relatively controllable; however, if the driver is driving on a mountain road with rain, fog, and numerous curves, the actual risk caused by the same state of mild fatigue will be significantly increased.

[0042] Based on this, the embodiments of the present invention introduce contextual data to numerically correct the driving risk index, and use environmental risk factors as adjustment quantities, so that the meaning of the corrected driving risk index changes from reflecting the driver's own risk level to reflecting the vehicle driving risk level under the combined effect of the driver's own state and the external environment.

[0043] Therefore, based on the range of values ​​of the revised driving risk index, the latest driving risk level of the vehicle in the current driving environment can be determined, which represents the driving risk level of the vehicle.

[0044] The higher the value of the revised driving risk index, the higher the corresponding driving risk level, indicating that the current driving risk of the vehicle is greater; the lower the value of the revised driving risk index, the lower the corresponding driving risk level, indicating that the current driving risk of the vehicle is smaller.

[0045] Step S108: Determine the vehicle's risk intervention strategy based on the latest driving risk level.

[0046] Risk intervention strategies represent a set of vehicle control measures pre-set for a specific level of driving risk. In practice, different levels of driving risk correspond to different levels of vehicle driving risk, and the intervention measures required will also differ in type, intensity, and scope.

[0047] In this embodiment of the invention, the risk intervention strategy is configured with at least one controllable actuator and corresponding operating parameters for the controllable actuator. The risk intervention strategy of this embodiment defines which controllable actuators the vehicle should activate under the current driving risk level, and with what operating parameters each controllable actuator should operate.

[0048] The controllable actuator represents a hardware unit on a vehicle whose physical output state can be controlled by electrical signals. For example, it may include visual or auditory prompting devices for alerting the driver, tactile feedback devices for applying physical warnings to the driver, and actuators for influencing the vehicle's motion. Operating parameters characterize the specific operating state of the controllable actuator when intervention is implemented.

[0049] In practical implementation, different driving risk levels correspond to different operating parameter configurations, which can be used to control each controllable actuator to operate in a manner appropriate to the current risk level. For example, for a loudspeaker, operating parameters may include the broadcast content and broadcast interval; for a vibration device, they may include the vibration mode and the time interval between adjacent vibrations; for a braking actuator, they may include the braking intervention timing and braking force.

[0050] By configuring the above operating parameters, each controllable actuator can perform intervention actions in a manner appropriate to the current risk level, helping the driver maintain safe control of the vehicle and avoid driving risks.

[0051] In risk intervention strategies, different driving risk levels correspond to different combinations of controllable actuators and configurations of operating parameters to ensure safe vehicle operation under current driving conditions. The higher the driving risk level, the more diverse the controllable actuators activated and the more direct the corresponding intervention methods.

[0052] For example, when the driving risk level is low, the corresponding risk intervention strategy can activate only a single type of controllable actuator to provide a gentle reminder to the driver, alerting them to their current driving status while avoiding interference with their normal operation. When the driving risk level is high, the corresponding risk intervention strategy can activate multiple controllable actuators to work together, applying intervention in a more direct manner to prompt the driver to regain stable control of the vehicle or for the system to intervene and maintain the vehicle's safe driving status.

[0053] Step S110: Control the controllable actuator according to the working parameters to intervene in the risk of the vehicle.

[0054] After determining the risk intervention strategy, the corresponding controllable actuators are controlled to perform intervention actions according to the working parameters of each controllable actuator in the strategy. This can help the driver maintain safe control of the vehicle and avoid driving risks caused by abnormal driver condition.

[0055] In practice, the intervention methods implemented can be gradually upgraded as the driving risk level increases. When the driving risk level is low, the driver is mainly reminded to pay attention to the current driving status through prompts, such as issuing prompts via speakers or displaying visual reminders through display devices, so that the driver can automatically regain attention and normal operation after perceiving the prompts. When the driving risk level increases, physical warnings can be superimposed on the prompts, such as applying tactile feedback to the driver through vibration devices to enhance the perceived intensity of the warnings and prompt the driver to pay attention to driving safety in a timely manner. When the driving risk level reaches a high level, if the driver fails to regain stable control, further intervention can be applied to the vehicle's motion state, such as implementing deceleration through the braking unit, in order to actively reduce the driving risk of the vehicle and maintain the vehicle's safe state under the current driving conditions.

[0056] Furthermore, based on the above embodiments, this invention also provides another vehicle control method, referring to... Figure 2 The method includes the following steps: Step S202: Monitor the driver's multi-dimensional monitoring data through multi-dimensional sensors inside the vehicle, and obtain the contextual data of the vehicle in the current driving environment.

[0057] Step S204: Determine the driver's driving risk index based on multi-dimensional monitoring data.

[0058] Based on the above embodiments, the collected multi-dimensional monitoring data may include the driver's facial feature data, physiological feature data, and voice feature data to reflect the driver's physiological and behavioral state. Facial feature data can be extracted using an infrared camera deployed in the cockpit and computer vision algorithms to obtain parameters such as the percentage of eyelid closure time, the duration and direction of gaze deviation from the road, and head posture angle, used to reflect the driver's fatigue level and degree of inattention from different perspectives. Physiological feature data can be extracted using an electrocardiogram sensor or photoelectric sensor integrated into the steering wheel to obtain parameters such as heart rate, time-domain indicators of heart rate variability, and blood oxygen saturation and its changing trends, used to reflect the driver's physiological abnormalities. Voice feature data can be extracted using an in-vehicle microphone and an emotion recognition model to obtain feature parameters reflecting changes in the driver's voice tone, used to help determine their emotional fluctuations.

[0059] In practice, the driving risk index for a driver can be determined through the following steps: 1) Determine the risk parameters corresponding to drivers based on multi-dimensional monitoring data.

[0060] The multi-dimensional monitoring data comes from different sensor channels, with each channel's data reflecting the driver's state from aspects such as facial features, physiological signs, and voice tone. Risk parameters are used to characterize the assessment results obtained for the driver's state under various risk dimensions after classifying the data from the above channels according to risk types. In one implementation, the above risk parameters include: fatigue risk value, disease risk value, distraction risk value, and emotional risk value.

[0061] The fatigue risk value characterizes the degree to which a driver's normal driving operation is affected by drowsiness. It can be derived from facial feature data such as eyelid closure frequency and head posture changes. For example, during long-term high-speed driving, the percentage of time the driver's eyelids are closed gradually increases, and the head nodding motion occurs frequently. At this time, the fatigue risk value increases accordingly, reflecting that the driver's ability to respond to sudden road conditions ahead has decreased due to drowsiness. For example, the fatigue risk value can be expressed as (R_fatigue), which can be calculated using the following formula: R_fatigue = 0.6 × f(PERCLOS) + 0.3 × g (nodding frequency) + 0.1 × h (driving duration).

