An EMB system brake clearance full life cycle online estimation method and system

Abstract

This invention discloses an online method and system for estimating the braking clearance throughout the entire lifecycle of an EMB system, relating to the field of braking technology for EMB systems in autonomous vehicles. The invention includes: acquiring motor current and motor angle signals of the electromechanical braking system during the braking clamping phase; constructing a current-angle regression model based on a Legendre orthogonal basis based on the nonlinear relationship between the motor current and motor angle signals; performing online robust estimation of the parameter vector of the current-angle regression model using a weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M-estimator and a two-stage estimation strategy; and mapping the parameter estimates to braking clearance estimates. This invention achieves high-precision braking clearance estimation without additional sensors based solely on motor current and angle signals through online self-learning. It maintains excellent robustness and stability under complex operating conditions and long-term operation, making it suitable for engineering applications in EMB systems.

Description

Technical Field

[0001] This invention relates to the field of braking technology for EMB systems in autonomous vehicles, specifically to an online estimation method and system for the entire lifecycle of braking clearance in EMB systems. Background Technology

[0002] Electro-Mechanical Braking (EMB) systems, as one of the core development directions of brake-by-wire technology, directly drive the brake actuator with an electric motor, replacing traditional hydraulic or pneumatic lines. This offers advantages such as fast response, high control precision, compact structure, and easy integration with the vehicle's overall control system. In EMB systems, brake clearance (the physical gap between the brake pads and the brake disc in the unbraked state) is a critical parameter. Accurate brake clearance estimation is essential for achieving efficient clamping force control, preventing drag torque, compensating for brake pad wear, ensuring braking safety, and improving energy recovery efficiency.

[0003] Currently, the industry mainly uses the following methods to estimate or detect the brake clearance of EMB systems: 1. Threshold-based detection methods, which use threshold triggering methods based on sudden changes in current, torque, or rotation angle, or judge by the cumulative stroke difference. These methods are difficult to adapt to the time-varying characteristics of parameters caused by factors such as brake pad wear, temperature changes, and system aging, and are prone to misjudgment under complex working conditions; 2. Offline fitting model-based methods require the introduction of force sensors or offline calibration to indirectly determine the brake clearance or contact position. This not only increases the system hardware cost and structural complexity, but also makes it difficult to achieve online updates of the brake clearance, and is not suitable for the long-term operation requirements of mass-produced electromechanical braking systems. Therefore, this invention proposes an online estimation method and system for the entire life cycle of braking clearance in an EMB system. Summary of the Invention

[0004] The purpose of this invention is to provide an online method and system for estimating the brake clearance throughout the entire life cycle of an EMB system. Based on motor current and rotation angle signals, it achieves high-precision estimation of the brake clearance without the need for additional sensors through online self-learning. It still has excellent robustness and stability under complex working conditions and long-term operation, and is suitable for engineering applications of EMB systems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an online estimation method for the entire life cycle of braking clearance in an EMB system, comprising the following steps: Acquire motor current signal of electromechanical braking system during braking clamping phase and motor rotation angle signal ; Based on the nonlinear relationship between motor current signal and motor rotation angle signal, a current-rotation angle regression model based on Legendre orthogonal basis is constructed. The current-rotation angle regression model includes a regression vector based on Legendre orthogonal basis functions and a parameter vector to be estimated. We employ a weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M-estimator to perform online robust estimation of the parameter vector of the current-turn angle regression model. We also utilize a two-stage recursive solution strategy to perform efficient and low-cost computation of the parameter estimation process, thereby obtaining the parameter vector estimates. Extrapolating the parameter vector estimates based on the zero-current condition maps the parameter estimates to brake clearance estimates. .

