Dynamic coordination control method for multi-mode soft rapid switching of power system

Abstract

The invention discloses a multi-mode soft rapid switching dynamic coordination control method for a power system, and relates to the technical field of coordination control, and the method comprises the steps: obtaining a driver physiological state signal and a steering wheel holding physical signal; calculating a physiological ready index, and calculating a physical control ready index; judging whether the physical control ready index is lower than a preset virtual holding threshold value or not; if the judgment result is yes, delaying execution of the power system mode switching request; and if the judgment result is no, dynamically adjusting the torque change rate and the switching time sequence of modal switching of the power system based on the grading result of the physiological ready index. According to the method, the virtual holding state is judged based on the grip space distribution characteristics, the delay execution and tactile feedback wake-up mechanism is combined, whether the limitation exists in the hand or not is judged only through the capacitance signal, accurate recognition and intervention of the dangerous state that the hand is on the steering wheel but has no actual control torque are achieved, and the hand-held state is more accurate. And the safety risk that the vehicle deviates from the lane is effectively avoided.

Description

Technical Field

[0001] This invention relates to the field of coordinated control technology, and in particular to a dynamic coordinated control method for multimodal, smooth, and rapid switching of power systems. Background Technology

[0002] With the rapid development of hybrid and electric vehicle technologies, multi-modal switching control of powertrain systems has become a key technology for improving vehicle energy efficiency and driving performance. In actual driving, vehicles need to frequently switch between pure electric drive, hybrid drive, and energy recovery modes depending on operating conditions. However, current technologies, when performing high-power power switching (such as rapid engine start-up or rapid drive mode transitions), typically rely solely on vehicle operating parameters such as vehicle speed and throttle opening, neglecting the impact of the driver's physiological state and actual control readiness on switching safety.

[0003] Current Hand-Off Detection (HOD) technology relies solely on the steering wheel's capacitive signal to determine hand presence, creating a dangerous blind spot: to deactivate warning lights, drivers may lightly place their hands on the steering wheel without applying effective control torque (i.e., a loose grip). If the system then performs a high-power switch (such as a sudden engine start), the vehicle may veer off course because the driver is not prepared to steer. Furthermore, while existing Driver Monitoring Systems (DMS) can detect fatigue (e.g., through eyelid closure time and head posture), they operate independently of the powertrain control and cannot dynamically adjust the smoothness of the transition process based on the driver's physiological readiness. More critically, current technology lacks the ability to distinguish between two complex states: physiological alertness with a lack of physical control (e.g., a driver who is alert but loosely grips the steering wheel to make a phone call) and physiological fatigue with tense physical control (e.g., a drowsy driver forcing themselves to grip the steering wheel tightly). This results in a failure to accurately match the switching control strategy to safety risks.

[0004] On the other hand, existing power mode switching control uses a fixed calibrated torque change rate, failing to dynamically adjust according to road adhesion conditions, driver cognitive load, and grip status. When the driver is in a loose grip or highly fatigued state, sudden torque shocks not only reduce ride comfort but may also lead to vehicle loss of control because the driver cannot apply effective steering force in time. Therefore, there is an urgent need for a control method that can integrate physiological and physical grip signals, identify the dangerous state of loose grip, and achieve dynamic coordination of smooth and rapid switching to solve the fatal blind spot problem in existing technologies where the hands are on the steering wheel but not yet ready to engage. Summary of the Invention

[0005] The purpose of this invention is to provide a dynamic coordinated control method for multimodal smooth and rapid switching of power systems, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a dynamic coordinated control method for multimodal smooth and rapid switching of a power system, comprising: The driver's physiological state signals and steering wheel grip physical signals are acquired. The physiological state signals include heart rate variability indicators and eye movement characteristics. The grip physical signals include total grip force, spatial distribution characteristics of grip force, and temporal rate of change of grip force. The physiological readiness index is calculated based on the physiological state signal, and the physical control readiness index is calculated based on the grip physical signal and the spatial distribution characteristics of the grip force. Determine whether the physical control readiness index is lower than a preset virtual grip threshold, and determine whether the grip force is in a virtual grip state based on the spatial distribution characteristics of the grip force; If the judgment result is yes, the execution of the power system mode switching request is delayed, and the haptic feedback wake-up program is started until the grip force reconstruction confirmation signal is detected; If the judgment result is negative, then based on the classification result of the physiological readiness index, the torque change rate and switching sequence of the power system mode switching are dynamically adjusted.

