An EMB-based vehicle wheel end clamping force closed-loop control system

Abstract

The application discloses a kind of vehicle wheel end clamping force closed-loop control systems based on EMB, and the application relates to vehicle control technical field, comprising: feedforward nominal acquisition module, the mapping relationship of brake pedal stroke and standard friction torque is set, and the mapping table of friction torque and wheel speed is established;Brake operation planning module, according to the pedal stroke data of last brake operation, brake process is divided into multiple subintervals, and the actual wheel end clamping force and friction torque of each subinterval are determined to calculate actual friction coefficient;Feedback control construction module, the feedback model of input as pedal stroke and actual friction coefficient, output as wheel end clamping force is constructed, and the loss function is the absolute difference value minimization of actual friction torque and standard friction torque;Closed-loop control module dynamically adjusts wheel end clamping force based on feedback model, to realize the brake effect of standard friction torque;The system improves the accuracy and reliability of vehicle braking.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, specifically to a closed-loop control system for vehicle wheel-end clamping force based on EMB. Background Technology

[0002] As a critical component of vehicle safety, the performance of the vehicle braking system directly affects vehicle safety and driving experience. With the continuous advancement of automotive technology, especially the gradual application of electric braking systems (EMB), traditional braking control methods face many challenges. Current braking systems typically rely on mechanical and hydraulic methods to control clamping force. While this method was widely used in early automobiles, its flexibility and adaptability are insufficient in modern, complex driving environments. Especially under high-speed driving, sudden braking, or complex road conditions, traditional braking systems struggle to make effective dynamic adjustments according to actual conditions.

[0003] Currently, many vehicle braking systems use a fixed coefficient of friction, set based on experimental data or empirical values, which often fails to reflect changes under actual driving conditions. For example, brake wear and environmental changes can affect vehicle braking performance, but traditional systems lack the ability to provide real-time feedback and adjustment, leading to fluctuations in braking performance and consequently impacting driving safety. Furthermore, during emergency braking, traditional braking systems may experience braking delays due to insufficient clamping force, increasing the risk of accidents. Therefore, a novel braking control scheme is urgently needed to improve the flexibility and reliability of vehicle braking systems.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a closed-loop control system for vehicle wheel-end clamping force based on EMB, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A closed-loop control system for vehicle wheel-end clamping force based on EMB, specifically including:

[0008] The feedforward nominal acquisition module is used to set the mapping relationship between brake pedal travel and standard friction torque, set up the laboratory environment for vehicle braking tests, analyze the attenuation of wheel speed under different friction torques, and establish the mapping relationship between friction torque and wheel speed attenuation through wheel speed attenuation data to form a brake feedforward mapping table.

[0009] The braking operation planning module is used to retrieve the brake pedal travel data of the previous complete braking operation. Based on the changes in brake pedal travel, the previous complete braking operation process is divided into several braking sub-sections. Based on the mapping relationship between friction torque and wheel speed, the actual wheel end clamping force and actual friction torque of each braking sub-section are determined, and then the actual friction coefficient is obtained.

[0010] The feedback control module is used to build a feedback model with the pedal travel and actual friction coefficient as inputs and the wheel end clamping force as output. Its loss function is to minimize the absolute difference between the friction torque obtained based on the actual friction coefficient and the actual wheel end clamping force and the standard friction torque corresponding to the pedal travel.

[0011] The closed-loop control module is used to dynamically adjust the wheel end clamping force output by the vehicle under different brake pedal strokes based on the feedback model, so that the vehicle generates a corresponding standard friction torque under each brake pedal stroke, thereby realizing the closed-loop control of the vehicle wheel end clamping force by EMB.

