Biofeedback-based active suspension control method, device, medium and program product

Abstract

The application provides a biofeedback-based active suspension control method, device, medium and program product. The method comprises the following steps: acquiring biofeedback signals and behavior state signals of a driver, vehicle driving state information, road excitation information and individual comfort preference parameters; determining driver physiological state parameters according to the biofeedback signals and determining driving intention parameters according to the behavior state signals, and determining a target control strategy of the active suspension in combination with the vehicle driving state information and the individual comfort preference parameters; matching the road excitation information with a working condition event library, determining target suspension control parameters based on preset suspension control parameters and the target control strategy when the matching is successful, and determining the target suspension control parameters online when the matching is unsuccessful; and generating and executing active suspension actuation instructions based on the target suspension control parameters, the vehicle driving state information and the road excitation information. The application can improve the individualized adaptation ability and working condition response ability of active suspension control.

Description

Technical Field

[0001] This application relates to the field of vehicle suspension control technology, and in particular to an active suspension control method, device, medium and program product based on biofeedback. Background Technology

[0002] With the development of vehicle intelligence and chassis electronic control technology, active suspension systems are increasingly being applied to vehicle attitude control and ride comfort control. Active suspension systems typically use electromagnetic, hydraulic, or electric actuators to actively adjust the forces, displacements, or damping characteristics between the vehicle body and the wheels to suppress pitch, roll, and vertical vibrations generated when the vehicle accelerates, brakes, turns, changes lanes, or traverses uneven road surfaces, thereby improving the vehicle's ride comfort, handling stability, and occupant comfort.

[0003] Existing active suspension control schemes typically use vehicle speed, vehicle acceleration, vehicle attitude, wheel bounce, road surface anticipation information, or vehicle control targets as inputs, generating suspension actuation commands based on preset damping coefficients, stiffness parameters, roof control, position servo control, or model predictive control. While these schemes can improve vehicle vibration and attitude changes to some extent, their control logic mainly focuses on closed-loop adjustment around the vehicle's motion state, lacking real-time perception and feedback of the driver's physiological state, tension level, driving intentions, and individual comfort preferences. Therefore, when different drivers have varying perceptions of suspension stiffness, body support, and vehicle response, using uniform calibration parameters or limited driving modes for suspension adjustment easily leads to homogenized suspension responses, failing to meet the personalized comfort needs of different drivers. In scenarios involving human-vehicle co-driving or intelligent driving intervention, there may be inconsistencies between the intelligent driving stability target, the driver's control intention, and the ride comfort target. Existing active suspensions struggle to dynamically coordinate with the driver's state in a timely manner, easily causing the driver to perceive a misalignment between vehicle attitude adjustments and their own driving expectations. Furthermore, for high-frequency or sudden road surface excitations such as speed bumps and bridge joints, if the suspension control still mainly uses rigid position tracking or fixed control curves for adjustment, it is easy to generate a strong mechanical shock, affecting the smoothness of the suspension execution process and the ride comfort.

[0004] Therefore, there is an urgent need in this field for an active suspension control scheme that can integrate driver biofeedback signals, driving intentions, vehicle driving status and road excitation information, so that the active suspension can perform personalized, low-latency and compliant control according to the driver's state and vehicle operating conditions. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application provides a biofeedback-based active suspension control method, device, medium, and program product, which at least solves the problem that existing active suspension control schemes lack real-time perception and dynamic adaptation of the driver's physiological state, driving intention, and individual comfort preferences, resulting in homogenized suspension response, insufficient human-vehicle coordination, and insufficient ride comfort.

[0006] To achieve the above objectives and other advantages, some embodiments of this application provide the following aspects:

[0007] In a first aspect, some embodiments of this application provide an active suspension control method based on biofeedback, comprising:

[0008] Acquire the driver's biofeedback signals and behavioral state signals, vehicle driving state information, road surface stimulus information, and individual comfort preference parameters corresponding to the driver;

[0009] The driver's physiological state parameters are determined based on the biofeedback signals, and the driving intention parameters are determined based on the behavioral state signals.

[0010] The target control strategy for the active suspension is determined based on the driver's physiological state parameters, the driving intention parameters, the vehicle driving state information, and the individual comfort preference parameters.

[0011] The road surface excitation information is matched with a preset working condition event database;

[0012] If a target operating condition event that meets the preset matching conditions is matched, the target suspension control parameters of the active suspension are determined according to the preset suspension control parameters corresponding to the target operating condition event and the target control strategy.

[0013] If no target operating condition event that meets the matching conditions is matched, the target suspension control parameters are determined online based on the target control strategy, the vehicle driving status information, and the road excitation information.

[0014] The active suspension actuation command is generated based on the target suspension control parameters, the vehicle driving state information, and the road excitation information. The active suspension actuator is then controlled to execute the active suspension actuation command to adjust the force and / or relative motion state between the vehicle body and the wheels.

[0015] Optionally, the biofeedback signal includes steering wheel grip force signal, heart rate signal, and heart rate variability signal;

[0016] The step of determining the driver's physiological state parameters based on the biofeedback signal includes:

[0017] The rate of change of steering wheel grip force is determined based on the changes in the steering wheel grip force signal within a preset time window;

[0018] The current heart rate value within a preset time window is determined based on the heart rate signal, and the relative change of the current heart rate value relative to the driver's baseline heart rate is calculated to obtain the heart rate deviation rate.

[0019] Based on the heart rate variability signal, a normalized heart rate variability parameter is determined, and the normalized heart rate variability parameter is transformed by a reciprocal to obtain the reciprocal of the heart rate variability.

[0020] The driver's physiological state parameters are obtained by fusing the steering wheel grip force change rate, the heart rate deviation rate, and the reciprocal of the heart rate variability.

[0021] In this alternative approach, the driver's physiological state is quantified by fusing the rate of change in steering wheel grip force, the rate of deviation in heart rate, and the inverse of heart rate variability. This approach can accurately characterize the driver's level of tension or physiological stress, avoiding the instability in state recognition caused by relying solely on a single heart rate signal.

[0022] Optionally, the behavioral state signal includes seat pressure distribution signal, eye movement signal, head posture signal, and steering wheel torque signal;

[0023] The step of determining driving intention parameters based on the behavioral state signal includes:

[0024] The driver's center of gravity offset parameters and torso tilt parameters are determined based on the seat pressure distribution signal.

[0025] The driver's gaze point movement direction and / or gaze area change parameters are determined based on the eye movement signals, and a first pre-aiming intention feature is determined based on the gaze point movement direction and / or gaze area change parameters.

[0026] The driver's head steering posture parameters are determined based on the head posture signal, and the second aiming intention feature is determined based on the head steering posture parameters;

[0027] The steering wheel torque direction and / or steering wheel torque change rate are determined based on the steering wheel torque signal, and the control confirmation intention feature is determined based on the steering wheel torque direction and / or the steering wheel torque change rate.

[0028] The first aiming intent feature, the second aiming intent feature, and the control confirmation intent feature are sequentially correlated and fused for consistency. The fusion result is then corrected based on the center of gravity offset parameter and the torso tilt parameter to obtain the driving intent parameter.

[0029] In this alternative solution, by performing time-series correlation and consistency fusion of multi-source behavioral state signals such as seat pressure, eye movement, head posture, and steering wheel torque, the driver's steering, lane change, or aggressive control intentions can be identified earlier and more accurately, enabling the active suspension to respond to the driver's posture before or at the initial stage of the actual control action.

[0030] Optionally, the step of determining the target control strategy of the active suspension based on the driver's physiological state parameters, the driving intention parameters, the vehicle driving state information, and the individual comfort preference parameters includes:

[0031] Construct a multi-party game decision-making model with the needs of manual driving operation, intelligent driving control, and passenger comfort as the players;

[0032] Based on the driver's physiological state parameters, determine the dynamic weight adjustment conditions of the multi-party game decision-making model;

[0033] The first benefit value corresponding to the manual driving control requirement is determined based on the driving intention parameters; the second benefit value corresponding to the intelligent driving control requirement is determined based on the vehicle driving status information; and the third benefit value corresponding to the ride comfort requirement is determined based on the individual comfort preference parameters.

[0034] Based on the dynamic weight adjustment conditions, a multi-objective game equilibrium solution is performed on the first benefit value, the second benefit value, and the third benefit value to obtain the dynamic control weight of each player under the current working condition.

[0035] The suspension target state is determined based on the dynamic control weights, and the game equilibrium solution containing the dynamic control weights and the suspension target state is determined as the target control strategy.

[0036] In this alternative approach, a multi-party game-theoretic decision-making model is constructed to address the needs of manual driving, intelligent driving, and passenger comfort. The control weights for each need are dynamically determined based on the driver's physiological state, driving intentions, vehicle status, and individual comfort preferences. This avoids using a fixed priority for suspension control among manual driving, intelligent driving, and comfort objectives, improving the coordination and adaptability of the target control strategy under different operating conditions.

[0037] Optionally, the step of determining the target suspension control parameters of the active suspension based on the preset suspension control parameters corresponding to the target suspension event and the target control strategy if a target operating condition event that meets the preset matching conditions is matched includes:

[0038] If a target operating condition event that meets the preset matching conditions is matched, the preset suspension control curve corresponding to the target operating condition event is invoked.

[0039] The personalized correction amount is determined based on the individual comfort preference parameters, and the personalized correction amount is superimposed on the baseline corresponding to the preset suspension control curve to obtain the preset suspension control parameters.

[0040] The preset suspension control parameters are modified according to the target control strategy to obtain the target suspension control parameters.

[0041] In this alternative approach, when a known operating condition event is matched, the response delay caused by complex online calculations can be reduced by calling the preset suspension control curve and making corrections based on individual comfort preferences, thus enabling a rapid response to typical operating conditions such as speed bumps and bridge joints.

[0042] Optionally, the step of generating active suspension actuation commands based on the target suspension control parameters, the vehicle driving state information, and the road surface excitation information includes:

[0043] Determine the target suspension state and target impedance parameters based on the target suspension control parameters;

[0044] The current vehicle body state and current wheel end state are determined based on the vehicle driving state information, and the estimated road surface force is determined based on the road surface excitation information;

[0045] The displacement control component is determined based on the deviation between the suspension target state and the current vehicle body state;

[0046] The force control component is determined based on the estimated force on the road surface and the current wheel end state;

[0047] Based on the target impedance parameters, the displacement control component and the force control component are subjected to force-position hybrid control to obtain the target driving force for each wheel;

[0048] The active suspension actuation command is generated based on the target power of each wheel.

[0049] In this alternative scheme, the displacement control component and the force control component are controlled by force-position hybrid control through the target impedance parameter. This can suppress changes in vehicle body posture while taking into account the wheel end's compliance with road excitation, thus reducing the mechanical impact caused by rigid position control.

[0050] Optionally, after the step of generating the active suspension actuation command based on the target power of each wheel, the method further includes:

[0051] Obtain the driver's heart rhythm information and determine the diastolic window based on the heart rhythm information;

[0052] Determine whether the active suspension actuation command is a vehicle posture adjustment command that meets the preset posture adjustment range conditions and allows for delayed execution;

[0053] If the active suspension actuation command is a vehicle posture adjustment command that satisfies the preset posture adjustment range condition and allows for delayed execution, then the execution time of the active suspension actuation command is adjusted to within the cardiac diastolic window, and the active suspension actuator is controlled to execute the active suspension actuation command within the cardiac diastolic window.

