A human-machine double preview coupling stable front-axle active suspension control method and system

Abstract

This invention discloses a human-machine dual-pre-aiming coupled and stable front axle active suspension control method and system, belonging to the field of vehicle active suspension control technology. This invention integrates driver control pre-aiming and machine road surface pre-aiming to construct a human-machine dual-pre-aiming coupled dynamic model, accurately characterizing the impact of coupled disturbances on the suspension, achieving quantitative determination of stability and hierarchical collaborative control. The system dynamically optimizes multi-objective control weights based on real-time operating conditions, prioritizing ride comfort in stable operating conditions and safety in emergency and dynamic operating conditions, achieving a synergistic improvement in comfort and stability. With zero hardware increments and lightweight algorithms, it can be directly integrated into the vehicle ECU, reusing onboard sensing and execution components, adapting to mass-produced solutions such as CDC and air suspension, and can be coordinated with ESP and ABS for overall vehicle ride quality, handling safety, and system robustness, possessing excellent engineering practicality and promotional value.

Description

Technical Field

[0001] This invention relates to the field of vehicle active suspension control technology, and in particular to a human-machine dual-pre-aiming coupled stable front axle active suspension control method and system. Background Technology

[0002] Front-mounted anti-collision active suspension relies on lidar and visual sensors to detect road elevation disturbances in advance. It can proactively adjust suspension damping and stiffness, effectively filtering road impacts and suppressing vehicle posture fluctuations. This is a core technology for improving the driving experience of intelligent vehicles. However, existing anti-collision suspension systems mostly only optimize road recognition and control algorithms for machine-based anti-collision, neglecting the coupling mechanism between driver-controlled anti-collision and machine-based road anti-collision. This results in the following significant technical defects in coupling stability: 1. Mismatch in pre-aiming timing can easily lead to control conflicts: There is a subjective lag and prediction deviation in the driver's steering, acceleration and deceleration control pre-aiming. The timing of human-machine pre-aiming is not synchronized and the predicted target conflicts can easily cause control command oscillations and abnormal fluctuations in vehicle posture. 2. Coupled instability exacerbates the risk of control failure: Under complex conditions such as aggressive driving and continuous bumps, the human-machine anticipation deviation is rapidly amplified, the system damping and stiffness adjustment changes frequently and abruptly, exacerbates the fluctuation of tire dynamic load, reduces handling stability, and may even cause safety risks such as suspension over-range and vehicle instability. 3. Lack of targeted coupling stability control mechanism: Existing technologies have not constructed a human-machine dual-aiming coupling dynamic model, and also lack a quantitative stability determination mechanism. The ability to resolve conflicts and coordinate control is insufficient, resulting in poor robustness and feasibility in engineering applications. Summary of the Invention

[0003] The purpose of this invention is to provide a human-machine dual-pre-aiming coupled and stable axle active suspension control method and system, to construct a human-machine dual-pre-aiming coupled dynamic model, to design a coupling stability quantitative judgment and hierarchical collaborative control mechanism, and to achieve deep collaboration between pre-aiming compensation and driving control, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A human-machine dual-pre-aiming coupled and stable axle front active suspension control method includes the following steps: Simultaneously collect multi-source vehicle operation and anti-aiming data, which includes driver control anti-aiming information, axle-front road machine anti-aiming information, and real-time vehicle operation status information; Based on the acquired multi-source vehicle operation and anti-targeting data, multi-source information is spatiotemporally aligned and standardized preprocessed to generate standardized multi-source control input data. Based on standardized multi-source control input data, driver control prediction features and axle-front road surface anti-aiming disturbance features are extracted to construct a human-machine dual anti-aiming coupled dynamic model. Define and calculate the human-machine anti-aiming coupling stability coefficient, and classify the coupling stability of the suspension system based on the numerical range of the coupling stability coefficient. Based on the results of the coupling stability classification, the multi-objective control weights of ride comfort, vehicle handling stability, and system coupling stability are dynamically optimized. A hierarchical collaborative control architecture is adopted, combined with optimized multi-objective control weights, to generate suspension damping and stiffness adjustment commands; Real-time acquisition of vehicle attitude and suspension operation feedback data, including vehicle vertical acceleration, suspension dynamic travel, tire dynamic load, and suspension actuator response status; The parameters and multi-objective control weights of the human-machine dual-aiming coupled dynamics model are iteratively updated based on feedback data.

[0005] Furthermore, the process of constructing the human-machine dual-aiming coupled dynamic model includes: Based on standardized multi-source control input data, the incremental driver control anti-aiming input and the incremental axle-front road machine anti-aiming control are calculated separately. Based on the vehicle's inherent parameters and suspension system parameters, the vehicle mass matrix, suspension damping matrix, and stiffness matrix are obtained. By combining the calculated incremental input of driver control anti-aiming, incremental control of machine anti-aiming on the road surface in front of the axle, vehicle mass matrix, suspension damping matrix and stiffness matrix, a dynamic correlation equation is established between the human-machine anti-aiming coupled disturbance force and vehicle displacement, suspension dynamic travel and tire dynamic load. The influence mechanism of human-machine anti-aiming timing deviation and control-road coupled disturbance on the stability of the suspension system is characterized by dynamic correlation equations, forming a human-machine dual anti-aiming coupled dynamic model.

[0006] Furthermore, the definition and calculation of the human-machine pre-aiming coupling stability coefficient includes: Based on standardized multi-source control input data and coupled dynamic model output, key coupling feature parameters are extracted and determined. These key coupling feature parameters include driver anti-aiming delay, machine anti-aiming delay, driver steering prediction deviation, and road surface elevation anti-aiming deviation. The driver's anticipation delay is compared with the machine's anticipation delay to obtain the difference between the human and machine anticipation delays. At the same time, the driver's steering prediction deviation and the road elevation prediction deviation are vector-synthesized to obtain the comprehensive prediction deviation magnitude. Based on the difference in human-machine pre-aiming delay and the magnitude of the comprehensive prediction deviation, a formula for calculating the human-machine pre-aiming coupling stability coefficient is constructed, and the human-machine pre-aiming coupling stability coefficient is calculated in real time. The formula for calculating the human-machine pre-aiming coupling stability coefficient is as follows: In the formula, The human-machine pre-aiming coupling stability coefficient; To help drivers anticipate the deviation between the steering angle and the actual steering angle; To determine the deviation between the machine's pre-aimed road surface elevation and the actual road surface elevation; Pre-aiming delay for the driver; Pre-aiming delay for the machine; Pre-aiming delay for the machine; The calculated coupling stability coefficient is compared with the preset grading threshold range. Based on the comparison results, the coupling stability of the suspension system is divided into stable state, mild instability state and severe instability state, and a coupling stability grading judgment result is generated. The calculated human-machine pre-aiming coupling stability coefficient is compared with the coupling stability grading threshold range one by one to determine the numerical range to which the coupling stability coefficient belongs. When the coupling stability coefficient is less than or equal to the first threshold, the suspension system is determined to be in a stable state, a stability determination result is generated, and a timing synchronization fine-tuning strategy is executed. When the coupling stability coefficient is greater than the first threshold and less than or equal to the second threshold, the suspension system is determined to be in a slightly unstable state, a slightly unstable determination result is generated, and a weight smoothing adjustment strategy is executed. When the coupling stability coefficient is greater than the second threshold, the suspension system is determined to be in a severely unstable state, a severely unstable determination result is generated, and a full conflict resolution strategy is executed.

[0007] Furthermore, the multi-objective control weights for dynamically optimizing ride comfort, vehicle handling stability, and system coupling stability include: Based on the results of the coupling stability classification, the human-machine pre-aiming coupling stability coefficient is obtained, and the initial configuration and dynamic adjustment rules of the multi-target control weights are determined. Based on dynamic adjustment rules, the weights of ride comfort, vehicle handling stability, and system coupling stability are calculated in real time, and the multi-objective control weights are optimized based on the calculation results. The optimized multi-objective control weights are input into the hierarchical collaborative control architecture for the coordinated regulation of suspension damping and stiffness. The optimized multi-objective control weights resolve human-machine anticipation timing deviations and coupling conflicts.

[0008] Furthermore, the hierarchical collaborative control architecture includes an upper coupling coordination layer and a lower control execution layer: The upper coupling coordination layer is used to obtain the coupling stability classification judgment result, dynamically match the corresponding weight allocation strategy according to the coupling stability classification judgment result, synchronize and coordinate the human-machine pre-aiming timing, and eliminate the timing deviation between human-machine operation and machine control. The lower control execution layer is used to receive the optimized multi-objective control weights, combine the standardized multi-source input data and the output results of the dynamic model, construct the control cost function using the model predictive control algorithm, and generate the corresponding suspension control commands based on the control cost function. The generated optimal control commands are sent to the suspension actuator in real time. The optimal control commands include suspension damping and stiffness control commands. The suspension actuator performs precise control of the front active suspension based on the suspension damping and stiffness control commands. The multi-objective control weights also include: The evolution rules of multi-objective control weights are obtained based on the coupling stability coefficient. The weights are then smoothly adjusted according to the evolution rules, and a lower limit protection threshold for coupling stability weights is set based on system safety requirements. The adjustment results are generated based on the trend of weight changes determined by the coupling stability coefficient. Among them, the weight of driving comfort decreases monotonically as the coupling stability coefficient increases, while the weights of handling stability and coupling stability increase monotonically as the coupling stability coefficient increases.

