Mpc-based brake force distribution method, system, device, medium and product
By using the MPC-based braking force distribution method, braking force is dynamically distributed, solving the problem that traditional braking systems cannot adapt to dynamic changes, and achieving faster response and more stable braking performance.
Patent Information
- Application Number
- CN202611506992.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-28
- Publication Date
- 2026-11-06
AI Technical Summary
Traditional braking systems cannot adapt to dynamically changing vehicle conditions and complex road conditions, leaving room for improvement in braking response speed and overall performance.
A braking force distribution method based on model predictive control (MPC) is adopted. By acquiring vehicle state data, a dynamic model is constructed, a reference trajectory of the objective function is set, and a multi-objective optimization problem is solved in the rolling time domain to dynamically distribute the braking force to each wheel.
It significantly shortens braking distance, improves the ability to respond to sudden working conditions, enhances braking stability and responsiveness, and adapts to various road conditions.
Smart Images
Figure CN123040792A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle braking control system technology, and in particular to a braking force distribution method, system, device, medium and product based on MPC. Background Technology
[0002] The performance of a vehicle's braking system is directly related to driving safety. Traditional braking systems use hydraulic or pneumatic transmission and are often equipped with anti-lock braking systems (ABS) and electronic brakeforce distribution (EBD). EBD typically distributes braking force between the front and rear axles based on a preset fixed-ratio curve, which provides basic protection but cannot adapt to dynamically changing vehicle conditions and complex road conditions.
[0003] With the development of drive-by-wire technology, electromechanical braking (EMB) systems have become a research direction for next-generation braking systems due to their advantages such as fast response, high control precision, and simple structure. To optimize EMB performance, existing research has proposed various intelligent control algorithms. These methods belong to "reactive" control, meaning they adjust based on the current error; therefore, their response speed and overall performance still have room for improvement when dealing with rapidly changing operating conditions. Summary of the Invention
[0004] This application provides a braking force distribution method, system, device, medium, and electronic product based on MPC.
[0005] In a first aspect, this application provides a braking force distribution method based on MPC, which includes: acquiring vehicle state data; wherein the vehicle state data includes a vehicle longitudinal speed signal, a vehicle yaw rate signal, and a pedal braking signal; Based on the total braking force determined by the pedal braking signal, a reference trajectory for the objective function is set; the reference trajectory is at least determined by the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio Characterized by at least two reference terms; Based on model predictive control, a multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain. The multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle's longitudinal velocity signal, the vehicle's yaw rate signal, and the reference terms in the reference trajectory. Based on the solved control quantity, the corresponding braking force is distributed to the EMB actuators configured on each wheel.
[0006] Optionally, the prediction model for MPC is determined based on the following steps: Construct the state-space equations for the vehicle dynamics model: The state-space equations include the following two items: Longitudinal dynamic equation: ; The yaw dynamics equation: ; in, It's about vehicle quality; It is the vehicle's longitudinal speed signal. It is a longitudinal acceleration signal It is the longitudinal braking force of the front left wheel; It is the longitudinal braking force of the front right wheel; It is the longitudinal braking force of the rear left wheel; It is the longitudinal braking force of the rear right wheel; It is the moment of inertia of the vehicle about its vertical axis; It is the vehicle yaw rate signal; It is a yaw rate acceleration signal; It is a signal indicating the distance between the left and right wheels on the front axle of the vehicle; It is a signal indicating the distance between the wheels on the left and right sides of the rear axle of the vehicle; The longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC.
[0007] Optionally, the longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC, including: by It is a state vector; and with To control the vectors, the longitudinal and yaw dynamic equations are rewritten in state-space form: Where A is the system matrix and B is the input matrix; The equations, rewritten in state-space form, are discretized using the forward Euler method to obtain the prediction model for MPC.
[0008] Optionally, the objective function of the multi-objective optimization problem includes at least two of the following: a first penalty term; a second penalty term; a third penalty term; and a fourth penalty term. The first penalty term is used to penalize the vehicle's longitudinal speed signal. and Deviation terms; The second penalty is used to penalize the vehicle's yaw rate signal. and Deviation; The third penalty term is used to penalize the control quantity during the k-th control cycle. Control quantity of the (k-1)th control cycle Deviation; The fourth penalty is used to penalize wheel slip ratio. and Deviation; The constraints of a multi-objective optimization problem include at least two of the following:
[0009] in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; ; It is the lower limit of the slip ratio; It is the upper limit of the slip ratio; For tire elliptical constraints; For longitudinal braking force; It is a lateral braking force; The road surface adhesion coefficient; This refers to the vertical force on the tire. Corresponds to the front right wheel; Corresponding to the front left wheel; Corresponding to the rear right wheel; Corresponding to the rear left wheel.
[0010] Optionally, the objective function of the multi-objective optimization problem is:
[0011] in, This is the first penalty item; This is the second penalty item; This is the third penalty item; This is the fourth penalty item; R, S, and P are weight matrices, respectively; Wheel slip ratio ; The wheel's angular velocity; The radius of the wheel; For prediction in the time domain; To control the time domain; k is the current control cycle; (k+ii) is the control period after ii control periods starting from the k-th control period.
[0012] Optionally, the multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain, including: In the rolling time domain, the multi-objective optimization problem of the prediction model of MPC is solved by an optimization solver; wherein, the optimization solver is configured with a maximum number of iteration steps.
[0013] Optionally, the corresponding braking force is distributed to the EMB actuators configured on each wheel according to the solved control quantity, including: The corresponding braking force is distributed to the EMB actuators configured on each wheel based on the first control variable in the solved optimal control sequence.
[0014] Optionally, vehicle status data is acquired, including: Raw data is collected through the sensor module; An extended Kalman filter is used, taking the original vehicle longitudinal velocity signal and the original yaw rate signal from the original data as inputs, and outputting the vehicle longitudinal velocity signal respectively. and vehicle yaw rate signal .
[0015] Optionally, based on the total braking force determined by the pedal braking signal, a reference trajectory for the objective function is set, including: Driver braking intention analysis is completed based on brake pedal opening signal: using pedal travel and pedal force as dual input parameters, a lookup operation is performed through a pre-calibrated mapping table to calculate the driver's desired total braking force of the vehicle. Based on the total braking force, a reference trajectory for the objective function is set.