[0006] Furthermore, based on the nonlinear relationship between the motor current signal and the motor rotation angle signal, a current-rotation angle regression model based on the Legendre orthogonal basis is constructed, as follows: (21) Collect the motor current of the electromechanical braking system during the braking clamping phase. and motor rotation angle With a sampling period of 0.005s, the following regression model is established using the nonlinear relationship between motor current and rotation angle: (1) in, This indicates the motor current during the loading phase. Indicates the motor rotation angle during the loading phase. Indicates the braking clearance. , , This indicates the nonlinear characteristics of the system; (22) Regarding the applied current Perform linear normalization to interval : (2) in, This indicates the preset maximum current value under actual measurement conditions. Auxiliary variables representing linear normalization; (23) Legendre orthogonal basis in the domain It has orthogonality, and the weight function The first four terms are in the form of an orthogonal polynomial as follows: (3) (24) Construct a new regression vector Define parameter vector Combined with formula The new regression model for current-rotation angle based on the Legendre orthogonal basis is as follows: (4) In the formula, This represents the vector of parameters to be estimated. Let T represent the regression vector based on Legendre orthogonal basis functions, and let T denote the transpose.

[0007] Furthermore, the cost function of the weighted multi-forgetting-factor recursive least squares algorithm is defined as: (5) (6) In the formula, The cost function representing the prediction error, This represents the actual motor rotation angle value. Represents the parameter vector estimated online. This represents the forgetting factor for each estimated parameter. The robust penalty function is defined as follows: (7) in, , , The threshold parameter, which represents the degree of outlier suppression, is set based on the quantiles of the standard normal distribution.

[0008] Furthermore, the adaptive M-estimator introduces a recursive adaptive variance estimation law, as follows: (8) in, Residual The estimated variance, The forgetting factor represents the estimated variance. Represents the finite sampling correction factor. Residual A sliding window.

[0009] Furthermore, define , Robust weights are introduced for updates related to braking clearance. The weight coefficients for the remaining parameter channels are set to 1, and the scalar information content is updated. and gain : (9) in, Represents the weighting coefficient, when hour, Other channels .

[0010] Therefore, according to the formula The parameters are updated in the following recursive form: (10).

[0011] Furthermore, a two-stage recursive solution strategy is used to perform efficient and low-cost computation of the parameter estimation process, as detailed below: Phase 1: Using the ( Step , The estimated value, i.e. As a known condition, recursively update ,Right now: (11) in, This is the estimate from the previous step. The gain of the first-stage estimator is defined as: (12) Phase Two: Use As a known condition, recursively update ,Right now: (13) in, The gain of the second-stage estimator is defined as: (14).

[0012] Furthermore, based on the zero-current condition, the parameter vector estimates are extrapolated to map the parameter estimates to brake clearance estimates: Obtain the current parameter estimate Then, normalize the auxiliary variables. Extrapolating to the theoretical zero current point, i.e. At that time, the estimated value of the braking clearance is obtained. Used for clearance compensation, force control, or condition monitoring: (15) In the formula, These are the estimated parameters based on the regression vectors of the Legendre orthogonal basis.

[0013] According to a second aspect of the present invention, the present invention provides an online estimation system for the entire life cycle of brake clearance in an EMB system, for implementing an online estimation method for the entire life cycle of brake clearance in an EMB system as described in the first aspect, comprising: The data acquisition module is used to acquire the motor current signal of the electromechanical braking system during the braking clamping phase. and motor rotation angle signal ; The model building module is used to construct a current-angle regression model based on the Legendre orthogonal basis based on the nonlinear relationship between the motor current signal and the motor rotation angle signal. The current-angle regression model includes a regression vector based on the Legendre orthogonal basis function and a parameter vector with estimation. The online parameter estimation module is used to perform online robust estimation of the parameter vector of the current-turn angle regression model by using a weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M-estimator. It also uses a two-stage recursive solution strategy to perform efficient and low-cost computation of the parameter estimation process and obtain the parameter vector estimate. The gap estimation module is used to extrapolate the parameter vector estimates based on the zero-current condition, mapping the parameter estimates to brake gap estimates. .