[0007] Preferably, acquiring the driver's physiological state signals and steering wheel grip physical signals specifically includes: Electrocardiogram (ECG) signals were collected using bioelectrodes embedded in the steering wheel, and the ratio of low-frequency to high-frequency power was extracted as the heart rate variability index. The duration of eyelid closure and the rate of pupil dilation are collected using in-vehicle vision sensors as the eye movement features; Real-time pressure distribution in the palm and fingertip areas is collected by a multi-zone pressure sensor array on the steering wheel, and the total grip force and the fingertip-palm pressure ratio in the spatial distribution characteristics of the grip force are calculated. The rate of change of grip strength over time is obtained by performing a differential operation on the total grip strength value within a time window. In addition, it acquires driving intention pre-operation signals, which include micro-motion pre-pressure signals of the accelerator pedal and brake pedal, as well as head posture scanning frequency signals.

[0008] Preferably, the physical control readiness index is calculated based on the grip physical signal and the spatial distribution characteristics of the grip force, specifically including: Calculate the grip strength quality factor Q, where Q is the ratio of the average pressure in the fingertip region to the average pressure in the palm base region; Calculate the basic physical readiness index, which is positively correlated with the total grip strength, negatively correlated with the left-right asymmetry of grip strength, and positively correlated with the time-series change rate of grip strength; The physical control readiness index is obtained by weighted coupling of the basic physical readiness index and the grip strength quality factor Q. The "virtual grip" state includes a grip strength quality factor Q that is less than a first preset ratio.

[0009] Preferably, calculating the physiological readiness index based on the physiological state signal specifically includes: Based on the sympathetic-parasympathetic balance in the heart rate variability index, the proportion of eyelid closure duration, and the stability of pupil dilation rate, a multi-dimensional physiological feature vector is constructed. The multidimensional physiological feature vector is mapped to the [0,1] interval to obtain the physiological readiness index; The dynamic adjustment of the torque change rate of the power system mode switching based on the grading results of the physiological readiness index includes: When the physiological readiness index is greater than a first threshold, a fast switching mode is allowed, and the torque change rate adopts a standard slope; When the physiological readiness index is between the first threshold and the second threshold, a gentle switching mode is executed, using an exponential torque ramp curve and a time constant that is negatively correlated with the physiological readiness index. When the physiological readiness index is less than the second threshold, a very gentle switching mode is executed, limiting the torque change rate to no more than 50% of the standard slope.

[0010] Preferably, the execution of the power system mode switching request is delayed, and the haptic feedback wake-up procedure is initiated until a grip strength reconstruction confirmation signal is detected, specifically including: A preset delay window is initiated, and a graded pulse vibration sequence is sent to the steering wheel actuator within the delay window. The frequency and amplitude of the pulse vibration sequence increase with time. Real-time monitoring of whether the grip force temporal change rate exceeds a preset grip rate threshold, and whether the fingertip-palm pressure ratio in the grip force spatial distribution characteristics is greater than a second preset ratio and continues to exceed the confirmation time; If the monitoring result is yes, it is determined that the grip strength reconstruction confirmation signal has been received, and the power system mode switching is performed, but the torque change rate is limited to no more than 60% of the standard slope; If the monitoring result is negative and the delay window times out, the mode switching request is suspended and the current power mode is maintained, while a visual warning is output to the driver.

[0011] Preferably, determining whether the physical control readiness index is lower than a preset virtual grasp threshold, and dynamically adjusting based on the grading result of the physiological readiness index, specifically includes: Establish a three-dimensional state discrimination coordinate system that includes the physiological readiness index, the physical control readiness index, and the driving intention pre-operation signal; Based on the three-dimensional state, the quadrant position in the coordinate system is determined, and the driver's control state is subdivided into the following four categories, with corresponding control strategies executed: Dual Readiness State: When the physiological readiness index is higher than the high readiness threshold, the physical control readiness index is higher than the actual grip threshold, and the driving intention pre-operation signal is present, it is determined that the virtual grip state is not met, and the standard fast switching mode is executed. Awake but holding state: When the physiological readiness index is higher than the high readiness threshold and the physical control readiness index is lower than the preset holding threshold, it is determined that the state is holding and the delayed execution power system mode switching request is executed. Fatigue but real grip state: When the physiological readiness index is lower than the low readiness threshold and the physical control readiness index is higher than the real grip threshold, it is determined that the person is not in a virtual grip state, but a gentle switching mode based on the physiological readiness index is executed. Dual Loss of Control State: When both the physiological readiness index and the physical control readiness index are below the corresponding safety threshold and there is no pre-operation signal of the driving intention, it is determined that the state is in a state of indecisive grip and safety inhibition is triggered, prohibiting the execution of the power system mode switching and initiating a safe stop request.

