A Controller Strategy Optimization Method for User Behavior Analysis

Abstract

The present application provides a method for optimizing the controller strategy for user behavior analysis, which relates to the technical field of controllers and includes: collecting user driving parameters; identifying and segmenting the speed risk level and speed change characteristics according to the driving speed to determine the speed adaptive interval; respectively evaluating the steering and braking risks of the speed adaptive interval according to the intensity and frequency of braking and steering; performing controller strategy matching analysis according to the stability risk coefficient to determine the optimization adjustment coefficient and adding it to the speed adaptive interval; matching the monitored current driving speed with the speed adaptive interval to obtain the optimization adjustment coefficient and adjusting the controller strategy. Through the present application, the technical problem in the prior art that the electric tricycle is prone to insufficient stability and rollover when turning at high speed, resulting in unstable driving, can be solved. According to the actual driving behavior of the user and the vehicle state, the control strategy is dynamically adjusted to improve the driving safety and stability.

Description

Technical Field

[0001] This application relates to the technical field of controllers, and in particular to a method for optimizing controller strategies for user behavior analysis. Background Art

[0002] As an important means of transportation, electric tricycles are widely used in fields such as short-distance transportation, urban distribution, and rural travel. With a simple structure, low cost, and convenient operation, they play an important role in the fields of logistics transportation, cargo handling, and individual travel. Electric tricycles usually adopt a layout form with front-wheel steering and rear-wheel drive, and can be divided into cargo-carrying type, passenger-carrying type, and comprehensive type according to different uses. However, due to the design of electric tricycles that needs to accommodate a relatively large cargo space or passenger area, their center of gravity is often higher than that of two-wheeled electric vehicles or ordinary cars, making them prone to instability when turning at high speeds. Especially when carrying cargo or passengers, the center of gravity may deviate from the center line, further increasing the risk of rollover.

[0003] The controller of an electric tricycle is one of the core electronic components, responsible for precisely controlling the speed, torque, and braking system of the electric motor to ensure the efficient and stable operation of the electric tricycle. However, most of the existing control strategies for electric tricycle controllers are based on fixed algorithms, usually without considering the dynamic factors in complex driving environments and lacking real-time response to the individual driving behaviors of drivers. Therefore, when facing emergencies or high-speed turning, it is easy to cause insufficient vehicle stability and increase the risk of rollover.

[0004] In summary, in the prior art, due to factors such as the structural characteristics, center of gravity distribution, and user operations of electric tricycles, there are technical problems that the vehicle is prone to insufficient stability and rollover when turning at high speeds, further affecting driving safety. Summary of the Invention

[0005] The purpose of this application is to provide a method for optimizing controller strategies for user behavior analysis to solve the technical problems in the prior art that due to factors such as the structural characteristics, center distribution, and user operations of electric tricycles, the vehicle is prone to insufficient stability and rollover when turning at high speeds, further affecting driving safety.

[0006] In view of the above problems, the present application provides a method for optimizing the controller strategy for user behavior analysis. Among them, the method for optimizing the controller strategy for user behavior analysis includes: collecting user driving parameters, where the user driving parameters include driving speed, the intensity and frequency of braking and steering; identifying and segmenting the speed risk level and speed change characteristics according to the driving speed, and determining the speed adaptive interval, where the speed adaptive interval includes the speed risk level and the speed change amplitude; respectively evaluating the risk of steering and braking for the speed adaptive interval according to the intensity and frequency of braking and steering to obtain the stability risk coefficient; performing controller strategy matching analysis according to the stability risk coefficient to determine the optimization adjustment coefficient, and adding the optimization adjustment coefficient to the speed adaptive interval to generate an interval optimization adjustment label; matching the monitored current driving speed with the speed adaptive interval to obtain the optimization adjustment coefficient corresponding to the adaptive interval, and adjusting the controller strategy according to the optimization adjustment coefficient.

[0007] The technical solution provided in the present application has at least the following technical effects or advantages:

[0008] By collecting user driving parameters, where the user driving parameters include driving speed, the intensity and frequency of braking and steering; identifying and segmenting the speed risk level and speed change characteristics according to the driving speed, and determining the speed adaptive interval, where the speed adaptive interval includes the speed risk level and the speed change amplitude; respectively evaluating the risk of steering and braking for the speed adaptive interval according to the intensity and frequency of braking and steering to obtain the stability risk coefficient; performing controller strategy matching analysis according to the stability risk coefficient to determine the optimization adjustment coefficient, and adding the optimization adjustment coefficient to the speed adaptive interval to generate an interval optimization adjustment label; matching the monitored current driving speed with the speed adaptive interval to obtain the optimization adjustment coefficient corresponding to the adaptive interval, and adjusting the controller strategy according to the optimization adjustment coefficient. That is to say, by collecting user driving parameters, comprehensively understanding the user's driving habits and behavior patterns, real-time evaluating the driving state of the vehicle, evaluating the risk of braking and steering during driving, determining the stability coefficient, selecting the most suitable control strategy for the current driving state, dynamically matching and adjusting the controller strategy to adapt to the user's driving characteristics and vehicle state, and improving the safety and stability of vehicle driving.

[0009] The above description is only an overview of the technical solution of the present application. In order to better understand the technical means of the present application, it can be implemented according to the content of the specification. In order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically exemplified below. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understandable through the following description of the specification. Brief Description of the Drawings

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0011] Figure 1 It is a schematic flow chart of a method for optimizing the controller strategy for user behavior analysis in the present application.

[0012] Figure 2 It is a schematic flow chart of determining the optimization adjustment coefficient in a method for optimizing the controller strategy for user behavior analysis in the present application. Detailed Description of the Embodiments

[0013] By providing a method for optimizing the controller strategy for user behavior analysis, the present application solves the technical problem in the prior art that due to factors such as the structural characteristics, center distribution, and user operations of electric tricycles, it is prone to insufficient stability and rollover during high-speed turning, further affecting driving safety. By collecting user driving parameters, comprehensively understanding the driving habits and behavior patterns of users, real-time evaluating the driving state of the vehicle, conducting risk assessments on braking and steering during driving, determining the stability coefficient, selecting the most suitable control strategy for the current driving state, dynamically matching and adjusting the controller strategy to adapt to the driving characteristics of users and the vehicle state, the driving safety and stability of the vehicle are improved.

[0014] Next, the technical solutions in the present application will be clearly and completely described with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described here. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of convenience of description, only the parts related to the present application are shown in the drawings rather than all of them.

