Real-time analysis method and system based on brake data

Abstract

The invention discloses a real-time analysis method and system based on brake data, and relates to the technical field of automobile braking, and the method comprises the following steps: mode determination: obtaining real-time operation data of a vehicle braking system, and identifying and determining a brake behavior mode of a driver; mode matching: setting a plurality of predefined brake system response models, and automatically matching the brake behavior mode of the driver with the brake system response models; and interactive adjustment: interacting the matched brake system response model with the simulation brake model, displaying a simulation result of the simulation brake model through a display interface, and adjusting the vehicle brake system according to the simulation result. The braking behavior mode of the driver is analyzed in real time, and the braking system response model is automatically adjusted, so that the braking performance of the vehicle better conforms to the current driving condition, and the driving safety is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile braking, and in particular to a real-time analysis method and system based on braking data. Background Art

[0002] With the rapid development of the automobile industry, the performance of the vehicle braking system directly affects driving safety. Although the traditional braking system has certain functions, it still has many shortcomings in terms of real-time data collection, processing speed and accuracy. Especially in complex driving environments, how to ensure the efficient response of the braking system has become an urgent problem to be solved.

[0003] In recent years, with the advancement of sensor technology and the improvement of data processing capabilities, real-time analysis of braking data has become possible. However, existing solutions are often limited by factors such as the accuracy of data collection and low processing efficiency, and it is difficult to fully meet the high performance requirements of modern vehicles, especially intelligent connected vehicles, for braking systems.

[0004] The Chinese invention patent with publication number CN114169083A relates to a method, device, equipment and medium for analyzing data of an automatic emergency brake system. The method includes labeling hardware-in-the-loop test data and real vehicle test data of the same data format; simulating the triggering and braking process of the vehicle's automatic emergency brake system, and obtaining the distance between the vehicle and the obstacle when it stops; judging the distance and the preset ideal distance between the obstacle and the obstacle when the vehicle stops; when the distance between the obstacle and the vehicle stops is less than the ideal distance between the obstacle and the obstacle when the vehicle stops, calculating the difference between the ideal distance between the obstacle and the obstacle when the vehicle stops, and solving the percentage of the ratio of the difference to the ideal distance between the obstacle and the obstacle when the vehicle stops. It solves the problems of hysteresis and low efficiency in the analysis of hardware-in-the-loop test data and real vehicle verification test data. The application has the effect of improving hysteresis and increasing efficiency.

[0005] In the actual operation of the above and similar brake system data analysis methods, different drivers have different driving behavior habits when driving the same vehicle. At the same time, due to the differences in the braking effect, deployment effect, hardware parameters, etc. of the brake components between different vehicles, the same braking behavior of the same driver when driving different vehicles may produce different braking effects. In other words, there is a certain running-in period between the driver and the vehicle, that is, during the running-in period, the driver cannot fully exert the optimal performance of the vehicle's braking system, and may even cause misoperation. The longer the running-in period, the higher the risk the driver faces during driving. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for real-time analysis of brake data to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a real-time analysis method based on brake data, the analysis method comprising:

[0008] Mode determination: Obtain real-time operating data of the vehicle's braking system to identify and determine the driver's braking behavior pattern;

[0009] Pattern matching: setting a plurality of predefined braking system response models, and automatically matching the driver's braking behavior pattern with the braking system response model;

[0010] Interactive adjustment: The matched braking system response model interacts with the simulated braking model, and the simulation results of the simulated braking model are displayed through the display interface, and the vehicle braking system is adjusted according to the simulation results, specifically:

[0011] When the simulation result is the same as the preset driving result of the driver, adjusting the vehicle braking system according to the braking parameters corresponding to the simulated braking model;

[0012] When the simulation result is different from the driver's preset driving result, the braking system response model performs adaptive parameter adjustment, and the adaptively adjusted braking system response model interacts with the simulated braking model until the simulation result is the same as the driver's preset driving result.

[0013] Furthermore, identifying and determining the driver's braking behavior pattern includes:

[0014] Acquire real-time operation data of the vehicle braking system, and extract statistical features, time domain features and frequency domain features from the real-time operation data;

[0015] The statistical features, time domain features and frequency domain features are classified by a recognition algorithm to determine the driver's braking behavior patterns: light braking, heavy braking and continuous braking.