[0012] Furthermore, the logic behind establishing the mapping relationship between brake pedal travel and standard friction torque is as follows: Obtain the ideal wheel-end clamping force generated by different brake pedal travels in the vehicle design plan; based on the ideal wheel-end clamping force, combined with the nominal radius of the brake disc and the nominal friction coefficient between the brake disc and brake pads, calculate the standard friction torque comprehensively through the definitions of friction force and torque. The specific formula used for this calculation is as follows:

[0013]

[0014] In the formula, For standard friction torque, The nominal coefficient of friction between the brake disc and the brake pads. The nominal radius of the brake disc. The ideal clamping force applied to the wheel end;

[0015] The specific method for establishing the mapping relationship between brake pedal travel and standard friction torque is as follows: multiple sets of mapping data between brake pedal travel and standard friction torque are obtained through the standard friction torque calculation method; a regression equation is constructed through a linear regression algorithm; regression analysis is performed on the randomly obtained mapping data set to determine the regression coefficients in the regression equation, thereby completing the establishment of the travel-standard friction torque mapping model; and the correspondence between different brake pedal travels and standard friction torque is determined through the travel-standard friction torque mapping model.

[0016] Furthermore, the laboratory environment is specifically a constant environment under standard ambient temperature, standard air humidity, and standard atmospheric pressure. The vehicle braking test is specifically set as follows: multiple sets of braking parameter combinations are preset, including friction torque and wheel braking initial speed. Vehicle braking tests are conducted under different braking parameter combinations, and the wheel speed and corresponding braking duration are obtained during the vehicle braking test. A model is constructed with the braking parameter combination and braking duration as inputs and the wheel speed under the corresponding braking duration as output, which is denoted as the friction torque-wheel speed mapping model.

[0017] Furthermore, the method for determining the braking sub-interval of the previous complete braking process is as follows: retrieve the brake pedal travel signal during the previous complete braking process, determine several monitoring points according to a fixed sampling interval during the complete braking process, and extract the pedal travel data of each monitoring point. Take the first monitoring point as the starting point of the first braking sub-interval, and traverse the subsequent monitoring points in chronological order until the absolute difference of the pedal travel between a monitoring point and the first monitoring point is not less than the preset maximum travel change threshold. The time span between the monitoring point and the first monitoring point is taken as a braking sub-interval, and the monitoring point is taken as the starting point of the next braking sub-interval. Traverse the subsequent monitoring points in chronological order, and iterate in this way to determine all braking sub-intervals in the previous complete braking operation process.

[0018] Furthermore, the actual wheel end clamping force of each brake sub-section is obtained by means of: extracting the actual wheel end clamping force at the corresponding moment of the previous complete braking process based on the starting time of each brake sub-section, and taking the actual wheel end clamping force at the starting time of the brake sub-section as the actual wheel end clamping force applied throughout the entire process of the brake sub-section.

[0019] The actual friction torque of each braking sub-section is obtained by: obtaining the wheel speeds at the start and end of the braking sub-section, taking the time span of the braking sub-section as the braking duration, and using the braking duration, the wheel speeds at the start and end of the braking sub-section, and the wheel speeds at the end of the braking sub-section, and inferring the actual friction torque of the braking sub-section in reverse based on the friction torque-wheel speed mapping model.

[0020] Furthermore, the weighting coefficient for calculating the curve differences between each braking sub-interval is determined based on the time span of each braking sub-interval. The specific formula used for this calculation is as follows:

[0021]

[0022] In the formula, Let be the weight coefficient of the i-th braking sub-interval. Let be the time span of the i-th braking sub-interval, where i is the index of the braking sub-interval. , This represents the total number of braking sub-intervals;

[0023] Based on the actual wheel end clamping force and actual friction torque of each brake sub-section, the actual friction coefficient of each brake sub-section is calculated. Then, based on the actual friction coefficient of each brake sub-section and its corresponding weighting coefficient, the actual vehicle friction coefficient is calculated. The specific formula used to calculate the actual friction coefficient is as follows:

[0024]

[0025] In the formula, This is the actual coefficient of friction of the vehicle. Let be the actual friction coefficient of the i-th braking sub-section, where the actual friction coefficient of the i-th braking sub-section is... The formula used for the calculation is:

[0026]

[0027] In the formula, The actual friction torque of the i-th braking sub-section is... is the actual wheel end clamping force of the i-th brake sub-section.