[0012] Preferably, it also includes a multimodal interlocking constraint mechanism, specifically including: The determination of whether the physical control readiness index is lower than the preset virtual grasp threshold further requires that the following conditions be met simultaneously: the heart rate variability index in the physiological state signal does not show stress-induced elevation characteristics; The method further includes state transition lag control: the transition from the dual-ready state to the virtual-holding state requires a first determination duration to trigger the delayed execution, while the recovery from the virtual-holding state to the dual-ready state only requires a second determination duration to release the delayed execution, wherein the first determination duration is longer than the second determination duration. During the execution of the power system mode switching, the current control strategy is locked until the switching is completed, and the switching is not interrupted due to a momentary drop in the physiological readiness index or the physical control readiness index.

[0013] Preferably, before step S2, a predictive switching constraint step is further included, specifically including: Based on the time series of the physiological state signals, Kalman filtering is used to predict the trend of the physiological readiness index within a preset prediction window in the future. When the physiological readiness index is predicted to fall below the safety threshold and the current physical control readiness index is below the medium readiness level, the range of switchable power system modes is restricted in advance, and the activation of power modes with high power requirements is prohibited. The advance restriction takes effect before the physiological readiness index actually falls below the safety threshold, forming a feedforward control barrier.

[0014] Preferably, it also includes an emergency safety override mechanism, specifically including: Real-time monitoring of vehicle emergency collision warning signals and automatic emergency braking pre-trigger signals; When the emergency collision warning signal or the automatic emergency braking pre-trigger signal is detected to be activated, all switching restrictions based on the physiological readiness index and the physical control readiness index are immediately released, and the power system mode switching to the maximum power output mode is forced to be executed at the maximum torque change rate. Record the triggering events of the emergency safety override mechanism and output a historical event reminder to the driver when the vehicle is started again.

[0015] Preferably, it also includes personalized adaptive learning steps, specifically including: Record the baseline distribution of physiological readiness index and the response pattern of physical control readiness index of a specific driver in a historical driving cycle; Based on the baseline distribution, the personalized offsets of the first threshold, the second threshold, and the preset virtual grip threshold are dynamically adjusted; For drivers who maintain a habit of holding their hands loosely for a long time, the trigger threshold of the haptic feedback wake-up program is gradually reduced, and their grip behavior pattern is optimized through progressive training.

[0016] The technical effects and advantages of this invention are as follows: This invention determines the state of a loose grip based on the spatial distribution characteristics of grip force, and combines delayed execution and tactile feedback wake-up mechanisms. It overcomes the limitation of judging the presence of a hand solely by capacitive signals, and achieves accurate identification and intervention in the dangerous state where the hand is on the steering wheel but there is no actual control torque. When the driver is detected to be in a loose grip state, the system delays the high-power power switch and initiates a progressive tactile wake-up until a grip force reconstruction confirmation signal is detected, effectively avoiding the safety risk of the vehicle deviating from the lane due to the driver not being ready to turn. This invention constructs a three-dimensional state discrimination coordinate system comprising a physiological readiness index, a physical control readiness index, and a pre-operation signal of driving intention. This system enables the differentiation and identification of two complex states: physiological alertness but lack of physical control (e.g., alert but holding the steering wheel loosely to answer a phone call) and physiological fatigue but with tense physical control (e.g., drowsy but forcing oneself to grip the steering wheel firmly). Corresponding control strategies are implemented for different quadrant states (allowing rapid switching between dual-readiness states, triggering delayed wake-up in the alert but loosely gripping state, and performing extremely smooth switching in the fatigued but firmly gripping state), achieving dynamic adaptive matching between the power switching strategy and the driver's actual readiness state. This invention addresses the shortcomings of existing technologies that use a fixed torque change rate, which cannot adapt to driver conditions, by dynamically adjusting the torque change rate based on the physiological readiness index (PRI) classification results (using a standard slope for rapid switching when PRI is high and limiting the torque change rate to ≤50% of the standard slope when PRI is low). Combined with a predictive switching constraint step based on Kalman filtering, this invention solves the problem of existing technologies where a fixed torque change rate cannot adapt to the driver's condition. While ensuring safety, the system can automatically select the most suitable switching smoothness according to the driver's physiological load state, avoiding discomfort and risks caused by torque shocks under high fatigue conditions, and providing rapid power response when the driver is in good condition, thus improving driving safety and ride comfort. This invention establishes a dual-protection system combining conventional fine control and emergency violation rules through a multimodal interlocking constraint mechanism and an emergency safety override mechanism. Multimodal cross-validation effectively avoids erroneous control strategy triggering caused by single sensor failure or transient interference, while the emergency override mechanism ensures that the system can immediately release restrictions to guarantee safety in extreme situations such as collision warnings, thus improving the reliability and safety of mode switching control in complex driving environments. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is the logic diagram for the three-dimensional state discrimination and refined control of the present invention. Detailed Implementation