[0015] Embodiment, please refer to the attachedFigure 1 , this application provides a method for optimizing the controller strategy of user behavior analysis. Specifically, the method for optimizing the controller strategy of user behavior analysis includes the following steps:

[0016] Step 1: Collect user driving parameters, where the user driving parameters include driving speed, braking and steering force and frequency.

[0017] Specifically, sensors on the vehicle (such as acceleration sensors, steering angle sensors, brake pressure sensors, etc.) are used to collect key data during driving in real time, including driving speed, braking and steering force and frequency, etc. The driving speed is obtained through in-vehicle GPS or wheel speed sensors; the braking force is measured by the pressure sensor on the brake pedal; the steering force is obtained through the force sensor of the steering system or the current signal of the electric power steering system, or by measuring the rotation angle and force of the steering wheel; the braking frequency and steering frequency are realized by detecting the timing of braking or steering operations, usually recorded by acceleration sensors or a dedicated clock synchronization system. The braking frequency refers to the frequency of the driver applying braking operations within a certain period of time, and the steering frequency refers to the frequency of the driver performing steering operations, reflecting the driver's driving habits (for example, frequent braking may indicate that the driver's operation is relatively intense or encounters more traffic interference). A higher steering frequency may mean that the driver is frequently changing lanes or turning, especially in urban traffic, which is usually closely related to driving habits, road conditions, and traffic flow. By analyzing the collected driving data, the driver's operation habits can be identified. For example, frequent hard braking may mean that the driver has no foresight when facing traffic conditions, while frequent high-speed steering may reflect a more aggressive driving style. By collecting data on driving speed, braking force, and steering behavior in real time, detailed information about the user's driving behavior can be obtained in a timely manner and responses can be made.

[0018] Step 2: Perform speed risk level and speed change characteristic identification and segmentation based on the driving speed, and determine the speed adaptive interval, where the speed adaptive interval includes speed risk level and speed change amplitude.

[0019] Specifically, conduct multiple tests on different vehicle models and collect vehicle speed test data. Perform speed stability risk level division, that is, conduct risk assessment on the stability of different vehicle models under different speed conditions to determine the corresponding speed risk levels. According to the risk division, determine multiple speed risk level intervals, including speed risk intervals and speed change amplitude intervals. Calculate the speed change amplitude based on the driving speed of the vehicle, and match and map the driving speed and speed change amplitude to multiple speed risk level intervals to determine the speed risk level corresponding to the current driving speed. The speed risk level division is to map and determine the corresponding risk level according to the driving speed of the vehicle in the speed risk interval, and match the driving speed with the corresponding speed range. The speed change characteristic identification is to map and determine the corresponding risk level according to the calculated speed change amplitude in the speed change amplitude interval, and identify which speed changes belong to relatively violent fluctuations.

[0020] Divide according to different risk levels and speed change amplitudes to ensure that the vehicle can maintain the best stability under different driving conditions, and obtain the speed adaptive interval. The speed adaptive interval is a dynamically adjusted interval, which is set according to the real-time driving speed of the vehicle, the driver's behavior, and the vehicle's stability requirements. The parameters within each interval may be different and will be automatically adjusted according to the current speed, vehicle condition, and external environmental factors to ensure the best driving experience and safety. The speed adaptive interval includes the speed risk level and the speed change amplitude. Among them, the speed risk level refers to the risk level divided according to the stability evaluation results within different speed ranges; the speed change amplitude refers to the change amount of the vehicle speed, indicating the severity of the vehicle's acceleration or deceleration. The shorter the time of strong speed change, the greater the change amplitude. By dividing the speed risk level and the speed change amplitude, adjust the control strategy in real time under different driving conditions to improve the stability of the vehicle during high-speed driving and sharp turns, thereby reducing the risks of rollover and loss of control.

[0021] Step 3: Respectively conduct steering and braking risk assessment on the speed adaptive interval according to the intensity and frequency of the braking and steering to obtain the stability risk coefficient.

[0022] Specifically, according to the braking and steering forces, obtain the operation duration of braking and steering. Using the operation force and operation duration of each braking or steering, obtain the operation intensity. According to the correspondence between the operation intensity and the frequency of braking and steering, obtain the operation frequencies of different operation intensities. Calculate the frequency discreteness of the operation frequencies of different operation intensities, and divide the corresponding characteristic evaluation levels according to the frequency discreteness using the preset level discrete interval values. According to the braking and steering force and frequency data, conduct an analysis of the user's behavior characteristics, associate the analyzed behavior characteristics with the user's driving mode, and determine the user's driving operation characteristics, namely conservative driving, steady driving, and aggressive driving.

[0023] According to the driving speed of the vehicle, the operation forces and frequencies of braking and steering, determine the corresponding speed adaptive interval, which is a dynamically changing range and will be continuously adjusted according to the speed, the intensity and frequency of the control behavior. Take the user's driving operation characteristics, the speed value and the speed change amplitude of the speed adaptive interval as input variables, input them into the risk prediction model to conduct risk predictions for different driving speeds within the speed adaptive interval respectively, and obtain the qualitative risk coefficients. Use the stability risk coefficients and the speed adaptive interval to automatically adjust the control strategy to adapt to different driving conditions and operation styles, thereby improving the stability and safety of the vehicle.

[0024] Step 4: Conduct a controller strategy matching analysis according to the stability risk coefficient, determine the optimization adjustment coefficient, and add the optimization adjustment coefficient to the speed adaptive interval to generate an interval optimization adjustment label.

[0025] Specifically, by analyzing historical data and conducting experiments, establish the control influence relationship between the steering stability risk, the braking stability risk and the controller strategy parameters, and determine which control parameters can be adjusted to solve the stability problems existing in braking and steering. Define a stability threshold to determine when to adjust the controller strategy parameters, which is a key indicator for measuring the safety of the vehicle. When the stability risk coefficient exceeds this threshold, trigger parameter adjustment to reduce the risk. Collect a large amount of historical data, calculate the stability risk coefficient of the vehicle under different driving conditions, and at the same time collect the strategy parameters (such as braking intensity, steering angle, etc.) of the control system and their changes during each experiment. Through the collected historical data, use the technique of fitting experimental data to establish the adjustment relationship between the stability risk coefficient and the controller strategy parameters, which is used to characterize the parameter adjustment amplitude matching the stability risk coefficient, that is, determine how much adjustment needs to be made to the controller strategy parameters according to the magnitude of the stability risk coefficient.

