Robust control method for decoupled chassis facing road adhesion disturbance

Abstract

The application relates to a kind of decoupled chassis robust control methods for road adhesion disturbance, comprising: based on the dynamics characteristics of each wheel drive, brake and steering independently controllable decoupled chassis system, through demand force reachability analysis, the realization ability of different control action combinations to target force is analyzed, and the control action decomposition feasible region of target force in ideal state is constructed;Calculate the influence of different longitudinal and lateral control actions on target force and the control accuracy of vehicle motion state under road adhesion disturbance, and construct the influence model of adhesion coefficient disturbance on the whole vehicle state under disturbance condition;Based on the framework of multi-objective optimization problem, the robust control algorithm for decoupled chassis is designed, and the redundant tire force control mechanism of decoupled steering chassis is used to generate control actions that minimize the influence of adhesion disturbance on the whole vehicle state.The application can fully utilize the redundant characteristics of decoupled chassis control action, and significantly improve the control robustness of vehicle under complex road adhesion conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control, in particular to a decoupled chassis robust control method for road adhesion disturbance. BACKGROUND

[0002] As a key environmental factor of vehicle dynamics control, the uncertainty of road adhesion condition has a decisive influence on the stability of vehicle control. Modern vehicle control systems can cope with control requirements under known adhesion conditions to a certain extent through tire force modeling and state estimation technology. However, due to the complexity and variability of the actual road environment, including sudden changes in road surface material, dry-wet state conversion and other factors, the adhesion coefficient has significant time-varying uncertainty and spatial difference.

[0003] Traditional control strategies based on the assumption of deterministic adhesion coefficient are difficult to adapt to dynamic changes in road conditions, especially in extreme conditions. The rapid change of adhesion coefficient will cause the vehicle state to change dramatically during the control process, making the traditional control method prone to failure. Therefore, how to fully utilize the control freedom of the decoupled chassis system to design a forward-looking robust control strategy to actively suppress disturbance propagation and achieve optimal tire force distribution is a key technical bottleneck to ensure the safety and reliability of vehicles in complex road conditions.

[0004] Chinese patent application publication No. CN110901647A discloses a vehicle road adhesion coefficient adaptive estimation method considering complex excitation conditions. Through the combination of whole vehicle single wheel dynamics and two-degree-of-freedom kinematics model and fuzzy reasoning method, the accuracy and robustness of adhesion coefficient estimation under complex excitation conditions are improved. Its technical solution has certain improvement effect for realizing rapid and stable estimation of adhesion coefficient in complex environment. However, due to the time lag characteristics of vehicle dynamics state, combined with the complexity and variability of actual road conditions, the current method still has certain lag phenomenon in adhesion coefficient estimation, and the estimation accuracy cannot be completely guaranteed.

[0005] In summary, there is no robust chassis dynamics control method designed for road adhesion disturbance to solve or partially solve the above problems. SUMMARY

[0006] The purpose of the present application is to overcome the defects of the prior art and provide a decoupled chassis robust control method for road adhesion disturbance to solve or partially solve the problem that traditional control strategies based on the assumption of deterministic adhesion coefficient are difficult to adapt to dynamic changes in road conditions.

[0007] The purpose of the present application can be realized by the following technical solutions:

[0008] The application provides a robust control method for a decoupled chassis facing road adhesion disturbance, comprising the following steps:

[0009] Based on the dynamics characteristics of the decoupled chassis system with independent control of each wheel driving, braking and steering, the implementation ability of different control action combinations on the target force is analyzed through demand force reachability analysis, and a control action decomposition feasible region of the target force in an ideal state is constructed.

[0010] In the control action decomposition feasible region, the influence of different longitudinal and lateral control actions on the target force and the control accuracy of the vehicle motion state under road adhesion disturbance is calculated, and an influence model of adhesion coefficient disturbance on the whole vehicle state in a disturbed working condition is constructed.

[0011] According to the influence model, a robust control algorithm for the decoupled chassis is designed based on a multi-objective optimization problem framework, the redundant tire force control mechanism of the decoupled steering chassis is used to control each wheel to generate control actions that minimize the influence of adhesion disturbance on the whole vehicle state, and the robust control of the chassis system under road adhesion disturbance is realized.

[0012] As a preferred technical solution, the process of constructing the control action decomposition feasible region comprises the following steps:

[0013] A tire longitudinal and lateral force model under ideal road adhesion state for the decoupled chassis system is established to represent the output characteristics of the tire force in the ideal state.

