EMB clamping force control method based on hybrid optimization strategy

Abstract

The invention belongs to the technical field of vehicle control, and relates to an EMB clamping force control method based on a hybrid optimization strategy, which comprises the following steps of: after modeling an actuating mechanism of an electronic mechanical braking system, automatically judging the stage of the system according to a comparison result of a target clamping force instruction and an actual clamping force feedback value, and controlling the clamping force of the system according to the comparison result. Comprising a brake clearance eliminating stage, a clamping force following control stage and a brake clearance forming stage after braking is finished, corresponding controllers are dynamically called according to all the stages, and controlled clamping force is output. According to the method, the response speed, the control precision and the robustness of the EMB system are comprehensively improved, buffeting of sliding mode control is effectively inhibited, discretization implementation of an algorithm is optimized, and the application requirement of a vehicle-mounted embedded system is met.

Description

Technical Field

[0001] This invention belongs to the field of vehicle control technology and relates to an EMB clamping force control method based on a hybrid optimization strategy. Background Technology

[0002] With the continuous development of advanced autonomous driving technology, the safety and responsiveness of vehicles in complex dynamic environments place higher demands on braking systems. Electro-Mechanical Braking (EMB), due to its millisecond-level response speed and high-precision control capabilities, has gradually replaced traditional hydraulic braking systems, becoming one of the core actuators in autonomous vehicles. However, the following technical challenges still exist in the practical application of EMB systems.

[0003] First, there's the chattering problem in sliding mode control: While sliding mode control boasts strong robustness, its inherent chattering issue leads to actuator wear and reduces system lifespan and stability. Existing methods offer limited effectiveness in suppressing chattering, failing to meet the high stability and reliability requirements of autonomous vehicles' braking systems. These issues severely limit the performance of EMB systems in high-level autonomous driving.

[0004] Second, there is a problem with the insufficient phased control strategy: The braking process of the EMB system includes three phases: gap elimination, clamping force following, and brake release. The dynamic characteristics and control requirements of each phase differ significantly. The gap elimination phase requires rapid establishment of the initial clamping force to ensure that the braking response time is minimized; the clamping force following phase requires precise control of the target clamping force to ensure the stability and safety of the vehicle, but it suffers from uncompensated nonlinear disturbances; the brake release phase requires the control system to smoothly release the clamping force to avoid vehicle swaying. However, existing technologies lack phased optimized control strategies for these phases, resulting in the system's inability to simultaneously achieve response speed, control accuracy, and robustness under complex operating conditions, making it difficult to fully meet the high-performance requirements of autonomous vehicles.

[0005] Third, there is the problem of insufficient nonlinear disturbance compensation: EMB systems exhibit significant nonlinearity, time-varying characteristics, and multivariable coupling. Since the actuators in EMB systems typically contain complex mechanical transmission chains, such as ball screws and reducers, the nonlinear characteristics of these transmission mechanisms introduce significant dynamic disturbances, further reducing the stability and accuracy of the control system. Furthermore, the dynamic friction characteristics between the brake pads and brake discs, the elastic deformation of the actuators, and clearance compensation issues also increase the difficulty of designing the control algorithm. Under emergency braking or complex operating conditions, traditional PID controllers, due to their fixed gain settings, struggle to adapt to system dynamic disturbances, resulting in insufficient control accuracy.

[0006] Fourth, the engineering implementation of control algorithms: The implementation of existing complex control algorithms on automotive embedded microcontrollers is limited by computing power, making it difficult to balance real-time performance and efficiency in practical applications. How to simplify and discretize high-performance control algorithms to ensure their adaptability to embedded systems is one of the current technical challenges. Summary of the Invention

[0007] To address the aforementioned technical problems in existing technologies, this invention proposes an EMB clamping force control method based on a hybrid optimization strategy, providing a safer and more efficient braking solution for high-level autonomous vehicles. The specific technical solution is as follows:

[0008] An EMB clamping force control method based on a hybrid optimization strategy, after modeling the actuator of the electromechanical braking system, automatically determines the stage of the system based on the comparison between the target clamping force command and the actual clamping force feedback value, including the brake gap elimination stage, the clamping force following control stage, and the brake gap formation stage after braking ends, and dynamically calls the corresponding controller according to each stage to output the controlled clamping force.

[0009] Furthermore, the modeling of the actuator mainly includes the construction of the drive motor model, drive motor load model, transmission mechanism model, and brake disc model.