[0054] If the active suspension actuation command does not meet the preset attitude adjustment range condition, or if the active suspension actuation command is not allowed to be delayed, then the active suspension actuation command will be executed at the original execution time.

[0055] In this alternative approach, by determining the diastolic window based on the driver's heart rhythm and executing vehicle posture adjustment commands that meet the conditions within that window, the abruptness and disjointedness caused by large-scale suspension posture adjustments to the driver can be reduced, and the coordination between active suspension actions and the driver's physiological rhythm can be improved, thereby enhancing the sense of synchronization between the driver and the vehicle.

[0056] Secondly, some embodiments of this application also provide an electronic device, the electronic device comprising:

[0057] One or more processors; and a memory storing computer program instructions that, when executed, cause the processors to perform the biofeedback-based active suspension control method as described above.

[0058] Thirdly, some embodiments of this application also provide a computer-readable storage medium having a computer program and / or instructions stored thereon, which, when executed by a processor, implement the biofeedback-based active suspension control method as described above.

[0059] Fourthly, some embodiments of this application also provide a computer program product, including a computer program and / or instructions, which, when executed by a processor, implement the biofeedback-based active suspension control method as described above.

[0060] Compared with existing technologies, the solution provided in this application obtains the driver's biofeedback signals and behavioral state signals, and determines the driver's physiological state parameters and driving intention parameters accordingly. This allows the active suspension control to incorporate the driver's current physiological reactions and control intentions, rather than relying solely on vehicle driving status or preset suspension modes. This improves the matching degree between active suspension control and the driver's real-time tactile needs. Determining the target control strategy based on the driver's physiological state parameters, driving intention parameters, vehicle driving status information, and individual comfort preference parameters allows the active suspension control parameters to dynamically change with the driver's state, control needs, and individual comfort preferences. This enables continuous and personalized parameter adjustment for different drivers, reducing the problems of coarse control granularity, abrupt mode switching, and insufficient individual adaptation in existing fixed-gear suspension modes. Regarding the matching results between road excitation information and the operating condition event database, when a target operating condition event is matched, the corresponding preset suspension control parameters are used to determine the target suspension control parameters, which helps improve the response speed under known operating conditions. When no target operating condition event is matched, the target suspension control parameters are determined online, which helps improve adaptability under unknown road conditions. Therefore, this application can improve the ride comfort, posture stability, and human-vehicle cooperative control effect of active suspension under different drivers, different road surface excitations, and different driving scenarios. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other implementation methods can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart illustrating an active suspension control method based on biofeedback provided in an embodiment of this application.

[0063] Figure 2 This is a timing diagram of the active suspension actuation command execution based on heart rhythm information provided in an embodiment of this application;

[0064] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0066] Some embodiments of this application relate to an active suspension control method based on biofeedback, see reference Figure 1 As shown, this embodiment may include the following steps:

[0067] Step S1: Obtain the driver's biofeedback signals and behavioral state signals, vehicle driving state information, road surface incentive information, and individual comfort preference parameters corresponding to the driver.

[0068] In some embodiments, the vehicle control system acquires driver-side data, vehicle-side data, and road-side data according to preset acquisition cycles or triggering conditions corresponding to various types of data. Driver-side data includes the driver's biofeedback signals, behavioral state signals, and individual comfort preference parameters corresponding to the current driver; vehicle-side data includes vehicle driving state information; and road-side data includes road surface stimulus information.

[0069] A vehicle control system can be implemented by a single control unit or by multiple on-board control units working collaboratively. When the vehicle control system is implemented by a single control unit, this control unit can acquire various multi-source data through the on-board communication network or sensor interface and execute subsequent active suspension control processes. When the vehicle control system is implemented collaboratively by multiple on-board control units, driver-side data, vehicle-side data, and road-side data can be acquired by the corresponding cockpit control unit, chassis control unit, vehicle control unit, or intelligent driving control unit, respectively, and converged via the on-board communication network to the target control unit for active suspension control decisions. The target control unit determines the active suspension actuation command based on the converged multi-source data and sends the active suspension actuation command to the active suspension actuator. The target control unit can be an active suspension controller, chassis domain controller, vehicle controller, or vehicle central computing unit.

[0070] In some embodiments, the biofeedback signal reflects the driver's physiological response state during the current driving process and may include at least one of steering wheel grip force signal, heart rate signal, and heart rate variability signal. The steering wheel grip force signal can be collected by a grip force sensor located in the steering wheel grip area, such as a capacitive grip force sensor or a pressure-sensitive grip force sensor, which reflects changes in the force with which the driver grips the steering wheel. The heart rate signal and heart rate variability signal can be collected by a photoplethysmography (PPG) sensor on the steering wheel, an onboard heart rate detection module, or a wearable device connected to the vehicle, corresponding to changes in the driver's heart rate level and heart rate rhythm.

[0071] Behavioral state signals reflect the driver's body posture, gaze changes, and control actions during the current driving process, and may include at least one of seat pressure distribution signals, eye movement signals, head posture signals, and steering wheel torque signals. Specifically, seat pressure distribution signals can be used to determine the driver's center of gravity shift and torso tilt; eye movement signals can be used to determine the direction of the driver's gaze point movement, changes in the gaze area, or the direction of gaze pre-aiming; head posture signals can be used to determine the driver's head turning angle, turning direction, or posture change trend; and steering wheel torque signals can reflect the direction and intensity of the driver's steering wheel control.

[0072] Vehicle driving status information describes the current motion and chassis status of the vehicle, and may include at least one of the following: vehicle speed, longitudinal acceleration, lateral acceleration, vertical acceleration, yaw rate, roll angle, pitch angle, steering wheel angle, steering wheel angular velocity, brake pressure, accelerator pedal opening, wheel speed, suspension travel, vehicle height, wheel end vertical speed, and vehicle driving mode. This vehicle driving status information may be provided by a vehicle inertial measurement unit, wheel speed sensors, steering system sensors, braking system sensors, suspension height sensors, or the vehicle controller.

[0073] Road surface excitation information indicates road disturbances that a vehicle is about to pass or is currently passing over. This information can include at least one of the following: road surface unevenness, road surface height changes, peak road impact values, road surface excitation frequency, wheel-end vertical acceleration, wheel speed fluctuation characteristics, suspension compression or rebound changes, preview road surface image features, speed bump recognition results, bridge joint recognition results, and pothole recognition results. Road surface excitation information can be obtained from a forward-looking camera, LiDAR, millimeter-wave radar, wheel-end acceleration sensors, suspension displacement sensors, wheel speed sensors, navigation map data, or a road event recognition module.

[0074] Individual comfort preference parameters correspond to the driver's personalized needs for active suspension control and may include at least one of the following: suspension stiffness preference, vertical vibration filtering preference, roll suppression preference, pitch suppression preference, steering support preference, historical suspension mode selection records, historical driving scores, and historical ride comfort feedback. Individual comfort preference parameters can be retrieved from local storage, the cockpit account management module, or the cloud account based on the driver's identity, or they can be learned from the driver's historical biofeedback signals, behavioral state signals, and suspension control feedback results.

[0075] In some embodiments, after acquiring the aforementioned multi-source data, time synchronization and validity verification can be performed on data from different sampling frequencies and sources. For example, using the system clock of the vehicle control system as a reference, timestamps are added to biofeedback signals, behavioral state signals, vehicle driving state information, and road excitation information, and data alignment is performed according to a preset time window. For data with missing samples, abnormal signals, or communication interruptions, the most recent valid value can be preserved, interpolation compensation, or degraded control methods can be used to reduce the impact of inconsistent timing of multi-source signals or abnormal data on subsequent active suspension control strategies.

[0076] Step S2: Determine the driver's physiological state parameters based on the biofeedback signals, and determine the driving intention parameters based on the behavioral state signals.

[0077] In some embodiments, after acquiring the biofeedback signal, the biofeedback signal can be filtered, denoised, and normalized. Feature parameters reflecting the driver's physiological state, such as tension level, fatigue level, heart rate change trend, or grip strength change trend, can be extracted from the processed biofeedback signal to obtain the driver's physiological state parameters. These parameters can be used to characterize the driver's physiological stress state or comfort sensitivity state during the current driving process.

[0078] Furthermore, temporal analysis and feature extraction can be performed on behavioral state signals to determine the driver's control intentions during the current driving process. Specifically, based on behavioral state signals such as the driver's body posture, gaze direction, head posture, and steering wheel operation status, it can be identified whether the driver intends to steer, change lanes, brake, accelerate, avoid road obstacles, or enhance vehicle posture support, thereby obtaining driving intention parameters. Driving intention parameters may include at least one of the following: driving intention type, driving intention direction, driving intention intensity, and intention confidence.

[0079] Step S3: Determine the target control strategy for the active suspension based on the driver's physiological state parameters, driving intention parameters, vehicle driving status information, and individual comfort preference parameters.

[0080] In some embodiments, driver physiological state parameters, driving intention parameters, vehicle driving state information, and individual comfort preference parameters are used as the basis for control strategy decisions to determine the target control strategy that the active suspension should currently adopt. When determining the target control strategy, the needs for handling stability, ride comfort, and individual preferences can be comprehensively considered to determine the priority, control intensity, target vehicle posture, target damping, or target support level of the active suspension control. Therefore, the target control strategy is no longer a simple selection of fixed comfort, standard, or sport modes, but a control strategy dynamically determined based on driver state, driving intention, vehicle state, and individual preferences.

[0081] Step S4: Match the road surface excitation information with the preset working condition event library.

[0082] In some embodiments, road disturbance features are extracted from road excitation information, and the extracted road disturbance features are matched with known operating condition events in a preset operating condition event library. Road disturbance features may include at least one of the following: road excitation amplitude, road excitation frequency, impact peak value, duration, wheel-end response features, preview image features, or road event recognition results. The preset operating condition event library may include typical operating condition events such as speed bumps, bridge joints, potholes, continuous wavy roads, manhole covers, road surface protrusions, or road surface settlement, as well as the event features and preset suspension control parameters corresponding to each operating condition event.

[0083] In some embodiments, the preset matching conditions may include at least one of a matching degree threshold condition, a feature range condition, a temporal consistency condition, and a confidence level condition. Specifically, the matching degree threshold condition may be that the matching degree between the road disturbance feature and the event feature corresponding to a known working condition event is greater than a preset matching threshold; the feature range condition may be that at least one of the road excitation amplitude, road excitation frequency, impact peak value, duration, or wheel-end response feature falls within a preset feature range corresponding to the known working condition event; the temporal consistency condition may be that the pre-aiming identification result and the wheel-end response feature satisfy a triggering timing relationship within a preset time window, or that the same disturbance pattern is continuously matched a preset number of times within a preset time window; and the confidence level condition may be that the identification confidence level corresponding to the road event identification result is greater than a preset confidence threshold.