[0014] Based on the decoupled chassis system, the redundant implementation characteristics of the target generalized force meeting the whole vehicle state control demand in the decoupled chassis system execution layer are analyzed, the tire longitudinal and lateral forces of each wheel are decomposed, and the different steering angles and slip rates meeting the whole vehicle longitudinal and lateral force demand are combined to form a tire force control multi-solution feasible region for the redundant chassis system by constructing the inverse function space of the tire longitudinal and lateral force model.

[0015] As a preferred technical solution, the tire longitudinal and lateral force model under ideal road adhesion state for the decoupled chassis system is:

[0016] ,

[0017] ,

[0018] wherein, 、 are the tire longitudinal force and lateral force, and are the corresponding tire longitudinal and lateral force models, is the tire vertical load, is the road adhesion coefficient, is the tire side slip angle, The tire slip ratio.

[0019] As a preferred technical solution, in the process of decomposing the tire longitudinal force of each wheel, the projection relationship of the tire force in the vehicle motion direction is:

[0020] ,

[0021] Among them, 、 respectively, the tire force along the longitudinal direction of the vehicle, the tire force along the lateral direction of the vehicle, 、 As the resultant force of the tire longitudinal force and the lateral force in the vehicle motion direction, is the wheel steering angle, is the vehicle mass center side slip angle, is the vehicle yaw rate, is the longitudinal vehicle speed, is the road adhesion coefficient, is the tire vertical load, is the tire slip ratio, 、 respectively, the defined single-wheel longitudinal and lateral force model in the vehicle coordinate system.

[0022] As a preferred technical solution, the process of constructing the influence model of the disturbed working condition of the adhesion coefficient disturbance on the vehicle state includes the following steps:

[0023] Quantify the sensitivity of different slip ratios and steering angle combinations to the tire force in the vehicle coordinate system through disturbance propagation in vehicle dynamics;

[0024] Construct a vehicle dynamics model including tire force, model the sensitivity of the vehicle state to the adhesion coefficient, and form an influence model of the adhesion coefficient disturbance on the vehicle state under the disturbed working condition.

[0025] As a preferred technical solution, the sensitivity of the tire force is calculated using the following formula:

[0026] ,

[0027] ,

[0028] Among them, 、 respectively, the sensitivity of the vehicle longitudinal and lateral force to the adhesion coefficient, 、 respectively, the tire longitudinal and lateral force model, is the wheel steering angle, is the vehicle mass center side slip angle, Let yaw rate be the vehicle's angular velocity. For longitudinal vehicle speed, The road surface adhesion coefficient, This refers to the tire slip ratio.

[0029] As a preferred technical solution, the vehicle dynamics model is modeled as follows:

[0030] ,

[0031] in, for longitudinal speed at any given moment The sideslip angle is the angle at the vehicle's center of gravity. Let yaw rate be the vehicle's angular velocity. , , These are the derivatives corresponding to the vehicle's state. , For the first Target tire force of each wheel in the longitudinal direction of vehicle movement and target tire force in the lateral direction of vehicle movement. These represent the front left, front right, rear left, and rear right wheels, respectively. The correlation coefficient for air resistance. For the overall vehicle quality, , These are the distances from the center of mass to the front and rear axles, respectively. , These are the track widths of the front and rear axles, Let represent the vehicle's moment of inertia.

[0032] As a preferred technical solution, the modeling of the sensitivity of the vehicle's state to the adhesion coefficient is as follows:

[0033] ,

[0034] ,

[0035] ,

[0036] in, , , They are respectively The longitudinal vehicle speed, center of gravity sideslip angle, and vehicle yaw rate at any given time affect the coefficient of adhesion. The degree of sensitivity, This represents the difference in tire force between the left and right sides of the front and rear axles along the longitudinal direction of the vehicle. They represent the front axle and the rear axle, respectively. These represent the front left, front right, rear left, and rear right wheels, respectively.

[0037] As a preferred technical solution, the process of controlling each wheel to generate minimal adhesion disturbance on the overall vehicle state includes:

[0038] A multi-objective optimization predictive controller is constructed, which includes state covariance and probability constraints. The control objectives include state tracking error, control input change, and disturbance sensitivity, wherein the disturbance sensitivity is calculated based on the influence model.

