Drive-by-wire chassis parameter optimization method and system based on driving behavior feedback

Abstract

The invention provides a drive-by-wire chassis parameter optimization method and system based on driving behavior feedback, and relates to the technical field of intelligent driving, and the method comprises the steps: collecting a driving behavior historical data set, and carrying out the feature extraction and style analysis; working condition table design and real vehicle driving test are carried out based on the multiple driving style feature clusters; parameter training fitting is carried out according to the multiple driving style-drive-by-wire chassis response data sets, and drive-by-wire chassis parameter control multiple channels are established; and target driver behavior data are monitored in real time, matching control analysis is carried out on multiple channels based on the drive-by-wire chassis parameters, and target drive-by-wire chassis parameters are determined to carry out driving feedback optimization control. The technical problem that in the prior art, drive-by-wire chassis parameter adjustment lacks flexibility, and man-machine co-driving conflicts are caused due to the fact that driving styles are not matched with chassis responses in the vehicle control right switching process is solved, drive-by-wire chassis parameter adjustment is conducted in combination with the driving styles, the man-machine co-driving conflicts are reduced, and driving safety is improved.

Description

A method and system for optimizing the parameters of a drive-by-wire chassis based on driving behavior feedback. Technical Field

[0001] This application relates to the field of intelligent driving technology, and in particular to a method and system for optimizing drive-by-wire chassis parameters based on driving behavior feedback. Background Technology

[0002] With the widespread adoption of autonomous driving, frequent switching between autonomous and manual driving has become commonplace. However, most current vehicle drive-by-wire chassis parameter adjustment mechanisms rely heavily on preset standard parameters, lacking the ability to dynamically adjust according to the driver's individual needs. Especially during the transition between autonomous and manual driving, a mismatch often exists between the driver's operating habits and the vehicle's response, causing the vehicle to fail to adapt to the driver's style in a timely manner, thus affecting the driver's experience and safety. Furthermore, the driver may encounter overreaction or sluggishness when taking over control, potentially leading to driving conflicts. This results in a significant disconnect between the autonomous and manual driving experiences, impacting the reliability of autonomous driving technology.

[0003] In summary, existing technologies suffer from technical problems such as a lack of flexibility in adjusting the parameters of the drive-by-wire chassis, leading to conflicts between human and machine driving due to a mismatch between driving style and chassis response during vehicle control switching, which further affects the driving experience. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for optimizing drive-by-wire chassis parameters based on driving behavior feedback, in order to solve the technical problem in the prior art that the lack of flexibility in adjusting drive-by-wire chassis parameters leads to human-machine co-driving conflicts during the vehicle control switching process due to the mismatch between driving style and chassis response, which further affects the driving experience.

[0005] In view of the above problems, this application provides a method and system for optimizing drive-by-wire chassis parameters based on driving behavior feedback.

[0006] Firstly, this application provides a method for optimizing steerable chassis parameters based on driving behavior feedback. This method is implemented through a steerable chassis parameter optimization system based on driving behavior feedback. The method includes: collecting historical driving behavior datasets; performing feature extraction and style analysis on the historical driving behavior datasets to obtain multiple driving style feature clusters; designing a driving cycle table and conducting real-vehicle driving tests based on the multiple driving style feature clusters to record multiple driving style-steerable chassis response datasets; training and fitting parameters based on the multiple driving style-steerable chassis response datasets to establish a multi-channel steerable chassis parameter control system, whereby the multi-channel steerable chassis parameter control system includes multiple driving style steerable chassis parameter control sub-channels; monitoring target driver behavior data in real time; performing matching control analysis on the target driver behavior data based on the multi-channel steerable chassis parameter control system to determine target steerable chassis parameters; and performing driving feedback optimization control based on the target steerable chassis parameters.

[0007] Optionally, the driving behavior history dataset is subjected to abnormal data cleaning and standardization to obtain a standard driving behavior history dataset; the driving behavior feature selection type is determined according to the driving behavior analysis objective; feature extraction calculation is performed on the standard driving behavior history dataset according to the driving behavior feature selection type to obtain a driving behavior associated feature dataset; driving style analysis is performed based on the driving behavior associated feature dataset to obtain multiple driving style feature clusters.

[0008] Optionally, a driving style type is preset, including aggressive, normal, and conservative types; based on the driving style type, the number of feature clusters is determined to be K=3, and feature data clustering centers are initialized based on the number of feature clusters; K-means clustering analysis is performed on the driving behavior associated feature dataset according to the feature data clustering centers to obtain feature data clustering results; driving style feature analysis is performed based on the feature data clustering results to obtain the multiple driving style feature clusters.

[0009] Optionally, a driving condition test dimension is designed, which includes a driving behavior dimension, a road condition dimension, a speed dimension, and a traffic condition dimension; parameters are selected and analyzed for each test dimension in the driving condition test dimension to determine the driving condition dimension test parameter thresholds; a test condition table is designed based on the multiple driving style feature clusters to generate a driving condition test parameter table; real vehicle driving tests are conducted based on the driving condition test parameter table, and the multiple driving style-drive-by-wire chassis response datasets are recorded.

[0010] Optionally, the response effects of the multiple driving style-drive-by-wire chassis response datasets are evaluated and filtered to obtain multiple usable driving style-drive-by-wire chassis response datasets; deep neural networks are used to train and fit parameters to the multiple usable driving style-drive-by-wire chassis response datasets to obtain multiple drive style drive-by-wire chassis parameter control sub-channels; the multiple drive style drive-by-wire chassis parameter control sub-channels are connected in parallel and integrated to establish the drive-by-wire chassis parameter control multi-channel.

[0011] Optionally, style matching analysis is performed on the target driver behavior data based on the multiple driving style feature clusters to determine the target driving style; the target driving style is matched and called with the multi-channel drive-by-wire chassis parameter control to obtain the target drive-by-wire chassis parameter control channel; parameter control analysis is performed on the target driver behavior data based on the target drive-by-wire chassis parameter control channel to determine the target drive-by-wire chassis parameters.