[0026] Adjust the parameter control coefficient in the control influence relationship using the adjustment relationship. The parameter control coefficient is part of the controller strategy parameters and determines the degree of adjustment of the controller to vehicle behavior. By adjusting the parameter control coefficient, an optimized adjustment coefficient is obtained, which is calculated based on the current stability risk coefficient and the preset adjustment relationship, aiming to optimize the stability and safety of the vehicle. For example, when the stability risk coefficient is low (such as 0.2), indicating that the current driving state is relatively safe, the controller can provide a more sensitive response, such as a higher steering angle and acceleration response; when the stability risk coefficient is high (such as 0.9), indicating that the current driving state has a higher risk, the controller needs to make a more conservative response, such as reducing the steering angle, decreasing the acceleration response, or strengthening the braking response, such as increasing the braking intensity by 20% and reducing the steering angle by 10%.

[0027] Add the optimized adjustment coefficient to the speed adaptive interval to generate an interval optimized adjustment label, which is used to identify the controller parameter adjustment strategy that should be applied within each speed interval, so that the controller can automatically select an appropriate control strategy according to the vehicle's real-time driving state and the user's driving behavior, adjust the controller parameters, and provide a more stable and safe driving experience. By performing controller strategy matching analysis based on the stability risk coefficient, generating the optimized adjustment coefficient and applying it to the speed adaptive interval, the control parameters are dynamically adjusted for each speed interval and stability risk coefficient, improving the vehicle's stability, especially in high-risk situations, and ensuring that the vehicle does not get out of control.

[0028] Step Five: Match the monitored current driving speed with the speed adaptive interval to obtain the optimized adjustment coefficient corresponding to the adaptive interval, and adjust the controller strategy according to the optimized adjustment coefficient.

[0029] Specifically, the current driving speed is monitored in real time through a speed sensor, and it is matched within the speed adaptation interval to determine the corresponding speed interval. The speed adaptation interval is determined based on the driving speed and the speed change amplitude of the vehicle, and each interval has a corresponding optimization adjustment coefficient. The matched interval corresponds to a set of optimization adjustment coefficients, and the preset optimization adjustment coefficients are extracted. The optimization adjustment coefficients are obtained by fitting and calculating historical driving data and a risk assessment model, and are specifically used to adjust the operating parameters of the controller. According to the obtained optimization adjustment coefficients, the controller strategy is adjusted, which means that according to the real-time driving state of the vehicle and the driving behavior of the user, the controller parameters are automatically adjusted to provide a more stable and safe driving experience. For example, if the current speed of the vehicle is 50 km / h and it is matched to the medium-speed interval; the optimization adjustment coefficients are extracted: braking intensity 1.2, steering sensitivity 0.8; the controller parameters are adjusted in real time, the braking intensity is increased from 100 to 120; the steering sensitivity is decreased from 1.0 to 0.8. When the strategy is not adjusted, the rollover probability of the vehicle within the medium-speed interval is 15%, and the driving stability score is 70%; after the strategy is adjusted, the rollover probability of the vehicle is reduced to 8%, and the driving stability score is increased to 85%. Dynamically adjusting the control strategy according to the real-time speed effectively reduces the stability risk caused by speed changes during driving, makes the handling performance of the vehicle in different speed intervals more in line with driving needs, and improves the overall comfort. According to the matching of the current driving speed and the speed adaptation interval, using the optimization adjustment coefficients to adjust the controller strategy realizes the dynamic optimization of vehicle control, ensures driving safety and comfort, and at the same time reflects the advantages of intelligent control.

[0030] Further, step two of the present application includes:

[0031] Obtain vehicle speed test data, where the vehicle speed test data has a vehicle model; perform a speed stability risk level division according to the vehicle speed test data to determine a speed risk level interval, including a speed risk interval and a speed change amplitude interval; calculate the speed change amplitude according to the driving speed to obtain the speed time series change amplitude; use the speed risk interval and the speed change amplitude interval to perform matching mapping on the driving speed and the speed time series change amplitude to obtain multiple speed risk level intervals; construct the speed adaptation interval according to the multiple speed risk level intervals.

[0032] Specifically, professional testing equipment, such as a speed recorder, is used to collect speed data of vehicles under specific test conditions. Each vehicle speed test data is marked with the corresponding vehicle model. Different models of vehicles vary in design, weight distribution, power system, etc., which affects their driving performance. The speed test data includes indicators such as acceleration performance, stability performance, and braking performance at different speeds of the vehicle. According to the vehicle speed test data, the stability under different speed conditions is evaluated. Through statistical analysis of the speed data, such as average speed, maximum speed, minimum speed, and speed fluctuation range, the risk level of speed is determined, including the speed risk interval and the speed change amplitude interval. The speed risk interval divides the speed range into several intervals, such as low speed, medium speed, high speed, etc., according to the stability performance of the vehicle at different speeds, and each interval corresponds to a different risk level. The speed change amplitude interval analyzes the change amplitude of the vehicle speed, such as the acceleration or deceleration rate, and divides it into different intervals, such as small amplitude, medium amplitude, large amplitude, etc., to evaluate the impact of speed change on stability.

[0033] Calculate the change amplitude of the driving speed, that is, the change amount of the vehicle speed within a certain time, which reflects the severity of the vehicle's acceleration or deceleration. By performing a difference calculation on the driving speed data of the vehicle (i.e., the speed difference between adjacent moments), the time-series change amplitude of the speed is obtained, that is, the change rate of the speed change over time. Match and map the real-time collected driving speed with the predefined speed risk interval, and match and map the time-series change amplitude of the speed calculated based on the driving speed with the predefined speed change amplitude interval to determine the risk level interval in which the current driving state of the vehicle is located. Matching and mapping is a data processing method that compares a set of data (such as the real-time monitored driving speed and speed change amplitude) with another set of predefined data intervals (such as the speed risk interval and speed change amplitude interval) to determine the category or level to which the data belongs. The multi-speed risk level interval refers to dividing the driving state of the vehicle into multiple risk levels according to different combinations of driving speed and speed change amplitude.

[0034] According to the multi-speed risk level interval, a speed adaptive interval is constructed, that is, a speed range that automatically adjusts the control strategy according to the real-time driving data of the vehicle (current speed and speed change amplitude). The speed adaptive interval allows for dynamic adjustment of the control strategy, enabling the vehicle to maintain optimal stability under different speed conditions. By real-time dividing the speed risk interval and speed change amplitude interval, the control strategy of the vehicle is accurately adjusted within different speed ranges, effectively reducing the stability risk during high-speed driving and reducing the possibility of rollover and loss of control. By adaptively adjusting the controller parameters, a smoother and more comfortable driving experience can be provided according to different driving speeds and behavior patterns.