[0016] Furthermore, the setting of the braking system response model includes:

[0017] By extracting statistical features, time domain features and frequency domain features from the real-time operation data of the vehicle braking system, the actual feature vector corresponding to the driver's braking behavior mode is determined;

[0018] Obtaining a model feature vector in each predefined braking system response model, and determining a vector distance between the actual feature vector and each model feature vector;

[0019] All vector distances are compared to determine the vector distance with the minimum value. The predefined braking system response model corresponding to the vector distance with the minimum value is the braking system response model corresponding to the braking behavior mode of the driver.

[0020] Furthermore, the vector distance between the actual feature vector and each model feature vector is specifically:

[0021]

[0022] in: is the vector distance, is the ith element in the actual eigenvector, is the i-th element in the model feature vector, is the index number of the element in the vector, is the number of elements in the vector, is the actual eigenvector, is the model feature vector.

[0023] Furthermore, when the brake system response model performs adaptive parameter adjustment, the adjustment is performed according to the driving state of the driver, specifically:

[0024] When the driver is in driving state:

[0025] The braking system response model is adjusted in an orderly manner according to the braking distance, braking time and braking smoothness;

[0026] When the driver is not in driving state:

[0027] The driver adjusts the braking distance, braking time and braking stability within a preset braking parameter adjustment range through an interactive interface.

[0028] Furthermore, the braking system response model is adjusted in an orderly manner according to the braking distance, braking time and braking stability, specifically:

[0029] When the braking distance is incorrect, the braking distance is adjusted, and the braking time and braking smoothness are not adjusted;

[0030] When the braking distance is correct but the braking time is incorrect, the braking time is adjusted, and the braking distance and braking stability are not adjusted;

[0031] When the braking distance and the braking time are correct but the braking stability is incorrect, the braking stability is adjusted, and the braking distance and the braking time are not adjusted.

[0032] Furthermore, adjusting the vehicle braking system according to the simulation result also includes:

[0033] Determine the driver's driving preference behavior based on real-time operating data of the vehicle's braking system;

[0034] The simulation result of the simulated braking model is matched with the driving preference behavior to determine the driving preference behavior corresponding to the simulation result, and the vehicle braking system is adaptively adjusted according to the preference data corresponding to the driving preference behavior.

[0035] Furthermore, the determination of the driver's driving preference behavior includes:

[0036] The operation data of the vehicle braking system is preprocessed, and the relationship between the data distribution and the driver's driving behavior is determined based on the preprocessed operation data, specifically:

[0037] By means of the distribution of braking pressure and braking time, the central tendency and the dispersion degree of the operating data are obtained to determine the braking habits of the driver;

[0038] Determine the factors affecting the driver's driving preference through the correlation between the braking data;

[0039] Through cluster analysis of braking data, different driving preference groups are identified;

[0040] The driver's braking habits, driving preference influencing factors and driving preference groups are comprehensively set to obtain comprehensive feature data, and the comprehensive feature data is used as input of a machine learning model to output the driver's driving preference behavior.

[0041] A real-time analysis system based on brake data uses any one of the real-time analysis methods based on brake data described above.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] Firstly, the present invention analyzes the driver's braking behavior pattern in real time and automatically adjusts the braking system response model, so that the vehicle's braking performance is more in line with the current driving conditions, thereby improving driving safety;

[0044] Second, the present invention allows the driver to adjust the braking parameters through an interactive interface according to his or her driving habits and preferences, and can also automatically match the corresponding braking system response model according to the adjusted braking parameters, thereby enhancing the driver's driving experience and making driving comfortable and personalized;

[0045] Third: The present invention can automatically identify the driver's braking behavior pattern and dynamically adjust the braking system parameters according to the actual situation, thereby helping to improve the response speed and accuracy of the braking system and reduce the impact of human errors;

[0046] Fourthly, the present invention can ensure that the braking system can provide the best braking effect under different driving conditions by analyzing the driver's driving preference behavior in real time and adjusting the braking system parameters, thereby improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of the real-time analysis method based on brake data of the present invention;