[0009] Furthermore, the parameters of the human-machine dual-aiming coupled dynamics model are iteratively updated based on feedback data, including: The current moment is taken as the endpoint, and the suspension dynamic travel time sequence data and tire dynamic load time sequence data within a preset time window are extracted based on the endpoint. Within a preset time window, determine the first moment corresponding to the maximum value of the suspension dynamic travel and the second moment corresponding to the maximum value of the tire dynamic load, respectively, and calculate the difference between the first moment and the second moment, and use the difference as the suspension-tire response phase difference. The driver control anti-aiming input increment and the axle-front road machine anti-aiming control increment are obtained at the current moment, and the suspension-tire response phase difference is compared with the preset first phase difference threshold and the preset second phase difference threshold, respectively, wherein the preset first phase difference threshold is less than the preset second phase difference threshold; When the suspension-tire response phase difference is less than the preset first phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be in-phase superposition type, and the current vehicle speed value is obtained based on the determination result. The corresponding first correction gain is obtained by querying the pre-stored in-phase gain mapping table based on the current vehicle speed value and the suspension-tire response phase difference. The absolute value of the summation of the driver's pre-aiming input increment and the pre-aiming control increment of the road surface machine in front of the axle is obtained, and the absolute value is divided by the preset reference force to obtain the in-phase superposition coefficient. At the same time, the product of the first correction gain and the in-phase superposition coefficient is determined, and the product is added to the value 1. The sum of the product and the value 1 is divided into intervals based on the preset lower limit and the preset upper limit to obtain the first stiffness correction coefficient. When the suspension-tire response phase difference is greater than the preset second phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be the anti-phase cancellation type, and the current vehicle speed value is obtained based on the determination result. The corresponding second correction gain is obtained by querying the pre-stored anti-phase gain mapping table based on the current vehicle speed value and the suspension-tire response phase difference. The absolute value is obtained by subtracting the driver's control pre-aiming input increment from the pre-aiming control increment of the road surface machine in front of the axle, and then dividing the absolute value by the preset reference force to obtain the anti-phase cancellation coefficient. At the same time, the product of the second correction gain and the anti-phase cancellation coefficient is determined, and the product is added to the value 1. The sum of the product and the value 1 is then divided into intervals based on the preset lower limit and the preset upper limit to obtain the second stiffness correction coefficient. When the suspension-tire response phase difference is greater than or equal to the preset first phase difference threshold and less than or equal to the preset second phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be the mid-position weak coupling type, and the original stiffness matrix remains unchanged. Obtain the original stiffness matrix in the human-machine dual-pre-aiming coupled dynamic model, and determine the diagonal elements of the original stiffness matrix, where the diagonal elements correspond to the vertical stiffness, lateral stiffness and longitudinal stiffness of the suspension, respectively. Based on the relationship between the suspension-tire response phase difference and the preset first phase threshold and second phase threshold, each diagonal element of the original stiffness matrix is ​​multiplied by the first stiffness correction coefficient or the second stiffness correction coefficient to obtain the corrected stiffness matrix. The original stiffness matrix in the human-machine dual-aiming coupled dynamic model is replaced by the modified stiffness matrix, thereby completing the iterative update of the stiffness parameters in the human-machine dual-aiming coupled dynamic model.

[0010] Furthermore, after calculating the human-machine pre-aiming coupling stability coefficient in real time, the process includes correcting and determining the coupling stability coefficient. When correction is needed, the coupling stability coefficient is corrected, and a coupling stability classification is performed based on the corrected coupling stability coefficient. The specific process is as follows: Obtain the coupling stability coefficient of the current control cycle and the coupling stability coefficient of the previous control cycle; The coupling stability coefficient of the current control cycle is compared with that of the previous control cycle to determine whether the coupling stability coefficient needs to be corrected. If the coupling stability coefficient of the current control cycle is less than or equal to the coupling stability coefficient of the previous control cycle, it is determined that no correction is needed for the coupling stability coefficient, and a coupling stability classification judgment is made based on the coupling stability coefficient of the current control cycle. Otherwise, calculate the change in the coupling stability coefficient between the current control cycle and the previous control cycle; Change in coupling stability coefficient With the preset stability threshold By comparing the coefficients, the smoothing suppression coefficient of the coupling stability coefficient for the current control cycle is determined; when If the coupling stability coefficient of the current control cycle is determined to be fluctuating normally within the allowable range, the smoothing suppression coefficient is set to 1. when If the coupling stability coefficient of the current control cycle is determined to be an abnormally rapid jump, then the smoothing suppression coefficient of the coupling stability coefficient of the current control cycle is calculated according to the following formula. ;in, The smoothing suppression coefficient represents the coupling stability coefficient of the current control cycle; The coupling stability coefficient of the current cycle is corrected based on the smoothing suppression coefficient of the coupling stability coefficient of the current control cycle; The coupling stability is graded and determined based on the modified coupling stability coefficient.

[0011] Furthermore, the real-time acquisition of vehicle attitude feedback data and suspension operation feedback data also includes: Real-time feedback data is spatiotemporally compared with standardized multi-source control input data, and closed-loop feedback correction is obtained based on the comparison results. The mass matrix, damping matrix, and stiffness matrix parameters of the human-machine dual-aiming coupled dynamic model are iteratively updated based on the closed-loop feedback correction. The initial baseline values ​​and evolution coefficients of the multi-objective control weights are updated synchronously and iteratively.

[0012] This invention provides another technical solution: a human-machine dual-pre-aiming coupled and stable front axle active suspension control system, comprising: The multi-source information perception module is used to acquire driver control preview information, axle-front road machine preview information and vehicle real-time operating status information, and to perform filtering and noise reduction, timestamp synchronization and alignment, and data normalization processing on the multi-source information, and to perform spatiotemporal alignment of the multi-source information to output standardized multi-source control input data. The coupled modeling module is used to extract driver control prediction characteristics and axle-front road pre-aiming disturbance based on standardized multi-source control input data, and to construct and iteratively update the human-machine dual pre-aiming coupled dynamic model. The stability determination module is used to calculate the computer-computer pre-aiming coupling stability coefficient based on the human-machine dual-aiming coupling dynamic model, and to perform a graded determination of coupling stability based on the coupling stability coefficient. The weight optimization module is used to dynamically optimize the control weights of multiple objectives based on the coupling stability classification results, thereby resolving human-machine pre-aiming timing deviations and coupling conflicts. The hierarchical control module is used to generate suspension damping and stiffness adjustment commands based on the optimized multi-objective control weights and a hierarchical collaborative control architecture. The suspension actuator module is used to receive and execute suspension damping and stiffness control commands, perform active suspension real-time adjustments, and provide feedback on vehicle attitude and suspension operating status data.

[0013] Furthermore, the multi-source information sensing module includes: The driver control perception unit is used to collect steering wheel angle, steering rate, throttle opening change rate, brake pressure change rate, and driver anti-aiming delay. The axle-front road surface pre-aiming sensing unit uses lidar or vision sensors to collect data on the axle-front road surface elevation, slope, bump level, and machine pre-aiming delay. The vehicle status acquisition unit is used to collect data on the vehicle's vertical acceleration, suspension travel, tire dynamic load, and vehicle speed.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention integrates driver control anti-aiming and axle-front road machine anti-aiming to construct a seven-DOF human-machine dual anti-aiming coupled dynamic model. It can accurately characterize the dynamic influence mechanism of anti-aiming timing deviation and control-road coupled disturbance on suspension attitude, tire dynamic load and system stability. It can realize quantitative judgment and hierarchical coordinated control of coupling stability, significantly improve the operational robustness and control reliability of the suspension system under complex working conditions, and completely eliminate safety hazards such as abnormal fluctuations in vehicle attitude and suspension over-range operation caused by coupling instability.

[0015] 2. This invention achieves dynamic adaptive optimization of multi-objective control weights based on the human-machine pre-aiming coupling stability coefficient. It can intelligently balance and smoothly switch between driving comfort, vehicle handling stability, and system coupling stability according to real-time operating conditions. Under stable driving conditions, it prioritizes driving comfort, while automatically enhancing handling stability and coupling safety under aggressive conditions such as emergency braking and continuous lane changes. At the same time, it avoids step changes in control commands through weight smooth evolution and lower limit protection mechanisms, effectively suppressing vehicle pitch, roll, and vertical impacts, reducing tire dynamic load fluctuations, and comprehensively improving vehicle driving quality, handling precision, and driving safety, achieving synergistic optimization of driving quality and handling performance in all scenarios.

