Heavy-duty carrying equipment chassis variable weight cooperative control method and system

Abstract

The application discloses a heavy carrying equipment chassis variable weight cooperative control method and system, and belongs to the technical field of vehicle dynamics control. The method comprises the following steps: acquiring state parameters of a vehicle, and determining a state error at a current time according to the state parameters; inputting the state error at the current time and a control increment to be solved into a state error equation constructed in advance to obtain a predicted state error; determining a current center of mass side slip angle according to the state parameters, and identifying the stability of the vehicle based on the current center of mass side slip angle and a center of mass side slip angle plane constructed in advance; updating each intelligent agent weight coefficient in a global cost function constructed in advance according to the stability identification result, solving the control increment to obtain an optimal control increment, with the minimum updated global cost function as the target; and the application can balance the path tracking accuracy and the lateral stability of the vehicle according to real-time states.

Description

Technical Field

[0001] This invention relates to a variable weight cooperative control method and system for heavy transport equipment chassis, belonging to the field of vehicle dynamics control technology. Background Technology

[0002] With the rapid development of heavy-duty transportation equipment towards larger size, higher mobility, and greater intelligence, its chassis systems have gradually evolved from traditional single-axle or dual-axle steering structures to complex systems combining multi-axle independent steering and distributed drive. Multi-axle steering chassis have significant advantages in improving vehicle mobility, load-bearing capacity, and handling stability, and have been widely used in engineering transportation. However, multi-axle systems exhibit obvious dynamic coupling and nonlinear characteristics, making the coordinated control of various steering actuators and drive units extremely difficult. Especially under complex operating conditions, the system is prone to problems such as control conflicts, increased trajectory deviations, and attitude instability.

[0003] Existing control methods mostly employ a centralized control architecture, using a single controller to achieve overall vehicle trajectory tracking and stability control. While these methods possess overall optimization capabilities, they suffer from high computational costs, poor real-time performance, and strong dependence on model accuracy and parameter matching, making it difficult to meet the rapid coordination requirements of multi-control unit systems. Furthermore, centralized control neglects the coupling constraints and collaborative relationships between subsystems, resulting in limited global performance. To address these issues, some studies have introduced hierarchical or distributed control strategies to reduce computational complexity, but these still lack adaptability in multi-axis coordination mechanisms and weight allocation strategies, making it difficult to balance trajectory accuracy and vehicle stability under different operating conditions.

[0004] Furthermore, traditional path tracking models generally assume a constant longitudinal vehicle speed to simplify controller design. However, in actual driving, vehicle speed dynamically changes due to factors such as drive, gradient, and air resistance; simultaneously, tire lateral stiffness fluctuates with the road adhesion coefficient. The nonlinearity of longitudinal vehicle speed and lateral stiffness causes the control model parameters to deviate from reality, thus affecting system robustness and control accuracy. Existing research has not yet been able to simultaneously consider the coupled impact of these two types of uncertainties on the dynamic cooperative performance of the chassis.

[0005] Therefore, in order to solve the above-mentioned technical problems, there is an urgent need for a variable weight collaborative control method and system for heavy transport equipment chassis. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a variable weight cooperative control method and system for heavy-duty transport equipment chassis. The method treats the front axle, middle axle, and rear axle steering systems and the direct yaw moment control system as four cooperative intelligent agents. A cooperative game mechanism is introduced to achieve coordination and optimization among the multiple intelligent agents, thereby achieving optimal overall vehicle performance. Based on phase plane analysis of the vehicle's stability boundary, an adaptive adjustment strategy for the game weights is designed, enabling the system to achieve a balance between path tracking accuracy and vehicle lateral stability according to the real-time status.

[0007] To achieve the above objectives, the present invention is implemented using the following technical solution:

[0008] In a first aspect, the present invention provides a variable weight cooperative control method for a heavy-duty transport vehicle chassis, comprising:

[0009] The vehicle's state parameters are acquired, and the state error at the current moment is determined based on the state parameters. The state parameters include yaw rate, longitudinal velocity, and steering angles and tire lateral stiffness of the front axle, center axle, and rear axle.

[0010] The current state error and the control increment to be solved are input into the pre-constructed interval type II fuzzy state error equation to obtain the predicted state error corresponding to the control increment. The interval type II fuzzy state error equation is constructed based on the premise variables, which are constructed according to the longitudinal velocity and the tire lateral stiffness of each axle.

[0011] The current centroid sideslip angle is determined based on the state parameters. The stability of the vehicle is then identified based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane to obtain the stability identification result.

[0012] The weight coefficients of each agent in the pre-constructed global cost function are updated based on the predicted state error and stability identification results. The control increment is solved with the goal of minimizing the updated global cost function to obtain the optimal control increment. The global cost function includes an error penalty term constructed based on the predicted state error.

[0013] The control increment is a joint decision vector composed of the control input increments of four cooperative agents: the front axle steering angle, the center axle steering angle, the rear axle steering angle, and the direct yaw moment. Each agent works together to solve the joint decision vector through cooperative game theory to obtain its own corresponding optimal control input increment.

[0014] Furthermore, the method for constructing the interval type II fuzzy state error equation includes:

[0015] Using longitudinal vehicle speed and tire lateral stiffness as premise variables, an interval type II fuzzy rule is established, and a type II fuzzy model is constructed through the interval type II fuzzy rule.

[0016] Discretize the type II fuzzy model to obtain a discrete-time state-space model;

[0017] Based on the discrete-time state-space model, a state error is defined, and an interval type II fuzzy state error equation is established to describe how the state error evolves with the discrete time step.

[0018] The interval type II fuzzy state error equation is expressed by the following equation:

[0019] ;

[0020] In the formula, and They represent the first The and the first State error at discrete moments Indicates the first Output error at each discrete time point For state error, , For state parameters, superscript Represents the transpose of a matrix. and For the vehicle's lateral velocity and yaw rate, This refers to the lateral displacement of the vehicle. For the horizontal swing angle, For reference only. , Indicates the reference lateral velocity. Indicates the reference yaw rate. Indicates the reference horizontal position. Indicates the reference heading angle. Indicates the first The discrete time... Fuzzy weights under fuzzy rules For the first The discrete state matrix of a fuzzy rule. For the first The first fuzzy rule Discrete input matrices of each agent For discrete output matrices, , , , It is the identity matrix. The dimension is The identity matrix, For discrete time steps, For the first The state matrix of the fuzzy rules For the first The first fuzzy rule The input matrix of each agent For the agent's serial number. For fuzzy rule numbers, , , and They represent the first The front axle angle control deviation, center axle angle control deviation, and rear axle angle control deviation at discrete moments; Indicates the first Direct yaw moment control deviation at discrete moments;

[0021] The discrete-time state-space model is represented by the following equation:

[0022] ;

[0023] In the formula, For the first A state vector at discrete moments. For the first A state vector at discrete moments. For the first The output vector at each discrete time point , , They are the first The front axle steering angle, center axle steering angle, and rear axle steering angle at discrete moments For the first The direct yaw moment at discrete moments.