[0048] Figure 2 A schematic diagram of a process flow for determining a braking system response model of the present invention;

[0049] Figure 3 A schematic diagram of a process for adjusting a vehicle braking system through a driving preference behavior according to the present invention;

[0050] Figure 4 A schematic diagram of a process for determining a driver's driving preference behavior according to the present invention;

[0051] Figure 5 A scatter plot of the braking time and the braking pressure of the present invention;

[0052] Figure 6 It is a line graph showing the relationship between the brake pressure and the frequency of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] In actual driving, due to the large differences in driving habits among different drivers, even the same driver may have significantly different driving experiences on different vehicles. This difference is mainly reflected in the running-in period between the driver and the vehicle's braking system, that is, in the initial use stage, the driver needs some time to adapt to the braking characteristics of the new car, during which there may be excessive or insufficient braking, thereby increasing the risk of traffic accidents. Therefore, it is necessary to shorten the running-in period and reduce the possibility of traffic accidents. The technical solution provided in this application collects and analyzes the operating data of the vehicle's braking system in real time, identifies the driver's braking behavior pattern, and dynamically adjusts the response parameters of the braking system according to the identified braking behavior pattern, which can not only help the driver adapt to the new vehicle braking system more quickly and reduce the running-in period, but also improve the overall safety performance of the vehicle to a certain extent.

[0055] Example 1

[0056] refer to Figure 1 and Figure 2 This embodiment provides a real-time analysis method based on brake data. The analysis method can be divided into three stages: pattern determination, pattern matching, and interactive adjustment. In this embodiment, the specific implementation process of the real-time analysis method based on brake data is as follows:

[0057] Step S1: Determine the type of sensor based on the characteristics and monitoring requirements of the vehicle's braking system. For example, the pressure sensor monitors the force of the brake pedal, the speed sensor monitors the speed of the vehicle, and the position sensor monitors the position change of the brake pedal. At the same time, the determined sensors are installed in the corresponding positions. For example, the pressure sensor is installed under the brake pedal to detect the pressure of the driver's brake pedal. The speed sensor is installed on the wheel hub or drive shaft to measure the speed change of the vehicle. The displacement sensor is installed on the brake pedal travel path to measure the travel length of the brake pedal. The time sensor records the time point when the braking action occurs, and calculates the braking frequency and duration in conjunction with other sensor data.

[0058] At the same time, the real-time operation data of the vehicle braking system is obtained through sensors to identify and determine the driver's braking behavior pattern. Specifically, it includes:

[0059] Step S1.1: Obtain real-time operating data of the vehicle braking system through sensors installed on the vehicle. Digital sensors output digital signals, and their sensor data can be directly read and transmitted. Analog sensors output analog signals, and their sensor data can be converted into digital signals by analog-to-digital converters before transmission. That is, according to the type of sensor, after selecting the data transmission path, the sensor data can be transmitted to the central processing unit through data transmission protocols (such as CAN bus, LIN bus and FlexRay).

[0060] Furthermore, in the central processing unit, after cleaning and standardizing the sensor data, the corresponding statistical features, time domain features and frequency domain features are extracted from it. The statistical features are to extract the statistical features such as the mean value, variance, maximum value and minimum value from the original data, the time domain features are to extract the time domain features such as the peak value, mean value and standard deviation of the signal, and the frequency domain features are to extract the frequency components as features by Fourier transforming the data.

[0061] Step S1.2: Classify the statistical features, time domain features and frequency domain features obtained in step S1.1, that is, perform data analysis through recognition algorithms (such as clustering algorithms, decision trees and support vector machines) to identify the driver's braking behavior pattern.

[0062] Specifically, the driver has the following braking behaviors during driving:

[0063] Brake pedal pressure: In an emergency, the driver usually applies greater pressure (for example, the pressure range is 70%-100%).

[0064] Braking frequency: When driving in the city, the number of braking times is more, and when driving on the highway, the number of braking times is less.