[0012] Optionally, vehicle driving control is performed using the target drive-by-wire chassis parameters, and the driving process is monitored in real time to obtain a vehicle driving state data stream; the effect of the vehicle driving state data stream is evaluated to obtain vehicle driving effect parameters; the target drive-by-wire chassis parameters are optimized based on the obtained vehicle driving effect parameters to obtain drive-by-wire chassis optimization parameters, and driving optimization control is performed using the drive-by-wire chassis optimization parameters.

[0013] Optionally, based on the obtained vehicle driving effect parameters, the target drive-by-wire chassis parameters are adjusted and analyzed to determine the drive-by-wire chassis parameter adjustment threshold; within the drive-by-wire chassis parameter adjustment threshold, parameter selection simulation and global effect optimization are performed to obtain the optimized parameters of the drive-by-wire chassis.

[0014] Optionally, driver control feedback evaluation parameters are collected through a driver feedback device installed in the vehicle; the drive-by-wire chassis optimization parameters are then optimized to meet specific requirements based on the driver control feedback evaluation parameters.

[0015] Secondly, this application also provides a drive-by-wire chassis parameter optimization system based on driving behavior feedback, used to execute the drive-by-wire chassis parameter optimization method based on driving behavior feedback as described in the first aspect. The drive-by-wire chassis parameter optimization system includes: a data acquisition module for acquiring historical driving behavior datasets, performing feature extraction and style analysis on the historical driving behavior datasets to obtain multiple driving style feature clusters; a driving style feature extraction module for designing a test schedule and conducting real-vehicle driving tests based on the multiple driving style feature clusters, recording multiple driving style-drive-by-wire chassis response datasets; a multi-channel parameter control module for parameter training and fitting based on the multiple driving style-drive-by-wire chassis response datasets to establish a multi-channel drive-by-wire chassis parameter control, wherein the multi-channel drive-by-wire chassis parameter control includes multiple driving style drive-by-wire chassis parameter control sub-channels; and a matching control analysis module for real-time monitoring of target driver behavior data, performing matching control analysis on the target driver behavior data based on the multi-channel drive-by-wire chassis parameter control, determining target drive-by-wire chassis parameters, and performing driving feedback optimization control through the target drive-by-wire chassis parameters.

[0016] One or more technical solutions provided in this application have at least the following beneficial effects:

[0017] By collecting historical driving behavior datasets, feature extraction and style analysis are performed on these datasets to obtain multiple driving style feature clusters. Based on these multiple driving style feature clusters, a driving cycle table is designed and real-vehicle driving tests are conducted, recording multiple driving style-drive-by-wire chassis response datasets. Parameter training and fitting are performed based on these multiple driving style-drive-by-wire chassis response datasets to establish a multi-channel drive-by-wire chassis parameter control system, which includes multiple sub-channels for driving style drive-by-wire chassis parameter control. Target driver behavior data is monitored in real time, and matching control analysis is performed on the target driver behavior data based on the multi-channel drive-by-wire chassis parameter control system to determine the target drive-by-wire chassis parameters. Driving feedback optimization control is then performed using these target drive-by-wire chassis parameters. In other words, by extracting features and analyzing styles from historical driving behavior datasets, the driving styles of drivers are identified. Based on different driving styles, operating condition tables are designed and real-vehicle driving tests are conducted to obtain drive-by-wire chassis response data corresponding to multiple driving styles. A multi-channel drive-by-wire chassis parameter control system is established, and matching control analysis is performed based on real-time monitored target driver behavior data to achieve personalized driving feedback optimization control, reduce human-machine co-driving conflicts, improve driving safety, and enhance the driving experience.

[0018] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 is a flowchart illustrating the method for optimizing the parameters of the drive-by-wire chassis based on driving behavior feedback in this application.

[0021] Figure 2 is a schematic diagram of the drive-by-wire chassis parameter optimization system based on driving behavior feedback in this application.

[0022] Figure labeling: Data acquisition module 11, driving style feature extraction module 12, multi-channel parameter control module 13, matching control analysis module 14. Detailed Implementation

[0023] This application provides a method and system for optimizing steerable chassis parameters based on driving behavior feedback. This addresses the technical problem in existing technologies where the lack of flexibility in adjusting steerable chassis parameters leads to human-machine co-driving conflicts during vehicle control switching due to mismatches between driving style and chassis response, further impacting the driving experience. By extracting features and performing style analysis on historical driving behavior datasets, the method identifies the driver's driving style and designs operating condition tables and conducts real-vehicle driving tests based on different driving styles. This yields steerable chassis response data corresponding to multiple driving styles, establishing a multi-channel steerable chassis parameter control system. Matching control analysis is then performed based on real-time monitored target driver behavior data to achieve personalized driving feedback optimization control, reducing human-machine co-driving conflicts, improving driving safety, and enhancing the driving experience.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to Figure 1. This application provides a method for optimizing steerable chassis parameters based on driving behavior feedback. The method is executed by a steerable chassis parameter optimization system based on driving behavior feedback. The method specifically includes the following steps: collecting a historical driving behavior dataset, performing feature extraction and style analysis on the historical driving behavior dataset, and obtaining multiple driving style feature clusters.

[0026] Furthermore, this application also includes the following steps: performing abnormal data cleaning and standardization on the driving behavior historical dataset to obtain a standard driving behavior historical dataset; determining the driving behavior feature selection type according to the driving behavior analysis objective; performing feature extraction calculation on the standard driving behavior historical dataset according to the driving behavior feature selection type to obtain a driving behavior associated feature dataset; and performing driving style analysis based on the driving behavior associated feature dataset to obtain multiple driving style feature clusters.

[0027] Furthermore, this application also includes the following steps: presetting driving style types, including aggressive, normal, and conservative types; determining the number of feature clusters K=3 according to the driving style type, and initializing feature data clustering centers based on the number of feature clusters; performing K-means clustering analysis on the driving behavior associated feature dataset according to the feature data clustering centers to obtain feature data clustering results; and performing driving style feature analysis based on the feature data clustering results to obtain the multiple driving style feature clusters.