[0024] Furthermore, the global cost function is expressed by the following equation:

[0025] ;

[0026] In the formula, For the first The objective function of each agent. In order to be in Predicting the future The output error vector at time t; In order to be in Predicting the future The control input increment of the i-th agent at time t; In order to be in Predicting the future The state error vector at time t. To predict the length of the time domain, Indicates the index of the cost item within the prediction time domain. , This is the terminal weight matrix. , Here is the state transition matrix. To control the input matrix, , These are the input matrices for the front axle angle, center axle angle, rear axle angle, and direct yaw moment, respectively. This represents the block diagonal operator. , , , , , For the longitudinal speed of the vehicle, and These represent the total mass of the vehicle and the moment of inertia of the vehicle's center of mass about the z-axis, respectively. , , These are the vertical distances from the front axle, middle axle, and rear axle to the vehicle's center of gravity, respectively. , , These are the lateral stiffness of the tires on the front, middle, and rear axles, respectively. and These are the positive definite weight matrices for the system state error and the control input, respectively. , , For the first The weight vector of each agent, Let be the weight matrix for the control input of the i-th agent. To control the input increment weight matrix, , For the first The equivalent output error weight matrix of each agent , For the first Control objective selection matrix for each agent Let be the weight matrix of the system state of the i-th agent. , , Let be the yaw angle error weight for the i-th agent. The weight of the yaw rate error of the i-th agent; The positive definite weight matrix of the system state error Weighted Euclidean distance For terminal weight matrix Weighted Euclidean distance; The positive definite weight matrix is ​​the result of control input. Weighted Euclidean distance.

[0027] Furthermore, the optimization problem of the global cost function is expressed by the following equation:

[0028] ;

[0029] in, For the first The minimum value of the objective function of each agent. For the first The sequence of prediction output errors at discrete time points; This is the state error prediction matrix; For the first The control input increment prediction matrix for each agent; For the first The discrete time... The sequence of control input increments for an agent in the prediction time domain. For the first Lower limit of control input increment for an agent For the first Upper limit of control input increment for each agent;

[0030] , , , ; For the first The discrete state transition matrix of the step. For the first The agent in the th... The discrete control input matrix of the step.

[0031] Furthermore, based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane, the vehicle's stability is identified, yielding stability identification results, including:

[0032] The position of the vehicle's current state point in the pre-constructed phase plane of the center of gravity sideslip angle is determined based on the current center of gravity sideslip angle and yaw rate, and the minimum vertical distance from the vehicle's current state point to the stability boundary is determined.

[0033] The stability weighting coefficient is determined based on the minimum vertical distance and the preset reference distance;

[0034] Based on the stability weight coefficients, the weights of the front, middle and rear axle steering agents and the weights of the direct yaw moment agent are determined, and each weight is used as the stability identification result.

[0035] The stability weighting coefficient is calculated by the following formula:

[0036] ;

[0037] in, For the first Stability weighting coefficients at discrete time points It is an exponential function. To adjust the sensitivity coefficient, For the first The minimum vertical distance from the vehicle's current state point to the stability boundary at each discrete time step. For the first The reference distance from the equilibrium point to the stability boundary at each discrete time step;

[0038] The weights of the front, center, and rear axle steering agents, as well as the weights of the direct yaw moment agent, are expressed by the following formula:

[0039] ;

[0040] in, , , The first The weights of the agent's front, middle, and rear axis steering at discrete moments. For the first The weights of the direct yaw moment agent at discrete moments. , , To smooth the constraints, For the first At the discrete time... The weight vector of each agent, , For the first At the discrete time... The weight vector of each agent.

[0041] Furthermore, the control increment is solved with the objective of minimizing the updated global cost function to obtain the optimal control increment, including:

[0042] The global cost function containing terminal weighting terms is transformed by an equivalent transformation to obtain the transformed global cost function.

[0043] The optimization problem is transformed based on the transformed global cost function to obtain the transformed optimization problem.

[0044] Based on the transformed optimization problem, a least squares formula is constructed, and the least squares formula is solved using the QR method to obtain the optimal control increment sequence. The first term of the optimal control increment sequence is taken as the optimal control increment.

[0045] The transformed global cost function is expressed by the following equation:

[0046] ;

[0047] in, This represents the transformed global cost function. For including the terminal weight matrix The prediction error weight matrix, ; For the first The control input incremental weight matrix of each agent. ; For the first The augmented prediction error vector at each discrete time point. ; For the first At the discrete time... The square of the peak value of the control input increment sequence of an agent in the prediction time domain; For the terminal weight matrix Prediction error weight matrix Weighted Euclidean distance; For the and The Euclidean distance weighted by the ratio;

[0048] The transformed optimization problem is expressed by the following equation:

[0049] ;

[0050] in, For the transformed optimization problem, Indicates the first Each agent solves for the equivalent prediction error term when the control input increment is calculated. , An augmented state error prediction matrix, used to characterize the current state error. Impact on augmented forecast error; Represents a set of four intelligent agents. , This indicates that there is no intelligent agent. set ; For the first discrete time The set of The sequence of control input increments for an agent in the prediction time domain; for The set of The augmented control input prediction matrix of each agent is used to characterize Impact on augmented forecast error;

[0051] The least squares expression is represented by the following formula:

[0052] ;

[0053] in, , , and Weight matrices and matrix factorization factor Euclidean distance;

[0054] The optimal control increment sequence is represented by the following formula:

[0055] ;

[0056] in, For the first At the discrete time... The optimal control increment sequence for each agent. For the first At the discrete time... The peak value of the control input increment sequence of an agent in the prediction time domain. For the first Augmented control input prediction matrix for each agent For the first The augmented least squares coefficient matrix of each agent is composed of the prediction error coefficient matrix and the control input increment coefficient matrix. ;

[0057] The optimal control increment is expressed by the following formula:

[0058] ;

[0059] in, For optimal control increment, It is the optimal solution matrix for controlling the input.

[0060] Secondly, the present invention provides a variable weight cooperative control system for a heavy-duty transport equipment chassis, comprising:

[0061] The first module is used to acquire the vehicle's state parameters and determine the state error at the current moment based on the state parameters. The state parameters include yaw rate, longitudinal speed, and steering angles and tire lateral stiffness of the front axle, center axle, and rear axle.

[0062] The second module is used to input the current state error and the control increment to be solved into the pre-constructed interval type II fuzzy state error equation to obtain the predicted state error corresponding to the control increment. The interval type II fuzzy state error equation is constructed based on the premise variables, which are constructed according to the longitudinal speed and the tire lateral stiffness of each axle.

[0063] The third module is used to determine the current centroid sideslip angle based on the state parameters, and to identify the stability of the vehicle based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane to obtain the stability identification result.

[0064] The fourth module is used to update the weight coefficients of each agent in the pre-constructed global cost function based on the predicted state error and stability identification results, and to solve the control increment with the goal of minimizing the updated global cost function to obtain the optimal control increment; the global cost function includes an error penalty term constructed based on the predicted state error;

[0065] The control increment is a joint decision vector composed of the control input increments of four cooperative agents: the front axle steering angle, the center axle steering angle, the rear axle steering angle, and the direct yaw moment. Each agent works together to solve the joint decision vector through cooperative game theory to obtain its own corresponding optimal control input increment.

[0066] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0067] Fourthly, the present invention provides a computer device, comprising:

[0068] Memory, used to store computer programs / instructions;

[0069] A processor for executing the computer program / instructions to implement the steps of any of the methods described above.