[0065] Braking duration: Multiple light brake applications in a short period of time indicate that the driver is slowing down. Heavy brake applications for a long period of time indicate that the driver is making an emergency stop.

[0066] Step S2: Set a plurality of predefined braking system response models, and automatically match the braking behavior pattern obtained in step S1.2 with the predefined braking system response model. In this embodiment, the characteristic data in the predefined braking system response model include: average braking pressure, standard deviation of braking pressure, maximum braking pressure, average value of braking duration, braking frequency and main frequency components of braking force. Specifically include:

[0067] Step S2.1: Determine the actual feature vector corresponding to the driver's braking behavior mode from the statistical features, time domain features and frequency domain features obtained in step S1.1. Specifically, they are:

[0068] Statistical characteristics: average brake pressure, standard deviation of brake pressure, maximum brake pressure, etc.

[0069] Time domain characteristics: average braking duration, braking frequency, etc.

[0070] Frequency domain characteristics: Spectral analysis results of brake pressure, such as main frequency components.

[0071] In the process of specific implementation, the above extracted features are as follows:

[0072] Average brake pressure: 40%.

[0073] Standard deviation of brake pressure: 10%.

[0074] Maximum brake pressure: 80%.

[0075] Average braking duration: 0.5s.

[0076] Braking frequency: 2 times / min.

[0077] The main frequency component of brake pressure: 2Hz.

[0078] That is to say, the actual feature vector extracted based on this feature is: [40, 10, 80, 0.5, 2, 2].

[0079] Step S2.2: Obtain a model feature vector in each predefined braking system response model, and determine a vector distance between the actual feature vector and each model feature vector.

[0080] In the process of specific implementation, this embodiment sets three predefined braking system response models, which correspond to three different braking behavior modes, namely model A (light pedal mode), model B (heavy pedal mode) and model C (continuous pedal mode), specifically:

[0081] Model A (Light-pedal mode):

[0082] Average brake pressure: 30%.

[0083] Standard deviation of brake pressure: 8%.

[0084] Maximum brake pressure: 70%.

[0085] Average braking duration: 0.6s.

[0086] Braking frequency: 1 time / min.

[0087] The main frequency component of brake pressure: 1Hz.

[0088] The extracted model feature vector A is: [30, 8, 70, 0.6, 1, 1].

[0089] Model B (Heavy pedal mode):

[0090] Average brake pressure: 60%.

[0091] Standard deviation of brake pressure: 15%.

[0092] Maximum brake pressure: 90%.

[0093] Average braking duration: 0.4s.

[0094] Braking frequency: 3 times / min.

[0095] The main frequency component of brake pressure: 3Hz.

[0096] The extracted model feature vector B is: [60, 15, 90, 0.4, 3, 3].

[0097] Model C (continuous pedaling model):

[0098] Average brake pressure: 40%.

[0099] Standard deviation of brake pressure: 10%.

[0100] Maximum brake pressure: 75%.

[0101] Average braking duration: 0.5s.

[0102] Braking frequency: 2 times / min.

[0103] The main frequency component of brake pressure: 2Hz.

[0104] The extracted model feature vector C is: [40, 10, 75, 0.5, 2, 2].

[0105] Furthermore, the characteristic data in the three different braking behavior modes of the above-mentioned model A (light-stepping mode), model B (heavy-stepping mode) and model C (continuous-stepping model) can be determined based on the actual driving record of the driver. That is, the brake pedal position (reflecting the degree to which the driver steps on the brake pedal), vehicle speed (the current speed of the vehicle), braking distance (the distance from the start of braking to complete stop), braking time (the time from the start of braking to complete stop), vehicle braking system intervention time (the time when the anti-lock braking system starts to intervene), brake pressure (pressure value in the brake system), brake frequency (the number of brakes per unit time) and the main frequency components of the brake pressure (the frequency components obtained by Fourier transform) are obtained from the actual driving record of the driver. In other words, the data obtained from the actual driving record of the driver can determine the corresponding average brake pressure, standard deviation of brake pressure, maximum brake pressure, average value of braking duration, brake frequency and main frequency components of brake pressure in different braking behavior modes.