[0016] Secondly, this application provides a brake force distribution system based on MPC, which includes: an electronic control unit and an EMB actuator; wherein, Electronic control unit, used for: Acquire vehicle status data; the vehicle status data includes vehicle longitudinal velocity signal, vehicle yaw rate signal, and pedal braking signal; Based on the total braking force determined by the pedal braking signal, a reference trajectory for the objective function is set; the reference trajectory is at least determined by the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio Characterized by at least two reference terms; Based on model predictive control, a multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain. The multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle's longitudinal velocity signal, the vehicle's yaw rate signal, and the reference terms in the reference trajectory. Based on the solved control quantity, the corresponding braking force is distributed to the EMB actuators configured on each wheel.
[0017] Optionally, the electronic control unit is also used for: Construct the state-space equations for the vehicle dynamics model: The state-space equations include the following two items: Longitudinal dynamic equation: ; The yaw dynamics equation: ; in, It's about vehicle quality; It is the vehicle's longitudinal speed signal; It is a longitudinal acceleration signal; It is the longitudinal braking force of the front left wheel; It is the longitudinal braking force of the front right wheel; It is the longitudinal braking force of the rear left wheel; It is the longitudinal braking force of the rear right wheel; It is the moment of inertia of the vehicle about its vertical axis; It is the vehicle yaw rate signal; It is a yaw rate acceleration signal; It is a signal indicating the distance between the left and right wheels on the front axle of the vehicle; It is a signal indicating the distance between the wheels on the left and right sides of the rear axle of the vehicle; The longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC.
[0018] Optionally, when the electronic control unit is configured to discretize the longitudinal and yaw dynamic equations of the vehicle dynamics model to obtain a prediction model of MPC, it is also specifically used for: by It is a state vector; and with To control the vectors, the longitudinal and yaw dynamic equations are rewritten in state-space form: Where A is the system matrix and B is the input matrix; The equations, rewritten in state-space form, are discretized using the forward Euler method to obtain the prediction model for MPC.
[0019] Optionally, the objective function of a multi-objective optimization problem includes at least two of the following: First penalty item; Second penalty item; Third penalty item; Fourth penalty item; The first penalty term is used to penalize the vehicle's longitudinal speed signal. and Deviation terms; The second penalty is used to penalize the vehicle's yaw rate signal. and Deviation; The third penalty term is used to penalize the control quantity during the k-th control cycle. Control quantity of the (k-1)th control cycle Deviation; The fourth penalty is used to penalize wheel slip ratio. and Deviation; The constraints of a multi-objective optimization problem include at least two of the following:
[0020] in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; ; It is the lower limit of the slip ratio; It is the upper limit of the slip ratio; For tire elliptical constraints; For longitudinal braking force; It is a lateral braking force; The road surface adhesion coefficient; This refers to the vertical force on the tire. Corresponds to the front right wheel; Corresponding to the front left wheel; Corresponding to the rear right wheel; Corresponding to the rear left wheel.
[0021] Here, by setting constraints for the multi-objective optimization problem, the physical limits of the actuator, the tire adhesion ellipse, and other constraints can be explicitly incorporated into the optimization problem, thereby always keeping the system within a safe and feasible range.
[0022] Optionally, the objective function of the multi-objective optimization problem is:
[0023] in, This is the first penalty item; This is the second penalty item; This is the third penalty item; This is the fourth penalty item; R, S, and P are weight matrices, respectively; Wheel slip ratio ; The wheel's angular velocity; The radius of the wheel; For prediction in the time domain; To control the time domain; k is the current control cycle; (k+ii) is the control period after ii control periods starting from the k-th control period.
[0024] Optionally, when the electronic control unit is configured to solve a multi-objective optimization problem of the prediction model of MPC in the rolling time domain, it is also specifically used for: In the rolling time domain, the multi-objective optimization problem of the prediction model of MPC is solved by an optimization solver; wherein, the optimization solver is configured with a maximum number of iteration steps.
[0025] Optionally, when the electronic control unit is configured to distribute the corresponding braking force to the EMB actuators configured on each wheel according to the solved control quantity, it is also specifically used for: The corresponding braking force is distributed to the EMB actuators configured on each wheel based on the first control variable in the solved optimal control sequence.
[0026] Optionally, the MPC-based brake force distribution system also includes a sensor module; when configured to acquire vehicle status data, the electronic control unit is also specifically used for: Raw data is collected through the sensor module; An extended Kalman filter is used, taking the original vehicle longitudinal velocity signal and the original yaw rate signal from the original data as inputs, and outputting the vehicle longitudinal velocity signal respectively. and vehicle yaw rate signal .
[0027] Optionally, when the electronic control unit is configured to set a reference trajectory for the objective function based on the total braking force determined by the pedal braking signal, it is also specifically used for: Driver braking intention analysis is completed based on brake pedal opening signal: using pedal travel and pedal force as dual input parameters, a lookup operation is performed through a pre-calibrated mapping table to calculate the driver's desired total braking force of the vehicle. Based on the total braking force, a reference trajectory for the objective function is set.
[0028] Thirdly, this application provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to implement the MPC-based braking force distribution method described in the first aspect.
[0029] Fourthly, this application provides a computer-readable storage medium that, when the instructions in the storage medium are executed by a processor corresponding to an electronic device, enables the electronic device to implement the MPC-based braking force distribution method described in the first aspect.
[0030] Fifthly, this application provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the MPC-based braking force distribution method as described in the first aspect.
[0031] The technical solution of this application has the following technical advantages over the prior art: The braking force distribution method based on MPC provided in this application has the following beneficial effects: Overall performance improvement: By solving multi-objective optimization problems, the braking distance can be significantly shortened while ensuring vehicle stability.
[0032] Proactive control and rapid response: By utilizing the predictive capabilities of the MPC predictive model, the algorithm can "anticipate" future state changes and take action in advance, effectively reducing control lag and improving the ability to respond to sudden operating conditions. Attached Figure Description
[0033] The preferred embodiments of this application will be described in detail below with reference to the accompanying drawings, which will help to understand the purpose and advantages of this application, wherein: Figure 1 A flowchart illustrating a braking force distribution method based on MPC provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of the MPC control unit provided in the embodiments of this application; Figure 3 A flowchart illustrating a braking force distribution method based on MPC provided in this application embodiment; Figure 4 A schematic diagram of a brake force distribution system based on MPC provided for an embodiment of this application; Figure 5 This is a schematic diagram of the hardware connection relationship of the electronic device for the MPC-based braking force distribution method provided in the embodiments of this application. Detailed Implementation
[0034] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0035] In the description of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0036] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.
[0037] Furthermore, the technical features involved in the different embodiments of this application described below can be combined with each other as long as they do not conflict with each other.