[0018] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides, for example Figures 1-2 The present invention relates to a dynamic coordinated control method for multimodal smooth and rapid switching of a power system. Before explaining this scheme, some terms will be explained in order to better understand the following explanation.

[0020] Physiological Readiness Index (PRI): A comprehensive index calculated based on physiological signals such as driver heart rate variability and eye movement characteristics. It reflects the driver's cognitive fatigue level and attention level. The value ranges from [0,1]. The higher the value, the more suitable the physiological state is for rapid power switching.

[0021] Physical Control Readiness Index (PCRI): A comprehensive index calculated based on steering wheel grip force, grip force distribution, and grip force change rate, reflecting the driver's actual readiness to control the steering wheel, with a value range of [0,1].

[0022] Grip strength quality factor Q: The ratio of the average pressure in the fingertip area to the average pressure in the palm base area, used to distinguish between a loose grip (palm resting but fingertips lacking strength) and a firm grip (fingers actively clenching).

[0023] "Holding the steering wheel loosely" state: The driver's hands are in contact with the steering wheel but the effective control torque applied is lower than the safety threshold. If a high-power power switch is made in this state, the driver may not be able to apply effective steering force in time.

[0024] See Figure 1 The flowchart illustrates a dynamic coordination control method for multimodal smooth and rapid switching of a power system provided by an embodiment of the present invention. The control method includes: Step S101: Acquire driver's physiological state signal and steering wheel grip physical signal.

[0025] In the specific implementation step S101, driver state data is collected in real time through a multimodal sensor network deployed in the vehicle. Physiological state signals include: Heart rate variability index: ECG signals are collected through embedded bioelectrodes in the steering wheel (such as capacitive ECG sensors), and the power ratio of low frequency (LF, 0.04-0.15Hz) to high frequency (HF, 0.15-0.4Hz) (LF / HF) is extracted as an index of sympathetic nerve activation. Eye movement characteristics: Eyelid closure duration (PERCLOS) and pupil dilation rate are collected using in-vehicle vision sensors (such as DMS cameras) at a sampling frequency of no less than 30fps.

[0026] The physical signals for holding the object include: Total grip force: Real-time pressure distribution in the palm and fingertip areas is collected by a multi-zone pressure sensor array on the steering wheel (such as a capacitive-piezoresistive composite sensor, arranged at the 3 o'clock and 9 o'clock positions on the steering wheel) to calculate the total grip force of both hands (0-100N range). Spatial distribution characteristics of grip force: Calculate the fingertip-palm base pressure ratio, i.e. grip force quality factor Q, where Q is the ratio of the average pressure in the fingertip region to the average pressure in the palm base region; Grip strength change rate over time: This is obtained by differentiating the total grip strength value within a time window (such as a 0.5s sliding window), reflecting the dynamic process of grip strength development.

[0027] In addition, driving intention pre-operation signals are also acquired, including accelerator / brake pedal pre-pressure signals collected by pedal microswitches (stroke resolution ≤2mm), and head posture scanning frequency (such as rearview mirror scanning frequency) collected by vision sensors.

[0028] Step S102: Calculate the physiological readiness index based on physiological state signals, and calculate the physical control readiness index based on grip physical signals and grip force spatial distribution characteristics.

[0029] The specific implementation of step S102 includes the following sub-steps: Calculate the Physiological Readiness Index (PRI): Based on the sympathetic-parasympathetic balance (LF / HF ratio), the percentage of eyelid closure time (PERCLOS, such as the percentage of time spent with eyes closed within 30 seconds) and the stability of pupil dilation rate (inverse variance) in the heart rate variability index, a multi-dimensional physiological feature vector is constructed. This multi-dimensional physiological feature vector is normalized and mapped to the [0,1] interval to obtain the physiological readiness index. For example, when LF / HF < 2, PERCLOS < 0.15, and pupil dilation is stable, PRI ≈ 0.8-1.0, indicating good physiological condition; when LF / HF > 4 and PERCLOS > 0.3, PRI < 0.4, indicating a state of high fatigue.