[0062] Wherein, f(PERCLOS) represents a piecewise function. In one implementation, when PERCLOS is below 0.15, this term is 0; when PERCLOS is between 0.15 and 0.3, this term increases linearly from 0 to 0.3; when PERCLOS exceeds 0.3, this term rapidly rises to 1. g(head nodding frequency) is used to calculate the number of times per minute that the head pitch angle is below a preset angle and lasts for more than a preset duration. When this frequency exceeds a certain number of times, such as 3 times / minute, the risk value of this term increases linearly. In addition, when h(driving duration) starts to increase linearly from 0 after continuous driving for more than a certain duration (such as 2 hours), it reaches its maximum value (such as 0.2) when the driving duration further increases (such as 4 hours).

[0063] The disease risk value characterizes the likelihood of a driver experiencing a sudden physiological abnormality and can be derived from the changing trends of physiological characteristic data such as heart rate variability and blood oxygen saturation. For example, when the time-domain index of heart rate variability deviates significantly from the driver's personal resting baseline, or when blood oxygen saturation shows a sustained decline, the disease risk value increases accordingly, reflecting the risk that the driver may lose control of the vehicle in a short period of time due to physiological abnormalities. This parameter can be represented as (R_disease) and can be calculated using the following formula: R_disease = max(R_HRV, R_SpO2), representing the larger of the two values.

[0064] R_HRV represents the physiological risk level of a driver due to abnormal heart rate variability. It is calculated based on the standard deviation of the normal sinus interval (SDNN) within the current time window using electrocardiogram (ECG) signals and compared with the driver's personal resting baseline. For example, the percentage deviation between the current 5-minute window SDNN and the driver's personal resting baseline is calculated. The larger the deviation (e.g., deviation >30%), or the higher the detected arrhythmia characteristics, the higher the value (e.g., 0.8 is a high-risk indicator). This indicates that the driver's current autonomic nervous system regulation is abnormal, and there may be driving risks due to stress, fatigue, or potential cardiovascular problems. R_SpO2 represents the physiological risk level of a driver due to decreased blood oxygen saturation, and can be determined based on the current value and trend of blood oxygen saturation. For example, when blood oxygen saturation remains below the normal range (e.g., below 94% for 30 seconds), the risk value can be increased linearly according to the rate of decrease. If blood oxygen saturation drops sharply in a short period of time (e.g., drops by more than 5% within 1 minute) and the drop exceeds the preset range, it indicates that the driver may have respiratory function limitations or circulatory disorders, and the body's oxygen supply capacity may be affected, thus posing a driving risk. In this case, a high value can be triggered directly.

[0065] Distraction risk value characterizes the degree to which a driver's attention deviates from the driving task, and can be derived from data such as gaze direction, duration of gaze deviation, and driving behavior. For example, when following another vehicle on urban roads, if a driver frequently takes their eyes off the road to check their phone, the standard deviation of steering wheel angle micro-motion will be significantly lower than the fluctuation level during normal cruising. In this case, the distraction risk value will increase accordingly, reflecting a blind spot in the driver's perception of vehicles in adjacent lanes. This parameter can be expressed as (R_distraction), and can be represented by the following formula: R_distraction = 0.5 × (percentage of time the driver's gaze deviates from the road) + 0.3 × (LDW trigger frequency) + 0.2 × (steering wheel angle standard deviation constant).

[0066] The percentage of time the driver's line of sight deviates from the road can be expressed as: the cumulative proportion of time in the past 30 seconds when the driver's line of sight is not in the area of ​​the road ahead (e.g., the deviation is determined by ±15° from the center of the windshield); the LDW (Lane Departure Warning) trigger frequency refers to the number of times the vehicle's lane departure warning system triggers a warning due to the driver's non-active lane change behavior within a preset statistical time window (e.g., 30 seconds); the standard deviation constant of the steering wheel angle can be expressed as: the comparison between the standard deviation of the steering wheel angle in the current time window (e.g., 10 seconds) and the historical average value of the straight road cruise. The greater the deviation between the two, the higher the risk value of this item.

[0067] The emotional risk value characterizes the degree to which a driver's driving judgment is affected by emotional fluctuations, and can be derived from data such as voice characteristics and facial expressions. For example, if there are significant fluctuations in voice tone or tense or angry facial expressions during driving, the emotional risk value increases accordingly, reflecting that the driver's judgment and operation may be interfered with by emotions. This parameter can be characterized as (R_emotion), and can be calculated using the following formula: R_emotion = 0.7 × (confidence level of facial expression tension / anger) + 0.3 × (agitation level of voice emotion). The higher the confidence level, the higher the emotional risk value.

[0068] Furthermore, multi-dimensional sensors can acquire driver behavior data via the vehicle's CAN bus, such as the standard deviation of steering wheel angle micro-motion and pedal operation non-stability. This behavior data can be cross-checked with facial feature data and physiological feature data. For example, when the optical camera detects that the driver's gaze is deviating from the line of sight and the standard deviation of steering wheel angle micro-motion is at an abnormally low value, it can more reliably determine that the driver is in a distracted or incapacitated state, reducing the false alarm rate caused by obstruction or misjudgment of a single sensing path.

[0069] By determining the aforementioned risk parameters, the driver's current state can be assessed from dimensions such as fatigue, illness, distraction, and emotions, thereby comprehensively judging whether the driver's physical and mental condition is suitable for continuing to perform driving tasks.

[0070] 2) Based on the driver's current driving environment and the driver's continuous driving time, determine the fusion weights corresponding to each parameter in the risk parameters.

[0071] After determining the aforementioned risk parameters, this embodiment of the invention further determines the weight of each risk parameter in the comprehensive assessment. The fusion weight is used to reflect the contribution of each risk dimension to the overall driving risk under the current driving environment and driving duration conditions. In one implementation, let the fusion weight corresponding to the fatigue risk value be α, the fusion weight corresponding to the distraction risk value be β, the fusion weight corresponding to the emotional risk value be γ, and the fusion weight corresponding to the disease risk value be δ. The sum of all weights is 1. Each weight can be preset according to needs, for example, α=0.4, β=0.3, γ=0.2, δ=0.1.

[0072] Furthermore, in actual driving scenarios, the same combination of risk parameters poses different threats to driving safety under different driving environments. For example, in congested urban traffic, where vehicle spacing is small and road conditions change frequently, the threat of distraction risk to driving safety is more prominent, and the fusion weight corresponding to the distraction risk value should be increased accordingly. In long-distance highway driving, where drivers drive continuously for longer periods, the cumulative effect of fatigue is more significant, and the fusion weight corresponding to the fatigue risk value should be increased accordingly.

[0073] This invention also dynamically determines the fusion weights of various risk parameters based on the driver's current driving environment and continuous driving duration, ensuring that the contribution of each risk parameter to the driving risk index is adapted to the current driving conditions. For example, when continuous driving time exceeds 3 hours, the threat of fatigue to driving safety increases significantly, so α can be increased by 0.1 from the baseline value, while β is decreased by 0.05, increasing the proportion of fatigue risk in the comprehensive assessment. When it is detected that the current driving is at night, the impact of fatigue is further aggravated due to poor lighting conditions and the driver's physiological rhythm being at its lowest point, so α can be increased by another 0.05.