[0038] Figure 1 This is a flowchart illustrating a brake force distribution method based on MPC, provided as an embodiment of this application. This MPC-based brake force distribution method comprises an electronic control unit (ECU). For example... Figure 1 As shown, the MPC-based braking force distribution method includes the following steps: Step 101: Obtain vehicle status data; The vehicle status data includes the vehicle longitudinal speed signal, the vehicle yaw rate signal, and the pedal braking signal. Step 102: Based on the total braking force determined by the pedal braking signal, set the reference trajectory of the objective function; The reference trajectory is at least through the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio Characterized by at least two reference terms; Step 103: Based on model predictive control, solve the multi-objective optimization problem of the prediction model of MPC in the rolling time domain; Among them, the multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle's longitudinal velocity signal, the vehicle's yaw rate signal, and the reference terms in the reference trajectory; Step 104: Distribute the corresponding braking force to the EMB actuators configured on each wheel according to the solved control quantity.
[0039] The braking force distribution method based on MPC provided in this application sets a reference trajectory for the objective function based on the total braking force determined by the pedal braking signal, and uses the reference terms of the reference trajectory as reference inputs for a multi-objective optimization problem. Then, based on MPC, the multi-objective optimization problem is solved in the rolling time domain. Utilizing the predictive function of the MPC predictive model, future state changes can be anticipated and actions taken in advance, effectively reducing control lag and improving the ability to respond quickly to sudden conditions. The multi-objective optimization problem, set based on the reference terms of the reference trajectory, improves braking stability and efficiency, and enhances vehicle handling. Then, braking force is distributed to the EMB actuators configured on each wheel according to the solution results, achieving dynamic control of the braking force of each wheel by the EMB actuators, realizing adaptive adjustment of braking force distribution. The MPC predictive model is solved dynamically in the rolling time domain, enhancing the algorithm's real-time performance and adaptability, realizing dynamic braking force distribution, adapting to various road conditions (such as slippery surfaces), reducing braking distance, and preventing wheel lock-up.
[0040] In this embodiment, the electronic control unit (ECU) is responsible for running the MPC (Model Predictive Control) algorithm of this application, i.e., the braking force distribution method based on MPC. The ECU needs to possess powerful floating-point arithmetic capabilities to handle complex optimization calculations. It receives data from various sensors via the CAN (Controller Area Network) bus, performs data fusion and state estimation, executes the MPC algorithm flow described below, and finally controls the EMB actuator through its Pulse Width Modulation (PWM) output module and drive circuit.
[0041] EMB (Electro-Mechanical Brake) actuator: One is installed on each wheel. The actuator is essentially an electromechanical actuator consisting of a brushless DC motor (BLDC), a reduction gear, and a ball screw. The PWM signal output from the electronic control unit is amplified by the drive circuit to control the motor's torque and speed, thereby precisely generating the required braking force. EMB actuators typically integrate force or position sensors for closed-loop force control.
[0042] The following is a detailed description of step 101 (i.e., acquiring vehicle status data; wherein the vehicle status data includes vehicle longitudinal speed signal, vehicle yaw rate signal and pedal braking signal).
[0043] In one example, step 101 includes: Raw data is collected through the sensor module; An extended Kalman filter is used to output the vehicle's longitudinal velocity signal, with the original vehicle speed and original yaw rate as inputs respectively. and vehicle yaw rate signal .
[0044] In this example, raw data is collected through sensor modules, including four wheel speeds, vehicle longitudinal / lateral speeds and yaw rate from GPS (Global Positioning System) / IMU (Inertial Measurement Unit), etc. An extended Kalman filter (EKF) is used, taking the original vehicle longitudinal velocity signal and the original yaw rate signal from the raw data as inputs, and outputting the vehicle longitudinal velocity signal respectively. and vehicle yaw rate signal .
[0045] Optionally, and The state components of the state vector .
[0046] In this embodiment of the application, the state vector is... This serves as the initial state for the prediction model of MPC. These are the known conditions of a multi-objective optimization problem, and these are the input conditions of a single-objective optimization problem. It is the output solution of a multi-objective optimization problem.
[0047] In one example, the sensor module may include, but is not limited to, the following types of sensors: Wheel speed sensors: One sensor is installed on each wheel to monitor the instantaneous rotational speed of each wheel. Optionally, a magnetoelectric or Hall effect wheel speed sensor is used, the output of which is a pulse signal proportional to the wheel speed, transmitted via hardwire to the input capture interface of the electronic control unit.
[0048] Vehicle speed sensor: A fusion positioning module based on Navigation Satellite Timing and Ranging Global Position System (GPS) / Inertial Measurement Unit (IMU) can be used. This module communicates with the electronic control unit via a Controller Area Network (CAN) bus to provide accurate longitudinal speed, lateral speed, and yaw rate.
[0049] Brake pedal sensor: Installed on the brake pedal assembly, it is used to detect the driver's braking intention and braking intensity, and outputs a 0-5V analog voltage signal to the analog-to-digital converter (ADC) interface of the electronic control unit.
[0050] The following describes step 102 (i.e., setting the reference trajectory of the objective function based on the total braking force determined by the pedal braking signal; the reference trajectory is at least determined by the longitudinal velocity reference). Yaw velocity reference and optimal slip ratio The at least two reference terms in the text are described in detail.
[0051] In this example, the ECU interprets the driver's braking intention based on the brake pedal sensor signal, and calculates the driver's desired total braking force based on the pedal travel and pedal force using a pre-calibrated lookup table method. Based on the total braking force, a reference trajectory for the objective function is set.
[0052] In braking scenarios, the main reference parameters for the reference trajectory include longitudinal velocity reference. Yaw velocity reference Optimal slip ratio At least two of them.
[0053] In one example, step 102 includes: Driver braking intention analysis is performed based on brake pedal opening signal: using pedal travel and pedal force as dual input parameters, a lookup operation is performed through a pre-calibrated mapping table to calculate the driver's desired total braking force; based on the total braking force, a reference trajectory for the objective function is set.
[0054] The following section provides a detailed explanation of step 103 (i.e., solving the multi-objective optimization problem of the MPC prediction model in the rolling time domain based on model predictive control; wherein the multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle longitudinal velocity signal, the vehicle yaw rate signal, and the reference terms in the reference trajectory).
[0055] In this step, corresponding penalty terms are set for each reference term in the reference trajectory to construct a multi-objective optimization problem; then, based on model predictive control, the multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain.
[0056] Following on from the previous text, the reference trajectory is at least through the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio At least two reference terms are represented in it.