[0035] Furthermore, step three of this application includes:

[0036] According to the force and frequency of braking and steering, the user's braking and steering behavior characteristics are analyzed to determine the user's driving operation characteristics; the user's driving operation characteristics and the speed value and speed change amplitude of the speed adaptation interval are used as input variables, and the risk prediction model is used to perform risk prediction on each speed adaptation interval to obtain the stability risk coefficient.

[0037] Specifically, the force of braking and steering refers to the force applied by the user when operating the brakes and steering during driving. For braking, the force determines the strength of the brakes; for steering, the force determines the steering angle or the speed of the steering. The frequency of braking and steering refers to the frequency of the user's braking and steering operations during driving, that is, the number of operations. Frequent braking or steering usually means that the driver is in a more complex driving environment, which may lead to higher operational risks. The force of each braking and steering operation performed by the user when operating the electric tricycle is obtained, and the duration is recorded, that is, the length of time the driver maintains the operation each time the brake or steering is performed. The operation intensity is calculated according to the operation force and operation duration of each braking or steering, and the operation intensity = force × duration. The intensity of each operation is analyzed with its frequency to obtain the operation frequency corresponding to different operation intensity levels. The calculation of the operation frequency is based on the number of times the driver brakes or turns at a certain intensity. Different operation intensities correspond to different operation frequencies. According to the obtained operation frequencies corresponding to different operation intensities, the discreteness of the operation frequency is calculated. Usually, the discreteness of the operation frequency is obtained by calculating the mean and standard deviation of the operation frequencies of different operation intensities and normalizing the standard deviation. According to the results of frequency discreteness, the driver's operating behavior is divided into three types: conservative driving, smooth driving, and aggressive driving, and the user's driving operation characteristics are obtained. Behavioral characteristic evaluation refers to the classification and evaluation of the driver's operating behavior based on characteristics such as frequency discreteness, which is usually divided into three types: conservative driving, smooth driving, and aggressive driving.

[0038] A risk prediction model is a mathematical or machine learning model that predicts potential risks based on input variables such as user behavior, vehicle status, environmental factors, etc. In the optimization of the controller strategy of an electric tricycle, the risk prediction model is mainly used to predict the potential stability risks during driving, especially safety issues such as rollover and out-of-control. By analyzing data such as driver behavior, vehicle speed, acceleration / braking characteristics, and steering operations, the safety of the current driving state is judged. Historical driving parameters are obtained, including driving speed, braking and steering force and frequency, and environmental factors. Data preprocessing is performed on the historical driving parameters, including data cleaning, normalization or standardization, feature extraction, etc. By analyzing the correlation between the characteristics of the historical driving parameters and the target variable (such as the stability risk coefficient), the most influential features are selected, including speed, speed change amplitude, braking force and frequency, steering force and frequency. If the number of features is too large, dimensionality reduction is performed on the data by methods such as principal component analysis (PCA) to reduce redundant data and improve the model training efficiency. According to the nature of the problem and the data characteristics, a suitable risk prediction model is selected, including logistic regression, decision tree, random forest, support vector machine, neural network, etc.

[0039] To predict the vehicle stability risk, a random forest model is selected because it can automatically select important features and has strong expressive ability for non-linear relationships. It is especially suitable for processing high-dimensional data and is not prone to overfitting. The collected data is divided into a training set and a validation set. Random forest is an ensemble learning method based on the decision tree model. By constructing multiple decision trees and integrating the prediction results of each decision tree, the overall prediction accuracy is improved. Each tree is generated by randomly sampling the training data, which helps to reduce overfitting. The training set is used to train the model, and the validation set is used to evaluate the model performance. A common ratio is that 80% of the data is used for training and 20% for validation. The selected algorithm (such as random forest) is used to learn from the training set, and the model parameters are adjusted to maximize the prediction accuracy. To ensure the representativeness of the training data and prevent biases caused by the contingency of data division, cross-validation is usually used. For example, k-fold cross-validation divides the data into k subsets. Each time, one subset is selected as the validation set, and the other k - 1 subsets are used as the training set. The training process is repeated k times. Finally, the model performance is evaluated by calculating the average value of each validation result.

[0040] The random forest algorithm learns based on the features and labels (such as the stability risk coefficient) in the training data. The construction steps for each tree are as follows: A part of the samples in the training set is randomly selected for training. Each time the decision tree is split, a specific number of features are randomly selected for partitioning, which helps increase the diversity among the trees and avoid overfitting. Each tree is trained by continuously splitting nodes until a stopping condition is met (such as the maximum depth of the tree, the number of samples in the node is less than a certain threshold, etc.). After training, the final stability risk prediction is obtained by integrating the prediction results of all trees. After training, the model needs to be evaluated on the validation set, calculating metrics such as accuracy, precision, or calculating the mean squared error with the actual values. According to the evaluation results, the hyperparameters in the random forest (such as the number of trees, the maximum depth of the trees, the minimum number of samples per tree, etc.) are adjusted, or new features are added, etc., to optimize the model performance and obtain the final risk prediction model. Suppose after evaluation, the accuracy of the model on the validation set is 85%, but it is inaccurate in some extreme situations (such as during sharp turns). Therefore, more data on extreme driving behaviors are added, or a different algorithm is selected for further optimization.

[0041] Taking the user's driving operation features and the speed values and speed change amplitudes within the speed adaptive interval as input variables, the risk prediction model combines all input variables (including speed, speed change amplitude, braking force and frequency, steering force and frequency, etc.) to evaluate the risk and generate a stability risk coefficient, indicating the likelihood of the vehicle losing control or rolling over under the current driving conditions. By analyzing the braking and steering behaviors of different drivers, personalized control strategies are provided for each driver, thereby achieving more accurate risk prediction and safety protection, predicting potential stability risks in a timely manner during driving, and taking measures in advance, such as adjusting vehicle control parameters, to improve the safety of the vehicle.

[0042] Furthermore, the present application further includes the following steps:

[0043] Obtain the operation durations of braking and steering; according to the braking and steering forces, use the operation force and operation duration of each braking or steering to obtain the operation intensity; according to the corresponding relationship between the operation intensity and the braking and steering frequencies, conduct operation intensity frequency analysis to obtain the operation frequencies of different operation intensities; according to the frequency discreteness of the operation frequencies of different operation intensities, conduct behavior feature evaluation to obtain the user's driving operation features.