[0106] In this embodiment, the calculation formula for the vector distance between the actual feature vector and each model feature vector is specifically:

[0107]

[0108] in: is the vector distance, is the ith element in the actual eigenvector, is the i-th element in the model feature vector, is the index number of the element in the vector, is the number of elements in the vector, is the actual eigenvector, is the model feature vector.

[0109] That is to say, the vector distances between the actual feature vector extracted by step S2.1 and the three model feature vectors extracted above are:

[0110] The vector distance between the model feature vector A is:

[0111]

[0112] The vector distance between the model feature vector B is:

[0113]

[0114] The vector distance between the model feature vector C is:

[0115]

[0116] That is, the vector distance between the actual feature vector and the model feature vector A is 1.438, the vector distance between the actual feature vector and the model feature vector B is 1.436, and the vector distance between the actual feature vector and the model feature vector C is 0.05.

[0117] Step S2.3: The three vector distances obtained in step S2.2 are compared collectively to determine the vector distance with the minimum value, wherein the predefined braking system response model corresponding to the vector distance with the minimum value is the braking system response model corresponding to the driver's braking behavior mode.

[0118] In the specific implementation process, among all the vector distances between the actual feature vector and the model feature vector, the vector distance between the actual feature vector and the model feature vector C has the smallest value. That is, among the three predefined braking system response models, the driver is executing model C (continuous pedaling model) at this time.

[0119] Step S3: The braking system response model matched in step S2.3 interacts with the simulated braking model, and the simulation results of the simulated braking model are displayed through the display interface, and the vehicle braking system is adjusted according to the simulation results. In this embodiment, in the process of displaying the simulation results of the simulated braking model, the simulation results can be displayed in the form of charts, animations, etc., so that the driver can see the response of the current vehicle braking system at a glance. The specific implementation can be specifically selected according to the simulation results, including:

[0120] When the simulation result is the same as the driver's preset driving result, the vehicle braking system can be directly adjusted according to the corresponding braking parameters in the braking system response model matched in step S2.3.

[0121] On the contrary, when the simulation result is different from the preset driving result of the driver, the brake system response model needs to be adaptively adjusted. It is worth noting that in the process of adaptive parameter adjustment in this embodiment, it is adjusted according to the driving state of the driver. At the same time, the adaptively adjusted brake system response model will interact with the simulated brake model until the simulation result is the same as the preset driving result of the driver.

[0122] Furthermore, when the driver is in driving state, the braking system response model is adjusted in an orderly manner according to the braking distance, braking time and braking smoothness, specifically:

[0123] When the braking distance is incorrect, only the braking distance is adjusted, and the braking time and braking smoothness are not adjusted.

[0124] When the braking distance is correct but the braking time is incorrect, only the braking time is adjusted, and the braking distance and braking smoothness are not adjusted.

[0125] When the braking distance and braking time are correct but the braking stability is incorrect, only the braking stability is adjusted, and the braking distance and braking time are not adjusted.

[0126] Furthermore, when the driver is not in the driving state, the driver adjusts the braking distance, braking time and braking smoothness within the preset braking parameter adjustment range through the interactive interface. It is worth noting that the braking parameter adjustment range in this embodiment is set according to the safety standard data, driving habit data and expert opinion data, that is, the maximum intersection data range determined from the three data sets of safety standard data, driving habit data and expert opinion data is the braking parameter adjustment range in this embodiment. Among them, the safety standard data is the data obtained from the vehicle manufacturer, relevant laws and regulations and industry standards, the driving habit data is the data obtained from actual driving records, user questionnaires and other channels, and the expert opinion data is the opinions and suggestions of professionals collected through interviews, questionnaires and other methods.

[0127] In the process of specific implementation, the initial sensor data is obtained as follows:

[0128] Braking distance: 8m.

[0129] Braking time: 1.5s.

[0130] Vehicle braking system intervention time: 0.25s in advance.

[0131] The driver's preset driving results are:

[0132] Braking distance: 7.5m.

[0133] Braking time: 1.5s.

[0134] Vehicle braking system intervention time: 0.25s in advance.