[0039] A probabilistic constraint model is established based on the vehicle state constraints and the degree of uncertainty. Based on the probabilistic constraint model and the multi-objective optimization predictive controller, each wheel is controlled to generate control actions that minimize the impact of adhesion disturbance on the vehicle state.

[0040] As a preferred technical solution, the optimization problem of the multi-objective optimization prediction controller is modeled as follows:

[0041] ,

[0042] in, and Prediction in the time domain and control in the time domain, respectively. Indicates the expectation. for The vehicle state in the vehicle dynamics model at any given time. The desired vehicle state to be tracked. This is the tracking weight matrix for the desired state. , and These are the weight matrices corresponding to longitudinal vehicle speed, sideslip angle, and yaw rate, respectively. This represents the trace operation. The covariance matrix characterizes the degree of uncertainty in the state. , These are the control quantity and the increment of the control quantity, respectively. For the adhesion coefficient The vehicle dynamics model These are state probability constraints. For state constraint boundaries, To allow for the probability of default,

[0043] The probabilistic constraint model is as follows:

[0044] ,

[0045] in, Here is the confidence coefficient. for The standard deviation of the state under the influence of the attached disturbance at any given time.

[0046] Compared with the prior art, the present application has at least one of the following beneficial effects:

[0047] (1) The present application calculates the sensitivity of the whole vehicle state to the adhesion coefficient based on disturbance sensitivity analysis, guides each wheel to produce a control action with the lowest sensitivity to the adhesion coefficient, and thus preferentially selects a control input combination with low sensitivity to adhesion fluctuations, while ensuring control accuracy and fully exploiting the tire force regulation potential of the decoupled chassis.

[0048] (2) The present application constructs a multi-objective optimization predictive controller including state covariance and probability constraints, the control objectives include state tracking error, control input variation and disturbance sensitivity, a probability constraint model is established based on whole vehicle state constraints and uncertainty level, which can adapt to stable control under various complex disturbance road adhesion conditions, and has excellent robustness and adaptability.

[0049] (3) The present application is based on a decoupled chassis system, through decoupled chassis dynamics control based on multi-objective optimization control, guiding the wheels to produce control actions with the lowest sensitivity to the adhesion coefficient, realizing robust control under road adhesion disturbance, through the characteristics of independent adjustment of each wheel angle and slip ratio of the decoupled chassis, combined with the tire force active distribution strategy, significantly reducing the disturbance of adhesion coefficient uncertainty to the whole vehicle state. Using a multi-objective optimization control framework, the state covariance constraint and probability constraint are explicitly considered, effectively suppressing the uncertainty propagation and avoiding the problem of control performance degradation caused by adhesion disturbance in traditional control methods.

[0050] (4) The propagation process of road adhesion disturbance is actively inhibited, and the vehicle's handling stability and safety are significantly improved. Its closed-loop optimization mechanism can adjust the control strategy in real time, and is perfectly compatible with the intelligent electric vehicle line control chassis system, providing a high-precision dynamics control solution for intelligent driving, and has important engineering application value. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 Flowchart of the decoupled chassis robust control method for road adhesion disturbance in the embodiment;

[0052] Figure 2 Schematic diagram of the effect of different control actions on the expected generalized force in the embodiment;

[0053] Figure 3 Schematic diagram of the longitudinal and lateral control action feasible region analyzed by taking the tire force in the whole vehicle longitudinal coordinate system as an example in the embodiment;

[0054] Figure 4 Schematic diagram of the adhesion coefficient sensitivity superimposed on the basis of Figure 3 . DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort should fall within the protection scope of the present application.

[0056] In view of the problems in the prior art, the embodiment provides a decoupled chassis robust control method for road adhesion disturbance, which is shown in Figure 1 The method comprises the following steps:

[0057] In step S1, based on the decoupled chassis system dynamics characteristics of independent control of wheel driving, braking and steering, the realization ability of different control action combinations to target force is analyzed through demand force reachability analysis, and a control action decomposition feasible region of target force in ideal state is constructed.

[0058] Specifically, step S1 can comprise steps S11-S12:

[0059] In step S11, a tire force model in an ideal road adhesion state for a decoupled chassis system is established to represent the output characteristics of tire force in an ideal state, specifically:

[0060] (1)

[0061] wherein, and are corresponding tire longitudinal and lateral force models, is a tire vertical load, is a road adhesion coefficient, is a tire slip angle, is a tire slip ratio.