[0044] Specifically, by monitoring the pressure change of the brake pedal or the response time of the braking system, record the time from the start to the end of braking to obtain the continuous operation time of braking. Similarly, use an angle sensor or a steering force sensor to detect the time when the driver turns the steering wheel, record the time from the start to the end of steering to obtain the continuous operation time of steering. According to the braking and steering forces, combined with the operation force and operation duration of each braking or steering operation, obtain the operation intensity. The operation intensity refers to the comprehensive intensity of braking or steering operations, usually determined by two factors: operation force and operation duration. The greater the operation intensity, the more significant the impact of the operation on the vehicle. For braking operations, the operation intensity can be calculated by braking force × braking duration; for steering operations, the operation intensity can be calculated by steering force × steering duration. If the user applies a large force during a braking operation and it lasts for a long time, then the overall intensity of this operation will be high.

[0045] The operation frequency refers to the number of operations occurring per unit time. By analyzing the frequencies of operations with different intensities, obtain the distribution of operation intensity and frequency. Calculate the frequencies of operations with different intensities per unit time. For example, within a certain time period, count the intensity of each operation and classify the operation intensities into different intervals, and then calculate the operation frequency within each interval. For each operation intensity, record the braking and steering frequencies of the user to obtain the operation frequencies under different operation intensities.

[0046] Using the frequency standard deviation formula, calculate the average value of the operation frequencies of different operation intensities, and then calculate the frequency standard deviation to obtain the intensity standard deviation, which reflects the fluctuation magnitude of the driver's operation intensity. Perform normalization processing on the intensity standard deviation. By calculating the ratio of the intensity standard deviation to the average value of the operation frequencies, obtain the normalized frequency variation coefficient, which is used to measure the relative dispersion degree of the data. Classify driving behaviors into different levels according to the frequency discreteness, such as low discreteness level, medium discreteness level, and high discreteness level. According to the characteristic evaluation levels, classify the driver's driving styles, such as conservative driving, smooth driving, and aggressive driving, which correspond to the characteristic evaluation levels respectively. The characteristic evaluation level is a quantitative evaluation of the user's operation behavior characteristics, while the user's driving operation characteristics are a qualitative description of the user's driving style. Conservative driving generally means that the user operates more cautiously during driving, tends to accelerate and decelerate slowly, and makes fewer sharp turns; smooth driving means that the user operates stably, accelerates and decelerates evenly, and turns smoothly; aggressive driving means that the user operates more boldly, tends to accelerate and decelerate quickly, and makes frequent sharp turns. By analyzing the operation intensity and frequency discreteness, accurately identify the driver's driving style (conservative, smooth, aggressive), timely identify the driver's aggressive driving behavior, and adopt corresponding control strategies to reduce safety risks (such as rollover, out of control, etc.) caused by the driver's overly intense operation.

[0047] Further, this application also includes the following steps:

[0048] According to the operation frequencies of different operation intensities, calculate the intensity standard deviation through the frequency standard deviation formula; perform normalization processing based on the intensity standard deviation to obtain the frequency discreteness, and the frequency discreteness is represented by the normalized frequency variation coefficient; divide the feature evaluation levels according to the frequency discreteness, and the feature evaluation levels are determined by matching the frequency discreteness with preset level discrete interval values, and the feature evaluation levels include a low discreteness level, a medium discreteness level, and a high discreteness level; determine the user driving operation characteristics according to the feature evaluation levels, and the user driving operation characteristics include conservative driving, stable driving, and aggressive driving, which respectively correspond to the feature evaluation levels.

[0049] Specifically, obtain the operation frequencies of different intensity levels, which represent the operation frequencies of a driver for different operation intensities (such as braking, steering) within a certain time period. According to the operation frequencies of different operation intensities, calculate the intensity standard deviation using the frequency standard deviation formula. The standard deviation is an index that measures the degree to which a set of data deviates from its average value and is used to measure the fluctuation range of the operation intensity frequency, that is, the stability of the driver's operation intensity. If the standard deviation is large, it indicates that the change in the operation intensity is large and the driving behavior may be more intense; if the standard deviation is small, it indicates that the operation intensity is relatively stable. Next, perform normalization processing on the intensity standard deviation. By calculating the ratio of the intensity standard deviation to the average value of the operation frequency, the normalized frequency variation coefficient, that is, the frequency discreteness, is obtained, which reflects the discreteness of the operation frequency. If the frequency discreteness is large, it means that the driver's operation behavior is relatively aggressive and intense operations are frequently performed; if it is small, it indicates that the operation frequency is relatively stable and the driver's behavior is relatively conservative. For example, the frequencies (unit: times / minute) of 5 operations are recorded as 1.2, 1.5, 1.3, 1.6, and 1.4 within a certain period of time. The calculated frequency average value is 1.4 times / minute. Subsequently, the average value of the squared deviations is calculated as [(1.2 - 1.4)² + (1.5 - 1.4)² + (1.3 - 1.4)² + (1.6 - 1.4)² + (1.4 - 1.4)²] / 5 = 0.02, the intensity standard deviation is 0.1414, and the frequency discreteness is 0.101. The operation frequency refers to the number of times a driver completes a certain operation (such as braking or steering) per unit time and reflects the operation activity of the driver. The frequency standard deviation is used to measure the fluctuation degree of the operation frequency. The normalized frequency variation coefficient is the ratio of the standard deviation to the mean value and is used to measure the relative discreteness of the data, which can exclude the influence of different mean values. The purpose is to convert the intensity standard deviation into a dimensionless index so that the discreteness between different data sets can be compared.

[0050] The operation characteristics of the driver are divided into different evaluation levels according to the frequency discreteness. The characteristic evaluation level is a quantitative evaluation of the user's operation behavior characteristics, reflecting the stability and consistency of the operation frequency under different operation intensities. The preset level discrete interval value is a set of predefined intervals used to map the frequency discreteness index to different characteristic evaluation levels. The preset level discrete interval value can be adjusted according to specific requirements and actual application scenarios. The frequency discreteness is divided into characteristic evaluation levels according to the preset level discrete interval value, including low discreteness level, medium discreteness level, and high discreteness level. The low discreteness level indicates that the operation frequency of the user under different operation intensities changes little, and the operation behavior is relatively conservative; the medium discreteness level indicates that the operation frequency of the user under different operation intensities has a certain change, and the operation behavior is relatively stable; the high discreteness level indicates that the operation frequency of the user under different operation intensities changes greatly, and the operation behavior is relatively unstable.