[0084] If the road surface disturbance features and the event features corresponding to a known working condition event meet preset matching conditions, then that known working condition event is identified as the target working condition event. If multiple known working condition events meet the preset matching conditions, then the known working condition event with the highest matching degree, the highest identification confidence, or the highest priority is identified as the target working condition event. If the road surface disturbance features and none of the known working condition events in the working condition event database meet the preset matching conditions, then the road surface disturbance pattern corresponding to the road surface excitation information is determined to be out of the working condition event database, belonging to an unknown working condition or an atypical working condition.

[0085] For example, when the features of the preview image indicate that there is a speed bump ahead, and the impact peak, duration and left and right wheel triggering sequence in the wheel end response features match the event features corresponding to the speed bump event in the working condition event library, and the corresponding matching degree is greater than the preset matching threshold, the speed bump event can be identified as the target working condition event, and the preset suspension control parameters corresponding to the speed bump event can be called in subsequent steps.

[0086] Step S5: If a target operating condition event that meets the preset matching conditions is matched, the target suspension control parameters of the active suspension are determined according to the preset suspension control parameters and target control strategy corresponding to the target operating condition event.

[0087] In some embodiments, when the road disturbance characteristics and a known operating condition event in the operating condition event library meet preset matching conditions, the known operating condition event is determined as the target operating condition event, and the preset suspension control parameters corresponding to the target operating condition event are invoked. The preset suspension control parameters may include at least one of target damping, target stiffness, target action force, target vehicle attitude adjustment, target impedance parameters, or preset suspension control curves. Simultaneously, the preset suspension control parameters are modified according to the target control strategy to obtain target suspension control parameters adapted to the current driver state, vehicle state, and individual comfort preferences. This allows for improved response speed under known operating conditions using preset parameters, while avoiding the direct application of fixed control parameters.

[0088] Step S6: If no target operating condition event that meets the matching conditions is matched, the target suspension control parameters are determined online based on the target control strategy, vehicle driving status information and road excitation information.

[0089] In some embodiments, when the road disturbance mode does not match the operating condition event library, the preset suspension control parameters stored in the operating condition event library are not invoked; instead, the online suspension decision-making process is initiated. In this process, the target control strategy is used as the control objective or weighting basis, and the target suspension control parameters corresponding to the current control cycle are calculated in real time, combining vehicle driving state information and road excitation information. Specifically, an online suspension decision model can be invoked. The online suspension decision model can be set in the target control unit used for active suspension control decision-making, and can be a rule mapping model, a multi-objective optimization model, a neural network model, or a combination thereof.

[0090] After receiving the target control strategy, the online suspension decision model analyzes the control objectives and weight allocations of the current active suspension control. Control objectives may include at least one of the following: ride comfort objectives, handling stability objectives, road impact suppression objectives, and individual comfort preference objectives. Weight allocations may include the weight values ​​corresponding to each control objective within the current control cycle. For example, when the target control strategy indicates a high demand for handling stability, the weights corresponding to vehicle posture support, roll suppression, or pitch suppression are increased; when the target control strategy indicates a high demand for comfort, the weights corresponding to vehicle vertical vibration suppression or road impact filtering are increased.

[0091] The vehicle control system can also extract vehicle state features based on vehicle driving status information and road disturbance features based on road excitation information. These vehicle state features may include at least one of the following: vehicle speed, vertical acceleration, roll angle, pitch angle, yaw rate, vehicle height, wheel speed, suspension travel, and wheel-end vertical acceleration.

[0092] In some embodiments, the online suspension decision model generates target suspension control parameters for the current control cycle online based on control objectives, weight allocation, vehicle state characteristics, and road disturbance characteristics. The target suspension control parameters may include target suspension states and / or target impedance parameters; wherein the target suspension states may include at least one of target vehicle body angular velocity, target vehicle body vertical acceleration, and target vehicle body pitch angle, and the target impedance parameters may include at least one of target stiffness parameters and target damping parameters.

[0093] In one specific implementation, the online suspension decision model can generate candidate parameters for comfort, handling stability, and shock suppression, respectively. These candidate parameters are then fused according to the weight allocation in the target control strategy to obtain the target suspension control parameters for the current control cycle. For example, comfort candidate parameters can focus on reducing vehicle vertical acceleration and pitch changes, handling stability candidate parameters can focus on suppressing roll and improving vehicle attitude support, and shock suppression candidate parameters can focus on reducing the transmission of peak road impacts to the vehicle body.

[0094] In some embodiments, the target suspension control parameters determined online can also be subject to boundary constraints and smoothing. Boundary constraints may include limiting the maximum and minimum values ​​of target stiffness parameters, target damping parameters, target dynamic force, or target vehicle body attitude adjustment; smoothing may include limiting the rate of change of target suspension control parameters between adjacent control cycles. When road excitation information is abnormal, vehicle driving state information is missing, or online calculation results exceed the safe range, the most recently effective control parameters, preset safe control parameters, or degraded control parameters can be used as the target suspension control parameters for the current control cycle.

[0095] In this way, when no target operating condition event is matched, the vehicle control system can get rid of its dependence on the preset parameters of known operating conditions and generate target suspension control parameters online based on the current target control strategy, the real-time status of the vehicle and the characteristics of road disturbances, thereby improving the adaptability of the active suspension to unknown or atypical operating conditions.

[0096] Step S7: Generate active suspension actuation commands based on target suspension control parameters, vehicle driving status information and road excitation information, and control the active suspension actuators to execute the active suspension actuation commands in order to adjust the force and / or relative motion state between the vehicle body and the wheels.

[0097] In some embodiments, regardless of whether the target suspension control parameters are determined via the known operating condition fast path in step S5 or the unknown operating condition online path in step S6, the vehicle control system can generate active suspension actuation commands based on the target suspension control parameters, vehicle driving state information, and road excitation information. These active suspension actuation commands may include at least one of actuation force commands, displacement adjustment commands, damping adjustment commands, stiffness adjustment commands, or motor drive commands. The vehicle control system sends the active suspension actuation commands to the active suspension actuators. The actuators adjust the forces and / or relative motion between the vehicle body and wheels according to the active suspension actuation commands, thereby reducing the transmission of road excitation to the vehicle body and improving vehicle posture stability and ride comfort. The active suspension actuators may include electromagnetic actuators, hydraulic actuators, air suspension actuators, linear motors, or other actuators capable of actively adjusting the forces and / or relative displacements between the vehicle body and wheels.

[0098] Compared with existing technologies, the solution provided in this application obtains the driver's biofeedback signals and behavioral state signals, and determines the driver's physiological state parameters and driving intention parameters accordingly. This allows the active suspension control to incorporate the driver's current physiological reactions and control intentions, rather than relying solely on vehicle driving status or preset suspension modes. This improves the matching degree between active suspension control and the driver's real-time tactile needs. Determining the target control strategy based on the driver's physiological state parameters, driving intention parameters, vehicle driving status information, and individual comfort preference parameters allows the active suspension control parameters to dynamically change with the driver's state, control needs, and individual comfort preferences. This enables continuous and personalized parameter adjustment for different drivers, reducing the problems of coarse control granularity, abrupt mode switching, and insufficient individual adaptation in existing fixed-gear suspension modes. Regarding the matching results between road excitation information and the operating condition event database, when a target operating condition event is matched, the corresponding preset suspension control parameters are used to determine the target suspension control parameters, which helps improve the response speed under known operating conditions. When no target operating condition event is matched, the target suspension control parameters are determined online, which helps improve adaptability under unknown road conditions. Therefore, this application can improve the ride comfort, posture stability, and human-vehicle cooperative control effect of active suspension under different drivers, different road surface excitations, and different driving scenarios.

[0099] Optionally, in some embodiments, the biofeedback signals include steering wheel grip force signals, heart rate signals, and heart rate variability signals. After acquiring the aforementioned biofeedback signals, the vehicle control system can first perform filtering, noise reduction, outlier removal, and time window truncation processing on the steering wheel grip force signals, heart rate signals, and heart rate variability signals to obtain valid biofeedback data within the same preset time window. The preset time window can be a sliding time window prior to the current control moment, such as a time window of 0.5S to 5S, or it can be dynamically adjusted according to the vehicle driving status or suspension control cycle.

[0100] Step S2, the step of determining the driver's physiological state parameters based on biofeedback signals, specifically includes:

[0101] Step S201: Determine the rate of change of steering wheel grip force based on the changes in the steering wheel grip force signal within a preset time window.

[0102] In some embodiments, multiple grip force sampling values ​​corresponding to the steering wheel grip force signal within a preset time window can be obtained, and the steering wheel grip force change rate can be determined based on the multiple grip force sampling values. For example, the steering wheel grip force change rate can be determined based on the ratio between the change in the steering wheel grip force signal within the preset time window and the time length, and the steering wheel grip force change rate can be normalized to obtain the normalized steering wheel grip force change rate.

[0103] In one specific implementation, the normalized rate of change of steering wheel grip force can be expressed as:

[0104]

[0105] in, This represents the normalized rate of change in steering wheel grip force. This indicates the change in steering wheel grip force within a preset time window. This indicates the duration of the preset time window. This represents the normalization function.

[0106] The greater the rate of change in steering wheel grip force, the more significant the change in the force with which the driver grips the steering wheel, and it can be used to reflect the driver's grip tension or changes in control intensity.

[0107] Step S202: Determine the current heart rate value within the preset time window based on the heart rate signal, and calculate the relative change of the current heart rate value relative to the driver's baseline heart rate to obtain the heart rate deviation rate.

[0108] In some embodiments, the current heart rate value can be determined based on the heart rate signal within a preset time window. The current heart rate value can be the average, median, or smoothed value of multiple heart rate samples within the preset time window. The driver's baseline heart rate can be the heart rate baseline value corresponding to the driver in a stable driving state, a resting state, or a historical low-stress state, or it can be dynamically updated based on the driver's historical heart rate data.

[0109] In one specific implementation, the heart rate deviation rate can be expressed as:

[0110]

[0111] in, Indicates heart rate deviation rate. This indicates the current heart rate value. This indicates the driver's baseline heart rate.

[0112] The larger the heart rate deviation rate, the more significant the increase in the driver's current heart rate relative to the baseline heart rate, which can be used to reflect the driver's physiological arousal level or stress level.

[0113] Step S203: Determine the normalized heart rate variability parameter based on the heart rate variability signal, and perform a reciprocal transformation on the normalized heart rate variability parameter to obtain the reciprocal of the heart rate variability.

[0114] In some embodiments, heart rate variability parameters can be extracted from the heart rate variability signal and normalized to obtain normalized heart rate variability parameters. These parameters may include the root mean square of the difference between adjacent heartbeat intervals, the standard deviation of heartbeat intervals, or other parameters reflecting the degree of fluctuation in heart rhythm. Typically, heart rate variability may decrease when a driver is under stress or tension; therefore, a reciprocal transformation can be used to establish a positive relationship between heart rate variability characteristics and the degree of physiological stress.

[0115] In one specific implementation, the reciprocal of heart rate variability can be expressed as:

[0116]

[0117] in, This represents the reciprocal of heart rate variability. This represents the normalized heart rate variability parameter. This indicates a preset correction amount, used to avoid calculation errors caused by a denominator that is zero or too small.