[0016] 3. This invention adopts a hierarchical collaborative control architecture and lightweight algorithm design. At the hardware level, it fully reuses the vehicle's perception sensors, ECU, and suspension actuators, eliminating the need for additional hardware equipment. This significantly reduces mass production costs and modification difficulty. At the same time, it can seamlessly collaborate with vehicle chassis control systems such as ESP and ABS, and is compatible with various mainstream active suspension actuators such as electromagnetic dampers and air springs. It features rapid control response, simple calibration process, low requirements for vehicle computing power, and strong engineering feasibility and platform adaptability. It can be widely applied to various passenger vehicles equipped with axle front anti-suspension active suspension. Attached Figure Description

[0017] Figure 1 This is a block diagram of the overall architecture of the human-machine dual-pre-aiming coupling stabilizing axle front active suspension system of the present invention; Figure 2 This is a flowchart of the human-machine dual-aiming coupling stability determination and hierarchical control process of the present invention; Figure 3 This is a topological diagram of the human-machine dual-aiming coupling dynamics model of the present invention; Figure 4 This is a flowchart illustrating the execution of the hierarchical collaborative MPC control algorithm of the present invention. Figure 5 This is a diagram showing the control timing and attitude curves under smooth driving conditions in Embodiment 1 of the present invention. Figure 6 This is a schematic diagram of suspension adjustment and vehicle posture under emergency braking conditions in Embodiment 2 of the present invention; Figure 7 This is a diagram showing the damping distribution and roll control of the left and right wheels under continuous lane change conditions in Embodiment 3 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figures 1-7 The present invention provides the following technical solutions: A human-machine dual-pre-aiming coupled and stable axle front active suspension control method includes the following steps: Simultaneously collect multi-source vehicle operation and anti-aiming data, which includes driver control anti-aiming information, axle-front road machine anti-aiming information, and real-time vehicle operation status information; Based on the acquired multi-source vehicle operation and anti-targeting data, filtering and noise reduction, timestamp synchronization and alignment and data normalization are performed to complete the spatiotemporal alignment and standardization preprocessing of multi-source information and generate standardized multi-source control input data. Based on standardized multi-source control input data, driver control prediction features and axle-front road surface anticipation disturbance features are extracted to construct a human-machine dual anticipation coupled dynamic model, which is used to characterize the dynamic correlation between human-machine anticipation coupled disturbance force and suspension attitude and tire dynamic load. Based on the human-machine dual-pre-aiming coupling dynamic model, parameters such as driver pre-aiming delay, machine pre-aiming delay, driver steering prediction deviation, and road elevation pre-aiming deviation are obtained. Based on the obtained parameters, the human-machine pre-aiming coupling stability coefficient is defined and calculated. The coupling stability of the suspension system is graded and judged based on the numerical range of the coupling stability coefficient. Based on the results of the coupling stability classification, the multi-objective control weights of driving comfort, vehicle handling stability and system coupling stability are dynamically optimized to resolve human-machine anticipation timing deviation and coupling conflict. A hierarchical collaborative control architecture is adopted, combined with optimized multi-objective control weights, to generate suspension damping and stiffness adjustment commands, thereby realizing coupled stability control of the front axle active suspension; Real-time acquisition of vehicle attitude and suspension operation feedback data, including vehicle vertical acceleration, suspension dynamic travel, tire dynamic load, and suspension actuator response status; Real-time feedback data is spatiotemporally compared with standardized multi-source control input data, and closed-loop feedback correction is obtained based on the comparison results. The mass matrix, damping matrix, and stiffness matrix parameters of the human-machine dual-aiming coupled dynamic model are iteratively updated based on the closed-loop feedback correction. Simultaneously, the initial baseline values ​​and evolution coefficients of the multi-objective control weights are iteratively updated to maintain dynamic matching between the model parameters and control weights and the real-time operating conditions. Through continuous iterative correction and real-time output adjustment, the suspension system achieves closed-loop adaptive stability control under all operating conditions, including smooth driving, emergency braking, and continuous lane changes, forming a closed-loop adaptive control system under all operating conditions.

[0020] In this embodiment, the standardization preprocessing includes filtering and noise reduction, timestamp synchronization and alignment, and data normalization. In this embodiment, the driver control preview information includes the steering wheel angle. angular rate Throttle opening change rate Braking pressure change rate Driver anticipation delay wait; In this embodiment, the road vehicle pre-aiming information includes the road surface elevation 0.5-5m in front of the axle. Road surface slope Bump level Machine aiming delay wait; In this embodiment, the vehicle state data includes the vehicle body vertical acceleration. Suspension travel Tire dynamic load Speed wait; In this embodiment, the constructed human-machine dual-pre-aiming coupled dynamic model, through multi-source information synchronous acquisition and spatiotemporal alignment preprocessing, can accurately characterize the coupling disturbance mechanism of driver control and machine road pre-aiming, upgrading the traditional qualitative description to a quantitative dynamic characterization. This provides core model support for coupling stability determination and conflict resolution control. Based on the human-machine pre-aiming coupling stability coefficient, real-time quantitative determination of coupling stability can be achieved, enabling rapid identification of the degree of human-machine pre-aiming mismatch, hierarchical resolution of conflicts, avoiding abrupt changes and oscillations in control commands, and ensuring that the suspension system remains stable and controllable under all operating conditions.

[0021] In this embodiment, the process of constructing the human-machine dual-aiming coupled dynamic model includes: Based on standardized multi-source control input data, the incremental values ​​of driver control anti-aiming input and axle-front road machine anti-aiming control are calculated separately. The incremental values ​​of driver control anti-aiming input and axle-front road machine anti-aiming control are obtained from the driver control anti-aiming information and axle-front road machine anti-aiming information, respectively. Based on the vehicle's inherent parameters and suspension system parameters, the vehicle mass matrix, suspension damping matrix, and stiffness matrix are obtained. By combining the calculated incremental driver control anti-aiming input, incremental machine anti-aiming control from the road surface in front of the axle, vehicle mass matrix, suspension damping matrix, and stiffness matrix, a dynamic correlation equation is established between the human-machine anti-aiming coupled disturbance force and vehicle displacement, suspension dynamic travel, and tire dynamic load. The dynamic correlation equation is as follows: In the formula, Represented as a vehicle mass matrix, it includes the mass of the body, suspension, and wheels. It is a diagonal matrix calibrated for the actual vehicle and determines the system's inertial characteristics. Represented as the system damping matrix, Represented as the system stiffness matrix, it jointly characterizes the passive vibration reduction and load-bearing support characteristics of the suspension. These are represented in sequence as vehicle displacement vector, velocity vector, and acceleration vector, which intuitively reflect the amplitude of the vehicle's attitude fluctuations. It represents the incremental input of the driver's control preview, which is generated by calculating real-time steering and acceleration / deceleration control data, reflecting the driver's subjective control intention; Represented as a control input mapping matrix, it converts dimensionless control signals into force inputs to the suspension system, enabling precise signal-to-vehicle interaction between the driver and vehicle. This is represented as the incremental control of machine road surface pre-aiming, which is generated by solving the elevation and slope data of the road surface in front of the axle, reflecting the prediction of objective road surface disturbances; Represented as a pre-aiming control mapping matrix, it converts sensor pre-aiming information into suspension control input, achieving seamless connection between vehicle and engine signals; This is represented as the human-machine anticipation coupling disturbance force, which is formed by the superposition of timing deviation and control-road conflict, and is the core cause of system instability and control oscillation. The influence mechanism of human-machine anti-aiming timing deviation and control-road coupled disturbance on the stability of the suspension system is characterized by dynamic correlation equations, forming a human-machine dual anti-aiming coupled dynamic model.

[0022] In this embodiment, the intrinsic relationship between human-machine anticipation timing deviation, control-road coupling disturbance, and suspension system stability is fully characterized. The core causes of system instability and control oscillation are clearly revealed. The driver's subjective control intention and the machine's objective road surface prediction information are uniformly transformed into force and control inputs that the suspension system can recognize, eliminating signal conversion deviation. The model is built based on the inherent parameters of the real vehicle, which fits the actual operating state of the vehicle, improving the authenticity and applicability of the dynamic description. It can accurately reflect the dynamic changes of human-machine coupling disturbance under different driving scenarios, providing a basis for adaptive suspension control under complex working conditions.

[0023] In this embodiment, defining and calculating the human-machine pre-aiming coupling stability coefficient includes: Based on standardized multi-source control input data and coupled dynamic model output, key coupling feature parameters are extracted and determined. These key coupling feature parameters include driver anti-aiming delay, machine anti-aiming delay, driver steering prediction deviation, and road surface elevation anti-aiming deviation. The driver's anticipation delay is compared with the machine's anticipation delay to obtain the difference between the human and machine anticipation delays. At the same time, the driver's steering prediction deviation and the road elevation prediction deviation are vector-synthesized to obtain the comprehensive prediction deviation magnitude. Based on the difference in human-machine pre-aiming delay and the magnitude of the comprehensive prediction deviation, a formula for calculating the human-machine pre-aiming coupling stability coefficient is constructed, and the human-machine pre-aiming coupling stability coefficient is calculated in real time. The formula for calculating the human-machine pre-aiming coupling stability coefficient is as follows: In the formula, The human-machine pre-aiming coupling stability coefficient, with a value range of... The smaller the value, the higher the human-machine pre-aiming fit and the stronger the system coupling stability. The deviation between the driver's predicted steering angle and the actual steering angle, measured in degrees, reflects the accuracy of the driver's control prediction. The greater the deviation, the higher the risk of instability. The deviation between the machine's pre-aimed road surface elevation and the actual road surface elevation, expressed in cm, reflects the accuracy of the sensor's pre-aiming. The larger the deviation, the higher the risk of pre-aiming failure. Driver anticipation delay, measured in milliseconds, refers to the lag time between the driver's anticipated action and the actual control, and is determined in real time by driving habits; Machine aiming delay, in milliseconds (ms), refers to the lag time between the sensor acquiring road surface information and the algorithm outputting control commands; it is an inherent hardware parameter. The calculated coupling stability coefficient is compared with the preset grading threshold range. Based on the comparison results, the coupling stability of the suspension system is divided into stable state, mild instability state and severe instability state, and a coupling stability grading judgment result is generated.

[0024] In this embodiment, generating the coupling stability classification result includes: Preset coupling stability grading threshold range α、β The calculated human-machine pre-aiming coupling stability coefficient is compared with the coupling stability classification threshold interval one by one to determine the numerical interval to which the coupling stability coefficient belongs. When K c ≤ α When the state is considered stable, a timing synchronization fine-tuning strategy is executed. when α <K c ≤ β When the situation is deemed to be in a state of mild instability, a weighted smoothing adjustment strategy is implemented. When K c > β When the situation is deemed to be in a state of severe instability, a full conflict resolution strategy is implemented. in, α The value range is 0.15~0.25. β The value range is 0.30~0.40.