[0135] Furthermore, in this embodiment, it is detected that the driver is currently in a driving state, and there is a difference between the simulation result and the preset driving result, that is, the braking distance is 8m>7.5m, and the braking time and the vehicle braking system intervention time are correct. At this time, the braking distance is automatically adjusted from 8m to 7.5m, and the simulation is performed again through the simulation braking model. At the same time, the new simulation results are as follows:

[0136] Braking distance: 7.5m.

[0137] Braking time: 1.5s.

[0138] Vehicle braking system intervention time: 0.25s in advance.

[0139] At this point, it indicates that the simulation results are consistent with the preset driving results, and the adaptive adjustment is completed.

[0140] Furthermore, in this embodiment, if it is detected that the driver is not currently in a driving state, the driver can adjust the parameters by himself through the interactive interface. That is, the driver can directly adjust the braking distance to 7.5m in the interactive interface, so that the simulation result displayed on the interactive interface is consistent with the preset driving result.

[0141] This embodiment also provides a real-time analysis system based on brake data, which uses the above-mentioned real-time analysis method based on brake data. The specific implementation process of the real-time analysis system based on brake data is as follows:

[0142] In the specific implementation process, the sensor data collected in real time by the driver while driving the vehicle are as follows:

[0143] Brake pedal position: 20%-80%.

[0144] Vehicle speed: 0-100km / h.

[0145] Braking distance: 5-15m.

[0146] Braking time: 1-3s.

[0147] Vehicle braking system intervention time: 0.1-0.5s in advance.

[0148] The following features can be extracted from the above sensor data:

[0149] Statistical characteristics:

[0150] Average braking distance: 10m.

[0151] Average braking time: 2s.

[0152] Average vehicle braking system intervention time: 0.3s in advance.

[0153] Time domain characteristics:

[0154] Brake pedal position change rate: 0.5% / s.

[0155] Speed ​​change rate: 5 .

[0156] Frequency domain characteristics:

[0157] Main frequency component: 2Hz.

[0158] Through the data analysis of the above statistical features, time domain features and frequency domain features, it can be seen that the driver's braking behavior mode is a continuous braking mode. The predefined braking system response model corresponding to the continuous braking mode is:

[0159] Braking distance: 7-9m.

[0160] Braking time: 1.3-1.8s.

[0161] Vehicle braking system intervention time: 0.15-0.35s in advance.

[0162] Furthermore, after adjusting the vehicle braking system parameters according to the above matching results, the parameters are:

[0163] Braking distance: 8m.

[0164] Braking time: 1.5s.

[0165] Vehicle braking system intervention time: 0.25s in advance.

[0166] Example 2

[0167] refer to Figure 3 and Figure 4 This embodiment provides a real-time analysis method based on brake data, and its specific implementation method is the same as that of Example 1. The difference is that the present invention is illustrated below in combination with the specific implementation method of this embodiment.

[0168] In this embodiment, the driving data of the driver is taken as an example, and the vehicle braking system is adjusted according to the simulation results of the simulated braking model. Different from the adjustment method in embodiment 1, in this embodiment, adaptive parameter adjustment is performed according to the driving preference behavior of the driver, as follows:

[0169] Step SB3.1: Determine the driver's driving preference behavior based on the real-time operating data of the vehicle's braking system (e.g., braking pressure, braking time, and braking distance). That is, use data mining and machine learning technology to conduct in-depth analysis of the collected braking data to extract the driver's driving preference characteristics. For example, through the distribution of braking pressure and braking time, determine whether the driver tends to brake suddenly or smoothly. Through the braking distance, evaluate the driver's ability to predict and control the braking distance.

[0170] It is worth noting that before extracting the driver's driving preference characteristics, the collected braking data needs to be preprocessed, including but not limited to cleaning (such as denoising, filling missing values) and normalization to improve the accuracy of the braking data.

[0171] In this embodiment, extracting the driver's driving preference features includes:

[0172] Step SB3.1.1: Preprocess the operating data of the vehicle braking system, and determine the relationship between the data distribution and the driver's driving behavior based on the preprocessed operating data, specifically:

[0173] In this embodiment, the central tendency and dispersion degree of the operation data are obtained through the distribution of the braking pressure and the braking time, and the braking habit of the driver is determined. That is, the central tendency in the braking process is determined by calculating the average braking pressure and the average braking time. The dispersion degree in the braking process is determined by calculating the standard deviation of the braking pressure and the standard deviation of the braking time.