[0118] when At lower levels, The level is relatively high, thus reflecting an increased level of physiological stress in drivers.

[0119] Step S204: The steering wheel grip force change rate, heart rate deviation rate, and the reciprocal of heart rate variability are fused to obtain the driver's physiological state parameters.

[0120] In some embodiments, the normalized steering wheel grip force change rate, heart rate deviation rate, and inverse of heart rate variability can be input into a preset physiological stress model. The preset physiological stress model can then fuse these multiple biofeedback features to obtain a sympathetic nerve excitation index, which can be used as a parameter of the driver's physiological state.

[0121] In one specific implementation, the pre-defined physiological stress model can be represented as:

[0122]

[0123] in, Indicates the sympathetic nervous system excitability index. This represents the normalized rate of change in steering wheel grip force. Indicates heart rate deviation rate. This represents the reciprocal of heart rate variability. , and These represent the corresponding fusion weights. The fusion weights can be determined based on experimental calibration, driver historical data, or vehicle default configuration, or can be dynamically adjusted based on driver identity, driving scenario, or vehicle driving status.

[0124] Using the methods described above, the rate of change in steering wheel grip force can reflect changes in driver grip tension or control intensity, the heart rate deviation rate can reflect the degree of increase in the driver's heart rate relative to a baseline state, and the reciprocal of heart rate variability can reflect the degree of physiological stress corresponding to a decrease in heart rate rhythm stability. Integrating these multiple biofeedback features can reduce misjudgments caused by noise, individual differences, or short-term fluctuations in individual biosignals, resulting in driver physiological state parameters that more stably reflect the driver's current physiological stress state.

[0125] In the solution provided in this embodiment, steps S201 to S204 fuse the steering wheel grip force change rate, heart rate deviation rate, and the reciprocal of heart rate variability as biofeedback features of different dimensions. This simultaneously reflects the driver's grip tension, relative increase in heart rate, and decrease in heart rate rhythm stability, thereby more accurately quantifying the driver's current physiological stress state. Compared to methods that judge the driver's state based solely on a single heart rate signal or a single grip force signal, this solution reduces misjudgments caused by individual differences, instantaneous noise, or short-term fluctuations in a single sensor signal. It improves the stability and reliability of the driver's physiological state parameters, allowing subsequent active suspension control strategies to be dynamically adjusted based on a more accurate understanding of the driver's physiological state to better suit the driver's current level of tension and sensory needs.

[0126] Optionally, in some embodiments, the behavioral state signals include seat pressure distribution signals, eye movement signals, head posture signals, and steering wheel torque signals. The vehicle control system can acquire the above-mentioned behavioral state signals within a preset recognition time window and align them according to the sampling frequency and timestamp of each behavioral state signal. For example, the seat pressure distribution signal can be collected by a pressure sensor array installed on the driver's seat, the eye movement signal can be collected by an in-vehicle infrared camera, the head posture signal can be identified by a driver monitoring camera, and the steering wheel torque signal can be collected by a steering wheel torque sensor. Based on the above multi-source behavioral state signals, the driver's driving intentions in driving scenarios such as steering, lane changing, swerving, or braking can be identified.

[0127] Step S2, the step of determining the driving intention parameters based on the behavioral state signal, specifically includes:

[0128] Step S205: Determine the driver's center of gravity offset parameters and torso tilt parameters based on the seat pressure distribution signal.

[0129] In some embodiments, the seat pressure distribution signal can be acquired by an array of pressure sensors disposed in the driver's seat cushion and / or backrest area. For example, the seat pressure distribution signal can be acquired by a 1024-point capacitor array at a sampling frequency of 100Hz. Based on the pressure distribution changes within a preset time window, the offset of the driver's pressure center relative to the reference sitting posture pressure center is determined, and a center of gravity offset parameter is obtained accordingly. The center of gravity offset parameter may include at least one of the center of gravity offset direction, center of gravity offset distance, and center of gravity offset speed. The driver's torso tilt parameter can also be determined based on the left-right pressure difference, the front-back pressure difference, and the pressure distribution changes in the backrest area. The torso tilt parameter may include at least one of the torso tilt direction, torso tilt angle, and torso tilt rate of change.

[0130] For example, when a driver's center of gravity shifts to one side before entering a curve or changing lanes, and the pressure on the corresponding side of the backrest increases, it can be determined that the driver has a pre-response related to that direction. When the driver's center of pressure shifts forward rapidly, it can be determined that the driver has a tendency to brake, avoid obstacles, or lean forward in tension. The above-mentioned center of gravity shift parameters and torso tilt parameters can be used as a basis for correcting the subsequent driving intention fusion results.

[0131] Step S206: Determine the driver's gaze point movement direction and / or gaze area change parameters based on the eye movement signal, and determine the first pre-aiming intention feature based on the gaze point movement direction and / or gaze area change parameters.

[0132] In some embodiments, eye-tracking signals can be acquired by an in-vehicle infrared camera at a preset sampling frequency, such as acquiring driver eye images or gaze direction at 60Hz. The positional changes of the driver's gaze point in the road ahead, inside a curve, adjacent lane, rearview mirror, or side area are determined based on the eye-tracking signals. Furthermore, gaze area change parameters are determined based on the gaze point movement direction, gaze point dwell time, gaze area switching frequency, and the angle of deviation of the gaze point relative to the vehicle's direction of travel.

[0133] A first anti-aiming intention feature is determined when the driver's gaze point moves from the center area ahead towards the inside of the curve, the target lane, or the direction of avoidance, and this movement trend continues to meet a preset time condition. For example, a gaze point moving towards the inside of the curve can anticipate the driver's steering intention; a gaze point moving towards the adjacent lane or the side / rear area can anticipate a lane change intention. The first anti-aiming intention feature may include at least one of anti-aiming intention direction, anti-aiming intention type, and anti-aiming intention confidence level.

[0134] Step S207: Determine the driver's head steering posture parameters based on the head posture signal, and determine the second aiming intention feature based on the head steering posture parameters.

[0135] In some embodiments, driver head posture signals can be acquired using a driver monitoring camera or in-vehicle image acquisition device, and the driver's head yaw angle, pitch angle, roll angle, and corresponding rate of change of these angles can be identified. Based on the direction, magnitude, and duration of the head yaw angle change, driver head steering posture parameters can be determined. When the driver's head continuously rotates to the left, right, inside a curve, or towards the target lane, a second anticipation intention feature can be determined. This second anticipation intention feature can be used to confirm the first anticipation intention feature determined by eye movement signals. For example, when eye movement signals indicate that the driver's gaze point is moving towards the left lane, and head posture signals indicate that the driver's head is turning to the left, the confidence level of a left lane change or left turn intention can be increased.

[0136] Step S208: Determine the steering wheel torque direction and / or steering wheel torque change rate based on the steering wheel torque signal, and determine the control confirmation intention feature based on the steering wheel torque direction and / or steering wheel torque change rate.

[0137] In some embodiments, the direction, amplitude, rate of change, and duration of torque applied to the steering wheel by the driver can be determined based on the steering wheel torque signal. The steering wheel torque direction reflects the driver's actual control direction, and the steering wheel torque rate of change reflects the urgency or intensity of the driver's control action. When the steering wheel torque direction is consistent with the direction indicated by the first or second pre-aiming intention feature, and the steering wheel torque amplitude or rate of change is greater than a preset threshold, a control confirmation intention feature can be determined. The control confirmation intention feature may include at least one of control direction, control intensity, and control confirmation confidence. For example, when both the eye movement signal and the head posture signal point to the left target area, and the steering wheel torque signal further characterizes the driver applying torque to the left, it can be determined that the left turn or left lane change intention has entered the control confirmation stage from the pre-aiming stage.

[0138] Step S209: Perform temporal correlation and consistency fusion on the first pre-aiming intention feature, the second pre-aiming intention feature, and the control confirmation intention feature, and correct the fusion result according to the center of gravity offset parameter and the torso tilt parameter to obtain the driving intention parameter.

[0139] In some embodiments, the first anticipation intention feature, the second anticipation intention feature, and the control confirmation intention feature can be temporally correlated according to the intention advance amount corresponding to the behavioral state signal. For example, eye movement signals can serve as earlier anticipation signals, which can anticipate the driver's intention before obvious steering wheel control actions occur; head posture signals can serve as auxiliary anticipation confirmation signals; and steering wheel torque signals can serve as confirmation signals that are closer to the actual control actions.

[0140] Further determine whether the first pre-aiming intention feature, the second pre-aiming intention feature, and the control confirmation intention feature meet the preset temporal relationship and directional consistency conditions within the preset recognition time window. If the intention directions of the first pre-aiming intention feature, the second pre-aiming intention feature, and the control confirmation intention feature are the same or similar, and the triggering order meets the temporal relationship of eye-tracking pre-aiming, head posture confirmation, and steering wheel torque control confirmation, then increase the confidence of the corresponding driving intention; if there is a directional conflict or abnormal triggering sequence among the intention features, then decrease the confidence of the corresponding driving intention, or use the intention feature with higher confidence as a candidate intention.

[0141] After obtaining the fusion result, the fusion result is corrected according to the center of gravity offset parameter and torso tilt parameter determined in step S205. For example, when the driver's center of gravity offset direction and torso tilt direction are consistent with the intention direction indicated by eye movement signal, head posture signal or steering wheel torque signal, the strength or confidence of the driving intention can be increased; when the center of gravity offset direction is inconsistent with the direction of other intention features, the confidence of the driving intention can be reduced, so as to reduce misjudgments caused by short-term eye movement, accidental steering wheel touch or non-driving related head movements.

[0142] The final driving intent parameters can include at least one of driving intent type, driving intent direction, driving intent intensity, and intent confidence. Driving intent type can include at least one of steering, lane changing, braking, acceleration, obstacle avoidance, or maintaining stable driving. Therefore, based on the temporal and consistency relationships between eye movements, head posture, steering wheel torque, and seat pressure distribution, driving intent can be identified before or in the early stages of the driver's actual control actions, and these driving intent parameters can be provided to subsequent active suspension target control strategy determination steps.

[0143] In the solution provided in this embodiment, by combining behavioral state signals such as seat pressure distribution, eye movement, head posture, and steering wheel torque through steps S205 to S209, the driver's driving intention can be identified from multiple dimensions, including body posture, eye gaze prediction, and actual control. By temporally correlated and fused the features of the predicted intention and the control confirmation intention, and by using center of gravity offset parameters and torso tilt parameters for correction, misjudgments caused by short-term eye gaze deviation, occasional head movements, or slight steering wheel disturbances can be reduced, improving the accuracy and reliability of driving intention parameters. Therefore, the active suspension can respond in advance before or at the initial stage of control actions such as steering, lane changing, avoidance, or braking, improving vehicle posture support and driver-vehicle coordination.

[0144] Optionally, in some embodiments, step S3 specifically includes:

[0145] Step S301: Construct a multi-party game decision-making model with manual driving operation requirements, intelligent driving control requirements, and passenger comfort requirements as the players.