[0028] Furthermore, the feedback model is specifically established through a convolutional neural network;

[0029] After each complete braking cycle, the wheel-end clamping force output by the vehicle at different brake pedal strokes is adjusted through a feedback model so that the vehicle always generates standard friction torque at each brake pedal stroke.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] This invention establishes a braking feedforward mapping table through a feedforward nominal acquisition module, and systematically analyzes and maps the relationship between brake pedal travel and standard friction torque, as well as the relationship between friction torque and wheel speed decay, providing a data-driven basis for the braking process.

[0032] Secondly, by dividing the previous complete braking operation into several braking sub-intervals through the braking operation planning module, and combining the relationship between friction torque and wheel speed, the actual wheel end clamping force and friction coefficient are analyzed. This allows for dynamic evaluation of the actual friction coefficient change during the braking process, thereby analyzing the fluctuation of friction braking performance in actual operation.

[0033] Furthermore, the feedback control module constructs a feedback model with the pedal travel and actual friction coefficient as inputs and the wheel end clamping force as output. Using the absolute difference between the friction torque and the standard value as the loss function, it can adjust the wheel end clamping force in real time during braking to ensure accurate output of friction torque. This allows for rapid response to changes in friction conditions during vehicle braking, significantly improving the dynamic adaptability and stability of the braking system and solving the problem of insufficient control accuracy caused by inaccurate friction models in traditional EMB systems. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0035] Figure 2 This is a distribution diagram of the weight coefficients for each braking sub-interval. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0037] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0038] Example:

[0039] Please see Figures 1-2 The present invention provides a technical solution:

[0040] A closed-loop control system for vehicle wheel-end clamping force based on EMB, specifically including:

[0041] The feedforward nominal acquisition module is used to set the mapping relationship between brake pedal travel and standard friction torque, set up the laboratory environment for vehicle braking tests, analyze the attenuation of wheel speed under different friction torques, and establish the mapping relationship between friction torque and wheel speed attenuation through wheel speed attenuation data to form a brake feedforward mapping table.

[0042] The logic behind establishing the mapping relationship between brake pedal travel and standard friction torque is as follows: Obtain the ideal wheel-end clamping force generated by different brake pedal travels in the vehicle's design plan. Based on this ideal wheel-end clamping force, combined with the nominal radius of the brake disc and the nominal friction coefficient between the brake disc and brake pads, and through the definitions of friction force and torque, calculate the standard friction torque using the following formula:

[0043]

[0044] In the formula, For standard friction torque, The nominal coefficient of friction between the brake disc and the brake pads. The nominal radius of the brake disc. The ideal clamping force applied to the wheel end;

[0045] During the vehicle design phase, designers determine the parameters of the braking system based on the vehicle's intended use and performance requirements. These specifications are typically detailed in the vehicle's technical documents or design manuals, from which the ideal wheel-end clamping force generated by different brake pedal travels is extracted.

[0046] The nominal radius of the brake disc is usually provided by the manufacturer in the product specifications. This can be found in relevant technical manuals or product brochures, or during the vehicle design phase, when detailed braking system design drawings, including the brake disc's geometric parameters, will be provided. The nominal radius of the brake disc should be listed in the design documents.

[0047] The friction coefficient of friction pads and brake discs can be tested experimentally under different temperature and pressure conditions to obtain more accurate friction coefficient values. The friction coefficient of friction pads and brake discs is usually provided by the manufacturer based on the material properties, especially the performance data under different working conditions.

[0048] The specific method for establishing the mapping relationship between brake pedal travel and standard friction torque is as follows: multiple sets of mapping data between brake pedal travel and standard friction torque are obtained through the standard friction torque calculation method; a regression equation is constructed through a linear regression algorithm; regression analysis is performed on the randomly obtained mapping data set to determine the regression coefficients in the regression equation, so as to complete the establishment of the travel-standard friction torque mapping model; and the correspondence between different brake pedal travels and standard friction torque is determined through the travel-standard friction torque mapping model.