[0062] In step S12, based on the decoupled chassis system, the redundant realization characteristics of a target generalized force meeting the vehicle state control demand in the decoupled chassis system execution layer in an ideal state are analyzed, the tire longitudinal and lateral forces of each wheel are decomposed, different steering angles and slip ratios meeting the vehicle longitudinal and lateral force demand are combined by constructing the inverse function space of the tire longitudinal and lateral force model, and a tire force control multi-solution feasible region for the redundant chassis system is formed.

[0063] Specifically, step S12 can comprise steps S121-S123:

[0064] In step S121, the multi-solution characteristics of the longitudinal and lateral forces of each wheel realizing the target generalized force in the decoupled chassis redundant configuration are analyzed.

[0065] When the vehicle control is performed, different tire longitudinal and lateral forces can be generated by each wheel to achieve the given target generalized force. Figure 2 When the vehicle control is performed, different tire longitudinal and lateral forces can be generated by each wheel to achieve the given target generalized force. When the vehicle control is performed, different tire longitudinal and lateral forces can be generated by each wheel to achieve the given target generalized force. When the vehicle control is performed, different tire longitudinal and lateral forces can be generated by each wheel to achieve the given target generalized force. When the vehicle control is performed, different tire longitudinal and lateral forces can be generated by each wheel to achieve the given target generalized force. When the vehicle control is performed, different tire longitudinal and lateral forces can be generated by each wheel to achieve the given target generalized force.

[0066] Step S122 determines the projection relationship of the tire force in the direction of vehicle motion.

[0067] For a certain wheel, the tire force in the direction of vehicle longitudinal motion and the tire force in the direction of vehicle lateral motion can be regarded as the projection of the resultant force generated by the tire longitudinal force and the lateral force in the direction of vehicle motion. Step S121 analyzes the decomposition form of the tire force of the whole vehicle, which is specifically:

[0068] (2)

[0069] Step S123 combines different steering angles and slip rates to meet the target vehicle longitudinal and lateral force requirements to form a tire force control multi-solution feasible region (i.e., control action decomposition feasible region) facing the redundant chassis system. Specifically:

[0070] By constructing the inverse function space of the tire force model, a multi-solution mapping interval can be constructed. Taking the tire force analysis in the longitudinal direction of the whole vehicle as an example, Figure 3 the change of the longitudinal force of the whole vehicle in the direction of the vehicle is given under the condition of given tire vertical load, adhesion coefficient, mass center side slip angle, yaw angular velocity and vehicle speed and other state parameters. It can be seen that when the vehicle control requirement is a certain target longitudinal force (blue line in the figure), there are different combinations of steering angle and slip rate, such as the intersection of the blue dashed line and the slip rate and steering angle change curve in the figure. Each group of points can theoretically output the expected force value.

[0071] Step S2 calculates the influence of different longitudinal and lateral control actions on the target force and the control accuracy of the vehicle motion state under road adhesion disturbance in the control action decomposition feasible region, and constructs a model of the influence of adhesion coefficient disturbance on the whole vehicle state under disturbance.

[0072] This step quantifies the sensitivity of the tire longitudinal and lateral force in the whole vehicle coordinate system to the road adhesion disturbance under the combined action of different slip rates and steering angles from the perspective of disturbance propagation in vehicle dynamics, and analyzes the sensitivity difference of different tire force distribution strategies to key vehicle states.

[0073] During driving, due to factors such as sensor signal quality and sudden changes in road conditions, the road adhesion coefficient is difficult to obtain accurately. The uncertainty of the adhesion coefficient will significantly affect the magnitude of the longitudinal and lateral forces of the tires, thus bringing potential instability factors to the overall vehicle motion.

[0074] Specifically, step S2 may include steps S21-S22:

[0075] Step S21: Through the propagation of disturbances in vehicle dynamics, quantify the sensitivity of different slip ratio and steering angle combinations to tire forces in the whole vehicle coordinate system.

[0076] Specifically, step S21 may include S211-S212:

[0077] Step S211: Calculate the sensitivity of longitudinal and lateral forces to the adhesion coefficient in the vehicle coordinate system.

[0078] Based on the longitudinal lateral force of the vehicle obtained from formula (2), the sensitivity of the longitudinal lateral force of the vehicle to the adhesion coefficient is calculated. and :

[0079] (3)

[0080] As can be seen from formula (3), under a given state, the wheel slip ratio and steering angle will jointly affect the sensitivity of the vehicle's directional tire force to road surface adhesion. Different combinations of control actions will show different responses to this adhesion disturbance.