[0146] In some embodiments, the multi-party game decision model can be stored as a program module in the memory of the target control unit and executed by the processor of the target control unit. The multi-party game decision model can be a multi-objective coordinated decision model based on benefit values ​​and dynamic weights. It uses manual driving control requirements, intelligent driving control requirements, and passenger comfort requirements as the players, and the dynamic control weights of each player under the current operating conditions as the variables to be solved. Specifically, the manual driving control requirements correspond to the driver's current control intention's requirements for vehicle posture support, steering response, braking support, or obstacle avoidance stability; the intelligent driving control requirements correspond to the vehicle's intelligent driving or assisted driving control requirements for trajectory stability, vehicle posture stability, and ride comfort; and the passenger comfort requirements correspond to the current driver's personalized comfort requirements for vertical vibration filtering, suspension stiffness, roll suppression, and pitch suppression. The dynamic control weights corresponding to each player can satisfy weights and constraints, for example, the sum of the weights corresponding to manual driving control requirements, intelligent driving control requirements, and passenger comfort requirements is 1.

[0147] Step S302: Determine the dynamic weight adjustment conditions of the multi-party game decision-making model based on the driver's physiological state parameters.

[0148] In some embodiments, the vehicle control system determines dynamic weight adjustment conditions based on driver physiological state parameters. These parameters may include a sympathetic nervous system excitation index, which reflects the driver's current level of tension or physiological stress. The sympathetic nervous system excitation index is compared with a preset threshold, and the weight adjustment direction among manual driving control requirements, intelligent driving control requirements, and passenger comfort requirements is determined based on the comparison result.

[0149] Specifically, when the sympathetic nerve excitation index is greater than the first threshold, it indicates that the driver may be in a state of high physiological stress, intense operation, or tension. In this case, the dynamic weight adjustment condition prioritizes manual driving control needs to enhance the impact of vehicle posture support, steering response, braking support, or obstacle avoidance stability on the target control strategy. When the sympathetic nerve excitation index is less than the second threshold and the intelligent driving function is activated, it indicates that the driver is in a state of low physiological stress and the vehicle possesses intelligent driving control capabilities. In this case, the dynamic weight adjustment condition prioritizes intelligent driving control needs to enhance the impact of trajectory stability, vehicle posture stability, and smooth control on the target control strategy. When the sympathetic nerve excitation index is between the first and second thresholds, the dynamic weight adjustment condition is determined to be a multi-demand balance adjustment, allowing manual driving control needs, intelligent driving control needs, and ride comfort needs to all participate in subsequent weight allocation.

[0150] The first threshold is greater than the second threshold. For example, the first threshold can be 0.7 and the second threshold can be 0.3; the first and second thresholds can also be dynamically adjusted based on vehicle calibration results, driver historical data, driving scenarios, or individual comfort preference parameters.

[0151] Step S303: Determine the first benefit value corresponding to the manual driving control requirements based on the driving intention parameters, determine the second benefit value corresponding to the intelligent driving control requirements based on the vehicle driving status information, and determine the third benefit value corresponding to the ride comfort requirements based on the individual comfort preference parameters.

[0152] In some embodiments, a first benefit value can be determined based on driving intention parameters. When the driving intention parameters indicate that the driver has a strong intention to steer, change lanes, brake, avoid obstacles, or perform aggressive maneuvers, the first benefit value corresponding to the manual driving control requirements can be increased; when the driving intention is weak or the confidence level of the intention is low, the first benefit value can be decreased.

[0153] The second benefit value can be determined based on vehicle driving status information. When the vehicle is in intelligent driving or assisted driving control state, or when the vehicle driving status indicates that there is a need for trajectory stabilization, roll suppression, pitch suppression, or vehicle ride comfort control, the second benefit value corresponding to the intelligent driving control need can be increased.

[0154] The third benefit value can be determined based on individual comfort preference parameters. The closer the current suspension control direction matches the current driver's individual comfort preference, or the stronger the demand for ride comfort under the current operating conditions, the greater the third benefit value corresponding to the ride comfort demand can be.

[0155] Step S304: Based on the dynamic weight adjustment conditions, perform multi-objective game equilibrium solution on the first benefit value, the second benefit value, and the third benefit value to obtain the dynamic control weight of each player under the current working condition.

[0156] In some embodiments, the first benefit value, the second benefit value, and the third benefit value are normalized according to the dynamic weight adjustment conditions, and a multi-objective game equilibrium is solved under the weight value range, weight and constraint, and control smoothness constraint to obtain the dynamic control weights corresponding to the manual driving operation requirements, intelligent driving control requirements, and ride comfort requirements under the current working conditions.

[0157] Specifically, under dynamic weight adjustment conditions prioritizing manual driving control needs, the influence of the first benefit value on the weight allocation result can be increased, thereby increasing the dynamic control weight corresponding to manual driving control needs. Under dynamic weight adjustment conditions prioritizing intelligent driving control needs, the influence of the second benefit value on the weight allocation result can be increased. Under multi-demand equilibrium adjustment conditions, the equilibrium weights among the three parties can be determined based on the relative magnitudes of the first, second, and third benefit values. The multi-objective game equilibrium solution can be achieved through at least one of Nash equilibrium solution, multi-objective optimization solution, or rule mapping solution. To avoid frequent jumps in the active suspension target control strategy, the rate of change of dynamic control weights within adjacent control cycles can be limited or smoothed.

[0158] Step S305: Determine the target state of the suspension based on the dynamic control weights, and determine the game equilibrium solution containing the dynamic control weights and the target state of the suspension as the target control strategy.

[0159] In some embodiments, candidate suspension target states corresponding to manual driving control requirements, intelligent driving control requirements, and ride comfort requirements can be fused according to the dynamic control weights corresponding to each player to obtain the suspension target state under the current operating condition. The suspension target state may include at least one of the target vehicle body angular velocity, target vehicle body vertical acceleration, and target vehicle body pitch angle. For example, when the dynamic control weight corresponding to manual driving control requirements is high, the suspension target state may focus more on roll suppression, pitch suppression, and vehicle attitude support; when the dynamic control weight corresponding to intelligent driving control requirements is high, the suspension target state may focus more on trajectory smoothness and vehicle motion stability; when the dynamic control weight corresponding to ride comfort requirements is high, the suspension target state may focus more on reducing vehicle body vertical acceleration and road impact transmission.

[0160] The game equilibrium solution, which includes dynamic control weights and the target suspension state, is determined as the target control strategy, and this strategy is then provided to the subsequent target suspension control parameter determination step. Therefore, active suspension control no longer employs fixed comfort, standard, or sport modes, but can dynamically determine the control strategy based on the driver's physiological state, driving intentions, vehicle status, and individual comfort preferences, achieving coordination between manual driving control, intelligent driving control, and ride comfort.

[0161] In the solution provided in this embodiment, through steps S301 to S305, the needs of manual driving, intelligent driving control, and ride comfort are treated as multi-party game decision objects. The weighting conditions of each need are dynamically adjusted based on the driver's physiological state parameters. This allows the active suspension control to move away from a fixed priority or fixed suspension mode, and instead dynamically coordinate based on the driver's current level of tension, driving intentions, vehicle driving status, and individual comfort preferences. By calculating the benefit values ​​corresponding to the needs of manual driving, intelligent driving control, and ride comfort respectively, and solving the multi-objective game equilibrium, a more reasonable trade-off can be formed between handling stability, smooth intelligent driving control, and ride comfort. This reduces the decrease in comfort or insufficient posture support caused by single-objective control, making the active suspension target control strategy more suitable for the current driving scenario and the driver's individual needs.

[0162] Optionally, in some embodiments, step S5 specifically includes:

[0163] Step S501: If a target operating condition event that meets the preset matching conditions is matched, the preset suspension control curve corresponding to the target operating condition event is invoked.

[0164] In some embodiments, when it is determined from road excitation information that the road disturbance mode and a known operating condition event in the operating condition event library meet preset matching conditions, the matched known operating condition event is identified as the target operating condition event, and the known operating condition fast control path is entered. The operating condition event library pre-stores multiple known operating condition events and their corresponding preset suspension control curves. The preset suspension control curve can be a suspension control curve pre-calibrated or learned based on the target operating condition event, and may include at least one of the following: baseline dynamic curve, target damping curve, target stiffness curve, target impedance curve, target vehicle attitude curve, or target suspension travel curve. The corresponding preset suspension control curve can be retrieved from the operating condition event library based on the event type of the target operating condition event, road excitation intensity, current vehicle speed, wheel trigger position, estimated arrival time, or left / right wheel trigger sequence.

[0165] For example, when the target operating condition event is a speed bump event, the corresponding preset suspension control curve can be retrieved from the operating condition event library based on the vehicle's current speed, speed bump height, or road surface excitation frequency; when the target operating condition event is a bridge joint event, the corresponding preset suspension control curve can be retrieved. By directly retrieving preset suspension control curves, the control delay caused by complex online inference or real-time optimization calculations can be reduced, improving the suspension response speed under known operating conditions.

[0166] Step S502: Determine the personalized correction amount based on the individual comfort preference parameters, and superimpose the personalized correction amount onto the baseline corresponding to the preset suspension control curve to obtain the preset suspension control parameters.

[0167] In some embodiments, a personalized correction amount is determined based on the individual comfort preference parameters corresponding to the current driver. The personalized correction amount may include at least one of the following: dynamic force correction amount, damping correction amount, stiffness correction amount, impedance correction amount, or vehicle posture target correction amount. Specifically, the suspension response preference of the current driver for a target operating condition event can be determined based on the individual comfort preference parameters. When the individual comfort preference parameters indicate that the driver prefers a softer, more comfortable ride, the impact support strength in the baseline dynamic force can be reduced, or the correction amount for vertical vibration filtering can be increased; when the individual comfort preference parameters indicate that the driver prefers a more dynamic support, the correction amounts corresponding to roll suppression, pitch suppression, or vehicle posture support can be increased. The personalized correction amount is superimposed on the baseline dynamic force corresponding to the preset suspension control curve to obtain preset suspension control parameters adapted to the current driver's individual comfort preference.

[0168] In one specific implementation, the baseline corresponding to the preset suspension control curve can be represented as the dynamic force. The personalized correction amount determined based on individual comfort preference parameters is expressed as: Then the target action force in the preset suspension control parameters can be expressed as:

[0169]

[0170] in, This indicates the target driving force in the preset suspension control parameters. Indicates the baseline as the driving force corresponding to the target operating condition event. This indicates the amount of personalized correction.

[0171] To prevent the personalized correction amount from causing the suspension to act too much or too little, the personalized correction amount can also be subject to amplitude constraints, rate of change constraints, or smoothing.

[0172] Step S503: Correct the preset suspension control parameters according to the target control strategy to obtain the target suspension control parameters.