[0010] Furthermore, the drive motor model is constructed using a permanent magnet synchronous motor. The AC electrical quantities of the motor, which vary with the rotor position in a three-phase stationary coordinate system, are equivalently transformed to a two-phase coordinate system that rotates synchronously with the rotor through Clarke transformation and Park transformation.

[0011] The drive motor load model is constructed using the Stribeck model;

[0012] The transmission mechanism model is constructed based on the angular displacement of the motor rotor, the lead of the ball screw, the linear displacement of the screw nut, and the reduction ratio of the planetary gear reducer, establishing the main kinematic transmission relationships.

[0013] The brake disc model is constructed based on the normal pressure applied to the surface of the brake disc by the brake pad on one side, the dynamic friction coefficient between the brake pad and the contact surface of the brake disc, and the effective radius of the braking force, to obtain the total braking torque output by the brake.

[0014] Furthermore, in the drive motor load model, the braking gap between the brake pad and the brake disc in the physical structure of the brake is compensated. When the linear displacement of the lead screw nut is greater than the braking gap, the brake pad begins to contact the brake disc and undergoes compression deformation, thereby generating a clamping force.

[0015] Furthermore, during the braking gap elimination phase, an initial clamping force is established through torque control, and a PI controller is used to correct the error between the actual clamping force value and the target value.

[0016] Furthermore, in the clamping force following control stage, a dual-loop cascaded sliding mode controller is adopted, including:

[0017] The outer ring clamping force tracking controller, as the main sliding diaphragm controller, is used to track the target clamping force and generate control variables;

[0018] The inner-loop disturbance suppression sliding mode controller, acting as a slave sliding mode controller, adjusts the actual control input according to the control variables and compensates for system disturbances in real time through the inner-loop disturbance observer.

[0019] Furthermore, the dual-loop cascaded sliding mode controller introduces the boundary layer method, using a continuous saturation function to replace the sign function in order to reduce chattering.

[0020] Furthermore, the outer ring clamping force tracking controller includes two stages for tracking the target clamping force:

[0021] In the first stage, a tracking control law is designed using a sliding surface to enable the system to quickly approach the target clamping force under dynamic working conditions.

[0022] In the second stage, Lyapunov stability theory is incorporated to maintain the global stability of the system.

[0023] Furthermore, during the braking gap formation stage after braking ends, the clamping force and displacement are dynamically adjusted by a multi-level PI controller, combining the target clamping force and the displacement feedback signal from the motor side.

[0024] Furthermore, the electromechanical braking system employs a signal conditioning module that performs coordinated scheduling and switching through multi-mode switching logic. This logic determines the stage of the system based on the comparison between the target clamping force command and the actual clamping force feedback value.

[0025] The present invention has the following beneficial effects:

[0026] Fast dynamic response and reduced braking delay: The control algorithm proposed in this invention, through optimization of the dual-loop control structure and sliding mode control strategy, enables the system to exhibit excellent dynamic response performance when achieving the target clamping force. Experiments show that the system's response time can be consistently maintained within 0.3 seconds, and even under conditions of rapidly changing target clamping force, it can still quickly follow the target force value, significantly reducing the response delay of the braking system. Compared with traditional control methods, the fast response performance of the algorithm in this invention can effectively reduce the time required for emergency braking, providing an important guarantee for the braking safety of vehicles in high-speed or complex dynamic environments. In addition, this characteristic can meet the stringent requirements of high-level autonomous vehicles for real-time performance and efficiency, buying more time for driving decisions in emergencies, thereby improving the overall system reliability and safety.

[0027] High-precision tracking with no overshoot and steady-state error: The control algorithm of this invention possesses precise target clamping force tracking capability, rapidly converging to the target value under dynamic conditions, with a steady-state error approaching zero. Through optimization of the sliding surface design and application of Lyapunov stability theory, the system not only achieves high-precision control throughout the tracking process but also completely avoids the overshoot problem caused by improper parameter adjustment in traditional control methods. This overshoot-free characteristic effectively reduces the impact of force overshoot on system stability, ensuring the smoothness and safety of the braking process. Furthermore, even under high dynamic loads or complex braking conditions, the algorithm of this invention can still maintain accurate target force tracking capability, significantly improving the control accuracy of electromechanical braking (EMB) systems and providing higher safety and comfort guarantees for advanced autonomous vehicles.