[0030] Calculate the Physical Control Readiness Index (PCRI): First, calculate the grip strength quality factor Q. For example, when the average fingertip pressure is 5N and the average palm heel pressure is 20N, Q=0.25, indicating a loose grip; when the average fingertip pressure is 15N and the average palm heel pressure is 10N, Q=1.5, indicating a firm grip.

[0031] The basic physical readiness index is calculated, which is positively correlated with the total grip strength (e.g., linearly normalized to [0,1]), negatively correlated with the left-right asymmetry of grip strength, and positively correlated with the temporal rate of change of grip strength (reflecting the degree of positivity in grip strength development).

[0032] The physical control readiness index is obtained by weighted coupling of the basic physical readiness index and the grip strength quality factor Q. For example, PCRI = 0.6 × basic index + 0.4 × Q (when Q > 1, it is counted as 1).

[0033] Step S103: Determine whether the physical control readiness index is lower than the preset virtual grip threshold, and determine whether it is in a virtual grip state based on the spatial distribution characteristics of grip force.

[0034] In the specific implementation of step S103, the determination of the virtual holding state includes: Determine if the PCRI level is below the preset virtual hold threshold (e.g., 0.3). Determine whether the grip strength quality factor Q is less than the first preset ratio (e.g., 0.3). If Q < 0.3, it is determined to be a loose grip. Multimodal interlocking: Further requirements include that the heart rate variability index in the physiological state signal does not show stress-induced increase characteristics (such as a sudden increase in LF / HF exceeding 50% of the baseline, excluding momentary relaxation of grip due to fright).

[0035] If PCRI < 0.3 and Q < 0.3 and HRV does not increase due to stress, the condition is determined to be in a false grasp state, and step S104 is executed; otherwise, step S105 is executed.

[0036] Step S104: If the judgment result is yes, then delay the execution of the power system mode switching request and start the haptic feedback wake-up program until the grip force reconstruction confirmation signal is detected.

[0037] The specific implementation of step S104 includes the following sub-steps: Initiate a preset delay window (e.g., 2-3 seconds). Within the delay window, send a graded pulse vibration sequence to the steering wheel actuator, with the frequency and amplitude of the pulse vibration sequence increasing over time. For example, initial stage: 5Hz, low amplitude (mild warning); middle stage: 10Hz, medium amplitude (stronger warning); final stage: 15Hz, high amplitude (strong warning).

[0038] Real-time monitoring of whether the rate of change of grip force over time exceeds the preset grip rate threshold (e.g., dF / dt>10N / s), and whether the fingertip-palm pressure ratio in the spatial distribution characteristics of grip force is greater than the second preset ratio (e.g., 0.5) and continues to exceed the confirmation time (e.g., 0.5 seconds).

[0039] If the monitoring result is yes, it is determined that a grip strength reconstruction confirmation signal has been received, and the power system mode switching is performed, but the torque change rate is limited to no more than 60% of the standard slope (to ensure safety redundancy).

[0040] If the monitoring result is negative and the delay window times out, the mode switching request is suspended and the current power mode is maintained. At the same time, a visual warning prompt is output to the driver (such as the instrument panel displaying "Please grip the steering wheel firmly").

[0041] Step S105: If the judgment result is negative, then based on the classification result of the physiological readiness index, dynamically adjust the torque change rate and switching sequence of the power system mode switching.

[0042] The specific implementation of step S105 includes: When the physiological readiness index is greater than the first threshold (e.g., 0.7), a fast switching mode is allowed, and the torque change rate adopts a standard slope (e.g., 1000 Nm / s).

[0043] When the physiological readiness index (PRI) is between the first threshold (0.7) and the second threshold (e.g., 0.4), a gentle switching mode is implemented, using an exponential torque ramp curve T(t) = T_target × (1 - e^(-λt)), where the time constant λ is negatively correlated with the physiological readiness index. For example, when PRI = 0.6, λ = 2; when PRI = 0.5, λ = 1.5.

[0044] When the physiological readiness index is less than the second threshold (0.4), a very gentle switching mode is executed, limiting the torque change rate to no more than 50% of the standard slope (e.g., ≤500Nm / s), while extending the switching time.

[0045] Prior to step S102, a predictive switching constraint step is also included: Based on the time series of physiological state signals (such as HRV data from the past 5-10 seconds), Kalman filtering is used to predict the trend of the physiological readiness index within a preset prediction window (such as 1-3 seconds). When the predicted physiological readiness index is about to fall below a safe threshold (such as 0.4) and the current physical control readiness index is below a medium readiness level (such as 0.5), the range of switchable power system modes is restricted in advance (such as prohibiting the start of the engine rapid start mode with high power demand), forming a feedforward control barrier.