[0070] Fifthly, the present invention provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0071] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0072] This invention provides a variable weight collaborative control method and system for heavy-duty transport equipment chassis. By establishing a trajectory tracking model through interval type II fuzzy modeling, it can effectively characterize the model uncertainty caused by changes in longitudinal vehicle speed and tire lateral stiffness, improve the system's adaptability to parameter disturbances, and ensure the stability and accuracy of the control process.

[0073] This invention provides a variable weight collaborative control method and system for heavy-duty transport equipment chassis. It regards the front, middle and rear axle steering systems and the direct yaw moment control system as four cooperative intelligent agents, and introduces a cooperative game mechanism to build a coordination and optimization framework among the intelligent agents. It can dynamically allocate control weights among multiple axes and realize the collaborative control of multiple execution units.

[0074] This invention provides a variable weight cooperative control method and system for heavy transport equipment chassis. By using phase plane analysis to adjust the game weight coefficients in real time, the control system can adaptively adjust the control strategy according to the vehicle's dynamic state and stability boundary, thereby simultaneously taking into account trajectory tracking accuracy and lateral attitude stability in complex road environments. Attached Figure Description

[0075] Figure 1 This is a flowchart illustrating a variable weight collaborative control method for a heavy transport vehicle chassis provided in an embodiment of the present invention.

[0076] Figure 2 This is a schematic diagram of the β–ω phase plane result in a variable weight cooperative control method for a heavy transport vehicle chassis provided in an embodiment of the present invention; Detailed Implementation

[0077] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0078] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0079] Example 1:

[0080] Figure 1 This is a flowchart of a variable weight cooperative control method for a heavy-duty transport vehicle chassis according to Embodiment 1 of the present invention. The variable weight cooperative control method for a heavy-duty transport vehicle chassis provided in this embodiment can be applied to a terminal and can be executed by a variable weight cooperative control system for the heavy-duty transport vehicle chassis. This system can be implemented by software and / or hardware and can be integrated into the terminal, such as any smartphone, tablet computer, or computer device with communication capabilities. See also... Figure 1 The method implemented in this way specifically includes the following steps:

[0081] The vehicle's state parameters are acquired, and the state error at the current moment is determined based on the state parameters. The state parameters include yaw rate, longitudinal velocity, and steering angles and tire lateral stiffness of the front axle, center axle, and rear axle.

[0082] The current state error and the control increment to be solved are input into the pre-constructed interval type II fuzzy state error equation to obtain the predicted state error corresponding to the control increment. The interval type II fuzzy state error equation is constructed based on the premise variables, which are constructed according to the longitudinal velocity and the tire lateral stiffness of each axle.

[0083] The current centroid sideslip angle is determined based on the state parameters. The stability of the vehicle is then identified based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane to obtain the stability identification result.

[0084] The weight coefficients of each agent in the pre-constructed global cost function are updated based on the predicted state error and stability identification results. The control increment is solved with the goal of minimizing the updated global cost function to obtain the optimal control increment. The global cost function includes an error penalty term constructed based on the predicted state error.

[0085] The control increment is a joint decision vector composed of the control input increments of four cooperative agents: the front axle steering angle, the center axle steering angle, the rear axle steering angle, and the direct yaw moment. Each agent works together to solve the joint decision vector through cooperative game theory to obtain its own corresponding optimal control input increment.

[0086] Furthermore, the method for constructing the interval type II fuzzy state error equation includes:

[0087] Using longitudinal vehicle speed and tire lateral stiffness as premise variables, an interval type II fuzzy rule is established, and a type II fuzzy model is constructed through the interval type II fuzzy rule.

[0088] Discretize the type II fuzzy model to obtain a discrete-time state-space model;

[0089] Based on the discrete-time state-space model, a state error is defined, and an interval type II fuzzy state error equation is established to describe how the state error evolves with the discrete time step.

[0090] The interval type II fuzzy state error equation is expressed by the following equation:

[0091] ;

[0092] In the formula, and They represent the first The and the first State error at discrete moments Indicates the first Output error at each discrete time point For state error, , For state parameters, superscript Represents the transpose of a matrix. and For the vehicle's lateral velocity and yaw rate, This refers to the lateral displacement of the vehicle. For the horizontal swing angle, For reference only. , Indicates the reference lateral velocity. Indicates the reference yaw rate. Indicates the reference horizontal position. Indicates the reference heading angle. Indicates the first The discrete time... Fuzzy weights under fuzzy rules For the first The discrete state matrix of a fuzzy rule. For the first The first fuzzy rule Discrete input matrices of each agent For discrete output matrices, , , , It is the identity matrix. The dimension is The identity matrix, For discrete time steps, For the first The state matrix of the fuzzy rules For the first The first fuzzy rule The input matrix of each agent For the agent's serial number. For fuzzy rule numbers, , , and They represent the first The front axle angle control deviation, center axle angle control deviation, and rear axle angle control deviation at discrete moments; Indicates the first Direct yaw moment control deviation at discrete moments;

[0093] The discrete-time state-space model is represented by the following equation:

[0094] ;

[0095] In the formula, For the first A state vector at discrete moments. For the first A state vector at discrete moments. For the first The output vector at each discrete time point , , They are the first The front axle steering angle, center axle steering angle, and rear axle steering angle at discrete moments For the first The direct yaw moment at discrete moments.

[0096] Furthermore, the global cost function is expressed by the following equation:

[0097] ;

[0098] In the formula, For the first The objective function of each agent. In order to be in Predicting the future The output error vector at time t; In order to be in Predicting the future The control input increment of the i-th agent at time t; In order to be in Predicting the future The state error vector at time t. To predict the length of the time domain, Indicates the index of the cost item within the prediction time domain. , This is the terminal weight matrix. , Here is the state transition matrix. To control the input matrix, , These are the input matrices for the front axle angle, center axle angle, rear axle angle, and direct yaw moment, respectively. This represents the block diagonal operator. , , , , , For the longitudinal speed of the vehicle, and These represent the total mass of the vehicle and the moment of inertia of the vehicle's center of mass about the z-axis, respectively. , , These are the vertical distances from the front axle, middle axle, and rear axle to the vehicle's center of gravity, respectively. , , These are the lateral stiffness of the tires on the front, middle, and rear axles, respectively. and These are the positive definite weight matrices for the system state error and the control input, respectively. , , For the first The weight vector of each agent, Let be the weight matrix for the control input of the i-th agent. The input increment weight matrix is ​​used to control the relative weights of the front axle steering angle increment, center axle steering angle increment, rear axle steering angle increment, and direct yaw moment increment in the final cost. , For the first The equivalent output error weight matrix of each agent , For the first Control objective selection matrix for each agent Let be the weight matrix of the system state of the i-th agent. , , Let be the yaw angle error weight for the i-th agent. The weight of the yaw rate error of the i-th agent; The positive definite weight matrix of the system state error Weighted Euclidean distance For terminal weight matrix Weighted Euclidean distance; The positive definite weight matrix is ​​the result of control input. Weighted Euclidean distance.