[0049] The specific steps include: different pedal travel can be set, such as multiple values ​​from 0 to the maximum travel; the standard friction torque for different pedal travels is calculated according to the standard friction torque calculation formula to form a dataset; the form of a linear regression model is selected according to the data characteristics; a portion of the dataset is randomly selected for training, and another portion is used for validation; the model is further validated using k-fold cross-validation to confirm its robustness; and the residuals between the predicted and actual values ​​are analyzed to ensure there is no systematic bias.

[0050] The laboratory environment is specifically a constant environment with standard ambient temperature, standard air humidity, and standard atmospheric pressure. Testing under standardized conditions ensures that conditions are as consistent as possible across different testing cycles and for different test samples. This helps reduce the impact of environmental factors on experimental results, making the data more comparable. Subsequent experiments need to be repeated under the same environmental conditions; standard environmental conditions serve as a benchmark, allowing experimental results to be verified and reproduced. Simultaneously, the constant environment with standard ambient temperature, standard air humidity, and standard atmospheric pressure reduces interference from environmental variables.

[0051] The vehicle braking test is specifically set as follows: multiple sets of braking parameter combinations are preset, including friction torque and wheel braking initial speed. Vehicle braking tests are conducted under different braking parameter combinations, and the wheel speed and corresponding braking duration are obtained during the vehicle braking test. A model is constructed with the braking parameter combination and braking duration as inputs and the wheel speed under the corresponding braking duration as output, which is denoted as the friction torque-wheel speed mapping model.

[0052] The friction torque-wheel speed mapping model is specifically constructed using an LSTM model, selecting an activation function and an optimization algorithm. The Tanh function is chosen as the activation function, and Adam is chosen as the optimization algorithm for the LSTM model. The formula for the Tanh function is:

[0053]

[0054] In the formula, Represents the Tanh function, with the independent variable... This represents the weighted sum of the neuron's inputs, that is, the result of the weighted sum of the inputs received by the neuron from the previous layer;

[0055] Simultaneously, the hyperparameters of the LSTM model are set, including: number of network layers, number of iterations, learning rate, batch size, number of training iterations, number of batches, and number of hidden layer neurons.

[0056] The network is set to a 3-layer structure, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training iterations is set to 100, and the number of hidden layer neurons is set to 32.

[0057] The model is trained using a combination of braking parameters and braking duration as inputs, wheel speeds at corresponding braking durations as labels, and mean squared error (MSE) as the loss function to constrain the model's prediction accuracy.

[0058] The braking operation planning module is used to retrieve the brake pedal travel data of the previous complete braking operation. Based on the changes in brake pedal travel, the previous complete braking operation process is divided into several braking sub-sections. Based on the mapping relationship between friction torque and wheel speed, the actual wheel end clamping force and actual friction torque of each braking sub-section are determined, thereby obtaining the actual friction coefficient.

[0059] The method for determining the braking sub-interval of the previous complete braking process is as follows: retrieve the brake pedal travel signal during the previous complete braking process, determine several monitoring points according to a fixed sampling interval during the complete braking process, and extract the pedal travel data of each monitoring point. Take the first monitoring point as the starting point of the first braking sub-interval, and traverse the subsequent monitoring points in chronological order until the absolute difference of the pedal travel between a monitoring point and the first monitoring point is not less than the preset maximum travel change threshold. The time span between the monitoring point and the first monitoring point is taken as a braking sub-interval, and the monitoring point is taken as the starting point of the next braking sub-interval. Traverse the subsequent monitoring points in chronological order, and iterate in this way to determine all braking sub-intervals in the previous complete braking operation process.

[0060] The complete braking operation specifically refers to the braking operation process that covers 80% to 100% of the total travel of the brake pedal.