[0028] High stability and elimination of high-frequency oscillations: This invention fundamentally solves the inherent high-frequency chattering problem of sliding mode controllers by introducing the boundary layer saturation function and dynamic gain optimization design of sliding mode control. Throughout the entire operation, the clamping force output curve of the system exhibits smooth and stable characteristics, without any high-frequency oscillations or fluctuations. Eliminating high-frequency chattering not only improves the robustness and stability of the system but also effectively reduces the mechanical wear and energy consumption of the actuator, significantly extending the service life of the system. In addition, this high stability enables the vehicle to have higher running smoothness during braking, avoiding vehicle swaying and passenger discomfort caused by braking oscillations, thereby further improving the user experience. In summary, the significant advantages of this invention in terms of stability fully meet the braking requirements under various complex working conditions and have broad engineering application value. Attached Figure Description

[0029] Figure 1 This is a diagram illustrating the process of transforming the three-phase physical model of the motor into a two-phase physical model using coordinate transformation in this embodiment.

[0030] Figure 2 These are the friction characteristic curves of the Coulomb viscosity model, Stribeck model, and LuGre dynamic model in this embodiment.

[0031] Figure 3 This is a model diagram of the transmission mechanism in this embodiment;

[0032] Figure 4 This is a block diagram of the composite closed-loop control strategy in this embodiment;

[0033] Figure 5 This is a schematic diagram of the ideal control law path of the outer ring clamping force tracking controller in this embodiment;

[0034] Figure 6 This is a schematic diagram of the control law path of the inner loop disturbance suppression sliding mode controller in this embodiment;

[0035] Figure 7 This is a graph showing the sign function and saturation function of this embodiment;

[0036] Figure 8 This is a schematic diagram of the saturation function in this embodiment;

[0037] Figure 9 This is a comparison diagram of the clamping force following effect of the dual-loop cascaded sliding mode control and PID control in this embodiment;

[0038] Figure 10 This is a simulation diagram of clamping force following under different target clamping forces in this embodiment. Detailed Implementation

[0039] To make the objectives, technical solutions, and technical effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0040] This embodiment provides an electromechanical braking (EMB) clamping force control method based on a hybrid optimization strategy. It is a method based on dual-loop cascaded sliding mode control (SMC). First, the actuator of the electromechanical braking system is modeled. Then, based on the clamping force feedback, different working stages of the system are determined: the braking gap elimination stage, the clamping force following control stage, and the braking gap formation stage after braking ends. Finally, for each stage, a composite closed-loop control strategy is designed and coordinated to achieve rapid response, high-precision tracking, and stable control of the desired clamping force, i.e., the target clamping force.

[0041] The modeling of the actuator mainly includes the construction of the drive motor model, drive motor load model, transmission mechanism model, and brake disc model.

[0042] The construction of the drive motor model is as follows:

[0043] In the overall design and modeling of the EMB system, the actuator consists of a high-response, high-precision drive motor and a mechanical transmission device, and its performance directly determines the response speed and control accuracy of the braking system. In autonomous driving scenarios, the stringent requirements for the braking system make permanent magnet synchronous motors (PMSMs) the ideal choice for drive motors in EMB systems due to their high power density, high efficiency, and excellent dynamic response characteristics.

[0044] Based on the above premises, this invention simplifies the complex, time-varying three-phase physical model of a motor using coordinate transformation theory. The core idea is to use Clarke and Park transformations to equivalently transform the AC electrical quantities that vary with the rotor position in the three-phase stationary coordinate system (ABC) to the dq coordinate system that rotates synchronously with the rotor. In this coordinate system, each electrical variable is represented as a DC quantity in steady state, thereby achieving model decoupling and dimensionality reduction, and simplifying control algorithm design. The coordinate transformation process is as follows: Figure 1 As shown.

[0045] like Figure 3 As shown, the construction of the transmission mechanism model is as follows:

[0046] The transmission mechanism takes the rotation angle θ of the electric drive unit as input and outputs the linear displacement s at the end of the brake actuator. The model's transfer operator mainly consists of a planetary gear reducer and a ball screw pair cascaded together. The planetary gear reducer serves as a torque amplification and speed decoupling module, while the ball screw pair is responsible for realizing the motion transformation from the rotational domain to the translational domain.