[0074] For example, when a driver is in congested urban traffic, frequent vehicle starts and stops, and the need to constantly monitor the surrounding road users, the risk of distraction is more pronounced. In this case, β can be increased by a corresponding amount from the baseline value, while α or γ can be appropriately decreased. When the driver is in a smooth highway environment and the driving time is short, the fusion weights of the various risk parameters can be maintained near the baseline value.

[0075] Furthermore, the fusion weights corresponding to each risk parameter can be obtained from driver sample data in the cloud. This driver sample data includes historical multi-dimensional monitoring data of multiple drivers under different driving environments, historical driving risk indices, and corresponding actual driving status results.

[0076] For example, the sample data can record the correlation between monitoring data on fatigue, distraction, emotional state, and illness of different drivers in different driving environments such as urban roads, highways, and mountain roads, and whether a dangerous event ultimately occurs. By analyzing the above sample data, the contribution of each risk parameter to actual driving risk under different driving environments and driving durations can be determined, thereby obtaining the fusion weight of each risk parameter under the current driving environment and driving duration. This method ensures that the fusion weight is set based on the statistical regularity of actual driving data, which can improve the adaptability of the driving risk index to different scenarios.

[0077] The aforementioned fusion weights can be continuously optimized using reinforcement learning algorithms. In practical implementation, the cloud data mentioned above can be used as training samples, with the fusion weights α, β, γ, and δ as policy parameters to be optimized. The success of the driving risk index in predicting the decline trend of the driver's state in the subsequent preset time period is used as the reward signal. The optimal configuration of the fusion weights under different driving scenarios is analyzed using reinforcement learning algorithms. For example, if the driving risk index calculated based on the current weights is high, and the driver's fatigue index continues to rise or distraction index continues to worsen in the subsequent time period, it indicates that the risk judgment is consistent with the actual trend of the driver's state, and a positive reward is given; if the driving risk index is high but the driver's state recovers to the normal range on its own in the subsequent time period, it indicates a false alarm, and a negative reward is given.

[0078] Reinforcement learning algorithms iterate repeatedly on a large number of anonymized samples, gradually adjusting the fusion weights so that the weight configurations for each driving scenario gradually approach the optimal values. The optimized fusion weights can be distributed to the vehicle via OTA to update the weight parameters in the local risk fusion model, enabling the vehicle to obtain a fusion weight configuration that better reflects the actual driving scenario and further improving the accuracy of the driving risk index in predicting changes in driver state.

[0079] 3) Based on the fusion weights corresponding to each parameter in the risk parameters, the fatigue risk value, disease risk value, distraction risk value and emotional risk value are fused and calculated to determine the driving risk index corresponding to the driver.

[0080] After obtaining the various risk parameters and their corresponding fusion weights, a fusion calculation is performed on each risk parameter based on the fusion weights to obtain the driver's corresponding driving risk index. In one implementation, this fusion calculation can be achieved by weighted summing of each risk value with its corresponding fusion weight. For example, refer to the following fusion formula: R = α × R_fatigue + β × R_illness + γ × R_distraction + δ × R_emotion.

[0081] This driving risk index reflects the comprehensive impact of a driver's own factors on driving safety under current multi-dimensional monitoring data. When risk parameters in multiple dimensions are at high levels simultaneously, the risk values ​​are superimposed during the fusion calculation, resulting in a corresponding increase in the driving risk index. When only a single dimension of risk parameters shows a slight abnormality while other dimensions of risk parameters are within the normal range, the driving risk index is at a low level.

[0082] Step S206: Based on the scenario data, determine the risk correction factor corresponding to the current driving environment.

[0083] The contextual data includes information on the current driving environment, such as road type, weather conditions, traffic conditions, and altitude. The risk correction factor characterizes the impact of the current driving environment on driving risk; when applied to the driving risk index, this factor ensures that the risk assessment results reflect changes in risk caused by environmental factors.

[0084] In practice, different environmental factors affect driving risks in different ways, and the degree of risk controllability varies accordingly. Road type information reflects the road geometry and design speed range of the current driving segment. For example, highways have wide lanes and straight lines, giving drivers more room to control the vehicle, and the risk is relatively controllable; while mountain roads have dense curves and limited visibility, resulting in a shorter reaction window for drivers to deal with emergencies, and a lower degree of risk controllability. Weather condition information reflects parameters that affect the vehicle's control boundaries, such as visibility and road surface adhesion. For example, in sunny weather, road surface adhesion is good, and the vehicle's braking and steering responses are within the normal range; in rainy or snowy weather, the road surface is slippery, visibility is reduced, the vehicle's control boundaries are significantly narrowed, and the risk is higher under the same driver conditions. Traffic condition information reflects the density and distribution of traffic participants on the current road. For example, in unobstructed sections of road, there is less traffic interference; in congested sections, vehicle spacing is small and stops and starts are frequent, requiring drivers to maintain a high level of attention continuously, and the consequences of distraction or delayed reaction are more serious. Altitude information reflects the potential impact of changes in oxygen content at high altitudes on the driver's physiological state, as well as the adaptability of the vehicle's powertrain to operating conditions in this environment.

[0085] Based on the above analysis, the correction amount can first be determined according to the risk controllability corresponding to road type information, weather condition information, traffic condition information, and altitude information. The correction amount for each factor reflects the degree to which that factor exacerbates or mitigates driving risk relative to the baseline driving environment.

[0086] Furthermore, a comprehensive correction amount corresponding to the scenario data is determined based on each correction amount. For example, the correction amounts are weighted and combined, and the weights can be determined based on the coupling relationship between the factors. Further, the preset base value of the correction factor can be adjusted based on the comprehensive correction amount to obtain the risk correction factor corresponding to the current driving environment. The base value of the correction factor can be determined based on the risk level under the baseline driving environment, and the adjustment direction and magnitude depend on the comprehensive correction amount. In one implementation, the base value of the correction factor K_base = 1.0 can be set, and the total correction amount can be expressed as: ΔK = ΔK_road + ΔK_weather + ΔK_traffic + ΔK_other.

[0087] The final risk correction factor K can be calculated using the following formula: K = max(0.5, min(2.0, K_base + ΔK)), where the upper and lower limits of K are used to restrict the correction factor within a reasonable range, avoiding distortion of the correction results due to extreme environmental parameters. The correction amount corresponding to each situational factor can be determined based on the direction and degree of its impact on driving risk.

[0088] Taking the correction value ΔK_road corresponding to the road type as an example: when the vehicle is traveling on a highway and the speed exceeds 100km / h, the severity of the accident consequences is relatively high under high-speed driving conditions, and ΔK_road can be taken as +0.25; when the vehicle is traveling on a mountain road with continuous curves, the frequent steering operations require high control precision from the driver, and ΔK_road can be taken as +0.20; when the vehicle is in a severely congested urban area, the overall speed is relatively low, and the severity of the accident consequences is relatively controllable, and ΔK_road can be taken as -0.10.