[0028] Specifically, driving behavior history datasets are collected through in-vehicle networks and sensors. These datasets are raw data sets reflecting driver operations and vehicle states, recorded over a long period. This includes time-series data such as steering wheel angle, pedal opening, vehicle speed, lateral acceleration, and yaw rate. The driving behavior history dataset undergoes anomaly cleaning, identifying outliers, missing values, and obvious errors in the raw data. These anomalies may be caused by momentary sensor malfunctions, signal transmission interference, or extreme driving events, such as a torque request signal momentarily jumping to 100% and then immediately returning to zero during cruise control. After data cleaning, to eliminate differences in the units of measurement between different features, the driving behavior history dataset is standardized to obtain a standard driving behavior history dataset with consistent metrics.

[0029] Define the driving behavior analysis objective, which is the goal set when conducting driving behavior analysis. For example, if the objective is to identify a driver's driving style, the analysis objective might include classifying the driver's driving behavior; if the objective is to assess a driver's safe driving habits, the analysis objective might include evaluating behaviors such as the frequency of emergency braking, acceleration smoothness, anticipatory braking use, efficient energy recovery utilization, or average energy consumption level. Based on the driving behavior analysis objective, determine the types of driving behavior features to be selected. For example, if the driving behavior analysis objective is acceleration behavior, the selected features might include full-load acceleration time, average torque utilization during acceleration, and typical gear selection strategies during acceleration. The selected features are predefined categories of features used to describe driving style.

[0030] Feature extraction calculations are performed on the standard driving behavior historical dataset according to the feature types selected for driving behavior to obtain a driving behavior associated feature dataset, which contains key information reflecting driver behavior and can reflect the driver's driving style, operating habits and behavioral patterns.

[0031] Pre-defined driving style types, that is, before cluster analysis, based on engineering experience and business objectives, conceptually define driving styles. These are typically categorized according to the driver's acceleration, braking, and steering behavior patterns, including aggressive, moderate, and conservative styles. Aggressive driving refers to drivers who tend to accelerate rapidly at low gears and high RPMs, brake late and sharply, steer quickly, and frequently change lanes and overtake; their overall driving style leans towards high energy consumption and high mechanical wear. Moderate driving refers to drivers with stable driving styles, relatively smooth acceleration and braking, focusing on gear and speed matching, and exhibiting a certain degree of foresight. Conservative driving refers to drivers with relatively mild behavior, avoiding rapid acceleration and emergency braking, making extensive use of predictive coasting, and favoring high gears and low RPMs; their overall driving style pursues low risk and low energy consumption.

[0032] Based on the driving style type, the number of feature clusters is determined to be K=3, that is, the data is divided into 3 clusters. Based on the number of feature clusters, the feature data cluster centers are initialized, that is, 3 samples are randomly selected as the initial cluster centers. Usually, three data points that are far apart from each other are selected as the initial feature data cluster centers to improve the convergence speed and effect.

[0033] K-means clustering analysis is performed on the driving behavior-related feature dataset based on the feature data cluster centers. This involves dividing the dataset into three clusters, ensuring that data points within each cluster are as similar as possible, while data points between different clusters are as different as possible. Each data point is assigned to the nearest cluster based on its distance from the cluster center. The cluster centers are then recalculated based on the data points in each cluster, and this process is iterated until convergence. After multiple iterations, the cluster centers are gradually adjusted until convergence, and ultimately each cluster represents a driving style. The feature data clustering result is the output of K-means clustering analysis, clearly indicating which cluster each sample in the dataset is assigned to.

[0034] Driving style feature analysis is performed based on the clustering results of feature data. The features of each cluster are analyzed to obtain the typical performance of each driving style, that is, to determine the driving style type within different clusters, such as aggressive, normal, and conservative. The data points within each cluster represent the driving behavior under that style, resulting in multiple driving style feature clusters.

[0035] For example, assume three driving style types are preset: aggressive, normal, and conservative. Three cluster centers are initialized, with initial center 1 at [70%, 0.6 m / s]. 3 [25° / s], center 2 is [50%, 0.35m / s] 3 [18° / s], center 3 is [35%, 0.2m / s] 3 The three indicators, [12° / s], correspond to the average drive torque request rate, the average braking deceleration build-up rate, and the steering angular velocity, respectively. The average braking deceleration build-up rate reflects the speed of braking response. K-means clustering analysis began iterating. After the first iteration, all data points were assigned to the nearest initial center, and the new centers of the three clusters were recalculated. After 5 iterations, the algorithm converged, and the centers of the three clusters no longer changed. The final feature data clustering results showed that 35 drivers were assigned to cluster A, 40 drivers to cluster B, and 25 drivers to cluster C. The coordinates of the center point of each final cluster were calculated, and the center point of cluster A was [72.5%, 0.65 m / s]. 3 The value of 28° / s indicates that this group of drivers is accustomed to high torque demands (average 72.5%) and seeks rapid braking response (deceleration build-up rate of 0.65 m / s²). 3 The cluster B has a relatively fast turning speed, thus it is an aggressive characteristic cluster; the center point of cluster B is [52.3%, 0.38m / s].3 [19° / s], all eigenvalues ​​are at an intermediate level, indicating a moderate shift timing, therefore it is a common type of feature cluster; the center point of cluster C is [38.8%, 0.22m / s]. 3 The value of 14° / s indicates that this group of drivers uses gentle control (average 38.8%) and is adept at using anticipatory driving to build up brake pressure smoothly (deceleration build-up rate is only 0.22 m / s). 3 ), and the shift is graceful, therefore it is a conservative feature cluster.

[0036] K-means clustering analysis avoids the bias of subjective judgment and can accurately discover the naturally existing group structure in the data. The resulting driving style feature clusters not only classify drivers, but more importantly, the coordinates of the center point of each cluster constitute a style profile of that type of driver.

[0037] Based on the aforementioned multiple driving style feature clusters, a test schedule was designed and real vehicle driving tests were conducted to record multiple driving style-drive-by-wire chassis response datasets.