[0047] In high-dynamic real-time control scenarios for autonomous driving, the nonlinear and time-varying characteristics of the transmission chain, such as meshing efficiency, contact stiffness, friction, and clearance, pose significant challenges to controller design. Therefore, in the initial modeling stage, this invention equates the multibody transmission system to a single-degree-of-freedom ideal kinematic mapping, temporarily ignoring the influence of the aforementioned nonlinear factors to establish its primary kinematic transmission relationships. This ideal kinematic model can be characterized by the following equation:

[0048] ,

[0049] Where θ is the angular displacement of the motor rotor, L is the ball screw lead, s is the linear displacement of the screw nut, and i is the reduction ratio of the planetary gear reducer.

[0050] The construction of the brake disc model is as follows:

[0051] The ultimate control output of the brake is the braking torque applied to the wheels. The product of the clamping force and the coefficient of friction between the materials is the tangential frictional force generated on the contact surface. This frictional force acts at a distance R from the center of the brake disc. b At the effective radius, a frictional torque is formed that opposes the rotation of the brake disc. Since the brake caliper clamps the disc from both sides, considering the frictional effect on both sides, the total braking torque output by the brake is... Mathematical modeling can be performed using the following formula:

[0052] ,

[0053] Among them, F n This indicates the normal force applied by a single brake pad to the surface of the brake disc; f b R is the dynamic friction coefficient between the brake pad and the brake disc contact surface. b The effective radius of the braking force.

[0054] The construction of the drive motor load model is as follows:

[0055] The classic Coulomb viscosity model, due to its overly simplified structure, cannot describe the transition characteristics from rest to motion, i.e., the Stribeck effect. While dynamic models such as LuGre offer higher accuracy, their parameter identification is complex and computationally intensive, making them unsuitable for the real-time requirements of autonomous driving scenarios. Figure 2 As shown.

[0056] Therefore, this embodiment selects the Stribeck model to describe the friction characteristics of the servo system. The Stribeck model achieves a good balance between model accuracy and engineering implementation complexity. It can comprehensively characterize the nonlinear friction characteristics of the system in the static, low-speed, and high-speed regions, and in particular, accurately describes the process of smoothly transitioning from maximum static friction to dynamic friction, thus effectively capturing the Stribeck effect, i.e., the negative damping characteristic where friction decreases with increasing speed at low speeds. At the same time, its model structure is relatively simple, facilitating parameter identification and real-time compensation calculations. Its specific expression is as follows:

[0057] ,

[0058] Among them, T f It is the total frictional torque, T s It is the maximum static friction torque (Stiction Torque), T c It is the Coulomb friction torque, T s >T c b is the viscous friction coefficient, which describes the portion of the frictional torque that increases linearly with speed at higher speeds. It is the mechanical angular velocity of the motor. δ is the Stribeck velocity, a small positive velocity threshold used to characterize the velocity range in which friction transitions from static friction to Coulomb friction. δ is an empirical index, and in this embodiment, δ=1.

[0059] In the physical structure of a brake, there is inevitably an initial gap between the brake pads and the brake disc, defined as the brake gap s0. This gap is a critical nonlinear element in the system and must be compensated for in the controller design. The linear displacement s of the lead screw nut must first overcome this gap; this process is ineffective and generates no braking force. The brake pads begin to contact the brake disc and undergo compressive deformation if and only if s > s0, thereby generating a clamping force F. n Therefore, the effective compressive displacement that can generate clamping force is defined. for:

[0060] ,

[0061] Clamping force F n With effective compressive displacement The relationship between them exhibits significant nonlinear stiffness characteristics. This nonlinearity mainly stems from two aspects: 1) the nonlinear elastic behavior of the brake pads themselves; and 2) the elastic deformation of load-bearing structural components such as brake calipers under enormous clamping forces. This complex nonlinear relationship can be approximated with high accuracy by a third-order polynomial model:

[0062] ,

[0063] in, These are the polynomial fitting coefficients characterizing the stiffness properties of the system.

[0064] like Figure 4 As shown, the design of the composite closed-loop control strategy is as follows:

[0065] During the braking gap elimination phase, the system quickly establishes the initial clamping force through basic torque control. A simple PI controller is used to correct the error between the actual clamping force value and the target value, with a focus on response speed and initial stability requirements.