[0014] According to a third aspect of the present invention, the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein when the processor loads and executes the computer program, it employs an online estimation method for the entire life cycle of braking clearance of an EMB system as described in the first aspect.

[0015] According to a fourth aspect of the present invention, the present invention provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform an online estimation method for the entire life cycle of braking clearance in an EMB system as described in the first aspect.

[0016] This invention has at least the following beneficial effects: 1. This invention requires no additional force sensor installation. It utilizes only the inherent motor current and angle signals of the EMB system and an advanced online learning framework to continuously obtain high-precision brake clearance estimates throughout the entire lifecycle. This invention demonstrates excellent adaptability and suppression of slow time-varying parameters caused by brake pad wear, temperature changes, and system aging, as well as sudden interferences such as measurement noise and abnormal data points, ensuring the stability and reliability of the estimation results.

[0017] 2. This invention reconstructs regression variables by introducing Legendre orthogonal basis functions, effectively overcoming the problem of multiple correlations of parameters in traditional multinomial regression, improving the condition number of the model, and thus significantly improving the numerical stability of the algorithm and the convergence speed of parameter estimation in a finite-precision computing environment.

[0018] 3. This invention integrates an adaptive M-estimator into a recursive least squares framework. By recursively estimating the statistical characteristics of the residuals, it dynamically adjusts the robust weights and adaptively suppresses outlier data. Faced with non-Gaussian noise and outlier interference introduced by sensor transient failures or road impacts, this mechanism can automatically reduce the contribution of outlier data to parameter updates, ensuring the continuity and reliability of the estimation results and significantly improving the robustness boundary under complex working conditions.

[0019] 4. The two-stage recursive solution strategy proposed in this invention simplifies the coupled identification problem that originally involved high-dimensional matrix inversion into a two-dimensional matrix operation sequence, reducing the computational complexity from quadratic law to linear order, eliminating inversion operations, significantly reducing the number of multiplication and addition instructions and storage resource consumption, fully meeting the real-time constraints of the millisecond-level scheduling cycle of the embedded controller, and providing computational redundancy for the rapid diagnosis and response required for the functional safety of EMB systems.