[0089] Taking the correction amount ΔK_weather corresponding to weather conditions as an example: In heavy rain or snow, the road surface adhesion coefficient is significantly reduced, and ΔK_weather can be taken as +0.20; in dense fog, visibility is greatly reduced, and the driver's perception distance of the road conditions ahead is severely limited, and ΔK_weather can be taken as +0.30.

[0090] Taking the correction value ΔK_traffic corresponding to traffic conditions as an example: In slow-moving sections of road, vehicles frequently start and stop and their speed fluctuates greatly, making drivers prone to anxiety or distraction. ΔK_traffic can be taken as +0.10.

[0091] In addition, certain special scenarios can be handled separately. For example, when the altitude exceeds 2,500 meters, changes in oxygen content in the high-altitude environment can affect the driver's physiological state. In this case, the threshold of the normal lower limit of blood oxygen saturation in the disease risk value assessment can be adjusted. For example, the threshold can be appropriately lowered for every 1,000 meters increase to adapt to normal physiological changes in the high-altitude environment. At the same time, ΔK_other can be set to moderately compensate for the correction factor.

[0092] For example, when a vehicle is traveling on a continuously bumpy road, the reliability of sensor data may decrease due to vibration interference. In this case, ΔK_other can be set to a negative value to make the correction factor more conservative, reducing the risk misjudgment caused by the decline in data quality. Through this method, the risk correction factor K is obtained by superimposing the correction amounts of each situational factor. Each correction amount independently reflects the degree of influence of the corresponding environmental factor on driving risk, and the superimposed correction factor comprehensively reflects the overall risk level of the current driving environment. Applying this correction factor to the driving risk index allows for the dynamic calibration of the risk assessment results by environmental factors.

[0093] Step S208: Determine the basic risk level corresponding to the driving risk index based on the threshold range in which the driving risk index falls among a number of preset dynamic thresholds.

[0094] For the above driving risk index, it can be compared with a plurality of preset dynamic thresholds, and the corresponding basic risk level is determined according to the threshold interval where the index is located. The plurality of dynamic thresholds divide the value range of the driving risk index into several intervals, each interval corresponding to a basic risk level, which can respectively reflect the risk level of the driver's own state within the current monitoring cycle.

[0095] In one embodiment, the basic risk levels at least include a normal level (L0), a mild risk level (L1), a moderate risk level (L2), a severe risk level (L3) and an emergency risk level (L4). The dynamic thresholds corresponding to each basic risk level can be expressed as T1, T2, T3, T4, so as to divide the value range of the driving risk index R into five intervals. The basic risk level corresponding to the driving risk index can be determined by comparison in the following manner: L0: R<T1; L1: T1≤R<T2; L2: T2≤R<T3; L3: T3≤R<T4; L4: R≥T4. In one embodiment, the corresponding dynamic thresholds can be set as: T1=0.25, T2=0.45, T3=0.70, T4=0.90.

[0096] Further, the dynamic thresholds can also be adaptively adjusted according to the driver's historical feedback data or personalized settings, so that the threshold setting can adapt to the driver's actual usage feedback and individual preferences. In one embodiment, the system can obtain the driver's operation records of canceling vehicle risk intervention during historical driving processes.

[0097] Specifically, when the driver frequently cancels the intervention within the risk index interval corresponding to a certain driving risk level, and the driving risk index does not continue to rise within a preset period after the cancellation, or does not trigger a higher-level determination, it indicates that the current threshold may be overly sensitive for this driver. At this time, the target adjustment value of the corresponding dynamic threshold can be determined according to the number of intervention cancellations for this driving risk level in the cancellation operation records, the determination threshold of this level is appropriately raised, and unnecessary intervention triggers are reduced. For example, if the driver frequently cancels the intervention of level L1 when the driving risk index R=0.4, the threshold T1 corresponding to L1 can be finely adjusted from 0.25 to 0.28, and the adjustment range can be set with an upper limit to avoid excessive deviation from the initial setting. In another embodiment, the system can also receive the sensitivity gear selected by the driver through the safety setting menu, and determine the adjustment direction and adjustment range of the plurality of dynamic thresholds according to the sensitivity gear.

[0098] Sensitivity levels can be represented as an overall shift of each dynamic threshold. For example, a driver can select a higher sensitivity level, in which case the dynamic thresholds are adjusted downwards, making it easier for the system to trigger risk level determinations and providing more frequent driving protection; or select a lower sensitivity level, in which case the dynamic thresholds are adjusted upwards, reducing the frequency of risk level triggers. Based on the sensitivity level selected by the driver and its corresponding adjustment direction and magnitude, the values ​​corresponding to multiple dynamic thresholds can be determined.

[0099] In this way, the dynamic threshold can be flexibly adjusted according to the driver's cancellation operation feedback or active sensitivity settings, so that the judgment result of the basic risk level is adapted to the driver's individual characteristics and usage preferences, reducing the interference caused by false alarms and improving the actual user experience of the system.

[0100] Step S210: Use the risk correction factor as the level correction coefficient to correct the level value of the basic risk level, and obtain the latest driving risk level of the vehicle.

[0101] After determining the basic risk level and risk correction factor, the risk correction factor can be used as a level correction coefficient to adjust the level value of the basic risk level. In one implementation, each basic risk level can be quantified into a corresponding level value. For example, the normal level L0, mild risk level L1, moderate risk level L2, severe risk level L3, and emergency risk level L4 can be quantified into values ​​from 0 to 4. The risk correction factor K is multiplied by the basic risk level value to obtain the corrected level value. Furthermore, the corrected level value is mapped back to the corresponding driving risk level, which is then used as the vehicle's latest driving risk level.

[0102] The following is a specific example to illustrate the correction process.

[0103] Assume the current basic risk level is determined to be L2 through the aforementioned steps, corresponding to a level value of 2. Simultaneously, according to step S206, the corresponding risk correction factor K is determined. For example, if the vehicle is currently traveling on a highway and encounters heavy rain, the correction amount corresponding to the road type is +0.25, and the correction amount corresponding to the weather conditions is +0.20. The basic correction factor value is 1.0, so K = 1.0 + 0.25 + 0.20 = 1.45. Multiplying the basic level value 2 by the correction factor K yields a corrected level value of 2 × 1.45 = 2.9, which, after rounding, is approximately 3. Mapping this back to the driving risk level, it corresponds to the final risk level L3. In this example, with the same driver's own condition, due to the combined environmental impact of the highway and heavy rain, the risk level rises from L2 to L3, demonstrating the aggravating effect of environmental factors on driving risk.

[0104] In this way, the revised driving risk level can reflect the vehicle driving risk level under the combined effect of the driver's own state and the external environment, so that subsequent risk intervention strategies can be configured based on the level judgment results that are closer to the actual driving conditions.