[0061] The actual wheel-end clamping force for each braking sub-section is obtained as follows: Based on the starting time of each braking sub-section, the actual wheel-end clamping force at the corresponding moment of the previous complete braking process is extracted. The actual wheel-end clamping force at the starting time of the braking sub-section is taken as the actual wheel-end clamping force applied throughout the entire braking sub-section. High-precision pressure or force sensors are used to monitor the wheel-end clamping force in real time. These sensors are usually installed in key parts of the braking system, such as near the brake calipers or brake discs. During vehicle braking tests, a data acquisition system, such as a data acquisition instrument or DAQ device, is used to record the signals output by the sensors in real time. A timestamp is added to the signal output by each sensor to facilitate time alignment of different parameters during subsequent analysis. The collected data is stored in a computer or data recording device, usually in the form of CSV, Excel, or a database.

[0062] The actual friction torque of each braking sub-section is obtained by: obtaining the wheel speeds at the start and end of the braking sub-section, taking the time span of the braking sub-section as the braking duration, and using the braking duration, the wheel speeds at the start and end of the braking sub-section, and the wheel speeds at the end of the braking sub-section, and inferring the actual friction torque of the braking sub-section in reverse based on the friction torque-wheel speed mapping model.

[0063] For each braking sub-section, the actual friction torque is derived using the starting and ending wheel speeds and the calculated braking duration through a friction torque-wheel speed mapping model. The starting and ending wheel speeds of each braking sub-section are extracted from the data table, and combined with the duration of the braking sub-section, the actual friction torque of each braking sub-section is solved using the friction torque-wheel speed mapping model.

[0064] The weighting coefficient for calculating the curve difference between each braking sub-interval is determined by the time span of each sub-interval. The specific formula used for the calculation is as follows:

[0065]

[0066] In the formula, Let be the weight coefficient of the i-th braking sub-interval. Let be the time span of the i-th braking sub-interval, where i is the index of the braking sub-interval. , This represents the total number of braking sub-intervals;

[0067] It should be noted that different braking sub-intervals may have different time spans. A longer time span usually means that more events occur within that interval, resulting in higher reliability for the analysis. Using the time span... Calculating weights can reasonably reflect the relative importance of each sub-interval in the entire analysis process;

[0068] The time span of each sub-interval Sum of the time spans of all sub-intervals Dividing the weights normalizes the weights, ensuring that the sum of the weights of all sub-intervals is 1, allowing the weights to be used as proportions for easy comparison and calculation.

[0069] Long-span braking sub-intervals may contain more information or changes, requiring them to be given higher weight in the calculation. This formula utilizes a larger... Value directly affects This allows these intervals to have a greater impact in further analysis or calculation. By taking into account the differences in time span, the importance of each braking sub-interval in the overall analysis can be reflected more accurately, thereby improving the accuracy of the final analysis results.

[0070] Based on the actual wheel end clamping force and actual friction torque of each brake sub-section, the actual friction coefficient of each brake sub-section is calculated. Then, based on the actual friction coefficient of each brake sub-section and its corresponding weighting coefficient, the actual vehicle friction coefficient is calculated. The specific formula used to calculate the actual friction coefficient is as follows:

[0071]

[0072] In the formula, This is the actual coefficient of friction of the vehicle. Let be the actual friction coefficient of the i-th braking sub-section, where the actual friction coefficient of the i-th braking sub-section is... The formula used for the calculation is:

[0073]

[0074] In the formula, The actual friction torque of the i-th braking sub-section is... is the actual wheel end clamping force of the i-th brake sub-section.

[0075] The feedback control module is used to construct a feedback model with the pedal travel and actual friction coefficient as inputs and the wheel end clamping force as output. Its loss function is to minimize the absolute difference between the friction torque obtained based on the actual friction coefficient and the actual wheel end clamping force and the standard friction torque corresponding to the pedal travel.

[0076] The feedback model is specifically established through a convolutional neural network, with the pedal travel and actual friction coefficient as inputs, and the wheel end clamping force that should be applied for the standard friction torque corresponding to the pedal travel as a label, and the model is trained accordingly.