[0038] Furthermore, this application also includes the following steps: designing driving condition test dimensions, which include driving behavior dimension, road condition dimension, speed dimension, and traffic condition dimension; performing parameter selection and analysis on each test dimension in the driving condition test dimensions to determine the driving condition dimension test parameter thresholds; designing a test condition table based on the multiple driving style feature clusters for the driving condition dimension test parameter thresholds to generate a driving condition test parameter table; and conducting real vehicle driving tests based on the driving condition test parameter table to record the multiple driving style-drive-by-wire chassis response datasets.

[0039] Specifically, the design of driving condition test dimensions involves considering different environmental factors used to describe vehicle and driver behavior during simulated or actual driving, comprehensively evaluating the performance of heavy commercial vehicles under various driving conditions. These dimensions include driving behavior, road conditions, speed, and traffic conditions. The driving behavior dimension describes the driver's operational behaviors, such as the frequency, intensity, and reaction speed of acceleration, braking, steering, shifting strategies, anticipatory coasting / braking, cruise control, and torque braking, directly corresponding to aggressive, normal, and conservative driving styles. The road conditions dimension describes road conditions, including road type, road smoothness, gradient, and curves, such as paved surfaces in storage yards, container terminal quay crane / yard crane areas, unpaved roads in mining areas, loading and unloading platforms, and internal road curves and slopes. The speed dimension relates to vehicle speed, typically related to the truck's acceleration capability, economical cruising speed, and downhill speed limits. The traffic conditions dimension describes the surrounding traffic environment, such as the density of other construction machinery and vehicles, pedestrian activity areas, interference from fixed facilities, and specific loading and unloading process requirements.

[0040] The parameters for each test dimension in the driving condition test were analyzed and selected to determine the threshold values ​​for the test parameters of the driving condition dimension, thus limiting the range of different test scenarios and ensuring that the experiment can be conducted within the range of actual driving conditions. For example, in the speed dimension, the threshold values ​​for the dimension test parameters were set as follows: ultra-low speed vehicle maneuvering 0 to 5 km / h, low-speed operation 5 to 20 km / h, and on-site transfer 20 to 40 km / h, with speed limits based on the scenario safety limits; in the road condition dimension, the threshold values ​​for the dimension test parameters were set as follows: the stockpile road surface is flat but may be slippery, with frequent right-angle turns; the mining area road surface is a combination of bumpy, gravel, and slope.

[0041] Based on multiple driving style characteristic clusters, a driving condition test parameter table is designed to set threshold values ​​for driving condition dimensions, generating a driving condition test parameter table. This table lists the parameter ranges under different driving conditions, describing the parameters of each dimension in different driving scenarios and providing specific standards for testing. In other words, different test standards are set for each dimension based on the driver's driving style. For example, for drivers with an aggressive driving style cluster, the focus is on designing test cases for overtaking slower vehicles, maintaining speed on mountain curves, and emergency obstacle avoidance; while for drivers with a conservative driving style cluster, the focus is on tests such as cruise control, braking management and gear shifting, and smooth following in congested traffic, thus generating a highly targeted driving condition test parameter table.

[0042] Based on the driving condition test parameter table, real-vehicle driving tests are conducted. Professional test drivers strictly follow the scenarios and style requirements specified in the driving condition test parameter table, while comprehensively recording all dynamic response signals of the vehicle, including the vehicle's feedback under acceleration, braking, steering, and other operations. For example, on a flat road in a storage yard, an aggressive driver requires approximately 8-12 seconds to accelerate a fully loaded heavy truck from 0 km / h to 30 km / h, and during emergency braking, the vehicle is required to establish a speed of no less than 6 m / s² within 0.8-1.2 seconds. 2 The peak deceleration of the former is higher, and the average deceleration gradient is also higher; while a conservative driver may choose a gentler deceleration to complete the same acceleration, taking longer, possibly up to 15-20 seconds, and using anticipatory braking earlier. Their braking process usually starts earlier and aims for smoothness, with peak deceleration mostly controlled between 2.5 and 3.5 m / s². 2 Within the specified range, and with a more linear deceleration process, the stability of the cargo is ensured to the greatest extent possible.

[0043] Through multi-dimensional and comprehensive operating condition design, the completeness and representativeness of multiple driving style-drive-by-wire chassis response datasets are ensured, covering real-world user scenarios. By deeply integrating driving style feature clusters with test conditions, data collection is no longer blind but highly targeted.

[0044] Based on the multiple driving style-drive-by-steer chassis response datasets, parameter training and fitting are performed to establish a multi-channel drive-by-steer chassis parameter control system, which includes multiple driving style drive-by-steer chassis parameter control sub-channels.

[0045] Furthermore, this application also includes the following steps: evaluating and filtering the response effects of the multiple driving style-drive-by-wire chassis response datasets to obtain multiple usable driving style-drive-by-wire chassis response datasets; using a deep neural network to train and fit the parameters of the multiple usable driving style-drive-by-wire chassis response datasets to obtain multiple drive style drive-by-wire chassis parameter control sub-channels; and performing parallel integration and identification of the multiple drive style drive-by-wire chassis parameter control sub-channels to establish the drive-by-wire chassis parameter control multi-channel.

[0046] Specifically, quality control and validity assessment were performed on multiple driving style-drive-by-wire chassis response datasets obtained from real-vehicle testing. Evaluation criteria included objective dynamic vehicle indicators such as lateral stability (roll angle, yaw rate), longitudinal pitch angle, load transfer rate, brake pressure stability, and whether the dynamic changes in suspension height were within safe and controllable ranges, as well as subjective driver evaluation scores (e.g., using a 5-point scoring system). Data resulting from testing errors, extreme driving maneuvers, or significantly undesirable parameter combinations were eliminated to ensure high-quality and representative training data. By setting objective indicator thresholds, such as a roll angle not exceeding 5 degrees and brake pressure fluctuation rate below 15%, and adopting high-scoring subjective evaluations, multiple usable driving style-drive-by-wire chassis response datasets were obtained from massive amounts of raw data, ensuring the purity and effectiveness of the training data. Evaluation standards included data accuracy, completeness, and vehicle response stability under different driving styles. The purpose of screening was to remove abnormal or unsuitable data, retaining the most representative and reliable response data.