[0173] In some embodiments, after obtaining the preset suspension control parameters, the preset suspension control parameters are further modified according to the target control strategy determined in step S3 to obtain the target suspension control parameters. Specifically, when the target control strategy represents a higher dynamic control weight corresponding to the manual driving control requirements, the vehicle body posture support, roll suppression, or pitch suppression in the preset suspension control parameters can be increased, so that the vehicle has stronger posture support during steering, lane changing, braking, or obstacle avoidance. When the target control strategy represents a higher dynamic control weight corresponding to the intelligent driving control requirements, the preset suspension control parameters can be modified to better meet the requirements of trajectory stability, vehicle body smoothness, and lateral and longitudinal motion coordination. When the target control strategy represents a higher dynamic control weight corresponding to the ride comfort requirements, the transmission of road impacts to the vehicle body can be reduced, and the vertical acceleration or pitch change of the vehicle body can be reduced.

[0174] In some embodiments, the modified suspension control parameters can be further subjected to boundary constraints and smoothing based on the actuation capability of the active suspension actuator, vehicle safety constraints, and changes in control parameters between adjacent control cycles to obtain the final target suspension control parameters. The target suspension control parameters may include at least one of the following: target actuation force of each wheel, target impedance parameter, target vehicle body angular velocity, target vehicle body vertical acceleration, and target vehicle body pitch angle.

[0175] In this way, under known operating conditions, it is possible to achieve rapid response using the preset suspension control curve, and also to adapt and modify it by combining the driver's individual comfort preferences and the current target control strategy, thus avoiding the simple mechanical application of the preset suspension control curve.

[0176] In the solution provided in this embodiment, through steps S501 to S503, when a target operating condition event is matched to the road disturbance pattern, the preset suspension control curve corresponding to the target operating condition event is directly invoked. This reduces the control delay caused by complex online calculations or real-time inference, and improves the response speed of the active suspension to known operating conditions such as speed bumps, bridge joints, and potholes. Simultaneously, by determining personalized correction amounts based on individual comfort preference parameters and superimposing them onto the baseline corresponding to the preset suspension control curve as power, it avoids different drivers using the exact same suspension control parameters under the same operating conditions, making the preset suspension control parameters more aligned with the driver's individual comfort needs. Therefore, it can achieve rapid response to known operating conditions while simultaneously considering the driver's individual comfort needs and the current vehicle control requirements.

[0177] Optionally, in some embodiments, step S7, the step of generating active suspension actuation commands based on target suspension control parameters, vehicle driving state information, and road excitation information, includes:

[0178] Step S701: Determine the target suspension state and target impedance parameters based on the target suspension control parameters.

[0179] In some embodiments, the target suspension control parameters can be analyzed to determine the target suspension state and target impedance parameters that the active suspension needs to achieve within the current control cycle. The target suspension state may include the target vehicle body angular velocity. Vertical acceleration of the target vehicle body and target vehicle pitch angle The values ​​of the aforementioned suspension target states can be determined based on the dynamic control weights included in the target control strategy. For example, when the target control strategy emphasizes manual driving needs, the control strength corresponding to the vehicle posture support target can be increased; when the target control strategy emphasizes ride comfort needs, the target vehicle vertical acceleration or the amplitude of vehicle posture changes can be reduced. The target impedance parameter is used to limit the response characteristics of the active suspension to the relative motion between the vehicle body and the wheels, enabling the active suspension to maintain compliance with road surface excitation while supporting the vehicle posture.

[0180] Step S702: Determine the current vehicle body state and the current wheel end state based on the vehicle driving state information, and determine the estimated road force based on the road excitation information.

[0181] In some embodiments, vehicle body inertial measurement information, vehicle body attitude information, suspension travel information, wheel speed information, and wheel-end acceleration information can be extracted from vehicle driving state information, and the current vehicle body state and the current wheel-end state can be determined based on the above information. Specifically, the vehicle body inertial measurement information may include at least one of the following: vertical acceleration, lateral acceleration, longitudinal acceleration, yaw rate, pitch rate, and roll rate; the vehicle body attitude information may include at least one of the following: vehicle height, pitch angle, and roll angle; the suspension travel information may include at least one of the following: suspension compression, suspension rebound, and suspension relative speed for each wheel; the wheel speed information may include at least one of the following: wheel speed, wheel speed change rate, and left / right wheel speed difference; and the wheel-end acceleration information may include the wheel-end vertical acceleration for each wheel.

[0182] Specifically, the vehicle body inertial measurement information and vehicle body attitude information can be filtered, transformed, and state estimated to obtain the current vehicle body state. The current vehicle body state can include at least one of the following: current vehicle body vertical acceleration, current vehicle body vertical velocity, current vehicle body height, current pitch angle, current pitch rate, current roll angle, current roll rate, and current yaw rate. For example, the current vehicle body vertical acceleration can be determined based on the vertical acceleration collected by the vehicle body inertial measurement unit; the current pitch angle and current pitch rate can be determined based on pitch angle sensing information or inertial measurement information; the current roll angle and current roll rate can be determined based on roll angle sensing information or inertial measurement information; and the current vehicle body height can be determined by combining the vehicle body height sensor information.

[0183] The current wheel-end state can also be determined based on suspension travel information, wheel-end acceleration information, and wheel speed information. The current wheel-end state can include at least one of the following: wheel-end vertical acceleration, wheel-end vertical velocity, wheel-end displacement, suspension compression, suspension relative velocity, wheel speed, wheel speed change rate, and tire dynamic load. For example, suspension compression can be determined based on suspension travel sensor signals for each wheel, and suspension relative velocity can be determined based on changes in suspension compression at adjacent sampling times; wheel-end vertical acceleration can be determined based on wheel-end acceleration sensor signals; wheel speed and wheel speed change rate can be determined based on wheel speed sensor signals; when it is necessary to determine wheel-end displacement or wheel-end vertical velocity, state estimation can be performed by combining vehicle height, suspension travel, and wheel-end vertical acceleration.

[0184] Furthermore, the estimated road surface force can be determined based on road surface excitation information. Specifically, the change in road surface height in front of the vehicle can be obtained from forward-looking cameras, lidar, millimeter-wave radar, or map data, and combined with the vehicle's current speed, wheel position, and estimated arrival time to determine the pre-aiming road surface input for each wheel; alternatively, the intensity of the current road surface disturbance on the wheel ends can be inferred from the wheel-end vertical acceleration, suspension compression, suspension relative speed, and wheel speed fluctuation characteristics.

[0185] In some embodiments, road excitation information can be converted into estimated road forces corresponding to each wheel based on wheel-end dynamics models, suspension dynamics models, or road surface observers. For example, the vertical force exerted by the road surface on the wheel end can be estimated based on changes in road height, wheel-end displacement, wheel-end velocity, tire equivalent stiffness, and tire equivalent damping; the force transmitted from the current road impact to the vehicle body can also be estimated based on wheel-end vertical acceleration, suspension travel changes, and tire dynamic load changes. The estimated road forces can include estimated road forces corresponding to the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, to reflect the intensity and timing of road disturbances experienced by different wheels.

[0186] In one specific implementation, the estimated force on the road surface can be expressed as:

[0187]

[0188] in, Indicates the first Estimated road surface forces corresponding to each wheel This indicates the equivalent stiffness of the tire. This indicates the equivalent damping of the tire. Indicates the first The relative displacement between the road surface input and the wheel end displacement at each wheel. Indicates the first The relative velocity between the road surface input velocity and the vertical velocity at the wheel end at each wheel. The above formula is only an example; in practical applications, the estimated road surface force can also be determined using a vehicle dynamics observer, Kalman filter, neural network estimation model, or lookup table mapping model.

[0189] In some embodiments, to improve the reliability of the road surface force estimation, the force estimation results obtained from the pre-aiming road surface input and the force estimation results obtained from the wheel-end sensor can be fused. For example, when the pre-aiming sensor detects a speed bump ahead and the wheel-end acceleration subsequently shows a corresponding impact peak, the confidence level of the road surface force estimation can be increased; when the pre-aiming identification result is inconsistent with the wheel-end response, the weight of the corresponding estimation result can be reduced, and the actual wheel-end response can be used as the main estimation basis.

[0190] Step S703: Determine the displacement control component based on the deviation between the suspension target state and the current vehicle body state.

[0191] Specifically, the vehicle angular velocity deviation can be determined based on the deviation between the target vehicle angular velocity and the current vehicle angular velocity; the vehicle vertical acceleration deviation can be determined based on the deviation between the target vehicle vertical acceleration and the current vehicle vertical acceleration; and the vehicle pitch attitude deviation can be determined based on the deviation between the target vehicle pitch angle and the current vehicle pitch angle. Displacement control components can be determined based on one or more of the above vehicle state deviations.

[0192] Displacement control components can be used to converge the current vehicle body state towards the target suspension state. For example, when the current vehicle body pitch angle deviates from the target vehicle body pitch angle, differential adjustment can be made through the displacement control components corresponding to the front and rear wheels to suppress vehicle body pitch; when the current vehicle body vertical acceleration deviates from the target vehicle body vertical acceleration, coordinated adjustment can be made through the displacement control components corresponding to the four wheels to reduce vehicle body vertical vibration; when the current vehicle body angular velocity deviates from the target vehicle body angular velocity, the trend of vehicle body attitude change can be adjusted through the displacement control components of the corresponding wheels.

[0193] In some embodiments, proportional control, proportional-integral-derivative control, or state feedback control can be used to output displacement control components for each wheel based on the deviation between the target vehicle body angular velocity, target vehicle body vertical acceleration, target vehicle body pitch angle, and the corresponding current vehicle body state. These displacement control components can be represented as target travel corrections, target displacement corrections, or target support force components related to vehicle body attitude convergence for each wheel, and are used to adjust the vehicle body attitude and vertical motion state in subsequent force-position hybrid control.

[0194] Step S704: Determine the force control component based on the estimated force on the road surface and the current wheel end state.

[0195] In some embodiments, the force control component may include at least one of a road disturbance feedforward component, a wheel-end feedback component, and a tire contact retention component. When the estimated road force indicates that the corresponding wheel is experiencing an upward road impact, the road disturbance feedforward component can be determined based on the estimated road force to preemptively offset or weaken the transmission of the road impact to the vehicle body. When the current wheel-end state indicates a large wheel-end vertical acceleration, rapid changes in suspension compression, or a large relative suspension speed, the wheel-end feedback component can be determined based on the wheel-end motion response to suppress wheel-end vibration and abrupt changes in suspension travel. When the wheel speed change rate or tire dynamic load change indicates a tendency for wheel to bounce or large fluctuations in contact load, the tire contact retention component can also be determined to reduce the risk of wheel lift-off or excessive fluctuations in tire dynamic load.

[0196] The road disturbance feedforward component can be determined based on the amplitude, direction, and timing of the estimated road force; the wheel-end feedback component can be determined based on the wheel-end vertical acceleration, suspension relative velocity, or suspension compression changes; and the tire contact patch component can be determined based on the wheel speed change rate, tire dynamic load changes, or wheel bounce judgment results. These components can be weighted and fused to obtain the force control components for the corresponding wheels.

[0197] In one specific implementation, for the first For each wheel, the estimated force exerted by the road surface can be expressed as: The vertical velocity at the wheel end is expressed as The relative speed of the suspension is expressed as The change in tire dynamic load is expressed as Then the first The force control component corresponding to each wheel can be determined based on at least one of the following parameters: estimated road surface force. Vertical velocity at wheel end Suspension relative speed and tire dynamic load variation For example, the feedforward compensation force can be determined based on the estimated force on the road surface, the feedback damping force can be determined based on the wheel-end vertical velocity and the relative velocity of the suspension, and the ground contact correction force can be determined based on the change in tire dynamic load. The above force components are then fused together to obtain the force control component.