[0099] Furthermore, the optimization problem of the global cost function is expressed by the following equation:

[0100] ;

[0101] in, For the first The minimum value of the objective function of each agent. For the first The sequence of prediction output errors at discrete time points; This is the state error prediction matrix; For the first The control input increment prediction matrix for each agent; For the first The discrete time... The sequence of control input increments for an agent in the prediction time domain. For the first Lower limit of control input increment for an agent , For the first Lower limit of single-step control input increment for an agent. For the first Upper limit of control input increment for an agent. , For the first Lower limit of single-step control input increment for an agent. , Indicates the first The peak value of each agent;

[0102] , , , ; For the first The discrete state transition matrix of the step. For the first The agent in the th... The discrete control input matrix of the step.

[0103] Furthermore, based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane, the vehicle's stability is identified, yielding stability identification results, including:

[0104] The position of the vehicle's current state point in the pre-constructed phase plane of the center of gravity sideslip angle is determined based on the current center of gravity sideslip angle and yaw rate, and the minimum vertical distance from the vehicle's current state point to the stability boundary is determined.

[0105] The stability weighting coefficient is determined based on the minimum vertical distance and the preset reference distance;

[0106] Based on the stability weight coefficients, the weights of the front, middle and rear axle steering agents and the weights of the direct yaw moment agent are determined, and each weight is used as the stability identification result.

[0107] The stability weighting coefficient is calculated by the following formula:

[0108] ;

[0109] in, For the first Stability weighting coefficients at discrete time points It is an exponential function. To adjust the sensitivity coefficient, For the first The minimum vertical distance from the vehicle's current state point to the stability boundary at each discrete time step. For the first The reference distance from the equilibrium point to the stability boundary at each discrete time step;

[0110] The weights of the front, center, and rear axle steering agents, as well as the weights of the direct yaw moment agent, are expressed by the following formula:

[0111] ;

[0112] in, , , The first The weights of the agent's front, middle, and rear axis steering at discrete moments. For the first The weights of the direct yaw moment agent at discrete moments. , , To smooth the constraints, For the first At the discrete time... The weight vector of each agent, , For the first At the discrete time... The weight vector of each agent.

[0113] Furthermore, the control increment is solved with the objective of minimizing the updated global cost function to obtain the optimal control increment, including:

[0114] The global cost function containing terminal weighting terms is transformed by an equivalent transformation to obtain the transformed global cost function.

[0115] The optimization problem is transformed based on the transformed global cost function to obtain the transformed optimization problem.

[0116] Based on the transformed optimization problem, a least squares formula is constructed, and the least squares formula is solved using the QR method to obtain the optimal control increment sequence. The first term of the optimal control increment sequence is taken as the optimal control increment.

[0117] The transformed global cost function is expressed by the following equation:

[0118] ;

[0119] in, This represents the transformed global cost function. For including the terminal weight matrix The prediction error weight matrix, ; For the first The control input incremental weight matrix of each agent. ; For the first The augmented prediction error vector at each discrete time point. ; For the first At the discrete time... The square of the peak value of the control input increment sequence of an agent in the prediction time domain; For the terminal weight matrix Prediction error weight matrix Weighted Euclidean distance; For the and The Euclidean distance weighted by the ratio;

[0120] The transformed optimization problem is expressed by the following equation:

[0121] ;

[0122] in, For the transformed optimization problem, Indicates the first Each agent solves for the equivalent prediction error term when the control input increment is calculated. , An augmented state error prediction matrix, used to characterize the current state error. Impact on augmented forecast error; Represents a set of four intelligent agents. , This indicates that there is no intelligent agent. set ; For the first discrete time The set of The sequence of control input increments for an agent in the prediction time domain; for The set of The augmented control input prediction matrix of each agent is used to characterize Impact on augmented forecast error;

[0123] The least squares expression is represented by the following formula:

[0124] ;

[0125] in, , , and Weight matrices and matrix factorization factor Euclidean distance;

[0126] The optimal control increment sequence is represented by the following formula:

[0127] ;

[0128] in, For the first At the discrete time... The optimal control increment sequence for each agent. For the first At the discrete time... The peak value of the control input increment sequence of an agent in the prediction time domain. For the first Augmented control input prediction matrix for each agent For the first The augmented least squares coefficient matrix of each agent is composed of the prediction error coefficient matrix and the control input increment coefficient matrix. The optimal control increment is expressed by the following formula:

[0129] ;

[0130] in, For optimal control increment, It is the optimal solution matrix for controlling the input.

[0131] The variable weight cooperative control method for heavy transport equipment chassis provided in this embodiment involves the following steps in its application process:

[0132] Step 1) Modeling the interval-type II fuzzy vehicle system;

[0133] Step 2) Construct a global cost function within a cooperative game theory framework;

[0134] Step 3) Establish the optimization problem of cooperative game theory;

[0135] Step 4) Adaptively adjust the relative weights of each agent based on the β–ω phase plane;

[0136] Step 5) Solve the coordination control law based on cooperative game theory;

[0137] Furthermore, the specific steps of step 1) are as follows:

[0138] Step 1.1) Establish a two-degree-of-freedom vehicle model with three-axis steering and direct yaw moment control:

[0139] The two-degree-of-freedom vehicle model neglects the role of the suspension during modeling. This model assumes the car body only undergoes planar motion parallel to the ground, that the left and right wheels on the same axle have the same rotation angle, and that the longitudinal velocity of the car is constant. The simplified two-degree-of-freedom vehicle model is shown in the following equation:

[0140] (1);

[0141] in For the longitudinal speed of the vehicle, and These are the vehicle's lateral velocity and yaw rate; The lateral velocity of the vehicle's center of gravity The first derivative with respect to time represents the instantaneous rate of change of the lateral velocity, i.e., the lateral acceleration of the vehicle. The vehicle's yaw rate The first derivative with respect to time represents the instantaneous rate of change of the yaw rate, i.e., the yaw acceleration. and These represent the total mass of the vehicle and the moment of inertia of the vehicle's center of mass about the z-axis, respectively. , , These are the vertical distances from the front axle, middle axle, and rear axle to the vehicle's center of gravity, respectively. , , These are the steering angles of the front axle, center axle, and rear axle, respectively. , , These are the lateral stiffness of the tires on the front, middle, and rear axles, respectively. This is the direct yaw moment.

[0142] Combined with the vehicle's lateral position and yaw angle The kinematic relationship can be obtained by the following dynamic equations under a centralized control strategy:

[0143] (2);

[0144] in, This refers to the lateral displacement of the vehicle. For vehicle displacement The first derivative with respect to time represents the instantaneous rate of change of the vehicle's displacement. For the horizontal swing angle, For the lateral angle The first derivative with respect to time represents the instantaneous rate of change of the yaw angle.

[0145] Step 1.2) Establish the state-space equations of the vehicle system:

[0146] Based on the different control variables, the centralized state-space equations of the vehicle are divided into four agents: the front axle, the middle axle, the rear axle steering agent, and the direct yaw moment agent.

[0147] (3);

[0148] in, for The state vector at time t, superscript Represents the transpose of a matrix. The state matrix; , , and They are respectively Front axle angle at time Central axis rotation angle Rear axle rotation angle and Direct yaw moment at moment The corresponding input matrix, This is the output matrix;

[0149] , , , , , , For dimension The identity matrix. Wherein, for State vector at time step Regarding time The first derivative is used to characterize the instantaneous rate of change of each state variable of the vehicle; This is the output vector of the vehicle system.