[0025] In this embodiment, by defining and calculating the human-machine anticipation coupling stability coefficient, the degree of human-machine anticipation matching and the system coupling stability state can be quantitatively characterized, enabling precise hierarchical judgment of the suspension system coupling stability. Based on different instability levels, corresponding control strategies are automatically matched to ensure that the control strategy is highly adapted to the actual working conditions. The stability coefficient is constructed based on key coupling characteristic parameters to reflect the impact of human-machine anticipation deviation and time delay differences on system stability. The hierarchical judgment mechanism avoids sudden changes and oscillations in control commands, ensuring that the suspension system is always in a safe and stable operating range, while improving the rationality and robustness of the control logic.

[0026] In this embodiment, the multi-objective control weights for dynamically optimizing ride comfort, vehicle handling stability, and system coupling stability include: Based on the results of the coupling stability classification, the human-machine pre-aiming coupling stability coefficient is obtained, and the initial configuration and dynamic adjustment rules of the multi-target control weights are determined. Based on dynamic adjustment rules, the weights of ride comfort, vehicle handling stability, and system coupling stability are calculated in real time, and the multi-objective control weights are optimized based on the calculation results. The optimized multi-objective control weights are input into the hierarchical collaborative control architecture for the coordinated regulation of suspension damping and stiffness. The optimized multi-objective control weights resolve human-machine anticipation timing deviations and coupling conflicts. The evolution rules of multi-objective control weights are obtained based on the coupling stability coefficient. The weights are smoothly adjusted according to the evolution rules to ensure that the multi-objective control weights meet the normalization constraints. The lower limit protection threshold of the coupling stability weights is set based on the system safety requirements to ensure that the weights evolve smoothly without step jumps. The adjustment results are generated based on the trend of weight changes determined by the coupling stability coefficient. Among them, the weight of ride comfort decreases monotonically as the coupling stability coefficient increases, while the weights of handling stability and coupling stability increase monotonically as the coupling stability coefficient increases. The optimization calculation expression for the multi-objective control weights is as follows: In the formula, The dynamic weight of ride comfort decreases monotonically as instability increases; The dynamic weight of vehicle handling stability increases monotonically with increasing instability. A hard lower limit of 0.2 is set for the dynamic weights of coupling stability to prevent the system from becoming completely unstable; w 10 , w 20 , w 30 The initial weights for the baseline conditions of smooth driving were obtained through actual vehicle calibration. k1, k2, and k3 are the human-machine pre-aiming coupling stability coefficients; k1, k2, and k3 are the evolution coefficients. gamma This is the lower limit protection threshold for the coupling stability weight.

[0027] In this embodiment, the hierarchical collaborative control architecture includes an upper coupling coordination layer and a lower control execution layer: The upper coupling coordination layer is used to obtain the coupling stability classification judgment result. Based on the coupling stability classification judgment result, the corresponding weight allocation strategy is dynamically matched, the human-machine pre-aiming timing is synchronously coordinated, the timing deviation between human-machine operation and machine control is eliminated, and the control command is accurately matched with the driver's operation intention and road conditions, providing a clear control direction for the lower layer control execution. The lower control execution layer is used to receive the optimized multi-objective control weights, combine the standardized multi-source input data and the output results of the dynamic model, construct the control cost function using the model predictive control algorithm, and generate the corresponding suspension control commands based on the control cost function. The generated optimal control commands are sent to the suspension actuator in real time. The optimal control commands include suspension damping and stiffness control commands. The suspension actuator performs precise control of the front active suspension based on the suspension damping and stiffness control commands, and completes the dynamic adjustment of damping and stiffness. The control cycle is no more than 10ms. The model predictive control cost function is as follows: In the formula, The optimal suspension control command can be obtained by finding the minimum value of the model predictive control cost function. The prediction step size is a positive integer, representing the number of steps the algorithm takes to predict the road ahead, balancing prediction accuracy and computational cost. For the first Predicting vehicle vertical acceleration and quantifying core indicators of ride comfort; For the first Predict the real-time value of suspension travel; It serves as a reference calibration value for the dynamic travel of the suspension and a preset threshold for the safety of the suspension structure to prevent damage caused by exceeding the range; For the first Predict the rate of change of vertical dynamic load on the tire and quantify the tire adhesion stability; To control the incremental penalty coefficient, it is a positive real number, which suppresses sudden changes in control commands and extends the lifespan of the actuator; The increment of the suspension control command represents the change in the adjustment amplitude between adjacent control cycles.

[0028] In this embodiment, the multi-objective control weights of driving comfort, vehicle handling stability, and system coupling stability are dynamically optimized based on the coupling stability determination results. This can accurately allocate the priority of control objectives in combination with real-time operating conditions, effectively resolve human-machine pre-aiming timing deviations and coupling conflicts, and the weights evolve smoothly according to preset rules and meet normalization constraints. At the same time, a lower limit protection threshold is set for the coupling stability weights to avoid control shocks caused by weight step jumps and ensure system operation safety. Furthermore, a hierarchical collaborative control architecture is adopted, with upper-level coupling and coordination and lower-level control execution. The upper layer can efficiently complete timing synchronization and conflict coordination to ensure that the control logic conforms to driving intentions and road conditions. The lower layer relies on model predictive control algorithms to solve for the optimal suspension control commands, which can quickly output accurate damping and stiffness control signals. The control response is rapid and can suppress command mutations and protect the suspension actuators, taking into account ride comfort, handling stability and system operation safety. Overall, it improves the control accuracy, response speed and all-condition adaptability of the front axle active suspension.

[0029] To verify the control effect and engineering feasibility of the proposed dual-preview coupling stable front axle active suspension control method under different real vehicle conditions, real vehicle tests were conducted under three typical conditions: smooth driving on a straight urban road, emergency braking, and continuous lane changing. Quantitative data was compared with traditional solutions to verify the improved coupling stability, ride comfort, and handling safety of the present invention. Specific embodiments are as follows: Example 1: Smooth driving on a straight urban road (speed 40km / h) 1. Implementation conditions Test vehicle: A compact family sedan equipped with a front-axle lidar-guided CDC active suspension system; Test conditions: straight asphalt road in the city, without potholes or bumps, constant speed of 40km / h, and smooth driving by the driver; Calibration parameters: Initial weights for steady-state operation , , Pre-aiming delay , MPC prediction step size Control the incremental penalty coefficient .

[0030] 2. Implementation Steps Actual data collected: Steering wheel deviation Machine pre-aiming at road surface elevation deviation Vehicle body vertical acceleration reference value Real-time value of suspension travel ; Substitute the actual collected data into the coupling stability coefficient formula for calculation: Judgment result: The system is in a stable operating condition, with a high degree of consistency between human-machine aiming timing and no coupling conflicts.

[0031] Coupling stability coefficient Substitute into the weight evolution formula to solve: (Dynamic weighting of ride comfort); (Dynamic weighting of vehicle handling stability); (Dynamic weights for coupling stability); Constraint verification: ,and It satisfies the requirements of weight normalization and hard constraints, and the weight transitions smoothly without step jumps.

[0032] The MPC control cycle is set to 10ms. Dynamic weights and preview information are substituted into the cost function, and the optimal control command is solved in a rolling manner. With the CDC electromagnetic damper damping adjusted to 0.6 kN·s / m, the suspension adjustment resulted in no oscillations or impacts.

[0033] Real vehicle test results: The vertical acceleration of the vehicle body is stable at 0.22m / s², which is 30% lower than the traditional fixed weight scheme, and the human-machine pre-aiming is synchronized and conflict-free throughout the process.

[0034] 3. Implementation Results Coupling stability: With a constant value of ≤0.2, no control conflicts, and no command oscillations, the probability of instability is reduced by 95% compared to the traditional single-aiming scheme; Ride quality: The vehicle's vertical acceleration is stable at 0.22m / s², improving comfort by 30% compared to the traditional fixed-weight scheme, with no bumps or jerks throughout the journey; Handling safety: Tire dynamic load fluctuation is reduced by 25%, and suspension dynamic travel is controlled within a safe range of 30-35mm, eliminating the risk of damage due to exceeding the limit. Engineering feasibility: Zero hardware increment, algorithm computing power consumption ≤30%, single calibration time ≤30min, perfectly compatible with mass-produced CDC and air suspension systems.

[0035] Example 2: Emergency braking control scenario (vehicle speed 60km / h, emergency braking deceleration ≥0.6g) 1. Implementation conditions Test vehicle: Same as Example 1, equipped with a front-axle lidar pre-aiming CDC active suspension system; Test conditions: straight paved road in the city, initial vehicle speed 60km / h, the driver suddenly brakes, the pressure of the brake master cylinder rises sharply, which is an aggressive handling condition with high risk of instability. Calibration parameters: Use the basic calibration coefficients of the whole vehicle, with anti-aiming delay. (Emergency control prediction delay shortened) MPC prediction step size Penalty coefficient (Enhance control smoothness).

[0036] 2. Implementation Steps Actual data collected: Steering wheel deviation There is a risk of brake slippage and road surface elevation deviation. Braking pressure change rate Peak vehicle pitch acceleration Suspension travel .

[0037] Substitute into the coupling stability coefficient formula for calculation: Judgment result: The system is in a state of severe instability, with misalignment of human-machine aiming timing and significant coupling conflicts, prompting the activation of a full conflict resolution strategy.

[0038] Will Substituting into the weight evolution formula, priority is given to ensuring manipulation stability and coupling stability: (Comfort priority has been reduced, and the demand for smoothness has been weakened). (The weight of handling stability has been increased to suppress pitch deviation); (Maximize the coupling stability weights to prevent system instability); Constraint verification: ,and It meets the hard constraint requirements, and the weights switch quickly and smoothly without any impact.