[0051] According to the characteristic evaluation level, the driving operation characteristics of the user, that is, the driving style of the user, are determined, including conservative driving, stable driving, and aggressive driving, which correspond to the characteristic evaluation levels respectively. The low discreteness level corresponds to conservative driving, the medium discreteness level corresponds to stable driving, and the high discreteness level corresponds to aggressive driving. Through the fluctuation analysis of the operation frequency, the driving style of the driver, such as conservative, stable or aggressive driving, can be accurately distinguished. The electric tricycle controller can adjust the control strategy according to the driving habits of the driver. By real-time monitoring the operation behavior of the driver, when the controller finds that the driver's operation is too aggressive, it automatically adjusts the response of the vehicle to avoid possible safety hazards (such as out of control, rollover, etc.).

[0052] Furthermore, the present application further includes the following steps:

[0053] The expression formula of the frequency standard deviation is: ; The frequency discreteness = , where is the frequency standard deviation, F i is the operation frequency of each intensity level, is the mean value of the operation frequency, and N is the number of operation intensity levels.

[0054] Specifically, the expression formula of the frequency standard deviation is , where is the frequency standard deviation, indicating the degree of data discreteness or the amplitude of fluctuation; Fi is the operation frequency of each intensity level, indicating the operation frequency corresponding to the i-th operation intensity level (such as low intensity, medium intensity, high intensity, etc.), reflecting the frequency of occurrence of a specific operation intensity (such as braking or steering) within a certain period of time; is the average of the operation frequencies, that is, the average of all operation intensity frequencies. Add up the operation frequencies of N intensity levels and divide by N to get the average value; N is the number of operation intensity levels, that is, the number of different operation intensity categories considered in the analysis process. First, calculate the deviation between each operation intensity frequency and the mean value, which reflects the difference between the operation frequency and the mean value. Then calculate the square of each deviation to avoid the influence of negative values on the calculation and at the same time amplify the influence of larger deviations. Squaring the deviations is to emphasize extreme values. Especially when the driver's behavior is relatively intense, frequent operations will be marked more significantly. By summing up the squares of all deviations and then dividing by the total number of operation intensity levels, the average value of the squared deviations is obtained, which is an index to measure the fluctuation of the operation frequency. Take the square root of the average squared deviation value to get the standard deviation, which measures the volatility of the operation frequency. The larger the standard deviation, the greater the change in the operation intensity and the more intense the driver's behavior; the smaller the standard deviation, the more stable the driver's operation.

[0055] Frequency discreteness = , that is, the ratio of the standard deviation to the mean value, is used to represent the relative discreteness of the data. Frequency discreteness is usually used to compare the degree of discreteness of different data sets. Especially when the means of the data vary greatly, the coefficient of variation can more effectively reflect the discreteness. The standard deviation is used to measure the degree of fluctuation of the driver's operation intensity frequency. A larger standard deviation indicates a larger fluctuation in the operation frequency, meaning that the driving behavior is more intense. Frequency discreteness normalizes by taking the ratio of the standard deviation to the mean value, so that even if the mean value of the operation frequency is small, the discreteness of the operation intensity can be effectively measured. A larger coefficient of variation indicates a larger fluctuation in the operation frequency and a more radical driver's behavior.

[0056] Further, as shown in the appendix Figure 2 It is shown that step four of this application includes:

[0057] Establish the control influence relationship between the steering stability risk, the braking stability risk and the controller strategy parameters; establish the adjustment relationship between the stability risk coefficient and the controller strategy parameters, and the adjustment relationship is obtained by fitting the experimental data with historical data with the stability threshold as the target, and is used to characterize the parameter adjustment amplitude matching the stability risk coefficient; use the adjustment relationship to adjust the parameter control coefficient in the control influence relationship to obtain the optimized adjustment coefficient.

[0058] Specifically, determine which controller strategy parameters can be adjusted to address the stability issues in braking and steering. The steering stability risk is an indicator for evaluating the impact of steering operations on vehicle stability, usually reflected in the risk of rollover or loss of control that may occur when the vehicle is steering, and is related to factors such as steering force, steering angle, steering frequency, and driving speed. Similar to the steering stability risk, the braking stability risk assesses the impact of the driver's braking operation on vehicle stability. Excessively sharp or frequent braking operations may cause the vehicle to lose control or roll over, especially when driving at high speeds. Controller strategy parameters refer to various parameters required by the vehicle controller when performing braking and steering, such as braking intensity, steering angle, acceleration limit, stability control intervention strength, etc., which directly affect the driving stability, response speed, and driving experience of the vehicle.

[0059] By optimizing the braking force distribution, according to the braking behavior characteristics of the user, adjust the braking force distribution between the front and rear wheels to optimize braking efficiency and stability. For example, if the user often performs emergency braking, the controller can increase the braking force on the rear wheels to reduce the risk of the front wheels locking. By adjusting the steering assist force, according to the steering behavior characteristics of the user, adjust the strength of the steering assist system. If the user tends to steer frequently or sharply, the controller can increase the steering assist force to provide a more stable steering experience. By controlling the power output, according to the current speed and the user's acceleration behavior, adjust the output power of the motor. When driving at high speeds or turning, the controller can limit the power output to prevent stability problems caused by excessive acceleration.

[0060] The stability threshold refers to the critical value of the stability risk coefficient that is considered safe under specific driving conditions. When exceeding this threshold, it indicates that the vehicle may face a greater risk of losing control and requires adjusting the controller strategy parameters to enhance the vehicle's stability. Experimental data fitting is to conduct experiments through a large amount of historical data to fit the relationship between the stability risk coefficient and the adjustment of controller strategy parameters. Defining a stability threshold is a key indicator for measuring vehicle safety. When the stability risk coefficient of the vehicle exceeds this threshold, it is considered that the vehicle has entered a dangerous driving state (such as steering operations, sudden braking, sharp turning at different speeds, or high-speed driving, etc.), and measures need to be taken to avoid accidents. Through the test data of the vehicle at different speeds, braking forces, and steering forces, record the stability risk coefficient and the corresponding controller strategy parameters (such as braking intensity, steering angle, etc.) in each case.

[0061] Use statistical methods (such as linear regression, non - linear regression, machine learning, etc.) to fit historical data to obtain the mathematical relationship between the stability risk coefficient and the controller strategy parameters. For example, if the risk coefficient is high, it may be necessary to adjust the braking strategy to reduce the braking response speed; if the risk coefficient is low, the sensitivity of vehicle response can be increased. If the relationship between the stability risk coefficient and the controller parameters is linear, find the best linear equation by fitting the data points through linear regression. If linear regression cannot accurately fit the data, use non - linear regression methods to obtain a more precise relationship. When the amount of data is large, use machine learning methods (such as decision trees, support vector machines, neural networks, etc.) to establish a more complex adjustment relationship model. The adjustment relationship is used to characterize the amplitude of parameter adjustment that matches the stability risk coefficient, meaning determining how much adjustment needs to be made to the controller strategy parameters according to the magnitude of the stability risk coefficient.

