Position-force feedback optimal control method and system for improving trafficability and wheel grounding performance of active suspension vehicle

Abstract

The invention discloses a position-force feedback optimal control method and system for improving the trafficability and wheel grounding performance of an active suspension vehicle, and belongs to the technical field of vehicle suspension control, and the method comprises the steps: building a two-degree-of-freedom 1 / 6 vehicle active suspension model and a valve control actuator system model; establishing a position impedance control model and obtaining expected displacement of real-time sprung mass; establishing a position-force feedback optimal control model of the vehicle active suspension system; and using a position-force optimal control strategy to realize optimal control of the vehicle active suspension system. According to the method, vibration transmission of the vehicle in the running process can be reduced, impact on the suspension system is relieved, the grounding capacity of the wheels is improved, the position and force information of the suspension system is monitored and fed back in real time through position-force feedback optimal control, and the smoothness and handling stability of the vehicle are improved.

Description

Technical Field

[0001] This invention relates to the field of active suspension control technology for automobiles, and in particular to a position-force feedback optimal control method and system for improving the passability and wheel contact performance of vehicles with active suspension. Background Technology

[0002] The suspension system, as the core force transmission connection between the vehicle body and the wheels, directly affects the vehicle's driving stability. An active suspension system can flexibly adjust the position, travel, and control force of the actuators according to road conditions, operating conditions, and load changes, ensuring that the suspension system always maintains optimal damping performance, thereby maintaining close contact between the wheels and the ground.

[0003] Furthermore, while methods such as fuzzy control, PID control, LQG control, and adaptive control perform well in improving vehicle ride comfort and are easy to operate, their effects on improving vehicle handling stability are not significant. When a vehicle encounters random road surface excitations and impact excitations from bumps (potholes), its suspension system cannot effectively reduce vehicle vibration, and the wheels lack ground contact, making it difficult to simultaneously improve vehicle ride comfort and handling stability.

[0004] To address the shortcomings of the above technologies, this invention proposes an active suspension system control method and system based on position-force feedback optimal control to synergistically enhance vehicle passability and wheel contact performance. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a position-force feedback optimal control method and system for improving the passability and wheel contact performance of vehicles with active suspension. The method monitors and feeds back the position and force information of the suspension system in real time through a position and force feedback controller. When performing position difference feedback control, impedance control is used to obtain the desired displacement of the sprung mass in real time, and the desired control force of the actuator is determined. Then, the force feedback optimal controller is used to realize the target force control of the active suspension, thereby further improving the passability and wheel contact capability of the vehicle.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A position-force feedback optimal control method for improving the handling and wheel contact performance of vehicles with active suspension includes the following steps:

[0008] S1. Establish a two-degree-of-freedom 1 / 6 vehicle active suspension model and a valve-controlled actuator system model;

[0009] S2. Establish a position impedance control model and obtain the expected displacement of the sprung mass in real time;

[0010] S3. Establish the position-force feedback optimal control model of the vehicle active suspension system, including the position state feedback control model of the vehicle active suspension system based on fuzzy control and PID control and the force feedback optimal control model based on dual-loop control.

[0011] S4. The position-force optimal control strategy is used to achieve optimal control of the vehicle active suspension system; the impedance control model in S2 outputs the desired acceleration of the impedance model based on the wheel dynamic load in S1, and obtains the tracking error of the vehicle active suspension system; the position state feedback model in S3 calculates the desired output control force of the vehicle actuator; the force feedback optimal control model in S3 outputs the output force of the vehicle actuator, thereby controlling the vehicle active suspension.