[0057] Here, the longitudinal velocity reference can be used. Yaw velocity reference and optimal slip ratio At least two of the terms are used to construct corresponding penalty terms to obtain a multi-objective optimization problem.
[0058] For example, using longitudinal velocity as a reference Yaw velocity reference and optimal slip ratio Construct corresponding penalty terms to obtain a multi-objective optimization problem.
[0059] When a system needs to strike a balance between "fast response" and "smooth execution," the rate of change of the control variable must be introduced into the multi-objective optimization problem. The penalty term is used to improve ride comfort, extend actuator life, and suppress high-frequency chattering, and can be implemented through a weight matrix. It can flexibly adapt to different working conditions.
[0060] In one example, the objective function of the multi-objective optimization problem includes at least two of the following: a first penalty term; a second penalty term; a third penalty term; and a fourth penalty term. The first penalty term is used to penalize the vehicle's longitudinal speed signal. and Deviation terms; The second penalty term is used to penalize the predicted yaw rate and... Deviation; The third penalty term is used to penalize the control quantity during the k-th control cycle. Control quantity of the (k-1)th control cycle Deviation; The fourth penalty is used to penalize wheel slip ratio. and Deviation; The constraints of a multi-objective optimization problem include at least two of the following:
[0061] in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; ; It is the lower limit of the slip ratio; It is the upper limit of the slip ratio; For tire elliptical constraints; For longitudinal braking force; It is a lateral braking force; The road surface adhesion coefficient; This refers to the vertical force on the tire. Corresponds to the front right wheel; Corresponding to the front left wheel; Corresponding to the rear right wheel; Corresponding to the rear left wheel.
[0062] Here, by setting constraints for the multi-objective optimization problem, the physical limits of the actuator, the tire adhesion ellipse, and other constraints can be explicitly incorporated into the optimization problem, thereby always keeping the system within a safe and feasible range.
[0063] Optionally, the objective function of the multi-objective optimization problem is:
[0064] in, This is the first penalty item; This is the second penalty item; This is the third penalty item; This is the fourth penalty item; R, S (uppercase), and P are the weight matrices, respectively; Wheel slip ratio ; The wheel's angular velocity; The radius of the wheel; For prediction in the time domain; To control the time domain; k is the current control cycle; (k+ii) is the control period after ii control periods starting from the k-th control period.
[0065] Here, the objective function is designed as a multi-objective weighted sum based on the MPC control principle, aiming to optimize multiple performance indicators simultaneously.
[0066] The first penalty term is the speed deviation from the target speed that is minimized, which is used to penalize the vehicle's longitudinal speed signal. Compared with reference speed The deviation. Weight matrix It is a positive definite matrix, and is adjusted through simulation to balance response speed and stability.
[0067] The second penalty is stability control, used to penalize excessive vehicle yaw rate signals. To prevent the vehicle from skidding or losing control. Weight matrix It is set according to the vehicle stability boundary.
[0068] The third penalty is control smoothness, used to penalize drastic changes in the control quantity. ,Right now: This is to ensure passenger comfort and actuator lifespan. Weight matrix .
[0069] The fourth penalty is for anti-lock braking system (ABS), which penalizes wheel slip ratio. With optimal slip ratio The deviation. The formula for calculating wheel slip ratio is: .
[0070] in, The wheel's angular velocity; The radius is the wheel radius.
[0071] Optimal slip ratio The weight matrix is dynamically adjusted based on the road surface adhesion coefficient. .
[0072] In one example, the prediction model for MPC is determined based on the following steps: Construct the state-space equations for the vehicle dynamics model: The state-space equations include the following two items: Longitudinal dynamic equation: ; The yaw dynamics equation: ; in, It's about vehicle quality; It is the vehicle's longitudinal speed signal; It is a longitudinal acceleration signal; It is the longitudinal braking force of the front left wheel; It is the longitudinal braking force of the front right wheel; It is the longitudinal braking force of the rear left wheel; It is the longitudinal braking force of the rear right wheel; It is the moment of inertia of the vehicle about its vertical axis; It is the vehicle yaw rate signal; It is a yaw rate acceleration signal; It is a signal indicating the distance between the left and right wheels on the front axle of the vehicle; It is a signal indicating the distance between the wheels on the left and right sides of the rear axle of the vehicle; The longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC.
[0073] It should be noted that this application uses a two-degree-of-freedom bicycle model as the prediction model for MPC. This model ensures accuracy while having a moderate computational burden, making it suitable for real-time applications. It assumes that the vehicle is traveling on a level road surface and ignores the influence of the suspension system on the vehicle. Under braking conditions, the braking force is much greater than air resistance and rolling resistance, so air resistance and rolling resistance are ignored.
[0074] In order to apply the continuous-time model to the discrete-time MPC, the embodiments of this application need to discretize it, specifically as follows: by It is a state vector; and with To control the vectors, the longitudinal and yaw dynamic equations are rewritten in state-space form: Where A is the system matrix and B is the input matrix; The equations, rewritten in state-space form, are discretized using the forward Euler method to obtain the prediction model for MPC.
[0075] The following derivation is performed on the formulas involved in discretization: by It is a state vector; and with To determine the control vector, the dynamic equations are rewritten in state-space form: ,include: First, the longitudinal dynamic equations and yaw dynamics equations Perform the following transformation: Longitudinal dynamics:
[0076] Yaw dynamics:
[0077] Then, according to the above transformed equation... ,as well as Substitute the form into Therefore, the equation in state form can be derived as follows: ; The equations, converted to state form, are discretized using the forward Euler method. The prediction model for MPC is obtained as follows: .
[0078] In one example, the multi-objective optimization problem of the prediction model for MPC is solved in the rolling time domain, including: In the rolling time domain, the multi-objective optimization problem of the prediction model of MPC is solved by an optimization solver; wherein, the optimization solver is configured with a maximum number of iteration steps.
[0079] In this example, the electronic control unit submits the multi-objective optimization problem in step 103 to the built-in optimization solver (such as qp (Quadratic Programming), OASES (Online Active Set Strategy), or OSQP (Operator Splitting Quadratic Programming Solver)) for solving. To ensure real-time performance, the solver needs to be configured with a fixed number of iterations, for example, the maximum number of iterations is set to 100.
[0080] Among them, qpOASES uses the Active Set algorithm, which is suitable for small-scale MPCs that require high efficiency in handling inequality constraints and scenarios that need frequent solutions with little structural change. OSQP uses the Alternating Direction Multiplier Method (ADMM), which is suitable for large-scale MPCs, embedded real-time control, and scenarios with high computational efficiency requirements.