[0046] See Figure 2 The diagram illustrates a three-dimensional state discrimination and refined control logic flowchart provided in an embodiment of the present invention, corresponding to the refined implementation of judging whether the physical control readiness index is lower than a preset virtual grasp threshold, and dynamically adjusting the judgment steps based on the grading results of the physiological readiness index: A three-dimensional state discrimination coordinate system is established, which includes the Physiological Readiness Index (PRI), the Physical Control Readiness Index (PCRI), and pre-operation signals of driving intention.

[0047] Based on the quadrant position in the coordinate system determined by the three-dimensional state, the driver's control state is subdivided into the following four categories: Dual Ready State: When PRI>0.7 (high ready threshold), PCRI>0.7 (real grip threshold) and there is a pre-operation signal of driving intention (such as pedal pre-press or head scan frequency>0.5Hz), it is determined that the virtual grip state is not met, and the standard fast switching mode is executed (torque change rate 100%, switching time <200ms).

[0048] Awake but holding state: When PRI>0.7 (awake) and PCRI<0.3 (holding threshold), it is determined to be in a holding state, and the delayed execution of the power system mode switching request is executed (step S104).

[0049] Fatigue but real grip state: When PRI<0.4 (low readiness threshold) and PCRI>0.7 (real grip), it is determined that it is not in a false grip state, but a gentle switching mode based on physiological readiness index grading (extremely gentle mode in step S105) is executed.

[0050] Dual loss of control state: When PRI<0.4, PCRI<0.3 (corresponding safety threshold) and there is no pre-operation signal of driving intention, it is determined to be in a false grip state and safety suppression is triggered, power system mode switching is prohibited and a safe stop request is initiated (such as requesting to pull over or turn on hazard lights).

[0051] Multimodal interlock constraint mechanism and state transition lag control: The transition from a dual-ready state to a virtual grasp state requires a first judgment duration (e.g., 2 seconds) before triggering delayed execution to prevent misjudgments caused by instantaneous signal fluctuations. Conversely, the recovery from the virtual grasp state to the dual-ready state only requires a second judgment duration (e.g., 0.5 seconds) to remove the delayed execution, ensuring rapid restoration of control. The first judgment duration being longer than the second judgment duration aligns with the ergonomic principle that state deterioration requires confirmation and recovery requires agility.

[0052] During the execution of mode switching in the power system (such as the clutch disengagement-engagement process), the current control strategy is locked until the switching is completed. Interruption of the switching due to a momentary drop in the physiological readiness index or physical control readiness index is prohibited to prevent the transmission system from getting stuck in a semi-engaged state.

[0053] Emergency safety override mechanism: The system monitors the vehicle's Emergency Collision Warning (FCW) and Automatic Emergency Braking Pre-Trigger (AEB) signals in real time. Upon detecting the activation of either the FCW or AEB signal, it immediately removes all switching restrictions based on physiological and physical control readiness indices, forcibly switching the powertrain mode to maximum power output mode at the maximum torque change rate (e.g., engine emergency start to provide maximum braking or acceleration). It records the triggering events of the emergency safety override mechanism and provides a historical event reminder to the driver upon the vehicle's next start.

[0054] Personalized adaptive learning steps: Record the baseline distribution of physiological readiness index (PRI) and physical control readiness index (PCI) response patterns of specific drivers in historical driving cycles. For example, if a driver's long-term PRI baseline is 0.6-0.8, the loose grip threshold can be set to 0.25 (lower than the general value of 0.3); or for drivers who maintain a loose grip habit for a long time, gradually reduce the trigger threshold of the haptic feedback wake-up program (e.g., gradually tighten it from Q<0.3 to Q<0.4), and optimize their grip behavior pattern through progressive training.

[0055] The overall control logic flow is illustrated with specific numerical examples: Step S301: Acquire physiological signals (HRV, PERCLOS) and grip signals (F_total, Q, dF / dt), and calculate PRI and PCRI.

[0056] Step S302: Determine if PRI < 0.4 and PCRI < 0.3 and the intent signal is missing? If yes, proceed to step S303 (double runaway → safe stop); if no, proceed to step S304.

[0057] Step S304: Determine if PCRI < 0.3 and Q < 0.3? If yes, proceed to step S305 (awake but with a loose grip → delayed + tactile arousal); if no, proceed to step S306.