[0105] Step S212: Determine the vehicle's risk intervention strategy based on the latest driving risk level.

[0106] After determining the latest driving risk level of a vehicle, a corresponding risk intervention strategy can be determined to translate the risk level into specific control measures that can be executed by the vehicle. This embodiment of the invention pre-constructs a mapping relationship between risk levels and intervention strategies, associating the configuration and operating parameters of controllable actuators corresponding to different risk levels, thus making the conversion from risk level to intervention strategy deterministic.

[0107] In this approach, the risk level, as the output of the assessment phase, can be directly mapped to a set of pre-set intervention actions, enabling the system to respond promptly to changes in risk. Simultaneously, the scope of implementing agencies and operational parameters configured for each risk level in the mapping relationship can be differentiated according to the varying degrees of risk, ensuring that the intensity and scope of intervention measures match the current risk level.

[0108] In one implementation, the mapping relationship between the established risk level and the intervention strategy can be represented by the following table:

[0109] The table above illustrates the mapping relationship between different driving risk levels and corresponding risk intervention strategies in embodiments of the present invention. The driving risk levels increase progressively from L0 to L4, with each level corresponding to a control strategy mode and a set of specific execution actions. These actions are identified by the domain controllers required to execute them. This mapping relationship reflects the correspondence between risk levels and intervention methods. For example, the lower the risk level, the more the intervention leans towards notification and alerting, involving fewer vehicle functional domains; conversely, the higher the risk level, the more the intervention leans towards active control, involving a wider range of vehicle functional domains.

[0110] For example, Level 0 involves no intervention, only data recording; Level 1 regulates the environment through the cabin's air conditioning, fragrance, audio, and lighting systems; Level 2 adds visual, auditory, and tactile cues from the cabin to Level 1; Level 3 extends intervention from the cabin to the chassis and powertrain domains, actively limiting vehicle functions through speed limits and enhanced driver assistance; Level 4 further extends intervention to the autonomous driving, body, and connectivity domains, with the system executing emergency takeover actions. Through this mapping relationship, the intervention strategies corresponding to each driving risk level form a progressively hierarchical structure in terms of intervention intensity, functional domain coverage, and degree of intervention in vehicle control.

[0111] In one implementation, the controllable actuator corresponding to each driving risk level configured in the aforementioned risk-strategy mapping table is determined from multiple functional domains of the vehicle based on the risk level indicated by that driving risk level. These multiple functional domains include the cockpit domain, chassis domain, powertrain domain, autonomous driving domain, body domain, and communication domain, with the controllable actuators in each functional domain undertaking intervention functions at different levels.

[0112] The controllable actuators in the cockpit domain mainly include the air conditioning controller, fragrance system, audio system, lighting system, instrument panel display, speakers, and seat vibration devices. Their intervention function focuses on providing environmental adjustments, visual cues, auditory cues, and tactile cues to the driver, aiming to encourage the driver to actively pay attention to their driving status through sensory cues. The controllable actuators in the chassis domain mainly include the braking unit and suspension controller. Their intervention function focuses on controlling the vehicle's motion, such as decelerating through the braking system or altering the vehicle's dynamic response through suspension adjustments.

[0113] The controllable actuators in the powertrain domain mainly include electric motor controllers or engine control units. Their intervention function focuses on regulating the vehicle's power output, such as limiting the maximum speed or acceleration response, to constrain the vehicle's operating range. The controllable actuators in the autonomous driving domain are mainly autonomous driving controllers. Their intervention function focuses on taking over vehicle control when the driver is unable to continue performing driving duties, executing preset safety actions such as emergency parking or lane-keeping. The controllable actuators in the body domain mainly include hazard light controllers, seatbelt pretensioners, and door lock controllers. Their intervention function focuses on issuing warning signals to external road users and implementing passive safety restraints on vehicle occupants.

[0114] The controllable actuators in the communication domain mainly include the vehicle-mounted emergency call module and the vehicle-to-everything (V2X) communication unit. Their intervention function focuses on sending vehicle location, vehicle status, and driver status information to external rescue platforms to trigger external rescue responses.

[0115] In the risk-strategy mapping table, the range of functional domains invoked for different driving risk levels gradually expands as the risk level increases.

[0116] When the driving risk level is low, the driver is still able to correct the state independently. The intervention strategy only calls the controllable actuators in the cockpit domain. The prompting purpose can be achieved through environmental adjustment or gentle reminders. There is no need to control the vehicle's motion state, so as to avoid interfering with the driver's normal driving.

[0117] When the driving risk level increases, sensory cues alone are no longer sufficient to effectively mitigate the risk. Intervention strategies extend the scope of controllable actuators from the cabin domain to the chassis and power domains, actively constraining the vehicle's operating speed and dynamic response.

[0118] When the driving risk level reaches a high level, the driver may no longer be able to perform safe driving duties. The intervention strategy further expands the scope of invocation to the autonomous driving domain, vehicle body domain and communication domain, and the system takes over vehicle control and triggers external rescue to minimize the risk of accident.

[0119] The operating parameters of the corresponding controllable actuator include output intensity parameters and action time parameters, which are determined according to the risk level indicated by the corresponding driving risk level. For example, the higher the risk level, the larger the output intensity parameter and the longer the action time parameter.

[0120] In one implementation, when the risk level is low, the seat vibration device can operate intermittently with a low vibration intensity to reduce interference to the driver while transmitting tactile alerts; when the risk level is high, the vibration intensity is increased to enhance the warning effect; and in the emergency takeover mode with the highest risk level, the braking actuator operates continuously with a high deceleration until the vehicle comes to a safe stop.

[0121] The differentiated configuration of the above-mentioned working parameters enables the same controllable execution agency to implement interventions with different intensities and durations at different risk levels, ensuring that the intensity of the intervention measures is commensurate with the current risk level.

[0122] Step S214: Control the controllable actuator according to the working parameters to intervene in the risk of the vehicle.

[0123] During vehicle operation, by monitoring the driver's state and driving environment, the corresponding driving risk level of the vehicle can be determined. Based on this, a risk intervention strategy can be determined, and the corresponding controllable actuators can be controlled according to the working parameters defined in the strategy to apply intervention actions to the vehicle that match the current risk level.

[0124] During vehicle operation, the driving risk level may change based on environmental or driver factors. For example, as driving time increases, driver fatigue gradually intensifies, and the driving risk level rises from L1 to L2; or if a sudden downpour occurs during the journey, environmental risks increase significantly, and the driving risk level rises due to environmental correction.

[0125] When the driving risk level changes, the risk intervention strategies corresponding to the old and new levels may differ in the controllable actuators and their operating parameters that need to be activated. If the actuators required by the new strategy are switched to the target operating parameters instantly, or if the actuators that are no longer needed are shut down directly, the sudden change in intervention actions may exceed the driver's expectations.