[0012] A further improvement to the technical solution of this invention lies in: In S1, firstly, a dynamic model of a two-degree-of-freedom 1 / 6 vehicle active suspension is established based on Newton's second law:

[0013]

[0014] In the formula, m s For the sprung mass; m u For unsprung mass; k s c s These represent the stiffness and damping of the suspension system, respectively; k t c t These are the tire's equivalent stiffness and damping, respectively; s z u z r These represent the sprung mass displacement, unsprung mass displacement, and road surface excitation displacement, respectively; F u The actual control force input for the actuator; F d The desired control force for the actuator;

[0015] Based on the above model, passive suspension lacks actuator control force compared to active suspension. The state variables for passive and active suspension are:

[0016] make Take the output quantity as

[0017] The state-space expression for the vehicle's active suspension system is:

[0018]

[0019] The expression for wheel dynamic load is:

[0020] Establish a nonlinear model of vehicle actuator dynamics:

[0021] With the piston rod extended, the flow continuity equation of the hydraulic cylinder is expressed as:

[0022]

[0023] When the piston rod retracts, the flow continuity equations for the rodless and rod-side chambers of the hydraulic cylinder are expressed as follows:

[0024]

[0025] In the formula, x p β represents the displacement of the hydraulic cylinder piston rod. e C represents the effective volumetric elastic modulus of the oil; a The internal leakage coefficient C of the hydraulic cylinder is represented. b A1 and A2 represent the external leakage coefficient of the hydraulic cylinder; A1 and A2 represent the effective areas of the rodless and rod-side chambers, respectively; V a V b q1 represents the initial volume of the rodless chamber and the rod chamber of the hydraulic cylinder; q2 represents the inlet flow rate and q1 represents the outlet flow rate of the rod chamber.

[0026] A further improvement to the technical solution of this invention lies in: In S2, the position impedance control model is:

[0027]

[0028] The transfer function model between wheel dynamic load and vehicle body vertical acceleration is as follows:

[0029]

[0030] In the formula, m d z represents the inertia of an ideal target. sd The desired displacement of the sprung mass is expressed as k. d c represents the ideal target stiffness; d F represents the ideal target damping; dl (s) and z sd (s) are respectively F dl and z sd Laplace transform; s 2 z sd (s) represents the Laplace transform of the vertical acceleration of the sprung mass.

[0031] A further improvement to the technical solution of this invention lies in: In S3, the displacement state tracking control model uses fuzzy control and PID control methods to calculate the position correction amount z of the sprung mass based on the dynamic load of the wheel. sd Simultaneously, combined with the spring-loaded mass displacement z measured by the displacement sensor s These two parameters are used together as inputs to the position controller to obtain the desired control force F. d for:

[0032]

[0033] In the formula, e(t) is the control deviation; F is the desired output F of the controlled object. d The difference between F(t) and the feedback value c(t); t is the time variable; F(t) is the controller output; K p The proportional gain coefficient, K, determines the system's response strength to the current deviation. i K is the integral gain coefficient used to eliminate steady-state error; d This is the differential gain coefficient used to suppress system overshoot and improve dynamic response.

[0034] A further improvement of the technical solution of the present invention is that: in S3, the inner loop control adopts linear feedback control, and the outer loop control adopts linear quadratic Gaussian control; the linear quadratic Gaussian optimal controller generates the target control force and transmits it as an input signal to the inner loop control; the inner loop performs feedback control and inputs the control signal into the vehicle model.

[0035] A further improvement to the technical solution of this invention lies in: In S3, the linear feedback control model used for the inner loop control is:

[0036]

[0037] In the formula, e1 = F d -F t For force error, k p F is the proportionality coefficient. t To control the force, F d For desired control.

[0038] A further improvement to the technical solution of this invention lies in: In S3, the linear quadratic Gaussian model used for the outer loop control is:

[0039]

[0040] In the formula, q1 represents the suspension dynamic deflection weighting coefficient; q2 represents the wheel dynamic load weighting coefficient; q3 represents the sprung mass acceleration weighting coefficient; q4 represents the unsprung mass acceleration weighting coefficient; and r represents the active control force weighting coefficient.

[0041] The above formula can be written in matrix form:

[0042]

[0043] In the formula, Q is a positive semi-definite matrix. R is a positive definite matrix.