[0058] Step S306: Determine if PRI < 0.4 but PCRI > 0.7? If yes, proceed to step S307 (fatigue but firm grip → extremely gentle switching, torque change rate ≤ 50%); if no, proceed to step S308.

[0059] Step S308: Determine if PRI > 0.7 and PCRI > 0.7 and the intent signal is present? If yes, proceed to step S309 (dual ready → fast switch); if no, proceed to step S310 (standard gentle switch).

[0060] As can be seen from the above embodiments, this solution fully considers multimodal factors such as the driver's physiological state, physical control readiness, and driving intention. During the power system mode switching process, it achieves dynamic coordination of virtual grip recognition, fatigue adaptation, and emergency response, minimizing the risk of vehicle loss of control due to insufficient driver preparation, while also taking into account switching response speed and ride comfort.

[0061] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dynamic coordinated control method for multimodal smooth and rapid switching of a power system, characterized in that, include: The driver's physiological state signals and steering wheel grip physical signals are acquired. The physiological state signals include heart rate variability indicators and eye movement characteristics. The grip physical signals include total grip force, spatial distribution characteristics of grip force, and temporal rate of change of grip force. The physiological readiness index is calculated based on the physiological state signal, and the physical control readiness index is calculated based on the grip physical signal and the spatial distribution characteristics of the grip force. Determine whether the physical control readiness index is lower than a preset virtual grip threshold, and determine whether the grip force is in a virtual grip state based on the spatial distribution characteristics of the grip force; If the judgment result is yes, the execution of the power system mode switching request is delayed, and the haptic feedback wake-up program is started until the grip force reconstruction confirmation signal is detected; If the judgment result is negative, then based on the classification result of the physiological readiness index, the torque change rate and switching sequence of the power system mode switching are dynamically adjusted.

2. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 1, characterized in that, Acquiring driver physiological state signals and steering wheel grip physical signals, specifically including: Electrocardiogram (ECG) signals were collected using bioelectrodes embedded in the steering wheel, and the ratio of low-frequency to high-frequency power was extracted as the heart rate variability index. The duration of eyelid closure and the rate of pupil dilation are collected using in-vehicle vision sensors as the eye movement features; Real-time pressure distribution in the palm and fingertip areas is collected by a multi-zone pressure sensor array on the steering wheel, and the total grip force and the fingertip-palm pressure ratio in the spatial distribution characteristics of the grip force are calculated. The rate of change of grip strength over time is obtained by performing a differential operation on the total grip strength value within a time window. In addition, it acquires driving intention pre-operation signals, which include micro-motion pre-pressure signals of the accelerator pedal and brake pedal, as well as head posture scanning frequency signals.

3. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 2, characterized in that, The physical control readiness index is calculated based on the grip physical signal and the spatial distribution characteristics of the grip force, specifically including: Calculate the grip strength quality factor Q, where Q is the ratio of the average pressure in the fingertip region to the average pressure in the palm base region; Calculate the basic physical readiness index, which is positively correlated with the total grip strength, negatively correlated with the left-right asymmetry of grip strength, and positively correlated with the time-series change rate of grip strength; The physical control readiness index is obtained by weighted coupling of the basic physical readiness index and the grip strength quality factor Q. The "virtual grip" state includes a grip strength quality factor Q that is less than a first preset ratio.

4. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 1, characterized in that, The physiological readiness index is calculated based on the aforementioned physiological state signals, specifically including: Based on the sympathetic-parasympathetic balance, the proportion of eyelid closure duration, and the stability of pupil dilation rate among the heart rate variability indicators, a multi-dimensional physiological feature vector is constructed. The multidimensional physiological feature vector is mapped to the [0,1] interval to obtain the physiological readiness index; The dynamic adjustment of the torque change rate of the power system mode switching based on the grading results of the physiological readiness index includes: When the physiological readiness index is greater than a first threshold, a fast switching mode is allowed, and the torque change rate adopts a standard slope; When the physiological readiness index is between the first threshold and the second threshold, a gentle switching mode is executed, using an exponential torque ramp curve and a time constant that is negatively correlated with the physiological readiness index. When the physiological readiness index is less than the second threshold, a very gentle switching mode is executed, limiting the torque change rate to no more than 50% of the standard slope.

5. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 1, characterized in that, The execution of the power system mode switching request is delayed, and the haptic feedback wake-up procedure is initiated until a grip strength reconstruction confirmation signal is detected, specifically including: A preset delay window is initiated, and a graded pulse vibration sequence is sent to the steering wheel actuator within the delay window. The frequency and amplitude of the pulse vibration sequence increase with time. Real-time monitoring of whether the grip force temporal change rate exceeds a preset grip rate threshold, and whether the fingertip-palm pressure ratio in the grip force spatial distribution characteristics is greater than a second preset ratio and continues to exceed the confirmation time; If the monitoring result is yes, it is determined that the grip strength reconstruction confirmation signal has been received, and the power system mode switching is performed, but the torque change rate is limited to no more than 60% of the standard slope; If the monitoring result is negative and the delay window times out, the mode switching request is suspended and the current power mode is maintained, while a visual warning is output to the driver.

6. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 2, characterized in that, Determining whether the physical control readiness index is lower than a preset virtual grasp threshold, and dynamically adjusting it based on the grading results of the physiological readiness index, specifically includes: Establish a three-dimensional state discrimination coordinate system that includes the physiological readiness index, the physical control readiness index, and the driving intention pre-operation signal; Based on the three-dimensional state, the quadrant position in the coordinate system is determined, and the driver's control state is subdivided into the following four categories, with corresponding control strategies executed: Dual Readiness State: When the physiological readiness index is higher than the high readiness threshold, the physical control readiness index is higher than the actual grip threshold, and the driving intention pre-operation signal is present, it is determined that the virtual grip state is not met, and the standard fast switching mode is executed. Awake but holding state: When the physiological readiness index is higher than the high readiness threshold and the physical control readiness index is lower than the preset holding threshold, it is determined that the state is holding and the delayed execution power system mode switching request is executed. Fatigue but real grip state: When the physiological readiness index is lower than the low readiness threshold and the physical control readiness index is higher than the real grip threshold, it is determined that the person is not in a virtual grip state, but a gentle switching mode based on the physiological readiness index is executed. Dual Loss of Control State: When both the physiological readiness index and the physical control readiness index are below the corresponding safety threshold and there is no pre-operation signal of the driving intention, it is determined that the state is in a state of indecisive grip and safety inhibition is triggered, prohibiting the execution of the power system mode switching and initiating a safe stop request.

7. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 6, characterized in that, It also includes a multimodal interlock constraint mechanism, specifically including: The determination of whether the physical control readiness index is lower than the preset virtual grasp threshold further requires that the following conditions be met simultaneously: the heart rate variability index in the physiological state signal does not show stress-induced elevation characteristics; The method further includes state transition lag control: the transition from the dual-ready state to the virtual-holding state requires a first determination duration to trigger the delayed execution, while the recovery from the virtual-holding state to the dual-ready state only requires a second determination duration to release the delayed execution, wherein the first determination duration is longer than the second determination duration. During the execution of the power system mode switching, the current control strategy is locked until the switching is completed, and the switching is not interrupted due to a momentary drop in the physiological readiness index or the physical control readiness index.

8. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 1, characterized in that, Prior to step S2, a predictive handover constraint step is also included, specifically including: Based on the time series of the physiological state signals, Kalman filtering is used to predict the trend of the physiological readiness index within a preset prediction window in the future. When the physiological readiness index is predicted to fall below the safety threshold and the current physical control readiness index is below the medium readiness level, the range of switchable power system modes is restricted in advance, and the activation of power modes with high power requirements is prohibited. The advance restriction takes effect before the physiological readiness index actually falls below the safety threshold, forming a feedforward control barrier.

9. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 1, characterized in that, It also includes an emergency safety override mechanism, specifically including: Real-time monitoring of vehicle emergency collision warning signals and automatic emergency braking pre-trigger signals; When the emergency collision warning signal or the automatic emergency braking pre-trigger signal is detected to be activated, all switching restrictions based on the physiological readiness index and the physical control readiness index are immediately released, and the power system mode switching to the maximum power output mode is forced to be executed at the maximum torque change rate. Record the triggering events of the emergency safety override mechanism and output a historical event reminder to the driver when the vehicle is started again.

10. The dynamic coordinated control method for multimodal smooth and rapid switching of a power system according to claim 1, characterized in that, It also includes personalized adaptive learning steps, specifically including: Record the baseline distribution of physiological readiness index and the response pattern of physical control readiness index of a specific driver in a historical driving cycle; Based on the baseline distribution, the personalized offsets of the first threshold, the second threshold, and the preset virtual grasp threshold are dynamically adjusted. For drivers who maintain a habit of holding their hands loosely for a long time, the trigger threshold of the haptic feedback wake-up program is gradually reduced, and their grip behavior pattern is optimized through progressive training.

Images