[0047] The deep neural network is designed with an input layer, multiple hidden layers, and an output layer. The input layer receives driver behavior features, such as acceleration, braking, and steering angle. The hidden layers extract complex patterns and features from the data. Assuming three hidden layers are designed, each containing 64 neurons, the ReLU activation function is used to help solve the vanishing gradient problem and improve training efficiency. The number of nodes in the output layer depends on the chassis control parameters to be predicted, such as the target steering assist torque curve parameters, the target stiffness and damping values ​​of the electronically controlled air suspension, the braking pressure adjustment coefficient, and the powertrain torque response mapping coefficient. The weights of the deep neural network, i.e., the parameters connected to each neuron, are randomly initialized to ensure that the network starts optimization from a different starting point each time it is trained. For each sample, the driving behavior features are first input into the network for forward propagation. The forward propagation process refers to the data being passed from the input layer through the hidden layers to the output layer, and through weighted calculations and the application of activation functions at each layer, a prediction result is finally obtained. The loss function is used to measure the difference between the network's prediction result and the actual target; the mean squared error (MSE) is the most commonly used loss function. During training, the network continuously optimizes its parameters to minimize the loss function, thereby improving prediction accuracy. Backpropagation is a crucial step in training neural networks. Through backpropagation, the neural network updates its parameters based on the gradient of the loss function, i.e., the derivative of the error with respect to each parameter. The network parameters are optimized using the gradient descent algorithm or its variants. The gradient descent algorithm calculates the gradient of the loss function and adjusts the weights and biases based on the gradient value to gradually reduce the loss function. The magnitude of each update is controlled by the learning rate. If the learning rate is too large, it may lead to non-convergence; if the learning rate is too small, training will be very slow. The training process typically requires multiple iterations. In each iteration, the network performs forward and backward propagation through all training samples and updates the parameters based on the gradient. After training, the trained models for each driving style are cross-validated and tested to ensure generalization ability on unknown data. After training, the three models with different driving styles, i.e., multiple driving style drive-by-wire chassis parameter control sub-channels, are connected in parallel and integrated to form a unified drive-by-wire chassis parameter control multi-channel. The multi-channel drive-by-wire chassis parameter control system includes multiple parallel sub-channels for different driving styles. It finds the optimal solution for each style in various typical and extreme heavy-duty truck operating conditions, such as heavy-load curves, braking, and lane changes. Each sub-channel is specifically designed for a driving style and can quickly and accurately adjust the vehicle chassis parameters under different driving situations. This provides heavy commercial vehicle drivers with a more personalized driving experience that adapts to their operating habits and balances safety, efficiency, and comfort.

[0048] Leveraging the powerful fitting capabilities of deep neural networks, it learns complex nonlinear relationships that far exceed what simple rules can describe, thereby finding the optimal solution for each style under various complex conditions. Each sub-channel corresponds specifically to a driving style and can quickly and accurately adjust vehicle chassis parameters in different driving scenarios, thus providing a more personalized, flexible, and safe driving experience.

[0049] Real-time monitoring of target driver behavior data; matching and analysis of target driver behavior data through multiple channels based on the drive-by-wire chassis parameters; determination of target drive-by-wire chassis parameters; and driving feedback optimization control through the target drive-by-wire chassis parameters.

[0050] Furthermore, this application also includes the following steps: performing style matching analysis on the target driver behavior data based on the multiple driving style feature clusters to determine the target driving style; matching and calling the target driving style with the multi-channel control of the steerable chassis parameters to obtain the target steerable chassis parameter control channel; and performing parameter control analysis on the target driver behavior data based on the target steerable chassis parameter control channel to determine the target steerable chassis parameters.

[0051] Specifically, vehicle sensors or other monitoring equipment are used to collect and monitor the target driver's behavior data in real time, including accelerator pedal opening and rate of change, brake pedal travel / brake pressure signals, steering wheel angle and angular velocity, current gear and shifting action, cruise control status, and torque braking usage, reflecting the driver's current driving habits and style. For example, key signals such as steering wheel rotation angular velocity, accelerator pedal opening gradient, brake pressure build-up rate, and gear shifting sequence are captured at a sampling frequency of 50 to 100 times per second.

[0052] Style matching analysis is performed on the target driver's behavior data based on multiple driving style feature clusters. The similarity between the target driver's behavior data and each driving style feature cluster is calculated, and the driving style feature cluster with the highest similarity is identified as the target driving style. For example, when a driver of a new energy heavy truck is performing container transfer in a yard, in order to quickly approach the target container position, the accelerator pedal opening is increased from 20% to 65% within 2 seconds. At the same time, in order to maintain straight driving in narrow passages or prepare for fine-tuning alignment, the steering wheel is continuously corrected with an average angular velocity of about 15° / s. By extracting the two real-time features, the accelerator pedal opening change rate and the average angular velocity of the steering wheel, it is found that the current feature vector is closest to the center point of the aggressive feature cluster in the closed scene [high accelerator pedal change rate, medium-high steering speed], with a similarity of 0.82. Therefore, the current target driving style is determined to be the aggressive operation type suitable for this scenario.

[0053] Based on the target driving style, the corresponding sub-channel for drive-by-wire chassis parameter control is retrieved from the multi-channel drive-by-wire chassis parameter control system to determine the target drive-by-wire chassis parameter control channel. Parameter control analysis is then performed on the target driver's behavior data based on this target drive-by-wire chassis parameter control channel. The target driver's behavior data is input into the target drive-by-wire chassis parameter control channel for parameter control analysis, and the target drive-by-wire chassis parameters are output. For example, the electro-hydraulic power steering characteristics are adjusted to sport mode to improve the central stiffness and return torque of the steering feel; the damping mode of the electronically controlled air suspension is adjusted to a high-damping state, and roll stiffness is increased to suppress body roll under heavy loads; simultaneously, the pressure response gradient of the braking system is adjusted to better match the driver's emergency braking expectations, thereby optimizing the vehicle's dynamic performance during acceleration, steering, and braking.