[0020] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the method described in this invention; Figure 2 This is a schematic diagram illustrating the implementation principle of the method described in this invention; Figure 3 This is a schematic diagram showing the verification and comparison results of the brake clearance estimation accuracy of the unpolished brake pads in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram showing the verification results of the brake clearance of a moderately worn brake pad in Embodiment 1 of the present invention; Figure 5 This is a schematic diagram illustrating the verification of brake clearance in a severely worn brake pad according to Embodiment 1 of the present invention. Figure 6 This is a schematic diagram illustrating the verification results of the two-stage recursive solution strategy in Embodiment 1 of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0023] Example 1: Please see Figures 1-6 This invention provides a technical solution: an online estimation method for the entire life cycle of braking clearance in an EMB system, comprising the following steps: S1. Acquire the motor current signal of the electromechanical braking system during the braking clamping phase. and motor rotation angle signal ; In motor current signal At the data acquisition level, this embodiment preferably configures a Hall effect current sensor or a current sampling circuit based on a shunt resistor on the power transmission path between the three-phase inverter bridge arm of the motor driver and the permanent magnet synchronous motor. The phase current acquired in real time is extracted by Clark transformation and Park transformation to obtain the quadrature-axis current component, which is then used as the motor current signal. The physical implementation of this signal is noteworthy. It is important to note that this current signal not only contains the effective component that generates electromagnetic torque, but also includes harmonic disturbances introduced by inverter nonlinearity, motor cogging effect, and dynamic load changes in the mechanical transmission chain. Therefore, this invention places particular emphasis on sampling synchronization and anti-interference capability in the signal acquisition stage. For motor rotation signal To obtain the absolute position signal, this embodiment exemplarily employs a rotary transformer or magnetoelectric position sensor mounted at the end of the motor shaft, with an output resolution preferably not less than fourteen bits for digital angle signal detection. In some alternative embodiments, an incremental photoelectric encoder combined with a zero-position reference signal can be used to accumulate the relative angle, but an additional position latching circuit is required to ensure timing consistency with the current signal. The motor angle signal... The physical meaning of this is the cumulative angular displacement of the motor rotor calculated from the moment the braking command is issued. The choice of this dimension is directly related to the preservation of the orthogonality of the subsequent Legendre polynomial in the domain [-1,1]. Regarding the sampling timing configuration, this embodiment requires the motor current signal With motor rotation angle signal Synchronous sampling must be completed within the same interrupt service cycle, with a preferred sampling period of 0.005 seconds. This value balances the computational load of the embedded controller with the dynamic characteristics capture requirements of the braking process. The synchronous sampling described here is not limited to strict consistency of hardware triggering times, but emphasizes achieving a one-to-one correspondence between the two sets of signals in time at the digital signal processing level through FIFO buffering or dual-port RAM mechanisms. This avoids systematic deviations in subsequent regression models caused by phase lag introduced by asynchronous sampling. S2. Based on the nonlinear relationship between the motor current signal and the motor rotation angle signal, a current-rotation angle regression model based on the Legendre orthogonal basis is constructed. The current-rotation angle regression model includes a regression vector based on the Legendre orthogonal basis function and a parameter vector with estimation. (S21) Utilizing the nonlinear relationship between motor current and rotation angle, the following regression model is established: (1) in, This indicates the motor current during the loading phase. Indicates the motor rotation angle during the loading phase. Indicates the braking clearance. , , This indicates the nonlinear characteristics of the system; (S22) For the applied current Perform linear normalization to interval : (2) in, This indicates the preset maximum current value under actual measurement conditions. Auxiliary variables representing linear normalization; (S23) Legendre orthogonal basis in the domain It has orthogonality, and the weight function The first four terms are in the form of an orthogonal polynomial as follows: (3) (S24) Construct a new regression vector Define parameter vector Combined with formula The new regression model for current-rotation angle based on the Legendre orthogonal basis is as follows: (4) In the formula, This represents the vector of parameters to be estimated. This represents the regression vector based on Legendre orthogonal basis functions, where T represents the transpose. S3. A weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M estimator is used to perform online robust estimation of the parameter vector of the current-turn angle regression model. A two-stage recursive solution strategy is used to perform efficient and low-cost computation of the parameter estimation process to obtain the parameter vector estimate. (S31) Since the EMB system is affected by factors such as wear, temperature and measurement noise during operation, the parameters in formula (4) are time-varying and lack stability. Therefore, a weighted multi-forgetting factor algorithm is established to deal with the above system characteristics. Based on the formula The cost function is defined as: (5) (6) in, The cost function representing the prediction error, This represents the actual motor rotation angle value. Represents the parameter vector estimated online. This represents the forgetting factor for each estimated parameter. The robust penalty function is defined as follows: (7) in, , , The threshold parameter, which represents the degree of outlier suppression, is set based on the quantiles of the standard normal distribution and is defined as follows: , ; (S32) The adaptive M-estimator introduces a recursive adaptive variance estimation law, as follows: (8) in, Residual The estimated variance, The forgetting factor represents the estimated variance. Represents the finite sampling correction factor. Residual Sliding window; (S33) Introduce robust weights for updates related to braking clearance. (Other parameters can be selected) ), and update scalar information. and gain : (9) in, Represents the weighting coefficient, when hour, Other channels ; Therefore, according to the formula The parameters are updated in the following recursive form: (10); (S34) A two-stage recursive solution strategy is used to perform efficient and low-cost computation of the parameter estimation process, as follows: Phase 1: Using the ( Step , The estimated value, i.e. As a known condition, recursively update ,Right now: (11) in, This is the estimate from the previous step. The gain of the first-stage estimator is defined as: (12) Phase Two: Use As a known condition, recursively update ,Right now: (13) in, The gain of the second-stage estimator is defined as: (14); S4. Extrapolate the parameter vector estimates based on the zero-current condition, mapping the parameter estimates to brake clearance estimates. ; Obtain the current parameter estimate Then, normalize the auxiliary variables. Extrapolating to the theoretical zero current point, i.e. At that time, the estimated value of the braking clearance is obtained. Used for clearance compensation, force control, or condition monitoring: (15) In the formula, These are the estimated parameters based on the regression vectors of the Legendre orthogonal basis.