[0039] The MPC control cycle is set to 10ms, the pre-aiming time domain is advanced to 150ms, and the emergency control command is solved by substituting it into the cost function. The CDC shock absorber damping was stiffened to 1.8kN·s / m, increasing the front suspension stiffness by 30%, which quickly suppressed body pitch, countered braking deviation, and eliminated human-machine anti-alignment timing deviation.

[0040] 3. Implementation Results Coupling stability: after modulation It quickly fell back to 0.28, turning from severe instability to a stable state, with no controlled oscillations and no suspension over-range. Handling safety: Brake pitch is reduced by 40%, tire vertical dynamic load fluctuation is reduced by 35%, and braking distance is shortened by 1.2m; Anti-instability capability: Compared with traditional solutions, the duration of human-machine coupling instability under emergency braking conditions is reduced by 85%, and there is no risk of vehicle body pitching or veering. Engineering compatibility: Emergency response delay ≤30ms, fully compatible with ESP and ABS collaborative control, with no hardware conflicts.

[0041] Example 3: Continuous lane change control scenario (vehicle speed 50km / h, 2 lane changes completed within 3 seconds) 1. Implementation conditions Test vehicle: Same as in Examples 1 and 2, equipped with a front-axle lidar pre-aiming CDC active suspension system; Test conditions: a straight multi-lane road in the city, a speed of 50 km / h, the driver completes two consecutive lane changes within 3 seconds, the steering wheel is turned back and forth at high frequency, which is a test condition with frequent human-machine pre-aiming mismatch and dynamic coupling disturbance. Calibration parameters: Pre-aiming delay , MPC prediction step size Penalty coefficient The upper limit of the suspension travel threshold is 40mm.

[0042] 2. Implementation Steps Actual dynamic data collected: Steering wheel angle deviation (Large deviation in high-frequency lane change prediction), road surface elevation deviation Peak steering wheel angular rate Body roll acceleration Peak suspension travel .

[0043] Substitute into the coupling stability coefficient formula for calculation: Judgment result: The system is in a state of mild instability, with dynamic misalignment between human and machine aiming. The weight smoothing adjustment and timing synchronization strategy is activated.

[0044] Will Substituting into the weight evolution formula, we can balance manipulation stability and coupling stability: (Comfort level moderately reduced); (Improved handling stability, reduced body roll); (Strengthening of coupling stability weights to ensure dynamic synchronization); Constraint verification: ,and The weights are dynamically and smoothly adapted to high-frequency lane-changing conditions.

[0045] The MPC control cycle is set to 10ms to correct human-machine anticipation timing deviation in real time. The damping difference between the left and right wheels of the CDC shock absorber is dynamically adjusted, with inner damping of 1.5kN·s / m and outer damping of 1.0kN·s / m to suppress body roll and wheel track deviation, ensuring that the anticipation trajectory is synchronized with the actual control.

[0046] 3. Implementation Results Coupling stability: throughout Stable within the 0.2~0.35 range, with no oscillations caused by control commands, and a human-machine anti-aiming dynamic synchronization rate ≥92%; Handling quality: Body roll angle reduced by 38%, no fishtailing or drifting sensation, tire grip retention rate ≥90%; Dynamic adaptability: Under high-frequency lane change conditions, the suspension response is lag-free and the anti-misalignment compensation is misaligned, improving comfort by 22% compared to traditional solutions; Reliability: The suspension dynamic travel is ≤38mm throughout, without reaching the safety threshold, and the actuator has no overload or overheating faults.

[0047] In summary, the verification results from three typical real-world driving scenarios—smooth driving on straight urban roads, emergency braking, and continuous lane changing—demonstrate that the proposed human-machine dual-pre-aiming coupled stability control method can accurately identify the system's coupled stability state under different driving scenarios, dynamically adapt control targets and adjustment strategies, and effectively solve problems such as human-machine pre-aiming conflicts, control command oscillations, and vehicle posture fluctuations. While ensuring system coupling stability, it significantly optimizes ride comfort and vehicle handling safety. Furthermore, it possesses engineering application advantages such as no hardware increments, lightweight algorithms, convenient calibration, and compatibility with mass-produced vehicles. This fully verifies the effectiveness, robustness, and practical engineering feasibility of the method, making it widely applicable to passenger vehicles equipped with axle-front pre-aiming active suspension and possessing high promotional and application value.

[0048] To better demonstrate the implementation of a human-machine dual-preview coupling stable axle front active suspension control method, this invention provides a human-machine dual-preview coupling stable axle front active suspension control system, comprising: The multi-source information perception module is used to acquire driver control preview information, axle-front road machine preview information and vehicle real-time operating status information, and to perform filtering and noise reduction, timestamp synchronization and alignment, and data normalization processing on the multi-source information, and to perform spatiotemporal alignment of the multi-source information to output standardized multi-source control input data. The coupled modeling module is used to extract driver control prediction characteristics and axle-front road pre-aiming disturbance based on standardized multi-source control input data, and to construct and iteratively update the human-machine dual pre-aiming coupled dynamic model. The stability determination module is used to calculate the computer-computer pre-aiming coupling stability coefficient based on the human-machine dual-aiming coupling dynamic model, and to perform a graded determination of coupling stability based on the coupling stability coefficient. The weight optimization module is used to dynamically optimize the control weights of multiple objectives based on the coupling stability classification results, thereby resolving human-machine pre-aiming timing deviations and coupling conflicts. The hierarchical control module is used to generate suspension damping and stiffness adjustment commands based on the optimized multi-objective control weights and a hierarchical collaborative control architecture. The suspension actuator module is used to receive and execute suspension damping and stiffness control commands, perform active suspension real-time adjustments, and provide feedback on vehicle attitude and suspension operating status data.

[0049] In this embodiment, the multi-source information sensing module includes: The driver control perception unit is used to collect steering wheel angle, steering rate, throttle opening change rate, brake pressure change rate, and driver anti-aiming delay. The axle-front road surface pre-aiming sensing unit uses lidar or vision sensors to collect data on the axle-front road surface elevation, slope, bump level, and machine pre-aiming delay. The vehicle status acquisition unit is used to collect data on the vehicle's vertical acceleration, suspension travel, tire dynamic load, and vehicle speed.

[0050] In this embodiment, the system is integrated into the vehicle electronic control unit (ECU). At the hardware level, it fully reuses the vehicle perception sensors and suspension actuators without any additional hardware increments. It can achieve coordinated control with the vehicle electronic stability program (ESP) and the anti-lock braking system (ABS). In this embodiment, the suspension execution module includes at least one of an electromagnetic damper (CDC) and an air spring, with an execution response time of no more than 50ms, and can dynamically adjust damping and stiffness to suppress vehicle pitch, roll and vertical fluctuations. In this embodiment, the system is adapted to multiple scenarios such as smooth driving, emergency braking, and continuous lane changing. By dynamically switching the weights of multiple objectives, it achieves synergistic optimization of driving comfort and vehicle handling stability. Under complex conditions, the system coupling stability coefficient meets the target rate of no less than 99%.

[0051] In the above embodiments, by using human-machine dual-pre-aiming coupling modeling, quantitative stability determination, dynamic weight optimization and hierarchical collaborative control, the problems of human-machine pre-aiming timing mismatch, coupling conflict and control oscillation are solved from the root. Under all working conditions, driving comfort, handling stability and system operation safety are improved at the same time. The hardware has strong reusability and is compatible with mass-produced vehicle platforms, and has outstanding engineering application value and large-scale promotion potential.

[0052] This embodiment provides a human-machine dual-look-ahead coupled and stable axle-front active suspension control method, which iteratively updates the parameters of the human-machine dual-look-ahead coupled dynamic model based on feedback data, including: The current moment is taken as the endpoint, and the suspension dynamic travel time sequence data and tire dynamic load time sequence data within a preset time window are extracted based on the endpoint. Within a preset time window, determine the first moment corresponding to the maximum value of the suspension dynamic travel and the second moment corresponding to the maximum value of the tire dynamic load, respectively, and calculate the difference between the first moment and the second moment, and use the difference as the suspension-tire response phase difference. The driver control anti-aiming input increment and the axle-front road machine anti-aiming control increment are obtained at the current moment, and the suspension-tire response phase difference is compared with the preset first phase difference threshold and the preset second phase difference threshold, respectively, wherein the preset first phase difference threshold is less than the preset second phase difference threshold; When the suspension-tire response phase difference is less than the preset first phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be in-phase superposition type, and the current vehicle speed value is obtained based on the determination result. The corresponding first correction gain is obtained by querying the pre-stored in-phase gain mapping table based on the current vehicle speed value and the suspension-tire response phase difference. The absolute value of the summation of the driver's pre-aiming input increment and the pre-aiming control increment of the road surface machine in front of the axle is obtained, and the absolute value is divided by the preset reference force to obtain the in-phase superposition coefficient. At the same time, the product of the first correction gain and the in-phase superposition coefficient is determined, and the product is added to the value 1. The sum of the product and the value 1 is divided into intervals based on the preset lower limit and the preset upper limit to obtain the first stiffness correction coefficient. When the suspension-tire response phase difference is greater than the preset second phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be the anti-phase cancellation type, and the current vehicle speed value is obtained based on the determination result. The corresponding second correction gain is obtained by querying the pre-stored anti-phase gain mapping table based on the current vehicle speed value and the suspension-tire response phase difference. The absolute value is obtained by subtracting the driver's control pre-aiming input increment from the pre-aiming control increment of the road surface machine in front of the axle, and then dividing the absolute value by the preset reference force to obtain the anti-phase cancellation coefficient. At the same time, the product of the second correction gain and the anti-phase cancellation coefficient is determined, and the product is added to the value 1. The sum of the product and the value 1 is then divided into intervals based on the preset lower limit and the preset upper limit to obtain the second stiffness correction coefficient. When the suspension-tire response phase difference is greater than or equal to the preset first phase difference threshold and less than or equal to the preset second phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be the mid-position weak coupling type, and the original stiffness matrix remains unchanged. Obtain the original stiffness matrix in the human-machine dual-pre-aiming coupled dynamic model, and determine the diagonal elements of the original stiffness matrix, where the diagonal elements correspond to the vertical stiffness, lateral stiffness and longitudinal stiffness of the suspension, respectively. Based on the relationship between the suspension-tire response phase difference and the preset first phase threshold and second phase threshold, each diagonal element of the original stiffness matrix is ​​multiplied by the first stiffness correction coefficient or the second stiffness correction coefficient to obtain the corrected stiffness matrix. The original stiffness matrix in the human-machine dual-aiming coupled dynamic model is replaced by the modified stiffness matrix, thereby completing the iterative update of the stiffness parameters in the human-machine dual-aiming coupled dynamic model.