[0077] The convolutional neural network described herein consists of an input layer, convolutional layers, pooling layers, fully connected layers, and an output layer, wherein the activation function in the convolutional layer is... function, The specific expression of the function is:

[0078]

[0079] in, Indicates the first Each corresponding convolutional layer Then it means the first In the corresponding convolutional layer, the th The first feature map One eigenvalue;

[0080] For the fully connected layer, the number of neurons in the fully connected layer is set to 32, the initial neural network learning rate is set to 0.001, and the number of training rounds is 100.

[0081] The specific method for training the feedback model is as follows: Collect the wheel-end clamping force required to generate standard friction torque under different friction coefficients and brake pedal travels. Map the friction coefficient, brake pedal travel, and wheel-end clamping force required to generate standard friction torque to form several training data sets, creating a training dataset. Use 70% of the training dataset as the training set and 30% as the validation set to train the model. Use the validation set data to evaluate model performance; commonly used regression performance metrics include mean absolute error and... Coefficient of determination, etc.

[0082] It should be noted that the loss function is set to evaluate the model's performance by calculating the difference between the predicted value and the actual standard friction torque. The loss function clearly shows the deviation between the model's predicted value and the standard value, which facilitates intuitive understanding and analysis of the model's performance. This ensures that the predicted wheel end clamping force of the output pedal stroke can compensate for the damage caused by the friction coefficient, thus minimizing the difference between the generated friction torque and the standard friction torque under the same pedal stroke.

[0083] The closed-loop control module is used to dynamically adjust the wheel end clamping force output by the vehicle under different brake pedal strokes based on the feedback model, so that the vehicle generates a corresponding standard friction torque under each brake pedal stroke, thereby realizing the closed-loop control of the vehicle wheel end clamping force by EMB.

[0084] After each complete braking cycle, the wheel-end clamping force output by the vehicle at different brake pedal strokes is adjusted through a feedback model so that the vehicle always generates standard friction torque at each brake pedal stroke.

[0085] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0086] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0087] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An EMB-based vehicle wheel end clamp load closed loop control system, characterized in that, Specifically comprising: The feedforward nominal acquisition module is used for setting the mapping relationship between the brake pedal stroke and the standard friction torque, setting the laboratory environment to perform the vehicle braking test, analyzing the wheel speed decay under different friction torques, and establishing the mapping relationship between the friction torque and the wheel speed decay through the wheel speed decay data to form the brake feedforward mapping table; The brake operation planning module is used for calling the brake pedal stroke data of the last complete brake operation, dividing the last complete brake operation process into a plurality of brake subintervals according to the brake pedal stroke change, determining the actual wheel end clamping force and the actual friction torque of each brake subinterval according to the mapping relationship between the friction torque and the wheel speed, and further obtaining the actual friction coefficient; The feedback control construction module is used for constructing a feedback model with the pedal stroke and the actual friction coefficient as the input and the wheel end clamping force as the output, and the loss function is the absolute difference between the standard friction torque corresponding to the pedal stroke and the friction torque obtained based on the actual friction coefficient and the actual wheel end clamping force; The closed-loop control module is used for dynamically adjusting the wheel end clamping force output by the vehicle under different brake pedal strokes based on the feedback model, so that the vehicle generates the corresponding standard friction torque under each brake pedal stroke, and the wheel end clamping force closed-loop control of the EMB vehicle is realized. The weight coefficient of each brake subinterval is determined according to the time span of each brake subinterval, and the specific formula for calculating the weight coefficient is: In the formula, is a weight coefficient of the i-th braking sub-interval, is a time span of the i-th braking sub-interval, where i is an index of the braking sub-interval, , is a total number of braking sub-intervals; The actual friction coefficient of each brake subinterval is calculated according to the actual wheel end clamping force and the actual friction torque of each brake subinterval, and the actual friction coefficient of the vehicle is calculated according to the actual friction coefficient of each brake subinterval and the weight coefficient of the corresponding brake subinterval. The specific formula for calculating the actual friction coefficient is: wherein is the actual friction coefficient of the vehicle, is the actual friction coefficient of the ith braking sub-interval, wherein the actual friction coefficient of the ith braking sub-interval is calculated as follows: The formula on which the calculation is based is: wherein is the actual friction torque of the i-th braking sub-interval, is the actual wheel end clamping force of the i-th braking sub-interval, is the nominal radius of the brake disc.