[0081] Step S212: Analyze the variation law of the sensitivity of the adhesion coefficient to different combinations of slip ratio and rotation angle.

[0082] like Figure 3 As shown, the sensitivity The process of change superimposed on Figure 3 The runtime space can be found, such as Figure 4 As shown, under the condition that the expectation force is constant In this case, comparing the sensitivity of multiple feasible operating points, there exists a wheel longitudinal force that both satisfies the target force requirement and minimizes the disturbance sensitivity. ) and wheel lateral force ( The combination is a combination of corresponding wheel slip ratios and sideslip angle control actions. This combination can provide the required desired longitudinal force of the whole vehicle while minimizing the sensitivity of the state to the disturbance of the adhesion coefficient.

[0083] Step S22: Construct a vehicle dynamics model including tire forces, model the sensitivity of the vehicle state to the adhesion coefficient, and form a model of the impact of adhesion coefficient disturbance on the vehicle state under disturbed working conditions.

[0084] The key dynamic response of a vehicle depends largely on the longitudinal and lateral forces generated by the tires. Through dynamic coupling, these tire forces directly determine the key motion states of the vehicle. Tire force disturbances propagate gradually in the system and eventually affect the overall control performance of the vehicle, and can lead to performance degradation or even instability.

[0085] Specifically, step S22 can include S221-S222:

[0086] Step S221, a three-degree-of-freedom model of the vehicle containing tire forces is established.

[0087] The established vehicle dynamics model is:

[0088] (4)

[0089] wherein, and are the distances from the center of mass to the front and rear axles, is the mass of the vehicle, is the air resistance related coefficient, and are the front and rear wheel tracks, and are the projections of the longitudinal and lateral forces of each wheel in the longitudinal and lateral directions of the vehicle.

[0090] Under the assumption that the load and other vehicle parameters are known, a vehicle dynamics model related to the adhesion coefficient can be established:

[0091] (5)

[0092] Step S222, the sensitivity of the vehicle state to the adhesion coefficient is calculated, and an influence model of the adhesion coefficient disturbance on the vehicle state under the disturbed condition is formed.

[0093] According to the chain rule of differentiation, the sensitivity of the vehicle state to the adhesion coefficient is calculated respectively :

[0094]

[0095]

[0096] (6)

[0097] wherein, is the difference between the combined tire forces of the front and rear axles on the left and right sides of the vehicle in the longitudinal direction of the vehicle.

[0098] It can be seen that the tire force of a single wheel The disturbance amplitude of the disturbance is directly related to the longitudinal speed of the whole vehicle and the disturbance degree of the mass center side angle, the yaw rate is positively related to the disturbance degree of the single wheel lateral force, and is positively related to the difference between the left and right wheel longitudinal forces.

[0099] Step S3, according to the influence model, a robust control algorithm for the decoupled chassis is designed based on a multi-objective optimization problem framework, the redundant tire force control mechanism of the decoupled steering chassis is used to control each wheel to generate a control action that minimizes the influence of the road adhesion disturbance on the whole vehicle state, and the chassis system robust control under the road adhesion disturbance is realized.

[0100] This step designs a multi-objective optimization controller that integrates state covariance constraints and probability constraints, and can significantly suppress the propagation of state uncertainty by independently adjusting the steering angle and slip rate of each wheel, so that the controller can maintain robustness under various adhesion uncertainty conditions.

[0101] Adhesion coefficient uncertainty can significantly affect key state variables of the vehicle, and traditional control strategies can easily lead to control result deviation, tire force imbalance, and even instability due to the neglect of adhesion coefficient disturbance uncertainty propagation. Step S3 designs a decoupled steering dynamics control algorithm based on multi-objective optimization, which guides each wheel to generate a control action with the lowest sensitivity to adhesion coefficient, and realizes robust control under road adhesion disturbance.

[0102] Specifically, step S3 includes steps S31-S32:

[0103] Step S31, a multi-objective optimization predictive controller including state covariance and probability constraints is constructed, and the control objectives include state tracking error, control input variation and disturbance sensitivity, wherein the disturbance sensitivity is calculated according to the influence model.