[0081] exist Figure 2 In this approach, a reference trajectory is used as the reference input for the MPC prediction model. Based on model predictive control, within the rolling time domain, an optimization solver solves the multi-objective optimization problem of the MPC prediction model, and the control output of the solver is distributed to the EMB actuators configured on each wheel to provide braking force. Furthermore, the actual output of the EMB actuators (i.e., the actual braking force) and / or the control output of the solver are used as the input feedback for the MPC prediction model in the next control cycle to correct the MPC prediction model, forming a "state-control-output" closed loop. This loop is used to correct model mismatch online, compensate for actuator dynamic delays, and improve prediction accuracy and control robustness. Here, the multi-objective optimization problem of the MPC prediction model is solved rollingly within the rolling time domain. The feedback input of the MPC prediction model can be continuously corrected based on the latest vehicle state data and / or the actual braking force of the EMB actuators, exhibiting good robustness to model errors and external disturbances (such as changes in road surface adhesion).
[0082] The following is a detailed description of step 104 (i.e., distributing the corresponding braking force to the EMB actuators configured on each wheel according to the solved control quantity).
[0083] In this step, the corresponding braking force can be distributed to the EMB actuators configured on each wheel according to the control quantity solved by the solver.
[0084] In one example, step 104 includes: assigning the first control quantity from the optimal control sequence output by the optimal solver to the corresponding braking force of the EMB actuators configured for each wheel.
[0085] For example, the solver outputs the optimal control sequence: .
[0086] The electronic control unit only processes the first control variable in the sequence. The output is sent to the EMB executor.
[0087] Here, the control variable includes the braking force of the EMB actuators configured for each wheel. This application utilizes model predictive control to solve the multi-objective optimization problem of the MPC prediction model in the rolling time domain, which is the execution method for dynamically allocating braking force. Through a closed-loop connection of "state update → optimization solution → execution → re-update", each braking force allocation is re-determined based on the current operating conditions (i.e., updated vehicle state data), thus outperforming traditional fixed-ratio allocation and improving vehicle driving safety and stability.
[0088] Among them, the optimal control sequence in " "" indicates that the variable is the optimal solution or the optimal value.
[0089] In one example, the MPC-based braking force distribution method can be executed by the MPC control unit built into the electronic control unit.
[0090] It should be noted that k is a positive integer; after each execution of steps 101 to 104, k+1 is executed to enter the next control cycle (k+1); in the next control cycle (k+1), the electronic control unit repeats steps 101 to 104. Based on the latest raw data measured by the sensors, it re-solves the optimization problem, thereby continuously and adaptively adjusting the braking force distribution strategy according to the latest changes in the vehicle state. In addition, the actual output of the EMB actuator (i.e., the actual braking force) and / or the control quantity output by the solver can be used as the input feedback of the MPC prediction model in the next control cycle to correct the MPC prediction model, forming a "state-control-output" closed loop, which is used to correct model mismatch online, compensate for actuator dynamic delay, and improve prediction accuracy and control robustness.
[0091] It should be noted that the embodiments of this application only use a four-wheeled vehicle as an example, and are not limited to four-wheeled vehicles, but can also be six-wheeled vehicles, eight-wheeled vehicles, etc.; as long as one EMB actuator is configured on each wheel, the braking force of each EMB actuator can be dynamically distributed through the MPC-based braking force distribution method provided in the embodiments of this application.
[0092] The following is combined Figure 2 and Figure 3 The method of the MPC-based braking force distribution method provided in this application embodiment is executed by the electronic control unit, and its operation flow is as follows: Step 1: Data Acquisition and State Estimation The electronic control unit reads all raw data in real time, including the four wheel speeds, vehicle longitudinal / lateral speeds and yaw rate from GPS / IMU, etc. Since the directly measured raw vehicle longitudinal speed signals and raw yaw rate signals may contain noise and delay, this implementation uses an extended Kalman filter (EKF) to optimally estimate the vehicle state data. The EKF takes the raw vehicle longitudinal speed signal and raw yaw rate signal from the raw data as inputs, and outputs the vehicle longitudinal speed signal respectively. and vehicle yaw rate signal This estimate This will be used as the initial state for the prediction model of MPC.
[0093] Step 2: Braking Intent Analysis and Reference Generation. The electronic control unit analyzes the driver's braking intent based on the brake pedal sensor signal. According to the pedal travel and pedal force, it calculates the total braking force desired by the driver using a pre-calibrated lookup table method. Define the reference trajectory for the objective function.
[0094] In braking scenarios, the primary reference is the longitudinal velocity reference. Yaw velocity reference Optimal slip ratio .
[0095] Step 3: Construct the rolling time-domain optimization problem In each control cycle k, MPC solves an optimization problem in a finite time domain.
[0096] The objective function is designed as a multi-objective weighted sum based on the MPC control principle, aiming to optimize multiple performance indicators simultaneously:
[0097] The first penalty term is to minimize the speed deviation from the target speed. This penalty applies to the vehicle's longitudinal speed signal. Compared with reference speed The deviation. Weight matrix It is a positive definite matrix, and is adjusted through simulation to balance response speed and stability.
[0098] The second penalty is for stability control. Excessive vehicle yaw rate signals are penalized. To prevent the vehicle from skidding or losing control. Weight matrix It is set according to the vehicle stability boundary.
[0099] The third penalty is for control smoothness. Drastic changes in the control quantity are penalized. ,Right now This is to ensure passenger comfort and actuator lifespan. Weight matrix .
[0100] The fourth penalty is for anti-lock braking system (ABS). It penalizes wheel slip ratio. (Lowercase) and optimal slip ratio The deviation.
[0101] The formula for calculating wheel slip ratio is: .
[0102] in, The wheel's angular velocity; The radius is the wheel radius.
[0103] Optimal slip ratio Adjust dynamically based on the road surface adhesion coefficient.
[0104] weight matrix .
[0105] Constraints:
[0106] in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; Ensure the tires always operate within a stable area of contact.
[0107] in, It is the lower limit of the slip ratio; It is the upper limit of the slip ratio; The tire is constrained by an ellipse.
[0108] Step 4: Optimize Problem Solving The electronic control unit submits the optimization problem (a standard constrained quadratic programming problem) constructed in step 3 to the built-in optimization solver (such as qpOASES or OSQP) for solving. To ensure real-time performance, the solver needs to be configured with a fixed number of iterations, for example, the maximum number of iterations is set to 100.