[0024] To verify the effectiveness and accuracy of the method described in this embodiment for braking clearance estimation in an EMB system, experimental verification was conducted. Figure 3 The results show that the standard RLS exhibits significant initial oscillations and biased steady-state estimations. EKF shows significant steady-state bias and a slow convergence rate, consistent with its sensitivity to noise covariance tuning. LeRLS and MFFRLS improve estimation performance due to enhanced regression conditions and multi-rate tracking, respectively; however, intermittent fluctuations and residual biases persist during transient intervals. In contrast, the proposed method strictly adheres to the reference value throughout the estimation process, demonstrating smooth convergence, minimal fluctuations, and strong robustness to disturbances. Quantitative results in Table 1 further confirm this, where the proposed method yields the lowest RMSE (0.09 rad) and RRMSE (0.27 × 10⁻³%).

[0025] To verify the advantages of the adaptive M-estimator under different brake pad wear conditions, tests were conducted using brake pads with moderate and heavy wear, such as... Figure 4 and Figure 5As shown. For comparison, the fixed M-estimate using a constant standard deviation was also evaluated. For both wear conditions, the fixed method showed increased bias and oscillations, indicating that the fixed residual statistics become suboptimal as the measured characteristics evolve over the brake pad's lifespan. In contrast, the proposed adaptive M-estimater updates... This results in smoother convergence and more accurate convergence to the reference value. The adaptive standard deviation also converges rapidly to a lower level. = 1.13 rad. Quantitatively, compared to the fixed method, under moderate wear conditions, the adaptive M estimator reduces RMSE from 1.05 rad to 0.14 rad and RRMSE from 2.25 × 10⁻⁶ rad. % decreased to 0.29× Similarly, under severe wear, RMSE and RRMSE decreased by approximately 51% and 50%, respectively.

[0026] The real-time applicability of the proposed two-stage estimation strategy is verified by comparing its estimation performance and computational complexity with that of the direct matrix inversion method. Figure 6 As shown, the estimation results produced by the two-stage estimator are almost identical to those of the direct matrix inversion method, with a maximum deviation of only 0.067 rad. The method proposed in this embodiment completely eliminates direct matrix inversion and significantly reduces the number of multiplications and additions, thus achieving a linear computational complexity of O(n). Although the two-stage method introduces additional adaptive threshold selection and square root operations, their computational cost is negligible. These results demonstrate that the proposed two-stage recursive structure significantly reduces the computational burden and is suitable for real-time operation of automotive-grade controllers.

[0027] In summary, this invention achieves high-precision online identification of brake clearance in EMB systems without the need for force sensors, avoiding the reliance on manual threshold setting and offline calibration in traditional methods. It constructs regression variables based on Legendre orthogonal basis functions, effectively suppressing parameter correlation and improving numerical stability. By introducing MFFRLS and an adaptive robust weighting mechanism, it enhances the tracking and suppression capabilities against time-varying parameters and abnormal noise. A two-stage recursive solution strategy reduces computational complexity. Finally, through a parameter mapping method based on zero input conditions, the online estimation results are directly converted into brake clearance quantities with clear physical meaning, achieving an engineering solution that is computationally efficient, highly interpretable, and meets the real-time control and functional safety requirements of EMB systems.