2. The EMB-based vehicle wheel end clamp load closed loop control system of claim 1, wherein: The logic for setting the mapping relationship between the brake pedal stroke and the standard friction torque is as follows: the ideal wheel end clamping force generated by different brake pedal strokes in the vehicle design plan is obtained, the standard friction torque is calculated by combining the ideal wheel end clamping force, the nominal radius of the brake disc, and the nominal friction coefficient between the brake disc and the brake friction plate, and the friction force and torque are defined. The specific formula for calculation is: wherein is the standard friction torque, is the nominal coefficient of friction between the brake disc and the brake linings, is the ideal wheel end clamp load applied; The specific method for establishing the mapping relationship between the brake pedal stroke and the standard friction torque is as follows: a plurality of mapping data groups of the brake pedal stroke and the standard friction torque are obtained through the standard friction torque calculation method, a regression equation is constructed through the linear regression algorithm, the regression coefficients in the regression equation are determined through regression analysis of randomly obtained mapping data groups, and the establishment of the stroke-standard friction torque mapping model is completed. The corresponding relationship between different brake pedal strokes and standard friction torques is determined through the stroke-standard friction torque mapping model.

3. The EMB-based vehicle wheel end clamp load closed loop control system of claim 2, wherein: The laboratory environment is specifically a constant environment under standard ambient temperature, standard air humidity and standard atmospheric pressure, and the vehicle braking test is specifically configured as follows: a plurality of braking parameter combinations are preset, the braking parameter combination includes friction torque and wheel braking initial rotating speed, vehicle braking tests are performed under different braking parameter combinations, wheel rotating speeds and corresponding braking durations during the vehicle braking tests are obtained, a model with input of braking parameter combination and braking duration and output of wheel rotating speed under corresponding braking duration is constructed, and the model is recorded as a friction torque-wheel rotating speed mapping model.

4. The EMB-based vehicle wheel end clamp load closed loop control system of claim 2, wherein: The method for determining the braking sub-interval in the last complete braking process is as follows: a brake pedal stroke signal in the last complete braking process is called, a plurality of monitoring points are determined at fixed sampling intervals in the complete braking process, pedal stroke data of the monitoring points are extracted, a first monitoring point is taken as a starting point of a first braking sub-interval, subsequent monitoring points are traversed in chronological order, and the process is iterated until a monitoring point appears for the first time, the absolute difference between the monitoring point and the first monitoring point is not less than a preset stroke maximum change threshold, the time span between the monitoring point and the first monitoring point is taken as a braking sub-interval, and the monitoring point is taken as a starting point of a next braking sub-interval.

5. The EMB-based vehicle wheel end clamping force closed-loop control system according to claim 4, characterized in that: The actual wheel end clamping force of each braking sub-interval is obtained, and the specific method is as follows: the actual wheel end clamping force at the corresponding time of the last complete braking process is extracted according to the starting time of each braking sub-interval, and the actual wheel end clamping force at the starting time of the braking sub-interval is taken as the actual wheel end clamping force applied during the entire process of the braking sub-interval; The actual friction torque of each braking sub-interval is obtained, and the specific method is as follows: the wheel rotating speeds at the starting point and the ending point of the braking sub-interval are obtained, the time span of the braking sub-interval is taken as the braking duration, the actual friction torque of the braking sub-interval is obtained by inversely reasoning based on the friction torque-wheel rotating speed mapping model, the braking duration, the wheel rotating speed at the starting point of the braking sub-interval and the wheel rotating speed at the ending point of the braking sub-interval.

6. The EMB-based vehicle wheel end clamp load closed loop control system of claim 5, wherein: The feedback model is specifically established by a convolutional neural network; After each complete braking is completed, the wheel end clamping force output by the vehicle under different brake pedal strokes is adjusted by the feedback model, so that the vehicle always generates a standard friction torque under each brake pedal stroke.

Images