[0054] By monitoring driver behavior data in real time and combining it with driving style matching analysis, the system dynamically identifies the driver's driving style and automatically adjusts the vehicle chassis response parameters, improving the coordination between the driver and the vehicle and reducing the mismatch between the driver and the vehicle's response. This provides a smoother driving experience that is in line with the handling characteristics of heavy trucks. In particular, when switching between autonomous driving and manual driving modes, it avoids driving conflicts and maladaptation caused by style mismatch.

[0055] Furthermore, this application also includes the following steps: performing vehicle driving control through the target drive-by-wire chassis parameters and monitoring the driving process in real time to obtain a vehicle driving state data stream; evaluating the effect of the vehicle driving state data stream to obtain vehicle driving effect parameters; optimizing the target drive-by-wire chassis parameters based on the obtained vehicle driving effect parameters to obtain drive-by-wire chassis optimization parameters, and performing driving optimization control through the drive-by-wire chassis optimization parameters.

[0056] Furthermore, this application also includes the following steps: analyzing and adjusting the target drive-by-wire chassis parameters based on the obtained vehicle driving effect parameters to determine the drive-by-wire chassis parameter adjustment threshold; performing parameter selection simulation and global effect optimization within the drive-by-wire chassis parameter adjustment threshold to obtain the optimized parameters of the drive-by-wire chassis.

[0057] Specifically, the vehicle's driving behavior is controlled by target drive-by-wire chassis parameters. This involves executing the target drive-by-wire chassis parameters while simultaneously collecting actual vehicle performance data to obtain a vehicle driving state data stream. This data stream includes vehicle motion data, chassis actuator status data, and vehicle-road interaction data. The effectiveness of this driving state data stream is evaluated by comparing the deviation between the actual vehicle response and the expected standard, assessing whether the control effect meets expectations. Vehicle driving performance parameters, derived from the performance evaluation results, are used to measure the interaction between the vehicle and the driver, such as acceleration response deviation, braking response deviation, steering response accuracy, and driving mode switching deviation.

[0058] The target drive-by-wire chassis parameters are analyzed and adjusted based on vehicle driving performance parameters. If the evaluation results are unsatisfactory, the currently used target drive-by-wire chassis parameters are diagnosed for problems and adjustment direction is determined. Adjustment thresholds for the drive-by-wire chassis parameters are determined based on empirical rules and safety regulations. For example, optimization of the torque response gradient of the electric drive system in the low-speed range is permitted, such as increasing the initial response speed, but the upper limit must not exceed the critical value that would cause the drive wheels to slip on wet and slippery yard surfaces; or appropriate optimization of the deceleration gradient of the braking system is permitted, such as increasing the response speed by 15%, but the maximum deceleration must not exceed the maximum safe deceleration threshold required to ensure stable parking of the vehicle on wet or sloped loading / unloading positions. All adjustments must prioritize ensuring operational efficiency, handling accuracy, and vehicle stability under low-speed heavy loads. The drive-by-wire chassis parameter adjustment thresholds are parameter adjustment boundaries set to ensure vehicle safety and basic performance, similar to a safe window that allows for optimization.

[0059] After determining the threshold values ​​for the steerable drive chassis parameters, simulations are performed on different control parameters to verify the impact of each parameter configuration on vehicle response. Ultimately, an optimal solution that simultaneously improves roll and vibration is found, resulting in optimized steerable drive chassis parameters. Through global effect optimization, the optimal combination of multiple control parameters is selected to ensure the vehicle's best performance under various driving conditions. After simulation and optimization, a set of optimal steerable drive chassis parameters is determined. These optimized parameters are then distributed to each steerable drive actuator in the steerable drive chassis, replacing the previous target steerable drive chassis parameters, thus completing an online self-improvement. For example, for emergency braking conditions of new energy heavy trucks within the 0-30 km / h speed range, under the current electric braking deceleration gradient calibration, the measured braking distance deviates from the driver's expectations. Based on this, an adjustment threshold is determined, and the braking system deceleration gradient can be optimized by increasing it by 10% to 35% from the current calibration. In a simulation environment, the following sets of parameters were quickly tested: Parameter A, with a 15% increase, predicted braking distance shortened by 8%, but vehicle pitch rate increased by 15%; Parameter B, with a 25% increase, predicted braking distance shortened by 15%, but vehicle pitch rate increased by 25%; Parameter C, with a 35% increase, predicted braking distance shortened by 22%, but vehicle pitch rate increased by 40% and affected the speed of recovery after pitch. After a global trade-off—balancing braking efficiency, safe distance, ride comfort, and vehicle stability—parameter B was selected as the optimal solution. This is because it significantly shortens braking distance and improves safety while keeping the vehicle attitude changes caused by braking impact within a range acceptable to most drivers and without affecting cargo stability. The optimized parameters were then applied to the braking and suspension systems. When the vehicle performed similar emergency braking again, the smoothness of the braking process was significantly improved while maintaining a safe distance.

[0060] By monitoring driver behavior data in real time, the vehicle's chassis parameters are dynamically adjusted. Through precise effect evaluation and global optimization, the vehicle provides optimal response in every driving phase. All parameter optimizations are performed within the control threshold of the online chassis parameters, fundamentally ensuring absolute safety in vehicle operation. Simultaneously, through global optimization, the vehicle's potential is continuously explored within a safety framework, improving core performance indicators such as braking efficiency and cornering stability, significantly enhancing driving safety, especially in various complex driving environments, reducing driving conflicts or maladaptations caused by parameter mismatches.

[0061] Furthermore, this application also includes the following steps: collecting driver control feedback evaluation parameters through a driver feedback device installed in the vehicle; and optimizing the drive-by-wire chassis optimization parameters based on the driver control feedback evaluation parameters to meet specific needs.