[0062] The control influence relationship refers to the connection between the controller strategy parameters and the vehicle's stability, response and other performances. The parameters of the controller (such as steering angle, braking force, etc.) directly affect the dynamic behavior and stability of the vehicle. The parameter control coefficient is the weight coefficient of each control parameter of the controller (such as steering, acceleration, braking, etc.) in the control influence relationship, which determines the degree of influence of each control parameter on the vehicle behavior. According to the established adjustment relationship, adjust the parameter control coefficient in the control influence relationship to obtain a new coefficient, that is, the optimized adjustment coefficient. The optimized adjustment coefficient is used to adjust the parameters of the controller in real - time to ensure that the vehicle maintains high stability under extreme conditions such as high - speed driving, sharp turning, and emergency braking. The controller continuously adjusts the controller strategy parameters according to the real - time monitored stability risk coefficient and historical test data. For example, when the stability risk coefficient is low, the controller will increase steering flexibility and improve acceleration response; when the stability risk coefficient is high, the controller will increase braking intensity and reduce the steering angle, etc., so as to ensure that the vehicle will not get out of control due to overly fast reactions. Dynamically adjust the controller strategy parameters according to the stability risk coefficient to ensure that the vehicle can maintain the best stability in different driving states. By continuously collecting test data and optimizing the model, the controller can adaptively adjust the controller parameters according to different operations of the driver and environmental conditions to meet more diverse driving needs.

[0063] Furthermore, the present application further includes the following steps:

[0064] Obtain the kinetic energy recovery strategy, conduct steering and braking correlation analysis on the kinetic energy recovery strategy to determine the associated kinetic energy recovery strategy; establish the influence relationship between the associated kinetic energy recovery strategy and the user's driving operation characteristics to obtain the recovery control strategy adjustment coefficient; according to the risk stability coefficient of the current speed adaptive interval, combine the recovery control strategy adjustment coefficient to adjust the control strategy of the kinetic energy recovery strategy.

[0065] Specifically, kinetic energy recovery is a technology that converts the kinetic energy of a vehicle during braking or deceleration into electrical energy for storage. The kinetic energy recovery strategy refers to the specific scheme for controlling the operation of the kinetic energy recovery system, including recovery efficiency, recovery mode, triggering conditions, etc. Conduct a correlation analysis of the kinetic energy recovery strategy with steering and braking, and study the impact of vehicle steering or braking on the kinetic energy recovery efficiency and effect, especially how these operations change the kinetic energy distribution and recovery potential of the vehicle. For example, the greater the braking force, the more kinetic energy that may be recovered; during steering, due to the inertia distribution of the vehicle, kinetic energy recovery may be restricted to a certain extent. During the braking process, the kinetic energy recovery system needs to work in coordination with the braking system. At low risk, the braking and kinetic energy recovery forces are linearly superimposed (for example, the working efficiency of the recovery system is 80%, and the mechanical braking only bears 20% of the deceleration demand); strengthen the recovery force, reduce the intervention of mechanical braking, and improve the energy recovery efficiency. At medium risk, optimize the ratio of kinetic energy recovery to mechanical braking, reduce the kinetic energy recovery force, and avoid the vehicle decelerating too quickly during emergency braking, which may cause the vehicle body to be unstable; at high risk, kinetic energy recovery may exacerbate the vehicle's deceleration instability, so its force needs to be reduced to the minimum or even disabled, relying entirely on mechanical braking to avoid the risk of skidding or rolling over.

[0066] During the vehicle steering process, especially in high-speed or sharp-turn scenarios, the adjustment of the kinetic energy recovery force is directly related to the lateral stability of the vehicle. At low risk, the kinetic energy recovery has little impact on the vehicle stability, and the kinetic energy recovery force remains unchanged (such as the recovery efficiency remains 70%). At medium risk, the vehicle may generate a certain degree of lateral instability during high-speed or medium-speed turning. The kinetic energy recovery force needs to be adjusted moderately to avoid an increase in lateral acceleration, and the recovery force is appropriately reduced to maintain the vehicle's smooth steering. At high risk, the vehicle is in a sharp-turn or high-speed steering state, and the kinetic energy recovery may significantly affect the lateral stability. It is necessary to temporarily disable the kinetic energy recovery and rely entirely on the mechanical braking and chassis stability control system to avoid instability or rollover caused by the recovery system.

[0067] Relate the user's driving operation characteristics (such as conservative driving, smooth driving, aggressive driving) to the recovery efficiency and mode of the associated kinetic energy recovery strategy to establish an impact relationship. Through historical driving data and driving behavior characteristics, use linear regression or machine learning models to fit the relationship between the user's operation characteristics and the kinetic energy recovery efficiency, and obtain the adjustment coefficient of the recovery control strategy for dynamically adjusting the working mode of the kinetic energy recovery system. Collect the actual operation data of the kinetic energy recovery system in different driving modes through experiments to establish a kinetic energy recovery strategy model. Use the user's driving operation characteristics (conservative, smooth, aggressive) as the input variables of the kinetic energy recovery strategy, set three driving scenarios (conservative, smooth, aggressive), record the vehicle deceleration stability, recovery efficiency, and user satisfaction data in each driving mode, quantify the impact of different driving characteristics on the kinetic energy recovery strategy, and obtain the relationship model between the recovery ratio and the driving characteristics.

[0068] Input the current user driving operation characteristics into the relationship model to calculate the adjustment coefficient of the current recovery control strategy. Match the speed range according to the real-time driving speed to obtain the risk coefficient. Dynamically adjust the kinetic energy recovery intensity based on two core inputs (risk stability coefficient and adjustment coefficient), and dynamically change the kinetic energy recovery intensity, recovery ratio or trigger condition to ensure vehicle stability and energy recovery efficiency. In high-risk scenarios such as high speed or sharp turns, dynamically reduce the kinetic energy recovery intensity to avoid vehicle instability; in low-risk scenarios (such as low-speed driving or coasting), enhance the kinetic energy recovery intensity to improve energy utilization and extend the cruising range. Adjusting the kinetic energy recovery strategy according to the risk stability coefficient and the adjustment coefficient of the recovery control strategy helps to improve energy utilization efficiency while ensuring driving stability, thereby extending the cruising range of the electric tricycle.