[0066] In the clamping force tracking control stage, a dual-loop cascaded sliding mode controller (SMC) is employed, comprising an outer-loop clamping force tracking controller and an inner-loop disturbance suppression sliding mode controller. The outer-loop clamping force tracking controller, acting as the master sliding mode controller, aims to eliminate the tracking error between the target force and the actual system output force. It dynamically generates an intermediate control variable based on the error, which can be considered as the desired control action under ideal, disturbance-free conditions. The inner-loop disturbance suppression sliding mode controller, acting as the slave sliding mode controller, ensures that the actual effective action applied to the controlled object accurately reproduces the outer-loop command. It generates the final physical control output by actively compensating for measurable lumped disturbances, thereby isolating the impact of disturbances on force tracking performance.

[0067] The outer ring clamping force tracking controller is designed to achieve precise tracking of the target clamping force. In the first stage, the control utilizes a sliding mode surface to design a tracking control law, rapidly approximating the target clamping force to ensure a fast response under dynamic conditions. In the second stage, a dynamic gain optimization strategy is combined to further improve steady-state accuracy, avoiding the overshoot problem common in traditional sliding mode control. Simultaneously, Lyapunov stability theory ensures the global stability of the system in complex dynamic environments, while enhancing the system's robustness to external disturbances. This design significantly improves the system's dynamic response performance and control accuracy, providing reliable technical support for the efficient operation of electromechanical braking (EMB) systems under complex dynamic conditions.

[0068] More specifically, the core of the outer ring clamping force tracking controller is force tracking. First, the system tracking error is defined. and its derivative:

[0069] ,

[0070] Among them, F n，tar It is the clamping force of the target, F n,act It is the actual clamping force;

[0071] To make the error Since both the sliding surface and its rate of change approach zero, a linear sliding surface is designed. :

[0072] ,

[0073] in It is an adjustable sliding surface coefficient. When the system state is constrained to the sliding surface... When this occurs, the error dynamics will follow the following first-order linear differential equation:

[0074] ,

[0075] Once the sliding mode is entered, the tracking error will converge to zero exponentially, with the convergence rate increasing from [previous value]. The size is determined by this. In order to drive the system state from any initial position to the sliding surface within a finite time... To maintain this stability, a suitable control law needs to be designed. This is equivalent to satisfying the Lyapunov stability condition, and the following Lyapunov candidate functions are selected:

[0076] ,

[0077] Its derivative is To ensure It employs a variation of the exponential law of convergence that combines a proportional term and a switching term:

[0078] ,

[0079] in, and These are the proportional approach gain and the constant velocity approach gain, respectively. Combining this with system dynamics, the outer-loop control can be derived. When the model is unknown or uncertain, control laws can be designed directly. To enforce the convergence law, such as Figure 5 As shown. The control law designed in this invention is in the form of:

[0080] ,

[0081] in, Ensure when When the size is large, the system can approach the sliding surface at a faster speed; The terms provide robustness against system uncertainties and external disturbances, and are the core guarantee of the robustness of sliding mode control.

[0082] The goal of the inner-loop disturbance suppression sliding mode controller is to enable the system to respond to disturbances... Immunization, ensuring effective control, is equivalent to external control. Define the inner loop error. for:

[0083] ,

[0084] Similarly, for the inner loop error Design sliding surface :

[0085] ,

[0086] in It is the inner ring sliding surface coefficient. When At that time, error will be by The determined rate exponent converges to zero, which means that the perturbation The impact was fully compensated.

[0087] Similar to the outer loop, the inner loop's final control output... Similarly, a control law combining proportional and switching terms is used, such as Figure 6 As shown, this is to achieve fast and robust perturbation suppression:

[0088] ,

[0089] in, and These are the proportional gain and switching gain of the inner loop, respectively. This control law drives... Approaching zero, thus forcing the system's actual behavior to conform to the expectations of the outer loop.

[0090] To address the inherent high-frequency chattering problem in sliding mode control, this invention proposes a chattering suppression technique based on a boundary layer saturation function. By replacing the sign function in the traditional sliding mode controller with a saturation function, the control output is significantly smoothed, reducing the amplitude of high-frequency vibrations. Simultaneously, this invention optimizes the sliding mode gain parameters based on the system's dynamic characteristics, ensuring a fast system response while suppressing chattering. Furthermore, this method achieves a balance between chattering suppression and response speed by dynamically adjusting the control gain parameters. This technique not only extends the lifespan of the actuator but also significantly improves the stability and reliability of the control system, enabling the EMB system to operate efficiently under high-frequency loads and complex operating scenarios.