[0053] In this embodiment, based on the difference between the suspension-tire response phase difference and the preset first phase threshold and second phase threshold, each diagonal element of the original stiffness matrix is ​​multiplied by either the first stiffness correction coefficient or the second stiffness correction coefficient to obtain the corrected stiffness matrix, specifically: When the suspension-tire response phase difference is less than the preset first phase difference threshold, that is, the current disturbance type is in-phase superposition, then each diagonal element of the original stiffness matrix is ​​multiplied by the first stiffness correction coefficient to obtain the corrected stiffness matrix. When the suspension-tire response phase difference is greater than the preset second phase difference threshold, that is, the current disturbance type is the anti-phase cancellation type, then each diagonal element of the original stiffness matrix is ​​multiplied by the second stiffness correction coefficient to obtain the corrected stiffness matrix. When the suspension-tire response phase difference is greater than or equal to the preset first phase difference threshold and less than or equal to the preset second phase difference threshold, the original stiffness matrix remains unchanged.

[0054] In this embodiment, dividing the obtained absolute value by a preset reference force to obtain the in-phase superposition coefficient and dividing the obtained absolute value by a preset reference force to obtain the out-of-phase cancellation coefficient include: The incremental driver control pre-aiming input is converted into the equivalent force on the suspension system through the control input mapping matrix. The incremental pre-aiming control of the road surface machine in front of the axle is also converted into the equivalent force on the suspension system through the pre-aiming control mapping matrix. Both have the dimension of force. The absolute value of the sum of the incremental driver control pre-aiming input and the incremental pre-aiming control of the road surface machine in front of the axle is taken to obtain the amplitude of the total coupled disturbance force. The amplitude of the total coupled disturbance force is divided by the preset reference force to obtain the dimensionless in-phase superposition coefficient. Similarly, the absolute value of the difference between the incremental driver control pre-aiming input and the incremental pre-aiming control of the road surface machine in front of the axle is taken to obtain the amplitude of the residual coupled force after cancellation. The amplitude of the residual coupled force is divided by the preset reference force to obtain the dimensionless anti-phase cancellation coefficient.

[0055] In this embodiment, the in-phase gain mapping table is pre-constructed based on bench test and real vehicle calibration data of the vehicle suspension system. The input of the in-phase gain mapping table is the current vehicle speed value and the suspension-tire response phase difference, and the output is the first correction gain. The vehicle speed value ranges from 0 km / h to the vehicle's maximum design speed, the suspension-tire response phase difference value ranges from 0 ms to the length of a preset time window, and the first correction gain value ranges from 0.5 to 1.5. In the in-phase gain mapping table, the first correction gain increases monotonically with increasing vehicle speed and decreases monotonically with increasing suspension-tire response phase difference. The inverting gain mapping table is also pre-constructed based on bench tests and real vehicle calibration data of the vehicle suspension system. The input of the inverting gain mapping table is the current vehicle speed value and the phase difference between the suspension and tire response, and the output is the second correction gain. The value range of the second correction gain is 0.8 to 1.2. In the inverting gain mapping table, the second correction gain increases monotonically with the increase of vehicle speed and increases monotonically with the increase of the phase difference between the suspension and tire response.

[0056] In this embodiment, the suspension-tire response phase difference refers to the difference between the first moment corresponding to the maximum value of the suspension dynamic travel and the second moment corresponding to the maximum value of the tire dynamic load, which is used to characterize the response hysteresis relationship between the suspension system and the tire.

[0057] In this embodiment, the preset first phase difference threshold refers to a pre-set phase difference critical value used to distinguish between in-phase superposition type disturbances and mid-position weak coupling type disturbances, which is determined through actual vehicle calibration.

[0058] In this embodiment, the preset second phase difference threshold refers to a pre-set phase difference critical value used to distinguish between mid-position weakly coupled disturbances and anti-phase canceling disturbances. This threshold is greater than the preset first phase difference threshold and is determined through actual vehicle calibration.

[0059] In this embodiment, the in-phase superposition type refers to the human-machine anti-coupling disturbance type when the suspension-tire response phase difference is less than a preset first phase difference threshold. Under the in-phase superposition type, the driver control anti-coupling disturbance and the machine road anti-coupling disturbance exhibit a superposition effect with the same direction.

[0060] In this embodiment, the anti-phase cancellation type refers to the human-machine anti-coupling disturbance type when the suspension-tire response phase difference is greater than a preset second phase difference threshold. Under the anti-phase cancellation type, the driver control anti-coupling disturbance and the machine road anti-coupling disturbance exhibit an anti-coupling effect with opposite directions.

[0061] In this embodiment, the first correction gain refers to the scaling factor used to calculate the stiffness correction coefficient under in-phase superposition disturbances, which is obtained from the in-phase gain mapping table.

[0062] In this embodiment, the second correction gain refers to the scaling factor used to calculate the stiffness correction coefficient under anti-phase cancellation perturbation, which is obtained from the anti-phase gain mapping table.

[0063] In this embodiment, the preset reference force refers to the pre-set reference force value used to convert the force value into a dimensionless coefficient. The reference force value is equal to the static load borne by a single wheel suspension under full load conditions. For example, the preset reference force value for a compact car is 4000N.

[0064] In this embodiment, the in-phase superposition coefficient refers to the dimensionless coefficient obtained by summing the driver's control aiming input increment and the road surface machine aiming control increment in front of the axle, taking the absolute value, and then dividing it by the preset reference force. It represents the relative amplitude of the total coupled disturbance force under in-phase superposition type disturbance.

[0065] In this embodiment, the anti-phase cancellation coefficient refers to the dimensionless coefficient obtained by subtracting the driver's control aiming input increment from the road surface machine aiming control increment in front of the axle, taking the absolute value, and then dividing it by the preset reference force. It represents the relative amplitude of the residual coupled disturbance force under the anti-phase cancellation type disturbance.

[0066] In this embodiment, the first stiffness correction coefficient refers to the value obtained by adding the product of the first correction gain and the in-phase superposition coefficient to the value 1, and then limiting it by a preset lower limit and a preset upper limit. This value is used to perform multiplication operations on the diagonal elements of the original stiffness matrix.

[0067] In this embodiment, the second stiffness correction coefficient refers to the value obtained by adding the product of the second correction gain and the antiphase cancellation coefficient to the value 1, and then limiting it by a preset lower limit and a preset upper limit. This value is used to perform multiplication on the diagonal elements of the original stiffness matrix.

[0068] In this embodiment, the mid-position weak coupling type refers to the human-machine pre-aiming coupling disturbance type when the suspension-tire response phase difference is greater than or equal to a preset first phase difference threshold and less than or equal to a preset second phase difference threshold. Under the mid-position weak coupling type, the original stiffness matrix remains unchanged.

[0069] In this embodiment, the original stiffness matrix refers to the uncorrected stiffness matrix at the current moment in the human-machine dual-pre-aiming coupled dynamic model, and its diagonal elements correspond to the vertical stiffness, lateral stiffness and longitudinal stiffness of the suspension, respectively.

[0070] In this embodiment, the corrected stiffness matrix refers to the new stiffness matrix obtained by multiplying each diagonal element of the original stiffness matrix with either the first stiffness correction coefficient or the second stiffness correction coefficient. This new stiffness matrix is ​​used to replace the original stiffness matrix to achieve iterative updates of the model parameters.

[0071] In this embodiment, the preset lower limit and the preset upper limit refer to the boundary values ​​that are preset to limit the sum of the product and the value 1. The preset lower limit is 0.8 and the preset upper limit is 1.5 to prevent the stiffness correction coefficient from exceeding the safe range.

[0072] In this embodiment, the purpose of adding the product to the value 1 is to normalize the stiffness correction coefficient based on the original stiffness. Specifically, when the product of the first correction gain and the in-phase superposition coefficient is zero, the first stiffness correction coefficient is 1, indicating that no correction is needed and the original stiffness matrix is ​​maintained. When the product is positive, the first stiffness correction coefficient is greater than 1, realizing positive incremental adjustment of stiffness. Similarly, the second stiffness correction coefficient is superimposed on the product of the second correction gain and the anti-phase cancellation coefficient based on 1, ensuring that the correction coefficient returns to the unit value when there is no disturbance input, thereby avoiding sudden changes in stiffness due to coefficient jumps.