[0150] Step 1.3) Fuzzy vehicle speed and tire lateral stiffness:

[0151] In traditional path tracking models, longitudinal velocity This is usually considered a fixed value to simplify controller design. However, in practical applications, It cannot remain constant; it will change over time. Meanwhile, the... Lateral stiffness of axle tire It is one of the key factors affecting vehicle driving stability; it is not a constant but changes with the road surface adhesion coefficient. Therefore, and Changes in these factors can lead to uncertainties in the parameters of the path tracking model. To address this issue, a fuzzy logic approach is introduced to capture these uncertainties in the system.

[0152] Assuming an uncertain longitudinal velocity satisfy: , express The minimum value, express The maximum value. Furthermore, based on the variation relationship of tire lateral stiffness, the first... Lateral stiffness of axle tire , No. Lateral stiffness of axle tire and the Lateral stiffness of axle tire It can be represented as:

[0153] (4);

[0154] in, , and These are the nominal lateral stiffnesses of the front axle, center axle, and rear axle, respectively. This represents a nonlinear function related to the road surface adhesion coefficient; The range of values ​​satisfies , and Representing variables respectively The minimum and maximum values.

[0155] Therefore, the three prerequisite variables are selected as follows: , , and They are respectively The first, second, and third preconditions for time, however, longitudinal velocity. Measurements typically contain unavoidable errors, making them difficult to obtain accurate readings in actual operation. The true value. In other words, , and Uncertainty exists in the membership functions of all.

[0156] To address this situation, an interval-type II fuzzy logic model is used to handle the uncertainty of membership functions. Therefore, the premise variables can be expressed as:

[0157] (5);

[0158] in, , express The first moment One prerequisite variable; specifically, , , ; and They represent The first moment Precondition variables The minimum and maximum values, and The corresponding membership functions are respectively and Its definition is as follows:

[0159] (6);

[0160] Furthermore, for variables (in The membership function of , with uncertainties ±Ω at its upper and lower boundaries, is expressed as:

[0161] (7);

[0162] in, Given a constant, and They represent Membership function at time The lower and upper boundaries; ,when hour, Indicates corresponding to Minimum value Membership function; when hour, Indicates corresponding to Maximum value The membership function.

[0163] Therefore, considering the uncertainty of system parameters, the vehicle path tracking model based on type II fuzzy logic can be expressed as:

[0164] Fuzzy rules: If Small or large, and Small or large, and If it is small or large, then:

[0165] (8);

[0166] The number of rules is 8. Local matrix. , , , and The corresponding matrices in the system state equations and their corresponding... It is derived. For the first The state matrix of the fuzzy rules For the first The first fuzzy rule The input matrix of each agent For agent serial number, For fuzzy rule numbers, .

[0167] also, Triggering intensity function of the p-th fuzzy rule at time p Defined as:

[0168] (9);

[0169] in, As the premise variable index, ; and They represent the first The first fuzzy rule Each premise variable corresponds to the lower and upper boundaries of the membership function; and They represent the first The lower and upper boundaries of the fuzzy rule trigger strength.

[0170] Therefore, the vehicle path tracking system can be represented by a type II fuzzy model as follows:

[0171] (10);

[0172] in, express Time of the first The fuzzy weights of the local model under a fuzzy rule are defined as follows:

[0173] (11);

[0174] in, for Time of the first The upper and lower trigger intensity adjustment function of the fuzzy rule has a value range of . Used to determine the lower boundary of the rule's trigger strength. With upper boundary The relative proportion in fuzzy weight calculation. When When, only the lower boundary of the trigger strength is used; when When, only the upper boundary of the trigger strength is used; when Then, a weighted combination of the two is performed. After normalization, the result is obtained. Time of the first Fuzzy weights corresponding to fuzzy rules , satisfy .

[0175] In short, the model integrates local linear models under different operating conditions through eight fuzzy rules, and uses type II fuzzy membership functions to handle the uncertainties of parameters such as vehicle longitudinal speed and tire lateral stiffness, thereby realizing fuzzy modeling of the path tracking system.

[0176] Step 1.4) Discrete-time state-space model:

[0177] To facilitate the solution of control variables by various intelligent agents, the continuous state-space equations need to be transformed into discrete form:

[0178] (12)

[0179] in, and They represent the first The and the first The vehicle system state vector at each discrete moment; Indicates the first The vehicle system output vector at each discrete time point. For the first The discrete state matrix of a fuzzy rule. For the first The first fuzzy rule Discrete input matrices of each agent To output a discrete matrix, during the discretization process, the matrix... From the continuous system matrix It is obtained by discretization using the Euler method; similarly... For the input matrix of a continuous system The discretization result, i.e. , It is the identity matrix. , , It is the identity matrix. The dimension is The identity matrix, The discrete time step;

[0180] Step 1.5) Establish the system's error equation:

[0181] To facilitate the design of the control strategy, the system error equation is first established based on the desired response of the vehicle state. The state error is defined. for Control increment for Then the error equation can be expressed as:

[0182] (13)

[0183] in, and They represent the first The and the first State error at each discrete moment; , and They represent the first Front axle, center axle, and rear axle rotation angle control deviations at discrete moments; Indicates the first Direct yaw moment control deviation at discrete moments; Indicates the first Output error at each discrete moment; For state parameters, superscript Represents the transpose of a matrix. and For the vehicle's lateral velocity and yaw rate, This refers to the lateral displacement of the vehicle. For the horizontal swing angle, For reference only. , Indicates the reference lateral velocity. Indicates the reference yaw rate. Indicates the reference horizontal position. Indicates the reference heading angle.

[0184] Step 2) Construct a global cost function within a cooperative game theory framework;

[0185] Step 2.1) Establish the cost function for each agent:

[0186] The front, center, and rear axle steering systems and the direct yaw moment control system are modeled as four cooperating agents, placed within a cooperative game framework as participants in the game. The individual payoff of each agent is measured by a corresponding cost function. The system achieves asymptotic stability as the control time domain approaches infinity. Therefore, the cost function of each agent can be expressed as:

[0187] (14)

[0188] in, For the first The cost function of the i-th agent at each discrete time step. This corresponds to four intelligent agents: the front axle steering agent, the center axle steering agent, the rear axle steering agent, and the yaw moment agent. In order to be in Predicting the future The output error vector at time t; In order to be in Predicting the future The control input increment of the i-th agent at time i. and Let be the weight matrices representing the system state and control input of the i-th agent, respectively, satisfying . , ; For control vector The One element; For the first The control objective selection matrix for each agent is used to select the output error that the agent is interested in. for The transpose of the matrix, superscript This represents the matrix transpose. When... (When the front, middle, and rear axles are turned, respectively) The performance indicators of the three axle steering agents are consistent, with the common control objective being to improve the vehicle's trajectory tracking accuracy. Therefore: , , ;when , This represents the performance index of the direct yaw moment agent, whose control objective is to enhance the lateral stability of the vehicle. Therefore: , , ;in This represents the target weight of the i-th agent with respect to its state x. ; This represents the control weight of the i-th agent for its control variable u. To facilitate the subsequent formula derivation, let... ,in, For the first The equivalent output error weight matrix of each agent is used to characterize the output error of the vehicle at the 1st moment. Weights in the cost function of each agent.