[0174] During the specific implementation process, the braking data shown in the following table is collected in real time:

[0175] Time (s) Brake pressure (%) Braking time (s) 1 35 0.6 2 40 0.5 3 38 0.55 4 36 0.6 5 37 0.5 6 39 0.55 7 35 0.6 8 41 0.5 9 38 0.55 10 36 0.6

[0176] According to the data in the above table, refer to Figure 5 and Figure 6 It can be seen that the average braking pressure is 37.30%, the average braking time is 0.55s, the standard deviation of braking pressure is 2.05%, and the standard deviation of braking time is 0.04s.

[0177] That is to say, the average braking pressure of the drivers is 37.30%, and the average braking time is 0.55 seconds, indicating that the pressure and time of the drivers during braking are relatively stable. The standard deviation of the braking pressure is 2.05%, and the standard deviation of the braking time is 0.04 seconds, indicating that the pressure and time of the drivers in different braking events fluctuate less and their braking habits are relatively consistent.

[0178] Furthermore, the influencing factors of the driver's driving preference are determined through the correlation between the braking data. That is, the correlation between different features is determined through the correlation coefficient, such as the correlation coefficient between braking pressure and braking time, which is specifically:

[0179]

[0180] in: is the correlation coefficient between braking pressure and braking time, is the braking pressure of the i-th braking event, is the average brake pressure of all braking events, is the braking time of the i-th braking event, is the average braking time of all braking events, is the total number of braking events.

[0181] Furthermore, different driving preference groups are identified through cluster analysis among braking data (such as K-means, DBSCAN, etc.).

[0182] Step SB3.1.2: The driver's braking habits, driving preference influencing factors and driving preference groups determined in the above step SB3.1.1 are comprehensively set to obtain comprehensive feature data, and the comprehensive feature data is used as input of a machine learning model (such as support vector machine, decision tree and random forest) to output the driver's driving preference behavior.

[0183] Step SB3.2: Match the simulation results of the simulated braking model with the driving preference behavior determined in step SB3.1.2, determine the driving preference behavior corresponding to the simulation results, and adaptively adjust the vehicle braking system according to the preference data corresponding to the driving preference behavior.

[0184] During the specific implementation process, the simulation results are as follows:

[0185] The simulated braking distance is: 8m.

[0186] The simulated braking time is: 1.5s.

[0187] The simulated vehicle braking system intervention time is: 0.25s in advance.

[0188] The machine learning model determines that the driver's driving preference behavior is "smooth braking", and the corresponding preference data is:

[0189] The simulated braking distance is: 7-9m.

[0190] The simulated braking time is: 1.3-1.8s.

[0191] The simulated vehicle braking system intervention time is: 0.15-0.35s in advance.

[0192] According to the preference data corresponding to the driving preference behavior, the braking system parameters are adjusted as follows:

[0193] The simulated braking distance is 8m, which is in line with the preferred data range (7-9m) and no adjustment is made.

[0194] The simulated braking time is 1.5s, which is in line with the preferred data range (1.3-1.8s) and no adjustment is made.

[0195] The simulated vehicle braking system intervention time is 0.25s in advance, which is in line with the preferred data range (0.15-0.35s in advance) and no adjustment is made.

[0196] That is, the simulation result is compared with the preferred data range, and when the simulation result is not within the preferred data range, the brake system parameters are adjusted to be within the preferred data range. If the above brake data are all within the preferred data range, the above brake data do not need to be changed.

[0197] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is limited by the attached embodiments and their equivalents.