[0062] Specifically, driver feedback parameters are collected through in-vehicle driver feedback devices, which are essentially data on the driver's subjective feelings about vehicle control. Driver feedback devices are in-vehicle equipment used to monitor and record driver feedback in real time. They typically include vibration feedback systems, seat feedback devices, haptic feedback systems, or biosensors, and can detect the driver's subjective feelings about vehicle control during driving. Driver feedback evaluation parameters are subjective and objective evaluation data related to vehicle control collected by driver feedback devices. They reflect the driver's perception of vehicle acceleration, braking, steering, and other control responses, such as steering force preference, braking smoothness rating, cornering roll perception, vehicle pitch control evaluation, and overall fatigue feedback.

[0063] Based on the collected driver control feedback evaluation parameters, the intent behind the feedback is analyzed and mapped to specific chassis parameter adjustment directions. For example, suppose that during a long-distance highway driving trip, the driver reports significant nose-diving during braking, leading to discomfort after frequent braking. In this case, the driver's need for brake balance and suspension control—namely, smoother braking body posture maintenance—is identified, and the brake pressure distribution and electronic air suspension damping calibration are adjusted accordingly. Based on the driver control feedback evaluation parameters, the drive-by-wire chassis optimization parameters are fine-tuned within a strict threshold conforming to heavy-duty truck safety and stability standards, and this process is recorded and integrated into the driver's personalized profile. During actual vehicle operation, driver control feedback is continuously received, and control parameters are dynamically adjusted. For example, each time the driver brakes, accelerates, or turns, the corresponding chassis parameters are adjusted based on real-time feedback to ensure consistency between driver operation and vehicle response.

[0064] For example, suppose a driver is driving on a bumpy road in a mining area, with the chassis parameters matched to their standard model. After navigating several right-angle turns in the yard or bumpy curves in the mining area, the driver feels the steering wheel's center of gravity is weak, and the steering feedback is blurry on bumpy surfaces, affecting precise control. Using the quick settings interface on the central control screen or steering wheel, the virtual steering feel preference is adjusted two levels from the default standard towards a more stable feel, recording the driver's handling feedback evaluation parameter as a steering force preference of +2 levels. Based on this instruction, immediate adjustment and analysis are performed, indicating that the current steering system's base assist curve corresponds to an equivalent force of approximately 8 N·m in the center zone. According to the internal mapping table, +2 levels correspond to increasing the center zone force by approximately 15%. After calculation within a safe threshold, the new target force is approximately 9.2 N·m. The corresponding steering assist characteristic curve is then generated and sent to the steering control system. Simultaneously, the lateral support stiffness of the electronically controlled suspension can be adjusted to maintain consistency between steering feel and vehicle dynamics. After the adjustment, the driver felt clearer and more stable steering feedback in subsequent similar curves, improving the accuracy of operation on complex road surfaces.

[0065] Adaptive optimization based on driver feedback dynamically adjusts the system according to the driver's real-time feedback and control needs, significantly improving the driving experience. The optimized control parameters ensure that the vehicle's acceleration, braking, and steering are consistent with the driver's expectations, reducing driver stress and improving driving safety. This forms a closed-loop feedback mechanism, enabling the vehicle to continuously optimize the chassis control system based on accumulated driver feedback data, thereby providing a personalized and intelligent driving experience.

[0066] In summary, the drive-by-wire chassis parameter optimization method based on driving behavior feedback provided in this application has the following beneficial effects: By collecting historical driving behavior datasets, feature extraction and style analysis are performed on the historical driving behavior datasets to obtain multiple driving style feature clusters; based on the multiple driving style feature clusters, a test schedule is designed and real-vehicle driving tests are conducted to record multiple driving style-drive-by-wire chassis response datasets; parameter training and fitting are performed based on the multiple driving style-drive-by-wire chassis response datasets to establish a multi-channel drive-by-wire chassis parameter control system, which includes multiple sub-channels for drive-style drive-by-wire chassis parameter control; target driver behavior data is monitored in real time, and matching control analysis is performed on the target driver behavior data based on the multi-channel drive-by-wire chassis parameter control system to determine the target drive-by-wire chassis parameters, and driving feedback optimization control is performed using the target drive-by-wire chassis parameters. In other words, by extracting features and analyzing styles from historical driving behavior datasets, the driving styles of drivers are identified. Based on different driving styles, operating condition tables are designed and real-vehicle driving tests are conducted to obtain drive-by-wire chassis response data corresponding to multiple driving styles. A multi-channel drive-by-wire chassis parameter control system is established, and matching control analysis is performed based on real-time monitored target driver behavior data to achieve personalized driving feedback optimization control, reduce human-machine co-driving conflicts, improve driving safety, and enhance the driving experience.

[0067] Example 2: Based on the same inventive concept as the drive-by-wire chassis parameter optimization method based on driving behavior feedback in Example 1, this application also provides a drive-by-wire chassis parameter optimization system based on driving behavior feedback. Referring to Figure 2, the drive-by-wire chassis parameter optimization system based on driving behavior feedback includes: a data acquisition module 11, used to collect historical driving behavior datasets, perform feature extraction and style analysis on the historical driving behavior datasets to obtain multiple driving style feature clusters; and a driving style feature extraction module 12, used to design a test schedule and conduct real-vehicle driving tests based on the multiple driving style feature clusters, recording multiple... Multiple driving style-drive-by-wire chassis response datasets; a multi-channel parameter control module 13, used to train and fit parameters based on the multiple driving style-drive-by-wire chassis response datasets to establish a multi-channel drive-by-wire chassis parameter control, wherein the multi-channel drive-by-wire chassis parameter control includes multiple driving style drive-by-wire chassis parameter control sub-channels; a matching control analysis module 14, used to monitor target driver behavior data in real time, perform matching control analysis on the target driver behavior data based on the multi-channel drive-by-wire chassis parameter control, determine target drive-by-wire chassis parameters, and perform driving feedback optimization control through the target drive-by-wire chassis parameters.