[0104] An example of whole vehicle control for adhesion coefficient uncertainty conditions is designed for a drive-by-wire chassis with redundant actuator configuration. Step S31 selects the steering angles of the two rear wheels of the decoupled rear wheel steering chassis system and , and the rotational speeds of the four wheels as control variables , in the objective function, in addition to the expected state tracking and control action size and fluctuation suppression, the whole vehicle state uncertainty caused by road disturbance is suppressed according to formula (6), the disturbance degree of each tire force is optimized through control action, and then the whole vehicle state fluctuation is reduced, and the designed optimization problem is:

[0105] (7)

[0106] Wherein, and are the prediction time domain and the control time domain, respectively, are the corresponding weight matrices, is the desired vehicle state to be tracked, including the desired vehicle speed and the desired yaw rate is the covariance matrix representing the degree of uncertainty of the state, which is represented by the degree of sensitivity of the whole vehicle state to the adhesion coefficient established by step S2, and can be obtained by a statistical sampling-based method, is the state constraint boundary, is the allowable violation probability.

[0107] Step S32, based on the whole vehicle state constraint and the degree of uncertainty, a probability constraint model is established, and based on the probability constraint model and the multi-objective optimization predictive controller, the control action of each wheel is controlled to minimize the influence of adhesion disturbance on the whole vehicle state.

[0108] The probability constraint is derived based on the statistical characteristics of the predicted state under adhesion uncertainty. Unlike the hard constraint assuming deterministic prediction, these chance constraints take into account the variability caused by road condition disturbances. By converting the probability boundary to a convex approximation, the control problem is still easy to handle within the framework of multi-objective optimization problem.

[0109] The probability constraint designed for formula (7) is approximated by Chebyshev inequality:

[0110] (8)

[0111] wherein, is the confidence coefficient, so as to convert the probability constraint into a convex constraint condition.

[0112] Compared with the prior art, the present application significantly reduces the disturbance of adhesion coefficient uncertainty on the whole vehicle state by actively adjusting the tire force distribution strategy, and can significantly improve the control robustness of the vehicle under complex working conditions.

[0113] The above describes only the specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any skilled person in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.​​​​

Claims

1. A decoupled robust chassis control method for road surface adhesion disturbances, characterized in that, Includes the following steps: Based on the dynamic characteristics of a decoupled chassis system where each wheel drive, braking, and steering can be independently controlled, this paper analyzes the ability of different combinations of control actions to achieve the target force through demand force reachability analysis, and constructs the feasible domain of control action decomposition for the target force under ideal conditions. Within the feasible domain of the control action decomposition, the influence of different longitudinal and lateral control actions on the target force and the control accuracy of the vehicle motion state under road surface adhesion disturbance is calculated, and a model of the influence of adhesion coefficient disturbance on the overall vehicle state under disturbed conditions is constructed. Based on the aforementioned influence model, a robust control algorithm for decoupled chassis is designed using a multi-objective optimization problem framework. By utilizing the redundant tire force control mechanism of the decoupled steering chassis, the algorithm controls each wheel to generate control actions that minimize the impact of adhesion disturbances on the overall vehicle state, thereby achieving robust control of the chassis system under road adhesion disturbances.

2. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 1, characterized in that, The process of constructing the feasible region of control action decomposition includes the following steps: Establish a tire longitudinal and lateral force model for a decoupled chassis system under ideal road adhesion conditions to characterize the output characteristics of tire force under ideal conditions; Based on the decoupled chassis system, this paper analyzes the redundancy realization characteristics of the target generalized force that meets the requirements of vehicle state control in the execution layer of the decoupled chassis system under ideal conditions. The longitudinal and lateral forces of each wheel are decomposed. By constructing the inverse function space of the tire longitudinal and lateral force model, different steering angles and slip ratios that meet the requirements of the vehicle longitudinal and lateral forces are combined to form a multiple feasible domain for tire force control of redundant chassis systems.

3. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 2, characterized in that, The tire longitudinal lateral force model for the ideal road adhesion state of the decoupled chassis system is as follows: ， ， in, , These are the longitudinal force and lateral force of the tire, respectively. and These are the corresponding tire longitudinal and lateral force models. For the vertical load of the tire, The road surface adhesion coefficient, This refers to the tire slip angle. This refers to the tire slip ratio.

4. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 2, characterized in that, In the process of decomposing the longitudinal and lateral forces of each wheel's tires, the projection relationship of the tire forces onto the vehicle's direction of motion is as follows: ， in, , These are the tire forces along the longitudinal direction of the vehicle's movement and the tire forces along the lateral direction of the vehicle's movement, respectively. , As a result of the longitudinal force of the tire and lateral force The projection of the resulting force onto the direction of vehicle motion. For the turning angle of the wheel, The sideslip angle is the angle at the vehicle's center of gravity. Let yaw rate be the vehicle's angular velocity. For longitudinal vehicle speed, The road surface adhesion coefficient, For the vertical load of the tire, For tire slip ratio, , These are the defined longitudinal and lateral force models of a single wheel in the vehicle coordinate system.

5. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 1, characterized in that, The process of constructing a model of the impact of adhesion coefficient disturbance on the vehicle state under disturbed operating conditions includes the following steps: By proliferating disturbances in vehicle dynamics, the sensitivity of different slip ratio and steering angle combinations to tire forces in the whole vehicle coordinate system is quantified. A vehicle dynamics model incorporating tire forces is constructed to model the sensitivity of the vehicle's state to the coefficient of adhesion, thus forming a model of the impact of adhesion coefficient disturbance on the vehicle's state under disturbed conditions.

6. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 5, characterized in that, The sensitivity of the tire force is calculated using the following formula: ， ， in, , These represent the sensitivity of the vehicle's longitudinal and lateral forces to the coefficient of adhesion, respectively. , These are the longitudinal and lateral force models of the tire, respectively. For the turning angle of the wheel, The sideslip angle is the angle at the vehicle's center of gravity. Let yaw rate be the vehicle's angular velocity. For longitudinal vehicle speed, The road surface adhesion coefficient, This refers to the tire slip ratio.

7. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 5, characterized in that, The vehicle dynamics model is modeled as follows: ， in, for longitudinal speed at any given moment The sideslip angle is the angle at the vehicle's center of gravity. Let yaw rate be the vehicle's angular velocity. , , These are the derivatives of the corresponding vehicle states. , For the first Target tire force of each wheel in the longitudinal direction of vehicle movement and target tire force in the lateral direction of vehicle movement. These represent the front left, front right, rear left, and rear right wheels, respectively. The correlation coefficient for air resistance. For the overall vehicle quality, , These are the distances from the center of mass to the front and rear axles, respectively. , These are the track widths of the front and rear axles, Let represent the vehicle's moment of inertia.

8. A decoupled chassis robust control method for road surface adhesion disturbances according to claim 5, characterized in that, The modeling of the sensitivity of the vehicle's overall condition to the adhesion coefficient is as follows: ， ， ， in, , , They are respectively The longitudinal vehicle speed, center of gravity sideslip angle, and vehicle yaw rate at any given time affect the coefficient of adhesion. The degree of sensitivity, This represents the difference in tire force between the left and right sides of the front and rear axles along the longitudinal direction of the vehicle. They represent the front axle and the rear axle, respectively. These represent the front left, front right, rear left, and rear right wheels, respectively.

9. The decoupled chassis robust control method for road surface adhesion disturbance as described in claim 1, characterized in that, The process of controlling each wheel to generate minimal adhesion disturbance on the overall vehicle state includes: A multi-objective optimization predictive controller is constructed, which includes state covariance and probability constraints. The control objectives include state tracking error, control input change, and disturbance sensitivity, wherein the disturbance sensitivity is calculated based on the influence model. A probabilistic constraint model is established based on the vehicle state constraints and the degree of uncertainty. Based on the probabilistic constraint model and the multi-objective optimization predictive controller, each wheel is controlled to generate control actions that minimize the impact of adhesion disturbance on the vehicle state.

10. A decoupled chassis robust control method for road surface adhesion disturbances according to claim 9, characterized in that, The optimization problem of the multi-objective predictive controller is modeled as follows: ， in, and Prediction in the time domain and control in the time domain, respectively. Indicates the expectation. for The vehicle state in the vehicle dynamics model at any given time. The desired vehicle state to be tracked. This is the tracking weight matrix for the desired state. , and These are the weight matrices corresponding to longitudinal vehicle speed, sideslip angle, and yaw rate, respectively. This represents the trace operation. The covariance matrix characterizes the degree of uncertainty in the state. , These are the control quantity and the increment of the control quantity, respectively. For the adhesion coefficient The vehicle dynamics model These are state probability constraints. For state constraint boundaries, To allow for the probability of default, The probabilistic constraint model is as follows: ， in, Here is the confidence coefficient. for The standard deviation of the state under the influence of the attached disturbance at any given time.

Images