Claims

1. A real-time analysis method based on brake data, characterized in that: The analysis method includes: Mode determination: Obtain real-time operating data of the vehicle's braking system to identify and determine the driver's braking behavior pattern; Pattern matching: setting a plurality of predefined braking system response models, and automatically matching the driver's braking behavior pattern with the braking system response model; Interactive adjustment: The matched braking system response model interacts with the simulated braking model, and the simulation results of the simulated braking model are displayed through the display interface, and the vehicle braking system is adjusted according to the simulation results, specifically: When the simulation result is the same as the preset driving result of the driver, adjusting the vehicle braking system according to the braking parameters corresponding to the simulated braking model; When the simulation result is different from the driver's preset driving result, the braking system response model performs adaptive parameter adjustment, and the adaptively adjusted braking system response model interacts with the simulated braking model until the simulation result is the same as the driver's preset driving result.

2. A real-time analysis method based on brake data according to claim 1, characterized in that: Identifying and determining the driver's braking behavior pattern, including: Acquire real-time operation data of the vehicle braking system, and extract statistical features, time domain features and frequency domain features from the real-time operation data; The statistical features, time domain features and frequency domain features are classified by a recognition algorithm to determine the driver's braking behavior patterns: light braking, heavy braking and continuous braking.

3. The method for real-time analysis based on brake data according to claim 1, characterized in that: The setting of the braking system response model includes: By extracting statistical features, time domain features and frequency domain features from the real-time operation data of the vehicle braking system, the actual feature vector corresponding to the driver's braking behavior mode is determined; Obtaining a model feature vector in each predefined braking system response model, and determining a vector distance between the actual feature vector and each model feature vector; All vector distances are compared to determine the vector distance with the minimum value. The predefined braking system response model corresponding to the vector distance with the minimum value is the braking system response model corresponding to the braking behavior mode of the driver.

4. The method for real-time analysis of brake data according to claim 3, characterized in that: The vector distance between the actual feature vector and each model feature vector is specifically: ,in: is the vector distance, is the ith element in the actual eigenvector, is the i-th element in the model feature vector, is the index number of the element in the vector, is the number of elements in the vector, is the actual eigenvector, is the model feature vector.

5. The real-time analysis method based on brake data according to claim 1 is characterized in that: When the brake system response model performs adaptive parameter adjustment, the adjustment is performed according to the driving state of the driver, specifically: When the driver is in driving state: The braking system response model is adjusted in an orderly manner according to the braking distance, braking time and braking smoothness; When the driver is not in driving state: The driver adjusts the braking distance, braking time and braking stability within a preset braking parameter adjustment range through an interactive interface.

6. The method for real-time analysis based on brake data according to claim 5, characterized in that: The braking system response model is adjusted in an orderly manner according to the braking distance, braking time and braking stability, specifically: When the braking distance is incorrect, the braking distance is adjusted, and the braking time and braking smoothness are not adjusted; When the braking distance is correct but the braking time is incorrect, the braking time is adjusted, and the braking distance and braking stability are not adjusted; When the braking distance and the braking time are correct but the braking stability is incorrect, the braking stability is adjusted, and the braking distance and the braking time are not adjusted.

7. The method for real-time analysis based on brake data according to claim 1, characterized in that: Adjusting the vehicle braking system according to the simulation results also includes: Determine the driver's driving preference behavior based on real-time operating data of the vehicle's braking system; The simulation result of the simulated braking model is matched with the driving preference behavior to determine the driving preference behavior corresponding to the simulation result, and the vehicle braking system is adaptively adjusted according to the preference data corresponding to the driving preference behavior.

8. The method for real-time analysis based on brake data according to claim 7, characterized in that: The determination of the driver's driving preference behavior includes: The operation data of the vehicle braking system is preprocessed, and the relationship between the data distribution and the driver's driving behavior is determined based on the preprocessed operation data, specifically: By means of the distribution of braking pressure and braking time, the central tendency and the dispersion degree of the operating data are obtained to determine the braking habits of the driver; Determine the factors affecting the driver's driving preference through the correlation between the braking data; Through cluster analysis of braking data, different driving preference groups are identified; The driver's braking habits, driving preference influencing factors and driving preference groups are comprehensively set to obtain comprehensive feature data, and the comprehensive feature data is used as input of a machine learning model to output the driver's driving preference behavior.

9. A real-time analysis system based on brake data, characterized in that: A real-time analysis method based on braking data as described in any one of claims 1-8 is used.

Images