[0198] In some embodiments, to avoid excessively strong instantaneous actuation of the actuator due to an excessively large force control component, the force control component can be subjected to amplitude limiting, rate of change limiting, and smoothing processing. Amplitude limiting can limit the maximum and minimum values ​​of the force control component, rate of change limiting can limit the rate of change of the force control component within adjacent control cycles, and smoothing processing can reduce the impact of sensor noise or road surface estimation errors on the force control component.

[0199] Step S705: Based on the target impedance parameters, perform force-position hybrid control on the displacement control component and the force control component to obtain the target driving force for each wheel.

[0200] In some embodiments, the target impedance parameter is used to define the target response characteristics corresponding to the relative motion between the vehicle body and the wheels, and may include at least one of a target stiffness parameter and a target damping parameter. The target stiffness parameter can be used to determine the support strength corresponding to the relative displacement change of the suspension, and the target damping parameter can be used to determine the damping strength corresponding to the relative velocity change of the suspension.

[0201] The displacement control component is used to converge the current vehicle body state towards the suspension target state, while the force control component is used to suppress the transmission of estimated road forces to the vehicle body and improve wheel-end compliance. The conversion relationship between the displacement control component and the displacement support force component can be determined based on the target impedance parameter, and the mixing coefficient corresponding to the force control component can be determined based on the target impedance parameter. When the target impedance parameter represents a higher target stiffness or a higher target damping, the displacement support force component formed after the displacement control component is converted from the target stiffness and target damping parameters can be increased, making the active suspension more focused on vehicle attitude support, pitch suppression, or vertical state convergence. When the target impedance parameter represents a lower target stiffness or a lower target damping, the displacement support force component can be decreased, and the mixing coefficient corresponding to the force control component can be increased, making the active suspension more focused on road impact filtering and wheel-end compliance response.

[0202] In one specific implementation, the first... The target corresponding to each wheel is represented by the following dynamic force:

[0203]

[0204] in, Indicates the first Each wheel corresponds to a target that provides power. Indicates the first The target stiffness parameters for each wheel Indicates the first The target damping parameters for each wheel Indicates the first Displacement control components corresponding to each wheel This represents the velocity control quantity obtained from the change of the displacement control component. Indicates the first Force control components corresponding to each wheel This represents the mixing coefficient determined based on the target impedance parameter.

[0205] The above formula is only an example. In practical applications, the target power of each wheel can also be determined by lookup table mapping, state feedback control, impedance controller or model predictive controller.

[0206] In some embodiments, the target action force corresponding to each wheel can be coordinated and constrained based on vehicle driving status information, target control strategy, and the actuation capability of the active suspension actuator. Coordinated allocation may include front-to-rear wheel action force distribution, left-to-right wheel action force distribution, and coupling coordination between vertical, pitch, and roll controls; boundary constraints may include maximum action force limits, minimum action force limits, action force change rate limits, suspension travel limits, and actuator power limits. Through the above processing, the target action force of each wheel can be obtained that satisfies both vehicle attitude control requirements and road impact filtering and actuator safety constraints.

[0207] Step S706: Generate active suspension actuation commands based on the target power of each wheel.

[0208] In some embodiments, active suspension actuation commands corresponding to the left front wheel, right front wheel, left rear wheel, and right rear wheel can be generated based on the target actuation force obtained in step S705 for each wheel. The active suspension actuation commands are used to drive the active suspension actuators at the corresponding wheels to generate the target actuation force, thereby adjusting the force and / or relative motion state between the vehicle body and the wheels. The active suspension actuation commands may include at least one of the following: wheel identification, target actuation force magnitude, actuation force direction, execution start time, duration, actuation force change rate, motor drive current, motor drive voltage, hydraulic valve opening, target pressure, or actuator target displacement.

[0209] In some embodiments, the target driving force corresponding to each wheel can be converted into the corresponding underlying driving quantity according to the type of active suspension actuator. When the active suspension actuator is a linear motor, the target driving force can be converted into motor drive current or motor drive voltage according to the thrust constant of the linear motor; when the active suspension actuator is an electro-hydraulic actuator, the target driving force can be converted into hydraulic valve opening, target oil pressure, or hydraulic cylinder driving force command; when the active suspension actuator is an air suspension actuator, the target driving force can be converted into air pressure regulation quantity or air valve control quantity.

[0210] The generated active suspension actuation commands can be transmitted via the vehicle communication network to the corresponding active suspension actuators or actuator controllers for each wheel. Upon receiving the active suspension actuation commands, the active suspension actuators or actuator controllers generate the corresponding actuation forces according to the target drive quantity and execution timing specified in the commands. The vehicle control system can also receive feedback from the actuators regarding the actual actuation forces, actual current, actual displacement, actual speed, or fault status, and use the feedback results to correct the actuation forces in subsequent control cycles.

[0211] In the solution provided in this embodiment, steps S701 to S706 determine the target suspension state and target impedance parameters based on the target suspension control parameters. These parameters are then combined with the current vehicle body state, current wheel end state, and estimated road surface force to generate active suspension actuation commands. This allows the active suspension control to simultaneously consider vehicle posture requirements, wheel end motion states, and road impact effects. Through force-position hybrid control of displacement and force control components, coordination between vehicle posture support and road impact filtering can be achieved, reducing issues such as insufficient posture support or stiff suspension action caused by a single control method. Therefore, the comprehensive adaptability of the active suspension actuation commands to driver state, vehicle state, and road disturbances can be improved.

[0212] Optionally, in some embodiments, after the step of generating active suspension actuation commands based on the target actuation force of each wheel, the active suspension actuation commands can be further subjected to heartbeat synchronization modulation. (Refer to...) Figure 2 As shown, the heartbeat synchronization modulation process specifically includes:

[0213] Step A1: Obtain the driver's heart rhythm information and determine the diastolic window based on the heart rhythm information.

[0214] In some embodiments, the vehicle control system can acquire the driver's heart rhythm information. This heart rhythm information can originate from an electrocardiogram (ECG) sensor, photoplethysmography (PPG) sensor, or other heart rate detection device in the steering wheel, seat, seatbelt, or a driver wearable device. The heart rhythm information can be filtered, peak detected, and timestamped to determine a baseline heartbeat time.

[0215] In one specific implementation, when the heart rhythm information is an electrocardiogram (ECG) signal, the reference time of the heartbeat can be determined by detecting the peak value of the R wave in the ECG signal. When the heart rhythm information is a pulse wave signal, the equivalent heart rate reference time can be calculated based on the pulse wave peak value or the characteristic point of the pulse wave rising edge. The vehicle control system can then use the heart rate reference time... And a preset diastolic offset, to determine the cardiac diastolic window. For example, it can be... The time window can be determined as the diastolic window corresponding to the current heartbeat cycle; it can also be dynamically adjusted based on the driver's real-time heart rate, heartbeat cycle length, or historical heart rhythm.

[0216] In some embodiments, when the diastolic window corresponding to the current cardiac cycle has been missed, the diastolic window corresponding to the next cardiac cycle can be determined based on the reference time of the next cardiac cycle.

[0217] Step A2: Determine whether the active suspension actuation command is a vehicle body attitude adjustment command that meets the preset attitude adjustment range conditions and allows for delayed execution.

[0218] In some embodiments, the determination of whether an active suspension actuation command belongs to a vehicle posture adjustment command that satisfies a preset posture adjustment range condition and allows for delayed execution can be made based on the control object corresponding to the active suspension actuation command, the change in actuation force, the posture adjustment target, the urgency of execution, and the maximum allowable delay time. The vehicle posture adjustment command may include an active suspension actuation command for adjusting the vehicle pitch angle, vehicle roll angle, vehicle height, or vehicle angular velocity. The preset posture adjustment range condition may include at least one of the following: the vehicle pitch angle adjustment amount is greater than a preset pitch threshold, the vehicle roll angle adjustment amount is greater than a preset roll threshold, the vehicle height adjustment amount is greater than a preset height threshold, the target change in actuation force is greater than a preset actuation force threshold, or the rate of change in actuation force is greater than a preset rate of change threshold. For example, when the roll adjustment amount corresponding to the active suspension actuation command is greater than a preset angle threshold, the active suspension actuation command can be considered to satisfy the preset posture adjustment range condition.

[0219] In some embodiments, whether delayed execution is permitted can be determined based on vehicle safety status, road disturbance type, command real-time performance level, and maximum permissible delay time. For example, active suspension actuation commands used for vehicle stability control, emergency obstacle avoidance, forced suppression of road impact peaks, or prevention of suspension bottoming out are generally not allowed to be delayed; active suspension actuation commands used for large but non-urgent vehicle pitch, roll, or height adjustments may be determined to be allowed to be delayed. For small-amplitude, high-frequency filtering commands, because they need to continuously and rapidly suppress road micro-vibrations, heartbeat window alignment is usually not performed, and they are executed according to the original execution time.

[0220] Step A3: If the active suspension actuation command is a vehicle posture adjustment command that meets the preset posture adjustment range conditions and allows for delayed execution, then the execution time of the active suspension actuation command is adjusted to within the diastolic window, and the active suspension actuator is controlled to execute the active suspension actuation command within the diastolic window.

[0221] In some embodiments, when the vehicle control system determines that the active suspension actuation command is a vehicle posture adjustment command that meets the preset posture adjustment range conditions and allows for delayed execution, the execution time of the active suspension actuation command can be adjusted according to the diastolic window determined in step A1. Specifically, if the diastolic window corresponding to the current cardiac cycle has not yet ended, and the execution time of the active suspension actuation command can be adjusted to within this window, then the active suspension actuation command is scheduled to be executed within the current diastolic window; if the current diastolic window has ended, or the remaining time in the current window is insufficient to complete the execution, then the active suspension actuation command is scheduled to be executed within the diastolic window corresponding to the next cardiac cycle.

[0222] In some embodiments, when adjusting the execution time, it can be determined whether the adjusted delay time exceeds the maximum allowable delay time corresponding to the active suspension actuation command. If it does not exceed the maximum allowable delay time, the actuation command is sent to the active suspension actuator according to the adjusted execution time; if it exceeds the maximum allowable delay time, the window alignment process can be canceled, and execution can proceed according to the original execution time or the most recent executable time. This avoids affecting the real-time performance and safety of vehicle attitude control by aligning with the diastolic window.

[0223] During execution, the active suspension actuator can generate the corresponding target force according to the adjusted execution time to complete the vehicle posture adjustment. Since the diastolic phase of the heart usually corresponds to the stage when the human body's perception of external impact or posture changes is relatively weak, scheduling the large-scale and delayed active suspension action commands to be executed within the diastolic window helps to reduce the abruptness to the driver caused by large-scale suspension movements.

[0224] Step A4: If the active suspension actuation command does not meet the preset attitude adjustment range condition, or if the active suspension actuation command is not allowed to be delayed, then the active suspension actuation command shall be executed at the original execution time.