[0189] Step 2.2) Establish the global cost function;

[0190] To meet the needs of goal interaction among participants in cooperative game theory, and to minimize the overall system cost rather than the individual benefit of a single agent, a global cost function of the following form is constructed. To achieve unified sharing of all control objectives:

[0191] (15)

[0192] in, , It is an intelligent agent The relative weight of individual gains, and , .

[0193] The shared objective cost function of the agents is:

[0194] (16)

[0195] in, For the first A shared objective cost function for each agent is used to comprehensively evaluate the overall vehicle state error and the first agent's shared objective cost function. Incremental control input for each agent; and These are the positive definite weight matrices for the system state error and the control input, respectively. , .

[0196] Since the optimization problem consisting of an infinite time-domain cost function cannot be solved directly, it needs to be decomposed into a finite-time domain term and a residual infinite-time domain term, as shown in the following expression:

[0197] (17)

[0198] in, The predicted time domain length is a pre-defined positive integer; The corresponding cost item within the finite prediction time domain, The corresponding cost term for the remaining infinite time domain.

[0199] The remaining infinite time-domain terms are transformed into a terminal penalty term, and its approximate calculation process is as follows:

[0200] (18)

[0201] in, The terminal weight matrix can be obtained from the discrete-time algebraic Riccati equation. The shared objective cost function of the participants can then be restated as follows:

[0202] (19)

[0203] Step 3) Establish the optimization problem of cooperative game theory;

[0204] Furthermore, step 3) specifically includes:

[0205] To solve cooperative game problems, it is necessary to predict the system state. Therefore, a global prediction model for the system is established, which is expressed as follows:

[0206] (20)

[0207] in , , , .

[0208] in, For the first The prediction output error sequence at discrete moments is derived from the future. The output error at each prediction time is composed of; This is the state error prediction matrix, used to characterize the current state error. Impact on prediction output error; For the first The control input increment prediction matrix of the i-th agent is used to characterize the i-th agent. The impact of the incremental control input of each agent on the prediction output error; For the first The sequence of control input increments for an agent in the prediction time domain.

[0209] To optimize the state error of the control system and the control input increments of the four participants, the control optimization problem is defined as follows:

[0210] (twenty one)

[0211] in, For the first The minimum value of the objective function of each agent. For the first Lower limit of control input increment for an agent , For the first Lower limit of single-step control input increment for an agent. For the first Upper limit of control input increment for an agent. , For the first The lower bound of the single-step control input increment for each agent, assuming that the output constraint of each actuator is limited only by its peak capability, i.e., , Indicates the first The peak value of each agent.

[0212] Step 4) Adaptively adjust the relative weights of the agent based on the β–ω phase plane;

[0213] Step 4.1) Determine the vehicle state zone based on the phase plane stability region division:

[0214] At sampling time k, the individual reward priority of the i-th agent is determined by its relative weight. Therefore, it is necessary to determine the relative weights of the i-th agent at sampling time k based on quantitative indicators. Dynamic adjustments are made to achieve a balance between path tracking performance and vehicle stability under different fault scenarios.

[0215] The sideslip angle is the angle at the vehicle's center of gravity, calculated based on the real-time longitudinal velocity of the vehicle. and the transverse velocity of the center of mass Calculated, i.e. ; This represents the vehicle's yaw rate. It is based on the sideslip angle calculated in real-time from the center of gravity. and yaw rate Build - The phase plane is used to characterize the current stability state of the vehicle. The β–ω phase plane can accurately reflect the stability state of the vehicle and can be used to assist in the design of dynamic weight adjustment strategies. Figure 2 The β–ω phase plane results are presented under specific conditions, with rads as the unit. The vehicle's longitudinal velocity is 20 m / s, the front wheel steering angle is 0°, and the road adhesion coefficient is 0.85. The red and green lines in the β–ω phase plane represent the vehicle's stability boundaries under these conditions, as shown in the figure; there are a total of four stability boundaries. The green line represents the limiting condition ω ≤ ω for calculating the vehicle's yaw rate. max The green line is obtained from phase trajectory analysis and saddle point location (the red dot represents the equilibrium point). Overall, the stable region changes with the longitudinal velocity v. x and front wheel steering angle δ f The decrease due to the increase is consistent with the fact that vehicles are more prone to instability under high-speed and large-angle conditions.

[0216] Assume the vehicle's state at time k is as follows: Figure 2 At the pink four-pointed star in the β–ω phase plane shown, the minimum perpendicular distance from this point to the four stability boundaries can be denoted as follows: , , , Meanwhile, the minimum vertical distance from the equilibrium point to the four stability boundaries is denoted as... , , , Therefore, the dynamic adjustment rules for the four relative weights can be set according to the above distance relationships as follows.

[0217] definition ,Right now This represents the minimum vertical distance from the vehicle's current state point to the stability boundary. In the β–ω phase plane, based on the stability boundaries of the vehicle under different longitudinal velocities and adhesion coefficients, the phase plane is divided into three regions:

[0218] (1) Region I (Stable Region): The vehicle state point is far from the stable boundary, satisfying:

[0219] (twenty two)

[0220] (2) Region II (buffer zone): The vehicle state is close to the stable boundary, satisfying:

[0221] (twenty three)

[0222] (3) Region III (Boundary Region): The vehicle state exceeds the stable boundary, satisfying:

[0223] (twenty four)

[0224] in, This is the minimum vertical distance from the vehicle's current state point to the stability boundary. This is the baseline distance from the equilibrium point to the boundary.

[0225] Step 4.2) Calculate the stability weight coefficients based on the distance mapping function:

[0226] Using a simple continuous mapping function, the distance Convert to stability weight coefficients :

[0227] (25)

[0228] in To adjust the sensitivity coefficient (typically taken as 3–5). When When it is large, When the value is close to 0, trajectory control dominates; when... When smaller, When the value is close to 1, stability control dominates.

[0229] Step 4.3) Dynamically allocate agent weights based on stability weights:

[0230] The weight vectors of the four agents are set as follows:

[0231] (26)

[0232] in, For the first At the discrete time... The weight vector of each agent, , , The first The weights of the agent's front, middle, and rear axis steering at discrete moments. For the first The weights of the direct yaw moment agent at discrete moments.

[0233] Based on stability weight The following continuous allocation relationship is adopted:

[0234] (27)

[0235] This ensures that when the vehicle is in a stable condition ( When the state approaches instability, the three-axis steering agent dominates, prioritizing path tracking; when the state approaches instability... When the yaw moment weight increases, lateral stability takes priority; a natural and smooth transition is achieved in the middle region.

[0236] All weights satisfy the normalization condition:

[0237] (28)

[0238] Weight change limits within each sampling period:

[0239] (29)

[0240] in To smooth out constraints and prevent frequent controller switching.