[0126] From perceiving the intervention signal and understanding the current situation to making corresponding operational adjustments, drivers need a certain psychological buffer time. If the intervention is applied too quickly, the driver may become tense or make mistakes due to insufficient time to adapt, thus introducing additional driving risks. Therefore, when the driving risk level changes, this step adopts a transitional adjustment method, allowing the operating parameters of the controllable actuator to gradually transition from the current value to the target value, rather than abruptly changing.

[0127] During the transition, the driver can perceive the gradually increasing or decreasing intervention signals, and has relatively ample time to understand the current risk situation and respond, thereby ensuring timely intervention while reducing the interference caused by sudden intervention changes to the driver's normal operation.

[0128] In one implementation, this transition adjustment can be achieved in a manner that describes how risk intervention can be carried out: 1) In response to changes in the latest driving risk level compared to the vehicle's previous driving risk level; 2) Adjust the working parameters of the controllable actuators according to the latest driving risk level risk intervention strategy, so as to intervene in the vehicle risk through the latest driving risk level risk intervention strategy.

[0129] When the driving risk level determined in the current period differs from that in the previous period, the change in the corresponding level may be due to driver-related factors such as increased driver fatigue, heightened inattention, or sudden physiological abnormalities. It may also be indirectly caused by changes in the driving environment (such as entering rainy or foggy roads, or navigating mountain curves), which alter the correction results of the driving risk index based on the contextual data. After determining the latest driving risk level, the target controllable actuators and their target operating parameters indicated by the risk intervention strategy corresponding to that level can be obtained. The currently operating controllable actuators and their current operating parameters are then compared with the aforementioned target controllable actuators and their target operating parameters to determine the differences between the two in terms of the scope of actuator activation and the values ​​of operating parameters.

[0130] In one implementation, to address the aforementioned differences, for mechanisms that exist in the target strategy but are not currently running, their operating parameters can be controlled to gradually increase from zero to the corresponding target operating parameters within a preset transition time; for mechanisms that are currently running but are no longer needed in the target strategy, their operating parameters can be controlled to gradually decrease from the current value to zero and then stop; for mechanisms that are currently running and still need to be retained in the target strategy, their operating parameters can be controlled to smoothly transition from the current value to the corresponding target operating parameters.

[0131] In this way, the switch between old and new intervention strategies is achieved through gradual changes in the working parameters of each institution, rather than a sudden jump at the moment of a change in level.

[0132] In practical implementation, the aforementioned control commands can be sent to each domain controller via the vehicle Ethernet for execution. Once the risk intervention strategy corresponding to the latest driving risk level is determined, the system can encapsulate the operating parameters of each controllable actuator into control command messages and send them via the vehicle Ethernet to the corresponding domain controllers in the cockpit, chassis, powertrain, autonomous driving, body, or communication domains. Each domain controller parses the messages and drives the corresponding controllable actuators to perform intervention actions according to the command parameters. Progressive adjustment commands for the operating parameters of each actuator are also continuously sent through this channel. The domain controllers can adjust the output state of the actuators in real time based on the commands updated frame by frame, thereby achieving a smooth transition from the current intervention state to the target intervention state.

[0133] The following specific example further illustrates the execution process of the above transition adjustment: In one implementation, the transition process from L1 to L2 is taken as an example: After entering L2 at time t0, the air conditioning and fragrance release mechanisms in the L1 strategy, which are currently running and need to be retained in the target strategy, continue to operate according to their original parameters. The newly added dashboard warning icons and voice prompts in L2 are mechanisms that exist in the target strategy but are not currently running. Specifically, during the first period from t0 to t0+5 seconds, the dashboard warning icons are displayed with 50% transparency and gradually increase to full display, while the voice prompts are played at 50% volume and gradually increase in volume. During the second period from t0+5 seconds to t0+10 seconds, the seat vibration device is activated with an initial intensity of 15% of the target intensity, and then gradually increases to 30% of the target intensity. After t0+10 seconds, all newly added L2 elements reach their target operating parameters. If the driving risk level drops back to L1 during the transition period, the transition adjustment stops, and the operating parameters of the newly added L2 elements are gradually reduced to zero within 2 seconds and then fade out.

[0134] Take the transition from L2 to L3 as an example: After determining that L3 has been entered at time t1, the warning reminder action of L2 is a reserved mechanism and continues to operate; the newly added speed limit instruction of L3 is a mechanism that exists in the target strategy but is not currently in operation. The target speed limit value is linearly reduced from the current speed to the target value within a preset transition time of 15 seconds, rather than jumping instantaneously, in order to avoid vehicle jerking caused by sudden changes in speed; the lateral control stiffness of the lane keeping assist function gradually increases to the target value within the transition time, and the voice prompt smoothly transitions from a first-level gentle tone to a second-level serious tone.

[0135] The following two typical driving scenarios illustrate the complete execution sequence of this solution in practical applications: In a long-distance fatigue driving scenario, the vehicle continues to travel on the highway. At time t0, the driving risk level is L0, and the system does not intervene, only recording data. At t0+10 minutes, the fatigue level intensifies, rising to L1. The air conditioning temperature is slowly adjusted to 22°C within 30 seconds, the fragrance is released for 5 seconds, and upbeat background music is played. At t0+25 minutes, fatigue further worsens, rising to L2. Following the aforementioned gradual transition mechanism, while maintaining L1 environmental adjustment, warning icons on the instrument panel gradually appear, and seat vibration gradually increases from low intensity. The driver is awakened after perceiving the gradually strengthening warning signals, and the driving state is improved.

[0136] In a sudden illness scenario, if a driver experiences palpitations, the risk value instantly spikes to 0.95, directly classifying the driver as Level 4. At this point, the gradual transition mechanism from Level 0 to Level 3 becomes inapplicable. The system skips the transition process and cancels the waiting phase, immediately completing a full set of Level 4 intervention actions within 5 to 8 seconds, including emergency parking, high-frequency flashing of hazard lights, automatic seat reclining, seatbelt pretensioning, and automatic dialing of emergency assistance, maximizing the driver's safety.

[0137] Through the above methods, this solution adaptively selects a gradual transition or immediate intervention approach based on the magnitude and urgency of the change in driving risk level. For changes between L0 and L3 levels, a phased gradual transition is adopted, with the intervention of newly added mechanisms gradually increasing in a subtle manner and the withdrawal of mechanisms gradually decreasing in a subtle manner, ensuring a smooth and continuous intervention process. For emergency scenarios that directly trigger L4 level, the transition process can be skipped and immediate execution can be carried out, completing vehicle control and driver protection in the shortest possible time, thereby ensuring driving safety while reducing interference to the driver's normal operation caused by sudden intervention changes.

[0138] Furthermore, this embodiment of the invention also outputs a prompt message to indicate that risk intervention is about to be performed; within a preset time after the prompt message is output, if a cancellation command is received from the driver, the transition adjustment is stopped, and the operating parameters of the already activated target controllable actuator are gradually reduced to zero.