[0109] Step 5: Output control quantity and rolling time domain The solver outputs the optimal control sequence: .
[0110] The electronic control unit only processes the first control variable in the sequence: The output is sent to the corresponding EMB executor.
[0111] Among them, the optimal control sequence in " "" indicates that the variable is the optimal solution or the optimal value.
[0112] In one example, the MPC-based braking force distribution method can be executed by the MPC control unit built into the electronic control unit.
[0113] It should be noted that k is a positive integer; after each execution of steps 1 to 5, k+1 is executed to enter the next control cycle (k+1); in the next control cycle (k+1), the electronic control unit repeats steps 1 to 5. Based on the latest raw data measured by the sensors, it re-solves the optimization problem, thereby continuously and adaptively adjusting the braking force distribution strategy according to the latest changes in the vehicle state.
[0114] The braking force distribution method based on MPC provided in this application ensures its effectiveness and robustness under various operating conditions through the following mechanism: Feedback correction: The state estimation in step 1 and the rolling time-domain mechanism in step 5 constitute a closed-loop feedback. This means that any control deviation caused by model inaccuracies, external disturbances (such as sudden road surface changes), or measurement noise will be detected and corrected in the next cycle.
[0115] Forward-looking optimization: By predicting the future state of the vehicle through a predictive model, the controller can "anticipate" potential risks (such as impending instability trends) and take corrective measures in advance, achieving proactive control and overcoming the lag of traditional reactive control.
[0116] Constraint protection: By explicitly handling all constraints through online optimization, the system ensures that all control commands generated at any time are within physically feasible and safe limits, fundamentally improving system security.
[0117] In summary, this application provides an efficient, stable, and reliable EMB braking force dynamic distribution solution by deeply integrating advanced model predictive control theory with vehicle dynamics. Those skilled in the art can modify the parameters and specific implementation methods based on the above teachings, but all such modifications should be included within the scope of protection of this application.
[0118] This application provides an MPC-based brake force distribution system for performing steps. Figure 1 The illustrated method is based on MPC (Multi-Process Control) for brake force distribution. The MPC-based brake force distribution system includes: Electronic control unit and EMB actuator; among which, Electronic control unit, used for: Obtain vehicle status data; The vehicle status data includes the vehicle longitudinal speed signal, the vehicle yaw rate signal, and the pedal braking signal. Based on the total braking force determined by the pedal braking signal, the reference trajectory of the objective function is set; The reference trajectory is at least through the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio Characterized by at least two reference terms; Based on model predictive control, the multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain; Among them, the multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle's longitudinal velocity signal, the vehicle's yaw rate signal, and the reference terms in the reference trajectory; Based on the solved control quantity, the corresponding braking force is distributed to the EMB actuators configured on each wheel.
[0119] Optionally, the electronic control unit is also used for: Construct the state-space equations for the vehicle dynamics model: The state-space equations include the following two items: Longitudinal dynamic equation: ; The yaw dynamics equation: ; in, It's about vehicle quality; It is the vehicle's longitudinal speed signal; It is a longitudinal acceleration signal; It is the longitudinal braking force of the front left wheel; It is the longitudinal braking force of the front right wheel; It is the longitudinal braking force of the rear left wheel; It is the longitudinal braking force of the rear right wheel; It is the moment of inertia of the vehicle about its vertical axis; It is the vehicle yaw rate signal; It is a yaw rate acceleration signal; It is a signal indicating the distance between the left and right wheels on the front axle of the vehicle; It is a signal indicating the distance between the wheels on the left and right sides of the rear axle of the vehicle; The longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC.
[0120] Optionally, when the electronic control unit is configured to discretize the longitudinal and yaw dynamic equations of the vehicle dynamics model to obtain a prediction model of MPC, it is also specifically used for: by It is a state vector; and with To control the vectors, the longitudinal and yaw dynamic equations are rewritten in state-space form: Where A is the system matrix and B is the input matrix; The equations, rewritten in state-space form, are discretized using the forward Euler method to obtain the prediction model for MPC.
[0121] Optionally, the objective function of a multi-objective optimization problem includes at least two of the following: First penalty item; Second penalty item; Third penalty item; Fourth penalty item; The first penalty term is used to penalize the vehicle's longitudinal speed signal. and Deviation terms; The second penalty is used to penalize the vehicle's yaw rate signal. and Deviation; The third penalty term is used to penalize the control quantity during the k-th control cycle. Control quantity of the (k-1)th control cycle Deviation; The fourth penalty is used to penalize wheel slip ratio. and Deviation; The constraints of a multi-objective optimization problem include at least two of the following:
[0122] in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; ; It is the lower limit of the slip ratio; and It is the lower limit of the slip ratio; For tire elliptical constraints; For longitudinal braking force; It is a lateral braking force; The road surface adhesion coefficient; This refers to the vertical force on the tire. Corresponds to the front right wheel; Corresponding to the front left wheel; Corresponding to the rear right wheel; Corresponding to the rear left wheel.
[0123] Optionally, the objective function of the multi-objective optimization problem is:
[0124] in, This is the first penalty item; This is the second penalty item; This is the third penalty item; This is the fourth penalty item; R, S, and P are weight matrices, respectively; Wheel slip ratio ; The wheel's angular velocity; The radius of the wheel; For prediction in the time domain; To control the time domain; k is the current control cycle; (k+ii) is the control period after ii control periods starting from the k-th control period.
[0125] Optionally, when the electronic control unit is configured to solve a multi-objective optimization problem of the prediction model of MPC in the rolling time domain, it is also specifically used for: In the rolling time domain, the multi-objective optimization problem of the prediction model of MPC is solved by an optimization solver; wherein, the optimization solver is configured with a maximum number of iteration steps.
[0126] Optionally, when the electronic control unit is configured to distribute the corresponding braking force to the EMB actuators configured on each wheel according to the solved control quantity, it is also specifically used for: The corresponding braking force is distributed to the EMB actuators configured on each wheel based on the first control variable in the solved optimal control sequence.
[0127] Optionally, the MPC-based brake force distribution system also includes a sensor module; when configured to acquire vehicle status data, the electronic control unit is also specifically used for: Raw data is collected through the sensor module; An extended Kalman filter is used, taking the original vehicle longitudinal velocity signal and the original yaw rate signal from the original data as inputs, and outputting the vehicle longitudinal velocity signal respectively. and vehicle yaw rate signal .