[0044] ρ is the weighting matrix of state variables; ρ = diag[q1q2q3q4]; n is the weighting matrix of control variables;

[0045] The optimal control force is expressed as:

[0046] F b =-KX

[0047] In the formula,

[0048] The linear quadratic Gaussian control feedback gain matrix K is determined by solving the following Riccati differential equation:

[0049]

[0050] The feedback gain matrix K is calculated by calling the LQR function in MATLAB:

[0051] K = LQR(A) s B s ,Q,R,M).

[0052] A position-force feedback optimal control system for improving the handling and wheel contact performance of vehicles with active suspension includes:

[0053] The road excitation module simulates the road unevenness excitation experienced by the vehicle during driving and transmits the road height information to the impedance control module.

[0054] The impedance control module, based on the road height information input from the road excitation module, combines impedance control to track wheel dynamic loads and obtain the desired sprung mass displacement.

[0055] The position status tracking control module generates a position correction amount using impedance control based on the actual displacement of the vehicle's active suspension. Subsequently, the fuzzy PID position controller adjusts the control force online in real time based on this position correction amount and transmits it to the force feedback optimal control module.

[0056] The force feedback optimal control module obtains the desired output force and the actual output force of the vehicle actuator, outputs the actual control quantity of the vehicle actuator, and drives the vehicle actuator to work.

[0057] The vehicle actuator and suspension system module outputs the actual control quantity of the actuator based on the force feedback optimal control module, which is then applied to the active suspension system to achieve real-time control of the vehicle's active suspension.

[0058] The technological advancements achieved by this invention due to the adoption of the above technical solutions are as follows:

[0059] 1. This invention employs a position-based impedance control method, which incorporates the contact force between the tire and the road surface into the control strategy, effectively reducing the impact of ground impact on the vehicle body's vibration, while improving the wheel's ground contact capability and handling stability.

[0060] 2. This invention monitors and provides feedback on the position and force information of the suspension system in real time through a position and force feedback controller. When performing position difference feedback control, impedance control is used to obtain the desired displacement of the sprung mass in real time, determine the desired control force of the actuator, and then use the force feedback LQG optimal controller to achieve active suspension target force control, thereby further improving the vehicle's passability and wheel contact capability.

[0061] 3. The position state tracking control proposed in this invention adopts fuzzy control + PID control, which can reduce the vibration transmission of the vehicle during driving, reduce the impact on the suspension system, improve the ground contact capability of the wheels, and improve the ride comfort and handling stability of the vehicle. Attached Figure Description

[0062] Figure 1 A flowchart of a position-force feedback optimal control method for improving the passability and wheel contact performance of vehicles with active suspension, provided by the present invention;

[0063] Figure 2 This is a schematic diagram of the vehicle active suspension system of the present invention;

[0064] Figure 3 This is a schematic diagram of the vehicle actuator system of the present invention.

[0065] Figure 4 This is a schematic diagram of the fuzzy + PID control method for position state tracking control according to the present invention;

[0066] Figure 5 This is a schematic diagram of the force feedback optimal control method of the present invention;

[0067] Figure 6 This is a comparison diagram of the vertical acceleration of the vehicle body on random road surfaces and obstacle road surfaces according to the present invention;

[0068] Figure 7 This is a comparison diagram of the dynamic deflection of the road suspension on random road surfaces and obstacle road surfaces according to the present invention;

[0069] Figure 8 This is a comparison diagram of wheel dynamic loads on random road surfaces and obstacle road surfaces according to the present invention. Detailed Implementation

[0070] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments:

[0071] like Figure 1 As shown, this invention provides a position-force feedback optimal control method for improving the passability and wheel contact performance of vehicles with active suspension, which includes the following steps:

[0072] S1. Establish a two-degree-of-freedom 1 / 6 vehicle active suspension model and a valve-controlled actuator system model;

[0073] like Figure 2 As shown, the dynamic model of a two-degree-of-freedom 1 / 6 vehicle active suspension is as follows:

[0074]

[0075] Based on the above model, the state variables for passive and active suspension are:

[0076] make Take the output quantity as

[0077] The state-space expression for the vehicle's active suspension system is:

[0078]

[0079]

[0080] The expression for wheel dynamic load is:

[0081]

[0082] In the formula, m s For the sprung mass; m u For unsprung mass; k s c s These represent the stiffness and damping of the suspension system, respectively; k t c t These are the tire's equivalent stiffness and damping, respectively; s z u z r These represent the sprung mass displacement, unsprung mass displacement, and road surface excitation displacement, respectively; F u The actual control force input for the actuator; F d The desired control force for the actuator.