[0241] Step 5) Solving the coordination control law for heavy transport equipment based on cooperative game theory:

[0242] To obtain the solution to the cooperative game, it can be achieved by simultaneously solving the coupled optimization problem involving all four participants. To facilitate the derivation of the analytical solution to the optimization problem, the original cost function, which includes terminal weighting terms, undergoes the following equivalent transformation to simplify the subsequent solution process:

[0243] (30)

[0244] in, This represents the transformed global cost function. For including terminal weight matrix The prediction error weight matrix, , This indicates the block diagonal operator; For the first The control input incremental weight matrix of each agent. ; For the first The augmented prediction error vector at each discrete time point. ; For the first At the discrete time... The square of the peak value of the control input increment sequence of an agent in the prediction time domain; For the terminal weight matrix Prediction error weight matrix Weighted Euclidean distance; For the and The weighted Euclidean distance of the ratio.

[0245] The optimization problem described above can be rephrased as follows:

[0246] (31)

[0247] in, For the transformed optimization problem, Indicates the first Each agent solves for the equivalent prediction error term when the control input increment is calculated. , An augmented state error prediction matrix, used to characterize the current state error. Impact on augmented forecast error; Represents a set of four intelligent agents. , This indicates that there is no intelligent agent. set ; For the first discrete time The set of The sequence of control input increments for an agent in the prediction time domain; for The set of The augmented control input prediction matrix of each agent is used to characterize The impact on augmented prediction error; Equation (31) indicates that, except for the first The contribution of other agents to the prediction error, in addition to the individual agent.

[0248] Constructed in least squares form as follows:

[0249] (32)

[0250] in, , , and Weight matrices and Matrix factorization factor, superscript Indicates matrix transpose. For Euclidean distance.

[0251] Solving the above equation using the QR method, we get:

[0252] (33)

[0253] In this context, the superscript * indicates the optimal solution, and \ is the solution symbol based on QR decomposition in MATLAB; For the first At the discrete time... The optimal control increment sequence for each agent. For the first At the discrete time... The peak value of the control input increment sequence of an agent in the prediction time domain. For the first Augmented control input prediction matrix for each agent For the first The augmented least squares coefficient matrix of each agent is composed of the prediction error coefficient matrix and the control input increment coefficient matrix. The optimal solution for the four agents can be obtained as follows:

[0254] (34)

[0255] in, For optimal control increment, It is the optimal solution matrix for controlling the input.

[0256] Example 2:

[0257] This embodiment provides a variable weight cooperative control system for heavy-duty transport equipment chassis, including:

[0258] The first module is used to acquire the vehicle's state parameters and determine the state error at the current moment based on the state parameters. The state parameters include yaw rate, longitudinal speed, and steering angles and tire lateral stiffness of the front axle, center axle, and rear axle.

[0259] The second module is used to input the current state error and the control increment to be solved into the pre-constructed state error equation to obtain the state error equation outputting the corresponding predicted state error according to the control increment. The state error equation is determined by an interval type II fuzzy model used to describe the uncertainty of vehicle longitudinal speed and tire lateral stiffness.

[0260] The third module is used to determine the current centroid sideslip angle based on the state parameters, and to identify the stability of the vehicle based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane to obtain the stability identification result.

[0261] The fourth module is used to update the weight coefficients of each agent in the pre-constructed global cost function based on the stability identification result, and to solve the control increment with the goal of minimizing the updated global cost function to obtain the optimal control increment; the global cost function includes an error penalty term constructed based on the predicted state error;

[0262] The control increment is a joint decision vector composed of the control input increments of four cooperative agents: the front axle steering angle, the center axle steering angle, the rear axle steering angle, and the direct yaw moment. Each agent works together to solve the joint decision vector through cooperative game theory to obtain its own corresponding optimal control input increment.

[0263] The specific functions of each module described above are explained in the relevant content of the method in Embodiment 1, and will not be repeated here.

[0264] Example 3:

[0265] This embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0266] Example 4:

[0267] This embodiment provides a computer device, including:

[0268] Memory, used to store computer programs / instructions;

[0269] A processor for executing the computer program / instructions to implement the steps of the method described in Embodiment 1.

[0270] Example 5:

[0271] This embodiment provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described in Embodiment 1.

[0272] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

[0273] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0274] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0275] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0276] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0277] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and not to limit its protection scope. Although this disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading this disclosure, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the pending claims.

Claims

1. A variable weight cooperative control method for a heavy-duty transport equipment chassis, characterized in that, include: The vehicle's state parameters are acquired, and the state error at the current moment is determined based on the state parameters. The state parameters include yaw rate, longitudinal velocity, and steering angles and tire lateral stiffness of the front axle, center axle, and rear axle. The current state error and the control increment to be solved are input into the pre-constructed interval type II fuzzy state error equation to obtain the predicted state error corresponding to the control increment. The interval type II fuzzy state error equation is constructed based on the premise variables, which are constructed according to the longitudinal velocity and the tire lateral stiffness of each axle. The current centroid sideslip angle is determined based on the state parameters. The stability of the vehicle is then identified based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane to obtain the stability identification result. The weight coefficients of each agent in the pre-constructed global cost function are updated based on the predicted state error and stability identification results. The control increment is solved with the goal of minimizing the updated global cost function to obtain the optimal control increment. The global cost function includes an error penalty term constructed based on the predicted state error. The control increment is a joint decision vector composed of the control input increments of four cooperative agents: the front axle steering angle, the center axle steering angle, the rear axle steering angle, and the direct yaw moment. Each agent works together to solve the joint decision vector through cooperative game theory to obtain its own corresponding optimal control input increment.

2. The variable weight cooperative control method for heavy transport equipment chassis according to claim 1, characterized in that, The method for constructing the interval type II fuzzy state error equation includes: Using longitudinal vehicle speed and tire lateral stiffness as prerequisite variables, an interval type II fuzzy rule is established, and a type II fuzzy model is constructed through the interval type II fuzzy rule; The type II fuzzy model is discretized to obtain a discrete-time state-space model; Based on the discrete-time state-space model, a state error is defined, and an interval type II fuzzy state error equation is established to describe how the state error evolves with the discrete time step. The interval type II fuzzy state error equation is expressed by the following equation: ； In the formula, and They represent the first The and the first State error at discrete moments Indicates the first Output error at each discrete time point For state error, , For state parameters, superscript Represents the transpose of a matrix. and For the vehicle's lateral velocity and yaw rate, This refers to the lateral displacement of the vehicle. For the horizontal swing angle, For reference only. , Indicates the reference lateral velocity. Indicates the reference yaw rate. Indicates the reference horizontal position. Indicates the reference heading angle. Indicates the first The discrete time... Fuzzy weights under fuzzy rules For the first The discrete state matrix of a fuzzy rule. For the first The first fuzzy rule Discrete input matrices of each agent For discrete output matrices, , , , It is the identity matrix. The dimension is The identity matrix, For discrete time steps, For the first The state matrix of the fuzzy rules For the first The first fuzzy rule The input matrix of each agent For the agent's serial number. For fuzzy rule numbers, , , and They represent the first The front axle angle control deviation, center axle angle control deviation, and rear axle angle control deviation at discrete moments; Indicates the first Direct yaw moment control deviation at discrete moments; The discrete-time state-space model is represented by the following equation: ； In the formula, For the first A state vector at discrete moments. For the first A state vector at discrete moments. For the first The output vector at each discrete time point , , They are the first The front axle steering angle, center axle steering angle, and rear axle steering angle at discrete moments. For the first The direct yaw moment at discrete moments.