[0128] Optionally, when the electronic control unit is configured to set a reference trajectory for the objective function based on the total braking force determined by the pedal braking signal, it is also specifically used for: Driver braking intention analysis is completed based on brake pedal opening signal: using pedal travel and pedal force as dual input parameters, a lookup operation is performed through a pre-calibrated mapping table to calculate the driver's desired total braking force of the vehicle. Based on the total braking force, a reference trajectory for the objective function is set.
[0129] In one example, an MPC-based brake force distribution system provided in this application embodiment includes: refer to Figure 4 The implementation of this application relies on a complete electromechanical braking (EMB) system. This system mainly includes a sensor module, an electronic control unit, and an EMB actuator. The data flow and control flow between these modules constitute a closed-loop control system.
[0130] Sensor module: Responsible for collecting vehicle dynamic parameters in real time.
[0131] Wheel speed sensors: One sensor is installed on each of the four wheels to monitor the instantaneous rotational speed of each wheel. This implementation can use either magnetoelectric or Hall effect wheel speed sensors. Their output is a pulse signal proportional to the wheel speed, which is transmitted via hardwire to the input capture interface of the electronic control unit.
[0132] Vehicle speed sensor (not shown): This is a vehicle status sensor and can employ a fusion positioning module based on Navigation Satellite Timing and Ranging Global Position System (GPS) / Inertial Measurement Unit (IMU). This module communicates with the electronic control unit via a Controller Area Network (CAN) bus, providing accurate longitudinal speed, lateral speed, and yaw rate.
[0133] Brake pedal sensor: A vehicle status sensor, installed on the brake pedal assembly, used to detect the driver's braking intention and braking intensity, and outputs a 0-5V analog voltage signal to the analog-to-digital converter (ADC) interface of the electronic control unit.
[0134] In this embodiment, the electronic control unit (ECU) is responsible for running the MPC (Model Predictive Control) algorithm of this application, i.e., the braking force distribution method based on MPC. The ECU needs to possess powerful floating-point arithmetic capabilities to handle complex optimization calculations. It receives data from various sensors via the CAN (Controller Area Network) bus, performs data fusion and state estimation, executes the MPC algorithm flow described below, and finally controls the EMB actuator through its Pulse Width Modulation (PWM) output module and drive circuit.
[0135] EMB (Electro-Mechanical Brake) actuator: One is installed on each wheel. The actuator is essentially an electromechanical actuator consisting of a brushless DC motor (BLDC), a reduction gear, and a ball screw. The PWM signal output from the electronic control unit is amplified by the drive circuit to control the motor's torque and speed, thereby precisely generating the required braking force. EMB actuators typically integrate force or position sensors for closed-loop force control.
[0136] This application provides a computer-readable storage medium storing program information. After reading the program information, the computer executes the steps of the MPC-based braking force distribution method described in any of the above schemes.
[0137] This application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the MPC-based braking force distribution method described in any of the above schemes.
[0138] This application also provides an electronic device, such as... Figure 5As shown, the electronic device includes at least one processor 31 and at least one memory 32. The at least one memory 32 stores program information. After reading the program information, the at least one processor 31 executes the MPC-based braking force distribution method described in any of the above method embodiments. The device may further include an input device 33 and an output device 34. The processor 31, memory 32, input device 33, and output device 34 can be communicatively connected. The memory 32, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The processor 31 executes various functional applications and data processing by running the non-volatile software programs, instructions, and modules stored in the memory 32, thereby implementing the MPC-based braking force distribution method provided in any of the above embodiments. The memory 32 may include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on the use of the MPC-based braking force distribution method, etc. Furthermore, memory 32 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 32 may optionally include memory remotely located relative to processor 31, which can be connected via a network to perform the MPC-based brake force distribution method. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. Input device 33 may receive user clicks and generate signal inputs related to user settings and function control of the MPC-based brake force distribution method. Output device 34 may include a display device such as a screen. When the one or more modules are stored in memory 32 and are run by the one or more processors 31, the MPC-based brake force distribution method in any of the above method embodiments is executed.
[0139] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this application.
Claims
1. A braking force distribution method based on MPC, characterized in that, The method includes: Acquire vehicle status data; wherein, the vehicle status data includes vehicle longitudinal speed signal, vehicle yaw rate signal, and pedal braking signal; Based on the total braking force determined by the pedal braking signal, a reference trajectory for the objective function is set; the reference trajectory is at least determined by the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio Characterized by at least two reference terms; Based on model predictive control, a multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain; wherein, the multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle's longitudinal velocity signal, the vehicle's yaw rate signal, and the reference terms in the reference trajectory. Based on the solved control quantity, the corresponding braking force is distributed to the EMB actuators configured on each wheel.
2. The braking force distribution method based on MPC according to claim 1, characterized in that, The prediction model for MPC is determined based on the following steps: Construct the state-space equations for the vehicle dynamics model: The state-space equations include the following two items: Longitudinal dynamic equation: ； The yaw dynamics equation: ； in, It's about vehicle quality. It is the vehicle's longitudinal speed signal. It is a longitudinal acceleration signal. It is the longitudinal braking force of the front left wheel. It is the longitudinal braking force of the front right wheel. It is the longitudinal braking force of the rear left wheel. It is the longitudinal braking force of the rear right wheel. It is the moment of inertia of the vehicle about its vertical axis. It is the vehicle yaw rate signal. It is the yaw rate acceleration signal. These are the distance signals between the left and right wheels on the front and rear axles of the vehicle, respectively. The longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC.
3. The braking force distribution method based on MPC according to claim 2, characterized in that, The discretization of the longitudinal and yaw dynamic equations of the vehicle dynamics model yields the MPC prediction model, including: by Let be the state vector, and by To control the vectors, the longitudinal and yaw dynamic equations are rewritten in state-space form: Where A is the system matrix and B is the input matrix; The equations, rewritten in state-space form, are discretized using the forward Euler method to obtain the prediction model for MPC.
4. The braking force distribution method based on MPC according to claim 1, characterized in that, The objective function of the multi-objective optimization problem includes at least two of the following: a first penalty term; a second penalty term; a third penalty term; and a fourth penalty term. The first penalty term is used to penalize the vehicle's longitudinal speed signal. and The deviation term, the second penalty term is used to penalize the vehicle yaw rate signal. and The deviation; the third penalty term is used to penalize the control quantity in the k-th control cycle. Control quantity of the (k-1)th control cycle The deviation; the fourth penalty term is used to penalize the wheel slip ratio. and Deviation; The constraints of the multi-objective optimization problem include at least two of the following: in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; , and These are the lower and upper limits of the slip ratio, respectively. For tire elliptical constraints; For longitudinal braking force; It is a lateral braking force; The road surface adhesion coefficient; This refers to the vertical force on the tire. These correspond to the front right wheel, front left wheel, rear right wheel, and rear left wheel, respectively.