[0083] like Figure 3 As shown, the dynamic model of the valve-controlled actuator is:

[0084] Specifically, the dynamic model of the valve-controlled actuator for piston rod extension and retraction includes:

[0085] When the piston rod extends, that is, when the valve core is displaced (x) v >0), the valve core moves upward. At this time, the oil inlet flow rate q1 in the rodless chamber and the oil outlet flow rate q2 in the rod chamber of the hydraulic cylinder are respectively:

[0086]

[0087] Define load flow q L and load pressure p L as follows:

[0088]

[0089] p L =p1-np2

[0090] The flow continuity equation for a hydraulic cylinder is expressed as:

[0091]

[0092] The valve core moves upward:

[0093]

[0094] The force balance equation for a hydraulic cylinder is:

[0095]

[0096] When the piston rod retracts, the flow continuity equations for the rodless and rod chambers of the hydraulic cylinder are expressed as follows:

[0097]

[0098] The valve core moves downward:

[0099]

[0100] in

[0101] The force balance equation for a hydraulic cylinder is:

[0102]

[0103] In the formula x v Indicates the displacement of the servo valve spool; p s Indicates the system oil supply pressure; p o Indicates the system return oil pressure; C d ρ represents the servo valve orifice flow coefficient; w represents the servo valve orifice area gradient; ρ represents the system oil density; p1 represents the oil pressure in the rodless chamber of the hydraulic cylinder; p2 represents the oil pressure in the rod chamber of the hydraulic cylinder; n represents the area ratio of the rod chamber to the rodless chamber of the hydraulic cylinder (n = A2 / A1); x p β represents the displacement of the hydraulic cylinder piston rod. e C represents the effective volumetric elastic modulus of the oil; a C represents the internal leakage coefficient of the hydraulic cylinder; b A1 and A2 represent the external leakage coefficient of the hydraulic cylinder; A1 and A2 represent the effective areas of the rodless and rod-side chambers, respectively; V a V b This indicates the initial volume of the rodless and rod-side chambers of the hydraulic cylinder; the total compression volume of the hydraulic cylinder. Leakage coefficient m t Indicates the equivalent load mass of the actuator; B p F represents the viscous damping coefficient of the piston and load. t It is expressed as the equivalent external load force of the actuator.

[0104] S2. Establish a position impedance control model and obtain the expected displacement of the sprung mass in real time;

[0105] Specifically, the position impedance control model is as follows:

[0106]

[0107] The transfer function model between wheel dynamic load and vehicle body vertical acceleration is as follows:

[0108]

[0109] In the formula, m d z represents the inertia of an ideal target. sd The desired displacement of the sprung mass is expressed as k. d c represents the ideal target stiffness; d F represents the ideal target damping; dl (s) and z sd (s) are respectively F dl and z sd Laplace transform; s 2 z sd (s) represents the Laplace transform of the vertical acceleration of the sprung mass.