[0225] In some embodiments, when the active suspension actuation command does not meet the preset attitude adjustment range condition, it indicates that the attitude adjustment range corresponding to the command is small, or it belongs to the high-frequency small-amplitude filtering control command. For such commands, the vehicle control system may not perform diastolic window alignment on their execution time, but instead send them to the active suspension actuator for execution according to the original execution time.

[0226] When an active suspension actuation command meets the preset attitude adjustment range conditions, but is a command that cannot be delayed, the vehicle control system will still execute the active suspension actuation command at the original execution time. Situations where delayed execution is not permitted include situations such as emergency braking, emergency obstacle avoidance, stability control intervention, suspension travel near its limit, abnormal tire contact with the ground, or road impact requiring immediate suppression. In these situations, the vehicle control system prioritizes vehicle driving safety and the real-time performance of suspension control.

[0227] In the solution provided in this embodiment, steps A1 to A4 determine the diastolic window based on the driver's heart rhythm information. The execution time of active suspension actuation commands that meet preset attitude adjustment range conditions and allow for delayed execution are adjusted only. This allows large-scale but non-urgent vehicle attitude adjustments to be performed within a time window where the driver's perception is relatively insensitive, thereby reducing the abruptness and discomfort caused by suspension actions and improving the coordination between active suspension actions and the driver's physiological rhythm. Simultaneously, for actuation commands that do not meet the attitude adjustment range conditions or do not allow for delayed execution, they are still executed at the original execution time, avoiding any impact on the real-time performance of road impact suppression, vehicle stability control, or emergency attitude adjustments due to waiting for the diastolic window.

[0228] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this application.

[0229] Furthermore, some embodiments of this application also provide an electronic device. The electronic device can be various forms of digital computer, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, etc. The electronic device can also be various forms of mobile devices, such as cellular phones, smartphones, wearable devices, and other similar computing devices.

[0230] The electronic device includes: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform a biofeedback-based active suspension control method as provided in any one or more of the above embodiments. Figure 3An exemplary structural diagram of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components are interconnected via different buses 1105 and can be mounted on a common motherboard or otherwise as required. The processors can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, multiple processors and / or multiple buses 1105 can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple electronic devices can be connected, each providing some of the necessary operations. The components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0231] The electronic device may further include an input device 1103 and an output device 1104. The processor 1101, memory 1102, input device 1103, and output device 1104 may be connected via a bus 1105 or other means. Figure 3 Taking the connection between China and Israel via bus 1105 as an example.

[0232] Input device 1103 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the electronic device, such as a touch screen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 1104 may include a display device, auxiliary lighting device (e.g., LED), and haptic feedback device (e.g., vibration motor). The display device may include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some embodiments, the display device may be a touch screen.

[0233] To provide interaction with the user, the electronic device can be a computer. The computer has: a display device (e.g., a cathode ray tube or LCD monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).

[0234] In this embodiment, a computer-readable medium stores a computer program / instruction, which, when executed by a processor, implements a biofeedback-based active suspension control method provided in any one or more of the above embodiments. This computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently and not be assembled into that device. The computer-readable medium carries one or more computer-readable instructions.

[0235] The memory 1102 can serve as a non-transitory computer-readable storage medium, used to store non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 1102, thereby implementing the program instructions / modules corresponding to the methods provided in any one or more of the embodiments described above in this application.

[0236] The memory 1102 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 1102 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 1102 may optionally include memory remotely located relative to the processor 1101, and these remote memories can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0237] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0238] Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory or other memory technologies, read-only optical discs, digital versatile optical discs or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0239] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0240] In the above embodiments, all or part of the implementation can be achieved through software, hardware, firmware, or any combination thereof. For example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device. In some embodiments, the software program of this application can be executed by a processor to implement the above steps or functions. Similarly, the software program of this application (including related data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drives, floppy disks, and similar devices. In addition, some steps or functions of this application can be implemented in hardware, for example, as circuitry that cooperates with a processor to perform the various steps or functions.

[0241] The computer program product provided in this application includes one or more computer programs / instructions. When executed by a processor, these computer programs / instructions generate, in whole or in part, the processes or functions described in this application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0242] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0243] The scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to encompass all variations falling within the meaning and scope of equivalents of the claims. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. Terms such as "first," "second," etc., are used only to distinguish descriptions and do not indicate any particular order, nor should they be construed as indicating or implying relative importance.

[0244] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily made by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A biofeedback-based active suspension control method, characterized in that, include: Acquire the driver's biofeedback signals and behavioral state signals, vehicle driving state information, road surface stimulus information, and individual comfort preference parameters corresponding to the driver; The driver's physiological state parameters are determined based on the biofeedback signals, and the driving intention parameters are determined based on the behavioral state signals. The target control strategy for the active suspension is determined based on the driver's physiological state parameters, the driving intention parameters, the vehicle driving state information, and the individual comfort preference parameters. The road surface excitation information is matched with a preset working condition event database; If a target operating condition event that meets the preset matching conditions is matched, the target suspension control parameters of the active suspension are determined according to the preset suspension control parameters corresponding to the target operating condition event and the target control strategy. If no target operating condition event that meets the matching conditions is matched, the target suspension control parameters are determined online based on the target control strategy, the vehicle driving status information, and the road excitation information. The active suspension actuation command is generated based on the target suspension control parameters, the vehicle driving state information, and the road excitation information. The active suspension actuator is then controlled to execute the active suspension actuation command to adjust the force and / or relative motion state between the vehicle body and the wheels.

2. The active suspension control method based on biofeedback according to claim 1, characterized in that, The biofeedback signals include steering wheel grip force signals, heart rate signals, and heart rate variability signals; The step of determining the driver's physiological state parameters based on the biofeedback signal includes: The rate of change of steering wheel grip force is determined based on the changes in the steering wheel grip force signal within a preset time window; The current heart rate value within a preset time window is determined based on the heart rate signal, and the relative change of the current heart rate value relative to the driver's baseline heart rate is calculated to obtain the heart rate deviation rate. Based on the heart rate variability signal, a normalized heart rate variability parameter is determined, and the normalized heart rate variability parameter is transformed by a reciprocal to obtain the reciprocal of the heart rate variability. The driver's physiological state parameters are obtained by fusing the steering wheel grip force change rate, the heart rate deviation rate, and the reciprocal of the heart rate variability.

3. The active suspension control method based on biofeedback according to claim 1, characterized in that, The behavioral state signals include seat pressure distribution signals, eye movement signals, head posture signals, and steering wheel torque signals; The step of determining driving intention parameters based on the behavioral state signal includes: The driver's center of gravity offset parameters and torso tilt parameters are determined based on the seat pressure distribution signal. The driver's gaze point movement direction and / or gaze area change parameters are determined based on the eye movement signals, and a first pre-aiming intention feature is determined based on the gaze point movement direction and / or gaze area change parameters. The driver's head steering posture parameters are determined based on the head posture signal, and the second aiming intention feature is determined based on the head steering posture parameters; The steering wheel torque direction and / or steering wheel torque change rate are determined based on the steering wheel torque signal, and the control confirmation intention feature is determined based on the steering wheel torque direction and / or the steering wheel torque change rate. The first aiming intent feature, the second aiming intent feature, and the control confirmation intent feature are sequentially correlated and fused for consistency. The fusion result is then corrected based on the center of gravity offset parameter and the torso tilt parameter to obtain the driving intent parameter.

4. The active suspension control method based on biofeedback according to claim 1, characterized in that, The step of determining the target control strategy of the active suspension based on the driver's physiological state parameters, the driving intention parameters, the vehicle driving state information, and the individual comfort preference parameters includes: Construct a multi-party game decision-making model with the needs of manual driving operation, intelligent driving control, and passenger comfort as the players; Based on the driver's physiological state parameters, determine the dynamic weight adjustment conditions of the multi-party game decision-making model; The first benefit value corresponding to the manual driving control requirement is determined based on the driving intention parameters; the second benefit value corresponding to the intelligent driving control requirement is determined based on the vehicle driving status information; and the third benefit value corresponding to the ride comfort requirement is determined based on the individual comfort preference parameters. Based on the dynamic weight adjustment conditions, a multi-objective game equilibrium solution is performed on the first benefit value, the second benefit value, and the third benefit value to obtain the dynamic control weight of each player under the current working condition. The suspension target state is determined based on the dynamic control weights, and the game equilibrium solution containing the dynamic control weights and the suspension target state is determined as the target control strategy.

5. The active suspension control method based on biofeedback according to claim 1, characterized in that, The step of determining the target suspension control parameters of the active suspension based on the preset suspension control parameters corresponding to the target suspension event and the target control strategy if a target operating condition event that meets the preset matching conditions is matched includes: If a target operating condition event that meets the preset matching conditions is matched, the preset suspension control curve corresponding to the target operating condition event is invoked. The personalized correction amount is determined based on the individual comfort preference parameters, and the personalized correction amount is superimposed on the baseline corresponding to the preset suspension control curve to obtain the preset suspension control parameters. The preset suspension control parameters are modified according to the target control strategy to obtain the target suspension control parameters.

6. The active suspension control method based on biofeedback according to claim 1, characterized in that, The step of generating active suspension actuation commands based on the target suspension control parameters, the vehicle driving state information, and the road surface excitation information includes: Determine the target suspension state and target impedance parameters based on the target suspension control parameters; The current vehicle body state and current wheel end state are determined based on the vehicle driving state information, and the estimated road surface force is determined based on the road surface excitation information; The displacement control component is determined based on the deviation between the suspension target state and the current vehicle body state; The force control component is determined based on the estimated force on the road surface and the current wheel end state; Based on the target impedance parameters, the displacement control component and the force control component are subjected to force-position hybrid control to obtain the target driving force for each wheel; The active suspension actuation command is generated based on the target power of each wheel.

7. The active suspension control method based on biofeedback according to claim 6, characterized in that, After the step of generating the active suspension actuation command based on the target power of each wheel, the method further includes: Obtain the driver's heart rhythm information and determine the diastolic window based on the heart rhythm information; Determine whether the active suspension actuation command is a vehicle posture adjustment command that meets the preset posture adjustment range conditions and allows for delayed execution; If the active suspension actuation command is a vehicle posture adjustment command that satisfies the preset posture adjustment range condition and allows for delayed execution, then the execution time of the active suspension actuation command is adjusted to within the cardiac diastolic window, and the active suspension actuator is controlled to execute the active suspension actuation command within the cardiac diastolic window. If the active suspension actuation command does not meet the preset attitude adjustment range condition, or if the active suspension actuation command is not allowed to be delayed, then the active suspension actuation command will be executed at the original execution time.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; and a memory storing computer program instructions that, when executed, cause the processors to perform the biofeedback-based active suspension control method as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program and / or instructions stored thereon, characterized in that, When the computer program and / or instructions are executed by the processor, they implement the biofeedback-based active suspension control method as described in any one of claims 1-7.

10. A computer program product comprising a computer program and / or instructions, characterized in that, When the computer program and / or instructions are executed by the processor, they implement the biofeedback-based active suspension control method as described in any one of claims 1-7.

Images