[0028] Example 2: This embodiment provides an online estimation system for the entire lifecycle of brake clearance in an EMB system, used to implement the online estimation method for the entire lifecycle of brake clearance in an EMB system described in Embodiment 1, including: The data acquisition module is used to acquire the motor current signal of the electromechanical braking system during the braking clamping phase. and motor rotation angle signal ; The model building module is used to construct a current-angle regression model based on the Legendre orthogonal basis based on the nonlinear relationship between the motor current signal and the motor rotation angle signal. The current-angle regression model includes a regression vector based on the Legendre orthogonal basis function and a parameter vector with estimation. The online parameter estimation module is used to perform online robust estimation of the parameter vector of the current-turn angle regression model by using a weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M-estimator. It also uses a two-stage recursive solution strategy to perform efficient and low-cost computation of the parameter estimation process and obtain the parameter vector estimate. The gap estimation module is used to extrapolate the parameter vector estimates based on the zero-current condition, mapping the parameter estimates to brake gap estimates. .

[0029] Example 3: This embodiment provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores the computer program capable of running on the processor. When the processor loads and executes the computer program, it adopts an online estimation method for the entire life cycle of braking gap of an EMB system as described in Embodiment 1.

[0030] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer, or a cloud server, and the terminal device includes, but is not limited to, a processor and a memory. For example, the terminal device may also include input / output devices, network access devices, and buses.

[0031] Furthermore, the processor can be a central processing unit (CPU). Of course, depending on the actual use, other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be used. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it in this regard.

[0032] Example 4: This embodiment provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform an online estimation method for the entire life cycle of braking clearance in an EMB system as described in Embodiment 1.

[0033] The computer program can be stored in a computer-readable medium. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or certain middleware. The computer-readable medium includes any entity or device capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the computer-readable medium includes, but is not limited to, the above-mentioned components.

[0034] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0035] For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on," "mounted on," "fixed to," or "set on" another element, it may be directly on the other element or there may be an intermediate element present. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be an intermediate element present. The terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible embodiments.

[0036] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0037] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

Claims

1. A method for online estimation of braking clearance throughout the entire lifecycle of an EMB system, characterized in that, Includes the following steps: Acquire motor current signal of electromechanical braking system during braking clamping phase and motor rotation angle signal ; Based on the nonlinear relationship between motor current signal and motor rotation angle signal, a current-rotation angle regression model based on Legendre orthogonal basis is constructed. The current-rotation angle regression model includes a regression vector based on Legendre orthogonal basis functions and a parameter vector to be estimated. We employ a weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M-estimator to perform online robust estimation of the parameter vector of the current-turn angle regression model. We also utilize a two-stage recursive solution strategy to perform efficient and low-cost computation of the parameter estimation process, thereby obtaining the parameter vector estimates. Extrapolating the parameter vector estimates based on the zero-current condition maps the parameter estimates to brake clearance estimates. .

2. The online estimation method for the entire life cycle of braking clearance in an EMB system according to claim 1, characterized in that: Based on the nonlinear relationship between motor current signal and motor rotation angle signal, a current-rotation angle regression model based on the Legendre orthogonal basis is constructed as follows: (21) Collect the motor current of the electromechanical braking system during the braking clamping phase. and motor rotation angle With a sampling period of 0.005s, the following regression model is established using the nonlinear relationship between motor current and rotation angle: （1） in, This indicates the motor current during the loading phase. Indicates the motor rotation angle during the loading phase. Indicates the braking clearance. , , This indicates the nonlinear characteristics of the system; (22) Regarding the applied current Perform linear normalization to interval : （2） in, This indicates the preset maximum current value under actual measurement conditions. Auxiliary variables representing linear normalization; (23) Legendre orthogonal basis in the domain It has orthogonality, and the weight function The first four terms are in the form of an orthogonal polynomial as follows: （3） (24) Construct a new regression vector Define parameter vector Combined with formula The new regression model for current-rotation angle based on the Legendre orthogonal basis is as follows: （4） In the formula, This represents the vector of parameters to be estimated. Let T represent the regression vector based on Legendre orthogonal basis functions, and let T denote the transpose.