[0110] S3. Establish the position-force feedback optimal control model of the vehicle active suspension system, specifically including the position state feedback control model and the force feedback optimal control model of the vehicle active suspension system;

[0111] S31. Establish a position state feedback control model for the vehicle's active suspension system;

[0112] like Figure 4 As shown, the position state feedback control adopts a fuzzy + PID control method, and calculates the position correction amount z of the sprung mass based on the dynamic load of the wheel. sd Simultaneously, combined with the spring-loaded mass displacement z measured by the displacement sensor s These two parameters serve as inputs to the position controller. This controller method optimizes the PID parameters online through a fuzzy inference module, preserving the accuracy of PID control while introducing the flexibility of fuzzy control. This effectively improves the dynamic response characteristics and robustness of the system, yielding the desired control force Fd as follows:

[0113]

[0114] In the formula, e(t) is the control deviation; F is the desired output F of the controlled object. d The difference between F(t) and the feedback value c(t); t is the time variable; F(t) is the controller output; K p The proportional gain coefficient, K, determines the system's response strength to the current deviation. i K is the integral gain coefficient used to eliminate steady-state error; d The differential gain coefficient is used to suppress system overshoot and improve dynamic response;

[0115] The core of fuzzy-PID control is to dynamically optimize PID control parameters through fuzzy rules, using the vehicle displacement error e and its rate of change e c As input variables, the output variable is the increment ΔK of the PID control parameters. p , △K i , △K d The implementation process is as follows:

[0116] A1. Fuzzification: Transforming the input variables e and e c The conversion from precise values ​​to fuzzy values ​​is achieved by mapping membership functions to fuzzy sets.

[0117] A2. Fuzzy Reasoning: Based on a preset fuzzy rule base, it performs logical judgments on input variables through a reasoning mechanism to generate fuzzy output;

[0118] A3. Defuzzification: Converting the fuzzy inference results into precise PID parameter increments ΔK using a defuzzification algorithm. p , △K i , △K d ;

[0119] A4. Parameter Adjustment: Update the PID control parameters in real time based on the incremental values, i.e.:

[0120]

[0121] Parameter in the formula: K p0 K i0 K d0 These represent the initial setpoints for PID control; q p q i q d These are the correction coefficients for the fuzzy controller.

[0122] A5. Control Output: Calculate the control force F using the optimized PID parameters. d It is then applied to the vehicle's active suspension system to achieve rapid tracking and adjustment of the vehicle body displacement correction.

[0123] The controller's input and output variables are described using seven fuzzy subsets as linguistic variables, defined as follows:

[0124] e,e c ={NB,NM,NS,ZO,PS,PM,PB}

[0125] The elements represent: NB (negative large), NM (negative medium), NS (negative small), ZO (zero), PS (positive small), PM (positive medium), and PB (positive large). The universe of discourse for the fuzzy variables is uniformly set to [-6, 6] to achieve standardized processing of input and output variables. Fuzzy inference adopts the Mamdani method, which is based on the "if-then" rule base. It generates fuzzy output through maximum-minimum composition operation, and then obtains precise control quantities through defuzzification.

[0126] Based on expert knowledge and practical engineering experience, the design of control rules follows these principles:

[0127] 1. When the vertical velocity and acceleration of the vehicle have opposite signs, it indicates that the system has a tendency to cancel each other out. In this case, the control input should be reduced appropriately to avoid over-adjustment.

[0128] 2. When the vertical velocity and acceleration have the same sign, it indicates that the system state is showing a continuous increasing or decreasing trend. In this case, the control input should be appropriately increased to suppress its divergence trend. Since the centroid method can effectively smooth the system output and improve the accuracy of the defuzzification results, this method is used to process the fuzzy output. The mathematical expression of the centroid method is:

[0129]

[0130] In the formula, C i To output the i-th element in the fuzzy output universe; u(C i ) is C i The corresponding membership value; n is the total number of output elements; G * This is the precise output value after defuzzification.

[0131] Based on the dynamic characteristics and interactions of proportional, integral, and derivative components in a control system, fuzzy rule sets for PID parameter self-tuning were designed. Among them, the proportional gain increment ΔK... p Integral gain increment ΔK i and differential gain increment ΔK d The fuzzy control rules are shown in Table 1 ΔK p Fuzzy control rules, Table 2 ΔK i Fuzzy control rules and Table 3 ΔK d As shown in the fuzzy control rules.