[0091] More specifically, in practical applications, sign functions The discontinuity of these switching events leads to high-frequency switching, or chattering, in the controller output. Chattering can wear down the actuator and potentially excite unmodeled high-frequency dynamics in the system. To address this issue, this invention employs the boundary layer method, using a continuous saturation function. To replace the symbolic function, such as Figure 7 and Figure 8 As shown.

[0092] The saturation function is defined as:

[0093] ,

[0094] in It is the thickness of the boundary layer. This indicates that within the boundary layer, the control action is linear, avoiding discontinuous switching and thus resulting in a smooth control output. The modified control law is:

[0095] .

[0096] In digital controllers, the above continuous-time algorithm needs to be discretized. Let the sampling period be... The current sampling time is k. The derivative of the error is approximated using a first-order backward difference:

[0097] ,

[0098] Substituting this approximation into the definitions of the sliding surface and the control law, we obtain the complete discrete-time control algorithm, including:

[0099] Outer loop discretization algorithm:

[0100] ,

[0101] Inner loop discretization algorithm:

[0102] .

[0103] During the braking gap formation stage after braking ends, a multi-variable collaborative control strategy is further introduced. Combining clamping force and displacement feedback signals, a multi-level PI controller is used to achieve dynamic collaborative adjustment of clamping force and displacement.

[0104] These three independent control strategies do not operate in isolation, but are coordinated and seamlessly switched through a well-defined multi-mode switching control logic. This logic automatically determines which stage the system should be in based on the comparison between the target clamping force command and the actual clamping force feedback value, thereby dynamically invoking the corresponding controller algorithm. The expression is as follows:

[0105] ,

[0106] F n,th This is the clamping force threshold.

[0107] This invention is based on an electromechanical braking (EMB) system architecture using a dual-loop cascaded sliding mode controller (SMC). It employs a modular control strategy, achieving rapid response and high-precision tracking of the target clamping force through outer-loop sliding mode control. The outer-loop control module constructs a target tracking control law based on the sliding surface, incorporating Lyapunov stability theory to ensure the system's global stability and rapid convergence performance under dynamic conditions.

[0108] Furthermore, the input target force is optimized in real time through a signal conditioning module to ensure the smoothness and stability of the control signal and avoid oscillations. The actuator section adopts a high-precision motor drive and a force sensor feedback closed-loop structure, further improving the system's control accuracy and dynamic response performance. Experimental verification has shown that this architecture can achieve efficient and stable clamping force control under various dynamic environments, fully demonstrating its application value in high-level autonomous vehicles.

[0109] The dual-loop cascaded sliding mode controller (SMC) proposed in this invention exhibits significant advantages in dynamic response performance, tracking accuracy, and system robustness, such as... Figure 9 As shown in the figure, experimental comparisons with traditional PID control methods reveal that the electromechanical braking (EMB) system employing the algorithm of this invention exhibits a faster dynamic response speed when the target clamping force changes. For example, during the change of clamping force from 0 to 20,000 Newtons, SMC control completes the response within 0.2 seconds, while PID control requires approximately 0.3 seconds. Furthermore, the dual-loop cascaded SMC algorithm can accurately track the target clamping force value, with a steady-state error approaching zero and a significantly smaller overshoot than PID control, demonstrating higher control accuracy. Simultaneously, by using an inner-loop disturbance observer to compensate for nonlinear disturbances in real time, the algorithm of this invention can maintain a smooth and stable control curve even under external disturbances and sudden input changes, avoiding the oscillation phenomenon common in PID control and significantly improving system robustness.

[0110] Experimental data fully demonstrate that the dual-loop cascaded sliding mode controller (SMC) proposed in this invention can meet the requirements of high-level autonomous vehicles for rapid response, high-precision control and high robustness of electromechanical braking systems, and has broad engineering application value.