[0139] In practical implementation, when a change in driving risk level triggers a transitional adjustment, the newly added controllable actuators gradually intervene from zero. Before executing the intervention, the system can output a warning message to the driver, informing them that the current risk level is about to change and the corresponding intervention measures will be upgraded. For example, this can be done through flashing dashboard icons, sound alerts, or voice announcements. Within a preset time after outputting this warning message, the system waits for driver feedback. During this period, if the driver actively issues a cancellation command through a preset method (such as double-clicking a custom steering wheel button or issuing a preset cancellation voice command), it indicates that the driver has a clear understanding of their driving status or believes that the current intervention upgrade is unnecessary. At this time, the system stops the current transitional adjustment and gradually reduces the operating parameters of the already activated target controllable actuators from their current values ​​to zero before exiting, and the vehicle returns to the intervention state corresponding to the previous driving risk level. If no cancellation command is received from the driver within the preset time, the transitional adjustment continues until all newly added elements reach their target operating parameters.

[0140] Through the above methods, drivers are given a certain degree of decision-making participation during the intervention escalation process, forming real-time collaboration between automatic judgment and human feedback. When a driver actively cancels the intervention after receiving an early warning, it indicates that they still possess autonomous judgment and control capabilities. At this point, the transitional adjustment can be terminated to avoid unnecessary intervention interfering with normal driving. Simultaneously, the system can also record this cancellation event for subsequent adaptive adjustments to dynamic thresholds, allowing the system's adaptability to individual driver preferences and driving habits to gradually improve with increased usage time.

[0141] Furthermore, driver and vehicle status data can be recorded before and after each risk intervention to optimize risk intervention strategies offline based on the driver and vehicle status data.

[0142] Each time a risk intervention is triggered, the system can encrypt and record and store the data across the entire chain before and after the intervention. The recorded content may include multi-dimensional monitoring data at the time of intervention triggering, driving risk index and corresponding driving risk level, working parameters of each controllable execution mechanism during the intervention process and the driver's cancellation operation record, as well as the trend of driver status changes within a preset period after the intervention.

[0143] By performing offline analysis on the above data, the effectiveness of intervention strategies corresponding to each risk level in real-world scenarios can be evaluated. For example, if, after a certain type of intervention strategy is triggered, the driver's fatigue index continues to decrease and the driving risk index returns to the normal range during the intervention period, it indicates that the strategy is effective. If, after a certain type of intervention strategy is triggered, the driver's cancellation rate remains high, it indicates that there is a discrepancy between the timing of the trigger or the intensity of the intervention and the driver's actual needs. In this case, the corresponding trigger threshold or operating parameters can be adjusted during offline optimization.

[0144] Through the above methods, risk intervention strategies can be continuously iterated as actual operational data accumulates, gradually improving the system's ability to adapt to individual driver characteristics and driving habits.

[0145] Furthermore, based on the above embodiments, this invention also provides a vehicle control device, referring to... Figure 3 The device includes: a data acquisition module 10, used to monitor multi-dimensional monitoring data of the driver through multi-dimensional sensors inside the vehicle, and acquire situational data of the vehicle in the current driving environment; a driving risk index determination module 20, used to determine the driver's driving risk index based on the multi-dimensional monitoring data; a driving risk level determination module 30, used to numerically correct the driving risk index based on the situational data, and determine the latest driving risk level of the vehicle based on the corrected driving risk index; an execution module 40, used to determine the risk intervention strategy of the vehicle based on the latest driving risk level; wherein, the risk intervention strategy is configured with at least one controllable actuator, and corresponding operating parameters for the controllable actuator; and a control module 50, used to control the operation of the controllable actuator according to the operating parameters to perform risk intervention on the vehicle.

[0146] The vehicle control device provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0147] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described... Figures 1-2 The steps of any of the methods shown. Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the above-described steps. Figures 1-2 The steps of any of the methods shown. This invention also provides a schematic diagram of an electronic device. Figure 4 A schematic diagram of the electronic device is shown, wherein the electronic device includes a processor 101 and a memory 100. The memory 100 stores computer-executable instructions that can be executed by the processor 101, and the processor 101 executes the computer-executable instructions to implement the above-mentioned... Figures 1-2 Any of the methods shown.

[0148] exist Figure 4 In the illustrated embodiment, the electronic device further includes a bus 102 and a communication interface 103, wherein the processor 101, the communication interface 103, and the memory 100 are connected via the bus 102. The memory 100 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk drive. Communication between this system network element and at least one other network element is achieved through at least one communication interface 103 (which can be wired or wireless), using the Internet, wide area network, local area network, metropolitan area network, etc. Bus 102 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, or an AMBA (Advanced Microcontroller Bus Architecture) bus. AMBA defines three types of buses: APB (Advanced Peripheral Bus), AHB (Advanced High-performance Bus), and AXI (Advanced eXtensible Interface). Bus 102 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4The diagram uses only a single double-headed arrow, but this does not imply a single bus or a single type of bus. Processor 101 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 101 or by instructions in software form. Processor 101 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor 101 reads information from the memory and, in conjunction with its hardware, completes the aforementioned tasks. Figures 1-2 Any of the methods shown.

[0149] The computer program product of the vehicle control method, device, and electronic device provided in this invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. Specific implementations can be found in the method embodiments and will not be repeated here. Those skilled in the art will understand that, for convenience and brevity, the specific working process of the system described above can be referred to the corresponding process in the preceding method embodiments, and will not be repeated here. If the function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0150] In the description of this invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Finally, it should be noted that the above embodiments are merely specific implementations of this invention, used to illustrate the technical solutions of this invention, and not to limit it. The scope of protection of this invention is not limited thereto. Although this invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in this invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this invention, and should all be covered within the scope of protection of this invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A vehicle control method, characterized in that, The method includes: Multi-dimensional monitoring data of the driver is monitored by multi-dimensional sensors inside the vehicle, and contextual data of the vehicle in the current driving environment is obtained. The driver's driving risk index is determined based on the multi-dimensional monitoring data; The driving risk index is numerically corrected based on the scenario data, and the latest driving risk level of the vehicle is determined based on the corrected driving risk index. The risk intervention strategy for the vehicle is determined based on the latest driving risk level; wherein the risk intervention strategy is configured with at least one controllable actuator, and the operating parameters corresponding to the controllable actuator; The controllable actuator is controlled according to the operating parameters to intervene in the risk of the vehicle.

2. The method according to claim 1, characterized in that, Controlling the controllable actuator to operate according to the stated operating parameters for risk intervention in the vehicle includes: In response to a change in the latest driving risk level compared to the previous driving risk level of the vehicle, the operating parameters of the controllable actuator are adjusted transitionally according to the risk intervention strategy of the latest driving risk level, so as to intervene in the risk of the vehicle through the risk intervention strategy of the latest driving risk level.