[0132] Table 1

[0133]

[0134] Table 2

[0135]

[0136]

[0137] Table 3

[0138]

[0139] S32 establishes a force feedback optimal control model;

[0140] Figure 5 The diagram shown is a flowchart of the force feedback optimal control model of the present invention. The force control system adopts a dual-loop structure. The inner loop control adopts linear feedback control, and the outer loop control adopts LQG (Linear Quadratic Gaussian) control. The LQG optimal controller generates the target control force and transmits it as an input signal to the inner control loop. The inner loop performs feedback control and inputs the control signal into the vehicle model.

[0141] The inner loop control uses a linear feedback control model, specifically:

[0142]

[0143] In the formula, e1 = F d -F t For force error, k p F is the proportionality coefficient. t To control the force, F d For desired control.

[0144] The LQG control model used in the outer loop control is as follows:

[0145]

[0146] In the formula, q1 represents the suspension dynamic deflection weighting coefficient; q2 represents the wheel dynamic load weighting coefficient; q3 represents the sprung mass acceleration weighting coefficient; q4 represents the unsprung mass acceleration weighting coefficient; and r represents the active control force weighting coefficient.

[0147] The above formula can be written in matrix form:

[0148]

[0149] In the formula, Q is a positive semi-definite matrix. R is a positive definite matrix. ρ is the weighted matrix of state variables; ρ = diag[q1q2q3q4]; n is the weighted matrix of control variables.

[0150] The optimal control force is expressed as:

[0151] F b =-KX

[0152] In the formula,

[0153] The LQG control feedback gain matrix K is determined by solving the following Riccati differential equation:

[0154]

[0155] The feedback gain matrix K is calculated by calling the LQR function in MATLAB:

[0156] K = LQR(A) s B s ,Q,R,M)

[0157] S4. The position-force optimal control strategy is used to achieve optimal control of the vehicle active suspension system; the impedance control model in S2 outputs the desired acceleration of the impedance model based on the wheel dynamic load in S1, and obtains the tracking error of the vehicle active suspension system; the position state feedback model in S3 calculates the desired output control force of the vehicle actuator; the force feedback optimal control model in S3 outputs the output force of the vehicle actuator, thereby controlling the vehicle active suspension.

[0158] Figure 6 The figure shows the time-domain simulation results of the vehicle's vertical acceleration under the random road surface active suspension of the present invention. It can be seen that the proposed control method effectively reduces the vehicle's vertical acceleration. Compared with the passive suspension, under Class C road surface excitation, the root mean square (RMS) acceleration of the vehicle is reduced by 42.58%; under a 0.11m raised road surface excitation, the RMS acceleration of the vehicle is reduced by 51.91%. This verifies the design requirements for reducing vehicle vibration.

[0159] Figure 7 The figure shows the time-domain simulation results of the dynamic deflection of the active suspension system for random road surfaces according to the present invention. It can be seen that the proposed control method effectively reduces the vertical acceleration of the vehicle body. Compared with the passive suspension, under Class C road surface excitation, the root mean square (RMS) acceleration of the vehicle body is reduced by 58.14%; under a 0.11m raised road surface excitation, the RMS acceleration of the vehicle body is reduced by 58.74%. This verifies the design requirement of reducing wheel dynamic load.

[0160] Figure 8The figure shows the time-domain simulation results of the wheel dynamic load of the random road surface active suspension of the present invention. It can be seen that the proposed control method effectively reduces the vertical acceleration of the vehicle body. Compared with the passive suspension, the root mean square acceleration of the vehicle body is reduced by 31.34% under Class C road surface excitation; and by 51.1% under 0.11m raised road surface excitation. This verifies the design requirement that the wheels have good ground contact capability.

[0161] The second aspect of this invention proposes a control system for a position-force feedback optimal control method to improve the passability and wheel contact performance of active suspension vehicles, comprising: a road excitation module, an impedance control module, a position state tracking control module, a force feedback optimal control module, and a vehicle actuator and suspension system module.