[0111] The control algorithm proposed in this invention exhibits superior performance in terms of fast response, high-precision tracking, and stable control under different target clamping forces, including 5000N, 10000N, 15000N, 20000N, 25000N, and 30000N. Figure 10 As shown in the figure. Experimental results show that the dynamic response time of the system is consistently within 0.3 seconds, enabling rapid establishment of the target clamping force and significantly shortening the braking response time, meeting the rapid response requirements of high-level autonomous vehicles for emergency braking. Furthermore, the actual clamping force value and the target value completely coincide in steady state, with a steady-state error close to zero and no overshoot or oscillation, significantly improving control accuracy and avoiding the force overshoot problem caused by overshoot in traditional methods. The system maintains consistent dynamic response characteristics and steady-state performance across different target clamping force ranges, demonstrating its good adaptability and robustness to various operating conditions. Throughout the response process, the clamping force curve exhibits no high-frequency oscillations or fluctuations, showcasing the system's high stability and further enhancing the reliability of the braking system and the smoothness of vehicle operation.

[0112] In summary, the control algorithm of this invention achieves comprehensive optimization of the electromechanical braking (EMB) system through fast response, accurate tracking and high stability, and has broad engineering application value.

[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any way. Although the implementation process of the present invention has been described in detail above, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An EMB clamping force control method based on a hybrid optimization strategy, characterized in that, After modeling the actuator of the electromechanical braking system, the system automatically determines the stage it is in based on the comparison between the target clamping force command and the actual clamping force feedback value. This includes the braking gap elimination stage, the clamping force following control stage, and the braking gap formation stage after braking ends. The system then dynamically calls the corresponding controller according to each stage and outputs the controlled clamping force.

2. The EMB clamping force control method as described in claim 1, characterized in that, The modeling of the actuator mainly includes the construction of the drive motor model, drive motor load model, transmission mechanism model, and brake disc model.

3. The EMB clamping force control method as described in claim 2, characterized in that, The drive motor model is constructed using a permanent magnet synchronous motor. The AC electrical quantities of the motor, which vary with the rotor position in a three-phase stationary coordinate system, are equivalently transformed to a two-phase coordinate system that rotates synchronously with the rotor through Clarke and Park transformations. The drive motor load model is constructed using the Stribeck model; The transmission mechanism model is constructed based on the angular displacement of the motor rotor, the lead of the ball screw, the linear displacement of the screw nut, and the reduction ratio of the planetary gear reducer, establishing the main kinematic transmission relationships. The brake disc model is constructed based on the normal pressure applied to the surface of the brake disc by the brake pad on one side, the dynamic friction coefficient between the brake pad and the contact surface of the brake disc, and the effective radius of the braking force, to obtain the total braking torque output by the brake.

4. The EMB clamping force control method as described in claim 3, characterized in that, In the drive motor load model, the braking gap between the brake pad and the brake disc in the physical structure of the brake is compensated. When the linear displacement of the lead screw nut is greater than the braking gap, the brake pad begins to contact the brake disc and undergoes compression deformation, thereby generating a clamping force.

5. The EMB clamping force control method as described in claim 1, characterized in that, During the braking gap elimination phase, an initial clamping force is established through torque control, and a PI controller is used to correct the error between the actual clamping force value and the target value.

6. The EMB clamping force control method as described in claim 1, characterized in that, In the clamping force following control stage, a dual-loop cascaded sliding mode controller is adopted, including: The outer ring clamping force tracking controller, as the main sliding diaphragm controller, is used to track the target clamping force and generate control variables; The inner-loop disturbance suppression sliding mode controller, acting as a slave sliding mode controller, adjusts the actual control input according to the control variables and compensates for system disturbances in real time through the inner-loop disturbance observer.

7. The EMB clamping force control method as described in claim 6, characterized in that, The dual-loop cascaded sliding mode controller introduces the boundary layer method, using a continuous saturation function to replace the sign function in order to reduce chattering.

8. The EMB clamping force control method as described in claim 6, characterized in that, The outer ring clamping force tracking controller tracks the target clamping force in two stages: In the first stage, a tracking control law is designed using a sliding surface to enable the system to quickly approach the target clamping force under dynamic working conditions. In the second stage, Lyapunov stability theory is incorporated to maintain the global stability of the system.

9. The EMB clamping force control method as described in claim 1, characterized in that, During the braking gap formation stage after braking ends, the clamping force and displacement are dynamically adjusted by a multi-level PI controller, combining the target clamping force and the displacement feedback signal from the motor side.

10. The EMB clamping force control method as described in claim 1, characterized in that, The electromechanical braking system employs a signal conditioning module, which coordinates and switches modes through multi-mode switching logic. This logic determines the stage of the system based on the comparison between the target clamping force command and the actual clamping force feedback value.

Images