3. The method according to claim 2, characterized in that, The operating parameters of the controllable actuator are transitionally adjusted according to the risk intervention strategy based on the latest driving risk level, including: Obtain the target controllable execution mechanism and its target operating parameters corresponding to the risk intervention strategy for the latest driving risk level; Based on the target controllable execution mechanism and its target operating parameters, determine the operating status adjustment target and operating parameter adjustment target of the controllable execution mechanism corresponding to the current risk intervention strategy; Within a preset transition period, the working state and working parameters of the controllable actuator are adjusted from their current values ​​to the working state and working parameters corresponding to the latest risk intervention strategy for the driving risk level, according to the working state adjustment target and the working parameter adjustment target.

4. The method according to claim 1, characterized in that, The multi-dimensional monitoring data includes: the driver's facial feature data, physiological feature data, and voice feature data; The driver's driving risk index is determined based on the multi-dimensional monitoring data, including: The risk parameters corresponding to the driver are determined based on the multi-dimensional monitoring data; the risk parameters include: fatigue risk value, disease risk value, distraction risk value and emotional risk value. Based on the driver's current driving environment and the driver's continuous driving time, the fusion weights corresponding to each parameter in the risk parameters are determined. Based on the fusion weights corresponding to each parameter in the risk parameters, the fatigue risk value, the disease risk value, the distraction risk value, and the emotional risk value are fused and calculated to determine the driving risk index corresponding to the driver.

5. The method according to claim 4, characterized in that, The fusion weights corresponding to each parameter in the risk parameters are obtained through the following methods: Acquire driver sample data from cloud data; wherein, the driver sample data includes historical multi-dimensional monitoring data of multiple drivers under different driving environments, historical driving risk index and corresponding actual driving status results; Based on the driver sample data, determine the fusion weights corresponding to the fatigue risk value, disease risk value, distraction risk value, and emotional risk value in the driving environment.

6. The method according to claim 1, characterized in that, The driving risk index is numerically corrected based on the scenario data, and the latest driving risk level of the vehicle is determined based on the corrected driving risk index, including: Based on the scenario data, determine the risk correction factor corresponding to the current driving environment; The basic risk level corresponding to the driving risk index is determined based on the threshold range in which the driving risk index falls among a number of preset dynamic thresholds. The risk correction factor is used as the level correction coefficient to correct the level value of the basic risk level, thereby obtaining the latest driving risk level of the vehicle.

7. The method according to claim 6, characterized in that, The contextual data includes information on road type, weather conditions, traffic conditions, and altitude of the current driving environment. Based on the contextual data, the risk correction factor corresponding to the current driving environment is determined, including: Based on the risk controllability corresponding to the road type information, weather condition information, traffic condition information, and altitude information, the corresponding correction amount is determined for each. Determine the comprehensive correction amount corresponding to the scenario data based on each of the aforementioned correction amounts; The preset base value of the correction factor is adjusted based on the comprehensive correction amount to obtain the risk correction factor corresponding to the current driving environment.

8. The method according to claim 1, characterized in that, Based on the latest driving risk level, determine the risk intervention strategy for the vehicle, including: Retrieve the pre-stored risk-policy mapping table; Based on the risk-strategy mapping table, the risk intervention strategy corresponding to the driving risk level is determined; wherein, the risk-strategy mapping table is configured with multiple driving risk levels, and the controllable actuator corresponding to each driving risk level is determined from multiple functional domains of the vehicle according to the risk level indicated by the driving risk level; The operating parameters of the controllable actuator include the output intensity parameter and / or the action time parameter of the controllable actuator; the output intensity parameter and the action time parameter are determined according to the risk level indicated by the corresponding driving risk level.

9. The method according to claim 8, characterized in that, The multiple functional domains include the cockpit domain, chassis domain, powertrain domain, autonomous driving domain, body domain, and communication domain.

10. The method according to claim 3, characterized in that, Within a preset transition period, according to the working state adjustment target and the working parameter adjustment target, the working state and working parameters of the controllable actuator are transitioned from their current values ​​to the working state and working parameters corresponding to the latest driving risk level risk intervention strategy. The working state and working parameters of the controllable actuator include: During the first transition period, the controllable actuators corresponding to the current risk intervention strategy continue to operate, and the target controllable actuators that need to be activated in the risk intervention strategy of the latest driving risk level are activated, and their working parameters are gradually increased from zero. During the second transition period following the first transition period, the operating parameters of the controllable actuators that are not needed in the current risk intervention strategy are gradually reduced to zero, and the operating parameters of the target controllable actuators are gradually increased to the corresponding target operating parameters.

11. The method according to claim 10, characterized in that, The method further includes: Output a notification message indicating that risk intervention is about to be performed; If a cancellation command is received from the driver within a preset time after the prompt message is output, the transition adjustment is stopped, and the operating parameters of the activated target controllable actuator are gradually reduced to zero.

12. The method according to claim 1, characterized in that, The method further includes: Record driver and vehicle status data before and after each risk intervention to optimize the risk intervention strategy offline based on the driver and vehicle status data.

13. The method according to claim 6, characterized in that, The basic risk levels include at least: normal level, mild risk level, moderate risk level, severe risk level, and emergency risk level; the multiple dynamic thresholds include the thresholds corresponding to each of the basic risk levels.

14. The method according to claim 13, characterized in that, The method further includes: Obtain the driver's cancellation operation records for the risk intervention of the vehicle during historical driving; Based on the number of cancellation interventions for the driving risk level in the cancellation operation record, determine the target adjustment value of the dynamic threshold corresponding to the driving risk level.

15. The method according to claim 6, characterized in that, The method further includes: Receives the sensitivity setting selected by the driver through the safety settings menu; Based on the sensitivity level, determine the adjustment direction and adjustment range of multiple dynamic thresholds; Based on the adjustment direction and the adjustment magnitude, the values ​​corresponding to the multiple dynamic thresholds are determined respectively.

16. A vehicle control device, characterized in that, The device includes: The data acquisition module is used to monitor the driver's multi-dimensional monitoring data through multi-dimensional sensors inside the vehicle, and to acquire the contextual data of the vehicle in the current driving environment. The driving risk index determination module is used to determine the driver's driving risk index based on the multi-dimensional monitoring data. The driving risk level determination module is used to numerically correct the driving risk index based on the scenario data, and determine the latest driving risk level of the vehicle based on the corrected driving risk index. An execution module is used to determine a risk intervention strategy for the vehicle based on the latest driving risk level; wherein the risk intervention strategy is configured with at least one controllable execution mechanism, and the corresponding operating parameters of the controllable execution mechanism; The control module is used to control the controllable actuator to operate according to the operating parameters in order to intervene in the risks of the vehicle.

17. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing machine-executable instructions that can be executed by the processor, the processor executing the machine-executable instructions to implement the vehicle control method according to any one of claims 1 to 15.

18. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores machine-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the vehicle control method according to any one of claims 1 to 15.

Images