[0069] In summary, the controller strategy optimization method for user behavior analysis provided by this application has the following technical effects:

[0070] By collecting user driving parameters, the user driving parameters include driving speed, braking and steering force and frequency; identify and segment the speed risk level and speed change characteristics according to the driving speed to determine the speed adaptive interval, and the speed adaptive interval includes speed risk level and speed change amplitude; conduct steering and braking risk assessment on the speed adaptive interval according to the braking and steering force and frequency respectively to obtain the stability risk coefficient; conduct controller strategy matching analysis according to the stability risk coefficient to determine the optimization adjustment coefficient, and add the optimization adjustment coefficient to the speed adaptive interval to generate an interval optimization adjustment label; match the monitored current driving speed with the speed adaptive interval to obtain the optimization adjustment coefficient corresponding to the adaptive interval, and adjust the controller strategy according to the optimization adjustment coefficient. That is to say, by collecting user driving parameters, comprehensively understand the user's driving habits and behavior patterns, real-time evaluate the vehicle's driving state, conduct risk assessment on braking and steering during driving, determine the stability coefficient, select the most suitable control strategy for the current driving state, dynamically match and adjust the controller strategy to adapt to the user's driving characteristics and vehicle state, and improve the safety and stability of vehicle driving.

[0071] The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0072] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application also intends to include these changes and modifications.

Claims

1. A controller strategy optimization method for user behavior analysis, characterized in that: include: Collecting user driving parameters, including driving speed, braking and steering strength and frequency; According to the driving speed, the speed risk level and the speed change feature are identified and segmented to determine a speed adaptive interval, wherein the speed adaptive interval includes the speed risk level and the speed change amplitude; According to the braking and steering forces and frequencies, respectively, the steering and braking risk assessment is performed on the speed adaptation interval to obtain a stability risk coefficient; Performing controller strategy matching analysis according to the stability risk coefficient, determining an optimization adjustment coefficient, and adding the optimization adjustment coefficient to the speed adaptation interval to generate an interval optimization adjustment label; The current driving speed is monitored and matched with the speed adaptive interval to obtain an optimization adjustment coefficient corresponding to the adaptive interval, and the controller strategy is adjusted according to the optimization adjustment coefficient.

2. A controller strategy optimization method for user behavior analysis according to claim 1, characterized in that: According to the driving speed, the speed risk level and speed change characteristics are identified and segmented, and the speed adaptation interval is determined, including: Acquiring vehicle speed test data, wherein the vehicle speed test data has a vehicle model; According to the vehicle speed test data, speed stability risk levels are divided and speed risk level intervals are determined, including speed risk intervals and speed change amplitude intervals; Calculate the speed change amplitude according to the driving speed to obtain the speed time series change amplitude; The speed risk interval and the speed change amplitude interval are used to match and map the driving speed and the speed time series change amplitude to obtain multiple speed risk level intervals; The speed adaptive interval is constructed according to the multiple speed risk level intervals.

3. The controller strategy optimization method for user behavior analysis according to claim 1, characterized in that: According to the braking and steering forces and frequencies, the steering and braking risk assessment is performed on the speed adaptation interval to obtain a stability risk coefficient, including: Analyze the user's braking and steering behavior characteristics according to the braking and steering strength and frequency, and determine the user's driving operation characteristics; The user's driving operation characteristics and the speed value and speed change amplitude of the speed adaptive interval are used as input variables, and risk prediction is performed on each speed adaptive interval respectively through a risk prediction model to obtain the stability risk coefficient.

4. A controller strategy optimization method for user behavior analysis as claimed in claim 3, characterized in that: According to the braking and steering strength and frequency, the user's braking and steering behavior characteristics are analyzed to determine the user's driving operation characteristics, including: Get the duration of braking and steering operations; According to the force of braking and steering, the operation intensity is obtained by using the operation force and operation duration of each braking or steering; According to the correspondence between the operation intensity and the frequency of braking and steering, an operation intensity frequency analysis is performed to obtain operation frequencies of different operation intensities; According to the frequency discreteness of the operation frequencies of the different operation intensities, a behavior characteristic evaluation is performed to obtain the user driving operation characteristics.

5. A controller strategy optimization method for user behavior analysis as claimed in claim 4, characterized in that: According to the frequency discreteness of the operation frequencies of the different operation intensities, a behavior characteristic evaluation is performed to obtain the user driving operation characteristics, including: According to the operation frequencies of the different operation intensities, the intensity standard deviation is calculated by using the frequency standard deviation formula; Performing normalization processing based on the intensity standard deviation to obtain the frequency discreteness, wherein the frequency discreteness is represented by a normalized frequency variation coefficient; Divide the characteristic evaluation level according to the frequency discreteness, the characteristic evaluation level is determined by matching the frequency discreteness with a preset discrete interval value, and the characteristic evaluation level includes a low discreteness level, a medium discreteness level, and a high discreteness level; The user driving operation characteristics are determined according to the characteristic evaluation level, and the user driving operation characteristics include conservative driving, smooth driving, and aggressive driving, which correspond to the characteristic evaluation levels respectively.

6. A controller strategy optimization method for user behavior analysis as claimed in claim 5, characterized in that: The frequency standard deviation formula is expressed as: ; The frequency discreteness = ,in, is the frequency standard deviation, The operating frequency for each intensity level, is the mean of the operation frequency, N is the total number of operation intensity levels, and i represents the i-th operation intensity level.

7. The controller strategy optimization method for user behavior analysis according to claim 1, characterized in that: According to the stability risk coefficient, controller strategy matching analysis is performed to determine the optimization adjustment coefficient, including: Establish the control influence relationship between steering stability risk, braking stability risk and controller strategy parameters; Establishing an adjustment relationship between the stability risk coefficient and the controller strategy parameter, wherein the adjustment relationship is obtained by fitting the test data using historical data with the stability threshold as the target, and is used to characterize the parameter adjustment range that matches the stability risk coefficient; The parameter control coefficient in the control influence relationship is adjusted using the adjustment relationship to obtain the optimized adjustment coefficient.

8. The controller strategy optimization method for user behavior analysis according to claim 5, characterized in that: Also includes: Obtaining a kinetic energy recovery strategy, performing steering and braking correlation analysis on the kinetic energy recovery strategy, and determining a correlated kinetic energy recovery strategy; Establishing the influence relationship between the associated kinetic energy recovery strategy and the user's driving operation characteristics, and obtaining the recovery control strategy adjustment coefficient; The kinetic energy recovery strategy is adjusted according to the risk stability coefficient of the current speed adaptation interval in combination with the recovery control strategy adjustment coefficient.

Images