[0162] The road excitation module can simulate the road unevenness excitation experienced by the vehicle during driving and transmit the road height information to the impedance control module.

[0163] The impedance control module uses the road height information input from the road excitation module and combines impedance control to track the wheel dynamic load to obtain the desired sprung mass displacement.

[0164] The position status tracking control module generates a position correction amount using impedance control based on the actual displacement of the vehicle's active suspension; then the fuzzy PID position controller adjusts the control force online in real time based on this position correction amount and transmits it to the force feedback optimal control module.

[0165] The force feedback optimal control module obtains the desired output force and the actual output force of the vehicle actuator, outputs the actual control quantity of the vehicle actuator, and drives the vehicle actuator to work.

[0166] The vehicle actuator and suspension system module outputs the actual control quantity of the actuator based on the force feedback optimal control module, which is then applied to the active suspension system to achieve real-time control of the vehicle's active suspension.

[0167] In summary, this invention monitors and provides feedback on the position and force information of the suspension system in real time through a position and force feedback controller. When performing position difference feedback control, impedance control is used to obtain the desired displacement of the sprung mass in real time, determine the desired control force of the actuator, and then use a force feedback optimal controller to achieve active suspension target force control, thereby further improving the vehicle's passability and wheel contact capability.

Claims

1. A position-force feedback optimal control method for improving the passability and wheel-ground performance of a main active suspension vehicle, characterized in that: The method comprises the following steps: S1, a two-degree-of-freedom 1 / 6 vehicle active suspension model and a valve-controlled actuator system model are established; S2, a position impedance control model is established and a real-time sprung mass desired displacement is obtained; S3, a position-force feedback optimal control model of the vehicle active suspension system is established, including a position state feedback control model of the vehicle active suspension system based on fuzzy control and PID control and a force feedback optimal control model based on double-loop control; S4, the optimal control of the vehicle active suspension system is realized by using the position-force optimal control strategy; the impedance control model in S2 outputs an impedance model desired acceleration according to the wheel dynamic load in S1, so as to obtain a vehicle active suspension system tracking error; the position state feedback model in S3 calculates a vehicle actuator desired output control force; the force feedback optimal control model in S3 outputs a vehicle actuator output force, so as to control the vehicle active suspension.

2. The position-force feedback optimal control method for improving the passability and the wheel-ground performance of an active suspension vehicle according to claim 1, characterized by: In S1, first, a two-degree-of-freedom 1 / 6 vehicle active suspension dynamics model is established according to Newton's second law: where m s is the sprung mass; m u is the unsprung mass; k s , c s are the suspension stiffness and damping, respectively; k t , c t are the tire equivalent stiffness and damping, respectively; z s , z u , z r are the sprung mass displacement, unsprung mass displacement, and road excitation displacement, respectively; F u is the actual control force input to the actuator; F d is the desired control force to the actuator; Based on the above model, the passive suspension lacks the actuator control force compared with the active suspension, and the state variables of the passive and active suspensions are: Let Take the output quantity as Then, the state space expression of the vehicle active suspension system is: The expression of the dynamic load of the wheel is: A vehicle actuator dynamics nonlinear model is established: When the piston rod extends, the flow continuity equation of the hydraulic cylinder is represented as: When the piston rod retracts, the flow continuity equations of the rodless cavity and the rod cavity of the hydraulic cylinder are represented as: where x p represents the displacement of the hydraulic cylinder piston rod; β e represents the effective volume of the oil elastic modulus; C a represents the leakage coefficient C b represents the leakage coefficient of the hydraulic cylinder; A1, A2 represents the effective area of the rodless cavity and the rod cavity; V a , V b represents the initial volume of the hydraulic cylinder rodless cavity and the rod cavity; q1 represents the oil flow and q2 the oil flow of the rod cavity.