[0073] The working principle and beneficial effects of the above technical solution are as follows: By real-time interception of suspension dynamic travel and tire dynamic load time series data, the suspension-tire response phase difference is calculated. Based on the comparison results of the phase difference with the preset first phase difference threshold and the preset second phase difference threshold, the human-machine pre-aiming coupling disturbance type is divided into in-phase superposition type, anti-phase cancellation type and mid-position weak coupling type. For different disturbance types, the correction gain is obtained by querying the mapping table based on vehicle speed and phase difference, and the dimensionless coefficient is obtained by combining the sum and difference calculation of the driver's control pre-aiming input increment and the machine pre-aiming control increment. The first stiffness correction coefficient or the second stiffness correction coefficient is generated. Then, the diagonal elements of the original stiffness matrix are multiplied to obtain the corrected stiffness matrix and replace the original model parameters. This realizes the dynamic iterative update of stiffness parameters and can adaptively adjust the suspension stiffness according to the real-time coupling disturbance characteristics, thereby improving the fitting accuracy of the human-machine dual pre-aiming coupling dynamic model to the real vehicle response.

[0074] This embodiment provides a human-machine dual-lookout coupling stability active suspension control method for the front axle. After calculating the human-machine lookout coupling stability coefficient in real time, the method includes correcting and determining the coupling stability coefficient. When correction is needed, the coupling stability coefficient is corrected, and a coupling stability classification is performed based on the corrected coupling stability coefficient. The specific process is as follows: Obtain the coupling stability coefficient of the current control cycle and the coupling stability coefficient of the previous control cycle; The coupling stability coefficient of the current control cycle is compared with that of the previous control cycle to determine whether the coupling stability coefficient needs to be corrected. If the coupling stability coefficient of the current control cycle is less than or equal to the coupling stability coefficient of the previous control cycle, it is determined that no correction is needed for the coupling stability coefficient, and a coupling stability classification judgment is made based on the coupling stability coefficient of the current control cycle. Otherwise, calculate the change in the coupling stability coefficient between the current control cycle and the previous control cycle; in, This indicates the change in the coupling stability coefficient between the current control cycle and the previous control cycle. This represents the coupling stability coefficient of the current control cycle; Indicates the current control cycle; This represents the coupling stability coefficient of the previous control cycle; Indicates the previous control cycle; Change in coupling stability coefficient With the preset stability threshold By comparing the coefficients, the smoothing suppression coefficient of the coupling stability coefficient for the current control cycle is determined; when If the coupling stability coefficient of the current control cycle is determined to be fluctuating normally within the allowable range, the smoothing suppression coefficient is set to 1. when If the coupling stability coefficient of the current control cycle is determined to be an abnormally rapid jump, then the smoothing suppression coefficient of the coupling stability coefficient of the current control cycle is calculated according to the following formula. ;in, The smoothing suppression coefficient represents the coupling stability coefficient of the current control cycle; The coupling stability coefficient of the current cycle is corrected based on the smoothing suppression coefficient of the coupling stability coefficient of the current control cycle; ;in, This represents the corrected coupling stability coefficient; The coupling stability is graded and determined based on the modified coupling stability coefficient.

[0075] In this embodiment, sensor noise, communication jitter, or extremely brief manipulation disturbances may cause the coupling stability coefficient to undergo a sudden change far beyond physical possibility within a single control cycle. If this sudden change value is used directly for classification, it may lead to misjudgment (e.g., misjudging a stable state as severe instability), thereby triggering unnecessary control command oscillations.

[0076] In this embodiment, the magnitude of the coupling stability coefficient directly reflects the risk of human-machine coupling instability. That is, the larger the coupling stability coefficient, the more severe the human-machine time delay mismatch is under the current prediction deviation, and the easier it is for control conflicts and attitude instability to occur.

[0077] In this embodiment, the change in the coupling stability coefficient between the current control cycle and the previous control cycle represents the growth rate of coupling instability risk within a single control cycle; by comparing it with a preset stability threshold, effective determination of abnormal rapid jumps is achieved. If the coupling stability coefficient of the current control cycle is determined to be fluctuating normally within the allowable range, it means that there may be real deterioration but the rate is reasonable, or there is only minor noise. In this case, no suppression is applied to the coupling stability coefficient. Otherwise, it indicates that the increase has exceeded the physically reasonable upper limit, and it is highly likely that the abnormal jump is caused by instantaneous noise or erroneous data. In this case, it is necessary to suppress the coupling stability coefficient to prevent abnormal coupling stability coefficients from directly entering the classification judgment stage. The preset stability threshold is a positive real number obtained in advance through actual vehicle calibration, representing the maximum allowable increase of the coupling stability coefficient within a single control cycle. Exceeding this value is judged as an abnormal rapid jump.

[0078] In this embodiment, the smoothing suppression coefficient takes a value in the range of (0, 1]. Therefore, it is used to compress the coupling stability coefficient after an abnormal jump. When indicates no suppression, when When the value is 0, it indicates inhibition, and the intensity of inhibition increases as the smoothing inhibition coefficient decreases.

[0079] The working principle and beneficial effects of the above technical solution are as follows: by comparing the stability coefficients coupled between adjacent periods and judging the threshold, it can effectively identify abnormal rapid jumps and implement smooth suppression corrections that are inversely proportional to the jump amount, filter out single-period noise disturbances, avoid graded misjudgments and control oscillations; when there is no deterioration or the fluctuation is within the allowable range, the original value is directly retained to ensure a sensitive response to the real risk trend; when the stability threshold is exceeded, adaptive compression is performed to ensure that the corrected coefficients take into account both the degree of deterioration and the stability of the output.

[0080] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A human-machine dual-pre-aiming coupled and stable axle front active suspension control method, characterized in that, Includes the following steps: Simultaneously collect multi-source vehicle operation and anti-aiming data, which includes driver control anti-aiming information, axle-front road machine anti-aiming information, and real-time vehicle operation status information; Based on the acquired multi-source vehicle operation and anti-targeting data, multi-source information is spatiotemporally aligned and standardized preprocessed to generate standardized multi-source control input data. Based on standardized multi-source control input data, driver control prediction features and axle-front road surface anti-aiming disturbance features are extracted to construct a human-machine dual anti-aiming coupled dynamic model. Define and calculate the human-machine anti-aiming coupling stability coefficient, and classify the coupling stability of the suspension system based on the numerical range of the coupling stability coefficient. Based on the results of the coupling stability classification, the multi-objective control weights of ride comfort, vehicle handling stability, and system coupling stability are dynamically optimized. A hierarchical collaborative control architecture is adopted, combined with optimized multi-objective control weights, to generate suspension damping and stiffness adjustment commands; Real-time vehicle attitude and suspension operation feedback data are acquired, and the parameters and multi-objective control weights of the human-machine dual-look-ahead coupled dynamic model are iteratively updated based on the feedback data.

2. The human-machine dual-pre-aiming coupled and stable axle-front active suspension control method as described in claim 1, characterized in that, The process of constructing the human-machine dual-aiming coupled dynamic model includes: Based on standardized multi-source control input data, the incremental driver control anti-aiming input and the incremental axle-front road machine anti-aiming control are calculated separately. Based on the vehicle's inherent parameters and suspension system parameters, the vehicle mass matrix, suspension damping matrix, and stiffness matrix are obtained. By combining the calculated incremental driver control anti-aiming input, incremental machine anti-aiming control on the road surface in front of the axle, vehicle mass matrix, suspension damping matrix and stiffness matrix, a dynamic correlation equation is established between the human-machine anti-aiming coupled disturbance force and the vehicle displacement, suspension dynamic travel and tire dynamic load, forming a human-machine dual anti-aiming coupled dynamic model.

3. The human-machine dual-pre-aiming coupled and stable axle-front active suspension control method as described in claim 1, characterized in that, The definition and calculation of the human-machine pre-aiming coupling stability coefficient includes: Based on standardized multi-source control input data and coupled dynamic model output, key coupling feature parameters are extracted and determined. These key coupling feature parameters include driver anti-aiming delay, machine anti-aiming delay, driver steering prediction deviation, and road surface elevation anti-aiming deviation. The driver's anticipation delay is compared with the machine's anticipation delay to obtain the difference between the human and machine anticipation delays. At the same time, the driver's steering prediction deviation and the road elevation prediction deviation are vector-synthesized to obtain the comprehensive prediction deviation magnitude. Based on the difference in human-machine pre-aiming delay and the magnitude of comprehensive prediction deviation, a formula for calculating the human-machine pre-aiming coupling stability coefficient is constructed, and the human-machine pre-aiming coupling stability coefficient is calculated in real time. The calculated coupling stability coefficient is compared with the preset grading threshold range. Based on the comparison results, the coupling stability of the suspension system is divided into stable state, mild instability state and severe instability state, and a coupling stability grading judgment result is generated. The calculated human-machine pre-aiming coupling stability coefficient is compared with the coupling stability grading threshold range one by one to determine the numerical range to which the coupling stability coefficient belongs. When the coupling stability coefficient is less than or equal to the first threshold, the suspension system is determined to be in a stable state, a stability determination result is generated, and a timing synchronization fine-tuning strategy is executed. When the coupling stability coefficient is greater than the first threshold and less than or equal to the second threshold, the suspension system is determined to be in a slightly unstable state, a slightly unstable determination result is generated, and a weight smoothing adjustment strategy is executed. When the coupling stability coefficient is greater than the second threshold, the suspension system is determined to be in a severely unstable state, a severely unstable determination result is generated, and a full conflict resolution strategy is executed.