3. The variable weight cooperative control method for heavy transport equipment chassis according to claim 2, characterized in that, The global cost function is expressed by the following equation: ； In the formula, For the first The objective function of each agent. In order to be in Predicting the future The output error vector at time t; In order to be in Predicting the future The control input increment of the i-th agent at time t; In order to be in Predicting the future The state error vector at time t. To predict the length of the time domain, Indicates the cost item number. , This is the terminal weight matrix. , Here is the state transition matrix. To control the input matrix, , These are the input matrices for the front axle angle, center axle angle, rear axle angle, and direct yaw moment, respectively. This represents the block diagonal operator. , , , , , For the longitudinal speed of the vehicle, and These represent the total mass of the vehicle and the moment of inertia of the vehicle's center of mass about the z-axis, respectively. , , These are the vertical distances from the front axle, middle axle, and rear axle to the vehicle's center of gravity, respectively. , , These are the lateral stiffness of the tires on the front, middle, and rear axles, respectively. and These are the positive definite weight matrices for the system state error and the control input, respectively. , , For the first The weight vector of each agent, Let be the weight matrix for the control input of the i-th agent. To control the input increment weight matrix, , For the first The equivalent output error weight matrix of each agent , For the first Control objective selection matrix for each agent Let be the weight matrix of the system state of the i-th agent. , , Let be the yaw angle error weight for the i-th agent. The weight of the yaw rate error of the i-th agent; The positive definite weight matrix of the system state error Weighted Euclidean distance For terminal weight matrix Weighted Euclidean distance; The positive definite weight matrix is ​​the result of control input. Weighted Euclidean distance.

4. The variable weight cooperative control method for heavy transport equipment chassis according to claim 3, characterized in that, The optimization problem of the global cost function is expressed by the following equation: ； in, For the first The minimum value of the objective function of each agent. For the first The sequence of prediction output errors at discrete time points; This is the state error prediction matrix; For the first The control input increment prediction matrix for each agent; For the first The discrete time... The sequence of control input increments for an agent in the prediction time domain. For the first Lower limit of control input increment for an agent For the first Upper limit of control input increment for each agent; , , , ; For the first The discrete state transition matrix of the step. For the first The agent in the th... The discrete control input matrix of the step.

5. The variable weight cooperative control method for heavy transport equipment chassis according to claim 4, characterized in that, The stability of the vehicle is identified based on the current centroid sideslip angle and a pre-constructed centroid sideslip angle phase plane, yielding stability identification results, including: The position of the vehicle's current state point in the pre-constructed phase plane of the center of gravity sideslip angle is determined based on the current center of gravity sideslip angle and yaw rate, and the minimum vertical distance from the vehicle's current state point to the stability boundary is determined. The stability weighting coefficient is determined based on the minimum vertical distance and the preset reference distance; Based on the stability weight coefficients, the weights of the front, middle and rear axle steering agents and the weights of the direct yaw moment agent are determined, and each weight is used as the stability identification result. The stability weighting coefficient is calculated by the following formula: ； in, For the first Stability weighting coefficients at discrete time points It is an exponential function. To adjust the sensitivity coefficient, For the first The minimum vertical distance from the vehicle's current state point to the stability boundary at each discrete time step. For the first The reference distance from the equilibrium point to the stability boundary at each discrete time step; The weights of the front, center, and rear axle steering agents, as well as the weights of the direct yaw moment agent, are expressed by the following formula: ； in, , , The first The weights of the agent's front, middle, and rear axis steering at discrete moments. For the first The weights of the direct yaw moment agent at discrete moments. , , To smooth the constraints, For the first At the discrete time... The weight vector of each agent, , For the first At the discrete time... The weight vector of each agent.

6. The variable weight cooperative control method for heavy transport equipment chassis according to claim 5, characterized in that, The optimal control increment is obtained by solving the control increment with the objective of minimizing the updated global cost function, including: The global cost function containing terminal weighting terms is transformed by an equivalent transformation to obtain the transformed global cost function. The optimization problem is transformed based on the transformed global cost function to obtain the transformed optimization problem. Based on the transformed optimization problem, a least squares formula is constructed, and the least squares formula is solved using the QR method to obtain the optimal control increment sequence. The first term of the optimal control increment sequence is taken as the optimal control increment. The transformed global cost function is expressed by the following equation: ； in, This represents the transformed global cost function. For including terminal weight matrix The prediction error weight matrix, ; For the first The control input incremental weight matrix of each agent. ; For the first The augmented prediction error vector at each discrete time point. ; For the first At the discrete time... The square of the peak value of the control input increment sequence of an agent in the prediction time domain; For the terminal weight matrix Prediction error weight matrix Weighted Euclidean distance; For the and The Euclidean distance weighted by the ratio; The transformed optimization problem is expressed by the following equation: ； in, For the transformed optimization problem, Indicates the first Each agent solves for the equivalent prediction error term when the control input increment is calculated. , An augmented state error prediction matrix, used to characterize the current state error. Impact on augmented forecast error; Represents a set of four intelligent agents. , This indicates that there is no intelligent agent. set ; For the first discrete time The set of The sequence of control input increments for an agent in the prediction time domain; for The set of The augmented control input prediction matrix of each agent is used to characterize Impact on augmented forecast error; The least squares expression is represented by the following formula: ； in, , , and Weight matrices and matrix factorization factor Euclidean distance; The optimal control increment sequence is represented by the following formula: ； in, For the first At the discrete time... The optimal control increment sequence for each agent. For the first At the discrete time... The peak value of the control input increment sequence of an agent in the prediction time domain. For the first Augmented control input prediction matrix for each agent For the first The augmented least squares coefficient matrix of each agent is composed of the prediction error coefficient matrix and the control input increment coefficient matrix. ; The optimal control increment is expressed by the following formula: ； in, For optimal control increment, It is the optimal solution matrix for controlling the input.

7. A variable weight cooperative control system for a heavy-duty transport equipment chassis, characterized in that, include: The first module is used to acquire the vehicle's state parameters and determine the state error at the current moment based on the state parameters. The state parameters include yaw rate, longitudinal speed, and steering angles and tire lateral stiffness of the front axle, center axle, and rear axle. The second module is used to input the current state error and the control increment to be solved into the pre-constructed interval type II fuzzy state error equation to obtain the predicted state error corresponding to the control increment. The interval type II fuzzy state error equation is constructed based on the premise variables, which are constructed according to the longitudinal speed and the tire lateral stiffness of each axle. The third module is used to determine the current centroid sideslip angle based on the state parameters, and to identify the stability of the vehicle based on the current centroid sideslip angle and the pre-constructed centroid sideslip angle phase plane to obtain the stability identification result. The fourth module is used to update the weight coefficients of each agent in the pre-constructed global cost function based on the predicted state error and stability identification results, and to solve the control increment with the goal of minimizing the updated global cost function to obtain the optimal control increment; the global cost function includes an error penalty term constructed based on the predicted state error; The control increment is a joint decision vector composed of the control input increments of four cooperative agents: the front axle steering angle, the center axle steering angle, the rear axle steering angle, and the direct yaw moment. Each agent works together to solve the joint decision vector through cooperative game theory to obtain its own corresponding optimal control input increment.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of the method according to any one of claims 1 to 6.

Images