3. The position-force feedback optimal control method for improving the passability and the wheel-ground performance of an active suspension vehicle according to claim 1, characterized by: In S2, the position impedance control model is: The transfer function model between the wheel dynamic load and the vehicle body vertical acceleration is: where m d represents the ideal target inertance; z sd represents the desired displacement of the sprung mass; k d represents the ideal target stiffness; c d represents the ideal target damping; F dl (s) and z sd (s) are the Laplace transforms of F dl and z sd respectively; s 2 z sd (s) is the Laplace transform of the sprung mass vertical acceleration.

4. The position-force feedback optimal control method for improving the passability and the wheel-ground performance of an active suspension vehicle according to claim 1, characterized by: In S3, the displacement state tracking control model uses fuzzy control and PID control methods to calculate the position correction amount z of the sprung mass from the wheel dynamic load sd while combining the measured sprung mass displacement z from the displacement sensor s These two parameters are used together as inputs to the position controller to obtain the desired control force F d F = k z + c v z In the formula, e(t) is the control deviation; F is the desired output F of the controlled object. d The difference between F(t) and the feedback value c(t); t is the time variable; F(t) is the controller output; K p The proportional gain coefficient, K, determines the system's response strength to the current deviation. i K is the integral gain coefficient used to eliminate steady-state error; d This is the differential gain coefficient used to suppress system overshoot and improve dynamic response.

5. The position-force feedback optimal control method for improving the mobility and wheel-ground performance of an active suspension vehicle according to claim 1, characterized by: In S3, the inner loop control adopts linear feedback control, and the outer loop control adopts linear quadratic Gaussian control; a linear quadratic Gaussian optimal controller generates a target control force as an input signal transmitted to the control inner loop; the inner loop executes feedback control and inputs the control signal into the vehicle model.

6. The position-force feedback optimal control method for improving the passability and the wheel-ground performance of an active suspension vehicle according to claim 5, characterized by: In S3, the linear feedback control model adopted by the inner loop control is: where e1=F d - F t is the force error, k p is the proportional coefficient, F t is the control force, F d is the desired control force.

7. The position-force feedback optimal control method for improving the passability and the wheel-ground performance of an active suspension vehicle according to claim 5, characterized by: In S3, the linear quadratic Gaussian model adopted by the outer loop control is: In the formula, q1 represents a suspension deflection weighting coefficient; q2 represents a wheel dynamic load weighting coefficient; q3 represents a sprung mass acceleration weighting coefficient; q4 represents a non-sprung mass acceleration weighting coefficient; r represents an active control force weighting coefficient; The above formula is written in matrix form as: wherein Q is a positive semi-definite matrix, R is a positive definite matrix, p is a state variable weighting matrix; p = diag [q1 q2 q3 q4]; n is a control variable weighting matrix; The optimal control force is represented as: F b = -KX In the formulae, The linear quadratic Gaussian control feedback gain matrix K is determined by solving the following Riccati differential equation: The feedback gain matrix K is calculated by calling the LQR function in MATLAB: K = LQR(A s , B s , Q, R, M).

8. A position-force feedback optimal control system for improving the vehicle passability and wheel-ground performance of an active suspension vehicle, characterized by: The position-force feedback optimal control method for improving the vehicle passability and wheel ground performance of the active suspension vehicle is used, and comprises: a road excitation module for simulating the road unevenness excitation received by the vehicle during driving and transmitting road height information to the impedance control module; an impedance control module for obtaining a desired sprung mass displacement by combining the impedance control with the wheel dynamic load tracking according to the road height information input by the road excitation module; a position state tracking control module for generating a position correction amount by using the impedance control according to the actual displacement of the vehicle active suspension; subsequently, a fuzzy PID position controller adjusts the control force in real time and on line according to the position correction amount, and transmits the control force to the force feedback optimal control module; The force feedback optimal control module acquires the expected output force of the vehicle actuator and the actual output force of the vehicle actuator, outputs the actual control quantity of the vehicle actuator, and drives the vehicle actuator to work. The vehicle actuator and the suspension system module controls the actual control quantity of the actuator output by the force feedback optimal control module, and acts on the active suspension system to realize real-time control of the active suspension of the vehicle.

Images