4. The human-machine dual-pre-aiming coupled and stable axle front active suspension control method as described in claim 1, characterized in that, The multi-objective control weights for dynamically optimizing ride comfort, vehicle handling stability, and system coupling stability include: Based on the results of the coupling stability classification, the human-machine pre-aiming coupling stability coefficient is obtained, and the initial configuration and dynamic adjustment rules of the multi-target control weights are determined. Based on dynamic adjustment rules, the weights of ride comfort, vehicle handling stability, and system coupling stability are calculated in real time, and the multi-objective control weights are optimized based on the calculation results. The optimized multi-objective control weights are input into the hierarchical collaborative control architecture for the coordinated regulation of suspension damping and stiffness.

5. The human-machine dual-pre-aiming coupled and stable axle-front active suspension control method as described in claim 4, characterized in that, The hierarchical collaborative control architecture includes an upper coupling and coordination layer and a lower control execution layer: The upper coupling coordination layer is used to obtain the coupling stability classification judgment result, dynamically match the corresponding weight allocation strategy according to the coupling stability classification judgment result, synchronize and coordinate the human-machine pre-aiming timing, and eliminate the timing deviation between human-machine operation and machine control. The lower control execution layer is used to receive the optimized multi-objective control weights, combine the standardized multi-source input data and the output results of the dynamic model, construct the control cost function using the model predictive control algorithm, and generate the corresponding suspension control commands based on the control cost function. The generated optimal control commands are sent to the suspension actuator in real time. The optimal control commands include suspension damping and stiffness control commands. The suspension actuator performs precise control of the front active suspension based on the suspension damping and stiffness control commands. The multi-objective control weights also include: The evolution rules of multi-objective control weights are obtained based on the coupling stability coefficient. The weights are then smoothly adjusted according to the evolution rules, and a lower limit protection threshold for coupling stability weights is set based on system safety requirements. The adjustment results are generated based on the trend of weight changes determined by the coupling stability coefficient. Among them, the weight of driving comfort decreases monotonically as the coupling stability coefficient increases, while the weights of handling stability and coupling stability increase monotonically as the coupling stability coefficient increases.

6. The human-machine dual-pre-aiming coupled and stable axle-front active suspension control method as described in claim 1, characterized in that, The parameters of the human-machine dual-aiming coupled dynamics model are iteratively updated based on feedback data, including: The current moment is taken as the endpoint, and the suspension dynamic travel time sequence data and tire dynamic load time sequence data within a preset time window are extracted based on the endpoint. Within a preset time window, determine the first moment corresponding to the maximum value of the suspension dynamic travel and the second moment corresponding to the maximum value of the tire dynamic load, respectively, and calculate the difference between the first moment and the second moment, and use the difference as the suspension-tire response phase difference. The driver control anti-aiming input increment and the axle-front road machine anti-aiming control increment are obtained at the current moment, and the suspension-tire response phase difference is compared with the preset first phase difference threshold and the preset second phase difference threshold, respectively, wherein the preset first phase difference threshold is less than the preset second phase difference threshold; When the suspension-tire response phase difference is less than the preset first phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be in-phase superposition type, and the current vehicle speed value is obtained based on the determination result. The corresponding first correction gain is obtained by querying the pre-stored in-phase gain mapping table based on the current vehicle speed value and the suspension-tire response phase difference. The absolute value of the summation of the driver's pre-aiming input increment and the pre-aiming control increment of the road surface machine in front of the axle is obtained, and the absolute value is divided by the preset reference force to obtain the in-phase superposition coefficient. At the same time, the product of the first correction gain and the in-phase superposition coefficient is determined, and the product is added to the value 1. The sum of the product and the value 1 is divided into intervals based on the preset lower limit and the preset upper limit to obtain the first stiffness correction coefficient. When the suspension-tire response phase difference is greater than the preset second phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be the anti-phase cancellation type, and the current vehicle speed value is obtained based on the determination result. The corresponding second correction gain is obtained by querying the pre-stored anti-phase gain mapping table based on the current vehicle speed value and the suspension-tire response phase difference. The absolute value is obtained by subtracting the driver's control pre-aiming input increment from the pre-aiming control increment of the road surface machine in front of the axle, and then dividing the absolute value by the preset reference force to obtain the anti-phase cancellation coefficient. At the same time, the product of the second correction gain and the anti-phase cancellation coefficient is determined, and the product is added to the value 1. The sum of the product and the value 1 is then divided into intervals based on the preset lower limit and the preset upper limit to obtain the second stiffness correction coefficient. When the suspension-tire response phase difference is greater than or equal to the preset first phase difference threshold and less than or equal to the preset second phase difference threshold, the current human-machine pre-aiming coupling disturbance type is determined to be the mid-position weak coupling type, and the original stiffness matrix remains unchanged. Obtain the original stiffness matrix in the human-machine dual-pre-aiming coupled dynamic model, and determine the diagonal elements of the original stiffness matrix, where the diagonal elements correspond to the vertical stiffness, lateral stiffness and longitudinal stiffness of the suspension, respectively. Based on the relationship between the suspension-tire response phase difference and the preset first phase threshold and second phase threshold, each diagonal element of the original stiffness matrix is ​​multiplied by the first stiffness correction coefficient or the second stiffness correction coefficient to obtain the corrected stiffness matrix. The original stiffness matrix in the human-machine dual-aiming coupled dynamic model is replaced by the modified stiffness matrix, thereby completing the iterative update of the stiffness parameters in the human-machine dual-aiming coupled dynamic model.

7. The human-machine dual-pre-aiming coupled and stable axle-front active suspension control method as described in claim 3, characterized in that, After calculating the human-machine pre-aiming coupling stability coefficient in real time, the process includes correcting and determining the coupling stability coefficient. When correction is needed, the coupling stability coefficient is corrected, and a coupling stability classification is performed based on the corrected coupling stability coefficient. The specific process is as follows: Obtain the coupling stability coefficient of the current control cycle and the coupling stability coefficient of the previous control cycle; The coupling stability coefficient of the current control cycle is compared with that of the previous control cycle to determine whether the coupling stability coefficient needs to be corrected. If the coupling stability coefficient of the current control cycle is less than or equal to the coupling stability coefficient of the previous control cycle, it is determined that no correction is needed for the coupling stability coefficient, and a coupling stability classification judgment is made based on the coupling stability coefficient of the current control cycle. Otherwise, calculate the change in the coupling stability coefficient between the current control cycle and the previous control cycle; Change in coupling stability coefficient With the preset stability threshold By comparing the coefficients, the smoothing suppression coefficient of the coupling stability coefficient for the current control cycle is determined; when If the coupling stability coefficient of the current control cycle is determined to be fluctuating normally within the allowable range, the smoothing suppression coefficient is set to 1. when If the coupling stability coefficient of the current control cycle is determined to be an abnormally rapid jump, then the smoothing suppression coefficient of the coupling stability coefficient of the current control cycle is calculated according to the following formula. ;in, The smoothing suppression coefficient represents the coupling stability coefficient of the current control cycle; The coupling stability coefficient of the current cycle is corrected based on the smoothing suppression coefficient of the coupling stability coefficient of the current control cycle; The coupling stability is graded and determined based on the modified coupling stability coefficient.

8. The human-machine dual-pre-aiming coupled and stable axle front active suspension control method as described in claim 1, characterized in that, The real-time acquisition of vehicle attitude and suspension operation feedback data also includes: Real-time feedback data is spatiotemporally compared with standardized multi-source control input data, and closed-loop feedback correction is obtained based on the comparison results. The mass matrix, damping matrix, and stiffness matrix parameters of the human-machine dual-aiming coupled dynamic model are iteratively updated based on the closed-loop feedback correction. The initial baseline values ​​and evolution coefficients of the multi-objective control weights are updated synchronously and iteratively.

9. A human-machine dual-preview coupling stable axle front active suspension control system, applied in the human-machine dual-preview coupling stable axle front active suspension control method as described in any one of claims 1-8, characterized in that, include: The multi-source information perception module is used to acquire driver control preview information, axle-front road machine preview information and vehicle real-time operating status information, and to perform filtering and noise reduction, timestamp synchronization and alignment, and data normalization processing on the multi-source information, and to perform spatiotemporal alignment of the multi-source information to output standardized multi-source control input data. The coupled modeling module is used to extract driver control prediction characteristics and axle-front road pre-aiming disturbance based on standardized multi-source control input data, and to construct and iteratively update the human-machine dual pre-aiming coupled dynamic model. The stability determination module is used to calculate the computer-computer pre-aiming coupling stability coefficient based on the human-machine dual-aiming coupling dynamic model, and to perform a graded determination of coupling stability based on the coupling stability coefficient. The weight optimization module is used to dynamically optimize the control weights of multiple objectives based on the coupling stability classification results, thereby resolving human-machine pre-aiming timing deviations and coupling conflicts. The hierarchical control module is used to generate suspension damping and stiffness adjustment commands based on the optimized multi-objective control weights and a hierarchical collaborative control architecture. The suspension actuator module is used to receive and execute suspension damping and stiffness control commands, perform active suspension real-time adjustments, and provide feedback on vehicle attitude and suspension operating status data.

10. The human-machine dual-pre-aiming coupled and stable front axle active suspension control system as described in claim 9, characterized in that, The multi-source information sensing module includes: The driver control perception unit is used to collect steering wheel angle, steering rate, throttle opening change rate, brake pressure change rate, and driver anti-aiming delay. The axle-front road surface pre-aiming sensing unit uses lidar or vision sensors to collect data on the axle-front road surface elevation, slope, bump level, and machine pre-aiming delay. The vehicle status acquisition unit is used to collect data on the vehicle's vertical acceleration, suspension travel, tire dynamic load, and vehicle speed.

Images