5. The braking force distribution method based on MPC according to claim 4, characterized in that, The objective function of the multi-objective optimization problem is: in, This is the first penalty item; This is the second penalty item; This is the third penalty item; This is the fourth penalty item; R, S, and P are the weight matrices; the wheel slip ratio is... ; The wheel's angular velocity; The radius of the wheel; For prediction in the time domain; For the control time domain; k is the current control period, and (k+ii) is the control period after ii control periods starting from the k-th control period.
6. The braking force distribution method based on MPC according to any one of claims 1-5, characterized in that, Solving the multi-objective optimization problem of the prediction model for MPC in the rolling time domain includes: In the rolling time domain, the multi-objective optimization problem of the prediction model of MPC is solved by an optimization solver; wherein the optimization solver is configured with a maximum number of iteration steps.
7. The braking force distribution method based on MPC according to any one of claims 1-5, characterized in that, Based on the solved control variables, corresponding braking forces are distributed to the EMB actuators configured on each wheel, including: The corresponding braking force is distributed to the EMB actuators configured on each wheel based on the first control variable in the solved optimal control sequence.
8. The braking force distribution method based on MPC according to claim 1, characterized in that, The acquisition of vehicle status data includes: Raw data is collected through the sensor module; An extended Kalman filter is used, with the original vehicle longitudinal velocity signal and the original yaw rate from the original data as inputs, to output the vehicle longitudinal velocity signal respectively. and vehicle yaw rate signal .
9. The braking force distribution method based on MPC according to claim 1, characterized in that, The step of setting a reference trajectory for the objective function based on the total braking force determined by the pedal braking signal includes: Driver braking intention analysis is completed based on brake pedal opening signal: using pedal travel and pedal force as dual input parameters, a lookup operation is performed through a pre-calibrated mapping table to calculate the driver's desired total braking force of the vehicle. Based on the total braking force, a reference trajectory for the objective function is set.
10. A braking force distribution system based on MPC, characterized in that, include: Electronic control unit and EMB actuator; among which, The electronic control unit is used for: Acquire vehicle status data; wherein, the vehicle status data includes vehicle longitudinal speed signal, vehicle yaw rate signal, and pedal braking signal; Based on the total braking force determined by the pedal braking signal, a reference trajectory for the objective function is set; the reference trajectory is at least determined by the longitudinal velocity reference. Yaw velocity reference and optimal slip ratio Characterized by at least two reference terms; Based on model predictive control, a multi-objective optimization problem of the prediction model of MPC is solved in the rolling time domain; wherein, the multi-objective optimization problem includes multiple penalty terms corresponding to the vehicle's longitudinal velocity signal, the vehicle's yaw rate signal, and the reference terms in the reference trajectory. Based on the solved control quantity, the corresponding braking force is distributed to the EMB actuators configured on each wheel.
11. The braking force distribution system based on MPC according to claim 10, characterized in that, The electronic control unit is also used for: Construct the state-space equations for the vehicle dynamics model: The state-space equations include the following two items: Longitudinal dynamic equation: ； The yaw dynamics equation: ； in, It's about vehicle quality. It is the vehicle's longitudinal speed signal. It is a longitudinal acceleration signal. It is the longitudinal braking force of the front left wheel. It is the longitudinal braking force of the front right wheel. It is the longitudinal braking force of the rear left wheel. It is the longitudinal braking force of the rear right wheel. It is the moment of inertia of the vehicle about its vertical axis. It is the vehicle yaw rate signal. It is the yaw rate acceleration signal. It is a signal indicating the distance between the wheels on the left and right sides of the front and rear axles of the vehicle; The longitudinal and yaw dynamic equations of the vehicle dynamics model are discretized to obtain the prediction model of MPC.
12. The braking force distribution system based on MPC according to claim 11, characterized in that, The electronic control unit is also specifically used for: by Let be the state vector, and by To control the vectors, the longitudinal and yaw dynamic equations are rewritten in state-space form: Where A is the system matrix and B is the input matrix; The equations, rewritten in state-space form, are discretized using the forward Euler method to obtain the prediction model for MPC.
13. The braking force distribution system based on MPC according to claim 10, characterized in that, The objective function of the multi-objective optimization problem includes at least two of the following: a first penalty term; a second penalty term; a third penalty term; and a fourth penalty term. The first penalty term is used to penalize the vehicle's longitudinal speed signal. and The deviation term, the second penalty term is used to penalize the vehicle yaw rate signal. and The deviation; the third penalty term is used to penalize the control quantity in the k-th control cycle. Control quantity of the (k-1)th control cycle The deviation; the fourth penalty term is used to penalize the wheel slip ratio. and Deviation; The constraints of the multi-objective optimization problem include at least two of the following: in, The maximum braking force provided to the EMB actuator; To prevent sudden changes in braking force at the boundary; , and These are the lower and upper limits of the slip ratio, respectively. For tire elliptical constraints; For longitudinal braking force; It is a lateral braking force; The road surface adhesion coefficient; This refers to the vertical force on the tire. These correspond to the front right wheel, front left wheel, rear right wheel, and rear left wheel, respectively.
14. The braking force distribution system based on MPC according to claim 13, characterized in that, The objective function of the multi-objective optimization problem is: in, This is the first penalty item; This is the second penalty item; This is the third penalty item; This is the fourth penalty item; R, S, and P are the weight matrices; the wheel slip ratio is... ; The wheel's angular velocity; The radius of the wheel; For prediction in the time domain; For the control time domain; k is the current control period, and (k+ii) is the control period after ii control periods starting from the k-th control period.
15. The MPC-based brake force distribution system according to any one of claims 10-14, characterized in that, When the electronic control unit is configured to solve a multi-objective optimization problem of the prediction model of MPC in the rolling time domain, it is also specifically used for: In the rolling time domain, the multi-objective optimization problem of the prediction model of MPC is solved by an optimization solver; wherein the optimization solver is configured with a maximum number of iteration steps.
16. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by at least one of the processors to enable at least one of the processors to perform the MPC-based braking force distribution method as described in any one of claims 1-9.
17. A storage medium, characterized in that, The storage medium stores computer instructions, which, when executed by the computer, are used to perform the MPC-based braking force distribution method as described in any one of claims 1-9.
18. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the MPC-based braking force distribution method as described in any one of claims 1-9.