3. The online estimation method for the entire life cycle of braking clearance in an EMB system according to claim 2, characterized in that: The cost function of the weighted multi-forgetting-factor recursive least squares algorithm is defined as: （5） （6） In the formula, The cost function representing the prediction error, This represents the actual motor rotation angle value. Represents the parameter vector estimated online. This represents the forgetting factor for each estimated parameter. The robust penalty function is defined as follows: （7） in, , , The threshold parameter, which represents the degree of outlier suppression, is set based on the quantiles of the standard normal distribution.

4. The online estimation method for the entire life cycle of braking clearance in an EMB system according to claim 3, characterized in that: The adaptive M-estimator introduces a recursive adaptive variance estimation law, as follows: （8） in, Residual The estimated variance, The forgetting factor represents the estimated variance. Represents the finite sampling correction factor. Residual A sliding window.

5. The online estimation method for the entire life cycle of braking clearance in an EMB system according to claim 4, characterized in that: definition , Robust weights are introduced for updates related to braking clearance. The weight coefficients for the remaining parameter channels are set to 1, and the scalar information content is updated. and gain : （9） in, Represents the weighting coefficient, when hour, Other channels . Therefore, according to the formula The parameters are updated in the following recursive form: （10）。 6. The online estimation method for the entire life cycle of braking clearance in an EMB system according to claim 5, characterized in that: A two-stage recursive solution strategy is used to perform efficient and low-cost computation of the parameter estimation process, as detailed below: Phase 1: Using the ( Step , The estimated value, i.e. As a known condition, recursively update ,Right now: （11） in, This is the estimate from the previous step. The gain of the first-stage estimator is defined as: （12） Phase Two: Use As a known condition, recursively update ,Right now: （13） in, The gain of the second-stage estimator is defined as: （14）。 7. The online estimation method for the entire life cycle of braking clearance in an EMB system according to claim 6, characterized in that: Extrapolating the parameter vector estimates based on the zero-current condition maps the parameter estimates to brake clearance estimates: Obtain the current parameter estimate Then, normalize the auxiliary variables. Extrapolating to the theoretical zero current point, i.e. At that time, the estimated value of the braking clearance is obtained. Used for clearance compensation, force control, or condition monitoring: （15） In the formula, These are the estimated parameters based on the regression vectors of the Legendre orthogonal basis.

8. An online estimation system for the entire lifecycle of brake clearance in an EMB system, used to implement the online estimation method for the entire lifecycle of brake clearance in an EMB system as described in any one of claims 1 to 7, characterized in that, include: The data acquisition module is used to acquire the motor current signal of the electromechanical braking system during the braking clamping phase. and motor rotation angle signal ; The model building module is used to construct a current-angle regression model based on the Legendre orthogonal basis based on the nonlinear relationship between the motor current signal and the motor rotation angle signal. The current-angle regression model includes a regression vector based on the Legendre orthogonal basis function and a parameter vector with estimation. The online parameter estimation module is used to perform online robust estimation of the parameter vector of the current-turn angle regression model by using a weighted multi-forgetting factor recursive least squares algorithm combined with an adaptive M-estimator. It also uses a two-stage recursive solution strategy to perform efficient and low-cost computation of the parameter estimation process and obtain the parameter vector estimate. The gap estimation module is used to extrapolate the parameter vector estimates based on the zero-current condition, mapping the parameter estimates to brake gap estimates. .

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor loads and executes the computer program, it employs an online estimation method for the entire life cycle of the braking clearance of an EMB system, as described in any one of claims 1 to 7.

10. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform an online estimation method for the entire life cycle of braking clearance in an EMB system as described in any one of claims 1 to 7.

Images