[0068] Furthermore, the data acquisition module 11 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used for: cleaning and standardizing the historical driving behavior dataset to obtain a standard historical driving behavior dataset; determining the driving behavior feature selection type according to the driving behavior analysis objective; performing feature extraction calculation on the standard historical driving behavior dataset according to the driving behavior feature selection type to obtain a driving behavior associated feature dataset; and performing driving style analysis based on the driving behavior associated feature dataset to obtain multiple driving style feature clusters.

[0069] Furthermore, the data acquisition module 11 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used for: presetting driving style types, including aggressive, normal, and conservative types; determining the number of feature clusters K=3 according to the driving style type, and initializing feature data clustering centers based on the number of feature clusters; performing K-means clustering analysis on the driving behavior associated feature dataset according to the feature data clustering centers to obtain feature data clustering results; and performing driving style feature analysis based on the feature data clustering results to obtain the multiple driving style feature clusters.

[0070] Furthermore, the driving style feature extraction module 12 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used for: designing driving condition test dimensions, which include driving behavior dimension, road condition dimension, speed dimension, and traffic condition dimension; performing parameter selection and analysis on each test dimension in the driving condition test dimensions to determine the driving condition dimension test parameter threshold; designing a test parameter table for the driving condition dimension test parameter threshold based on the multiple driving style feature clusters to generate a driving condition test parameter table; and conducting real vehicle driving tests based on the driving condition test parameter table to record the multiple driving style-drive-by-wire chassis response datasets.

[0071] Furthermore, the multi-channel parameter control module 13 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used to: evaluate and filter the response effects of the multiple driving style-drive-by-wire chassis response datasets to obtain multiple usable driving style-drive-by-wire chassis response datasets; use a deep neural network to train and fit the parameters of the multiple usable driving style-drive-by-wire chassis response datasets to obtain multiple drive-style drive-by-wire chassis parameter control sub-channels; and perform parallel integration and identification of the multiple drive-style drive-by-wire chassis parameter control sub-channels to establish the drive-by-wire chassis parameter control multi-channel.

[0072] Furthermore, the matching control analysis module 14 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used for: performing style matching analysis on the target driver behavior data based on the multiple driving style feature clusters to determine the target driving style; matching and calling the target driving style with the drive-by-wire chassis parameter control multi-channel to obtain the target drive-by-wire chassis parameter control channel; and performing parameter control analysis on the target driver behavior data based on the target drive-by-wire chassis parameter control channel to determine the target drive-by-wire chassis parameters.

[0073] Furthermore, the matching control analysis module 14 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used for: performing vehicle driving control through the target drive-by-wire chassis parameters and monitoring the driving process in real time to obtain a vehicle driving state data stream; evaluating the effect of the vehicle driving state data stream to obtain vehicle driving effect parameters; optimizing the target drive-by-wire chassis parameters based on the obtained vehicle driving effect parameters to obtain drive-by-wire chassis optimization parameters, and performing driving optimization control through the drive-by-wire chassis optimization parameters.

[0074] Furthermore, the matching control analysis module 14 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used to: analyze and regulate the target drive-by-wire chassis parameters based on the obtained vehicle driving effect parameters to determine the drive-by-wire chassis parameter regulation threshold; and perform parameter selection simulation and global effect optimization within the drive-by-wire chassis parameter regulation threshold to obtain the drive-by-wire chassis optimization parameters.

[0075] Furthermore, the matching control analysis module 14 in the drive-by-wire chassis parameter optimization system based on driving behavior feedback is also used to: collect driver operation feedback evaluation parameters through a driver feedback device installed in the vehicle; and perform adaptive demand optimization of the drive-by-wire chassis optimization parameters based on the driver operation feedback evaluation parameters.

[0076] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The drive-by-wire chassis parameter optimization method and specific examples based on driving behavior feedback in Embodiment 1 of Figure 1 are also applicable to the drive-by-wire chassis parameter optimization system based on driving behavior feedback in this embodiment. Through the foregoing detailed description of the drive-by-wire chassis parameter optimization method based on driving behavior feedback, those skilled in the art can clearly understand the drive-by-wire chassis parameter optimization system based on driving behavior feedback in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0077] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0078] Obviously, those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for optimizing drive-by-wire chassis parameters based on driving behavior feedback, characterized in that, include: A historical driving behavior dataset is collected, and feature extraction and style analysis are performed on the historical driving behavior dataset to obtain multiple driving style feature clusters; Based on the multiple driving style feature clusters, a test schedule was designed and a real vehicle driving test was conducted to record multiple driving style-drive-by chassis response datasets. Based on the multiple driving style-drive-by-the-wheel-drive response datasets, parameter training and fitting are performed to establish a multi-channel drive-by-the-wheel-drive parameter control system, which includes multiple sub-channels for drive-style drive-by-the-wheel-drive parameter control. Real-time monitoring of target driver behavior data; matching and analysis of target driver behavior data through multiple channels based on the drive-by-wire chassis parameters; determination of target drive-by-wire chassis parameters; and driving feedback optimization control through the target drive-by-wire chassis parameters.

2. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 1, characterized in that, The process involves obtaining multiple driving style feature clusters, including: performing abnormal data cleaning and standardization on the historical driving behavior dataset to obtain a standard historical driving behavior dataset; determining the driving behavior feature selection type based on the driving behavior analysis objective; performing feature extraction calculation on the standard historical driving behavior dataset according to the selected driving behavior feature type to obtain a driving behavior associated feature dataset; and performing driving style analysis based on the driving behavior associated feature dataset to obtain multiple driving style feature clusters.

3. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 2, characterized in that, Based on the driving behavior associated feature dataset, driving style analysis is performed to obtain multiple driving style feature clusters, including: a preset driving style type, which includes aggressive, normal, and conservative driving styles; based on the driving style type, the number of feature clusters is determined to be K=3, and feature data clustering centers are initialized based on the number of feature clusters; K-means clustering analysis is performed on the driving behavior associated feature dataset according to the feature data clustering centers to obtain feature data clustering results; driving style feature analysis is performed based on the feature data clustering results to obtain the multiple driving style feature clusters.

4. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 1, characterized in that, Recording multiple driving style-drive-by-wire chassis response datasets includes: designing driving condition test dimensions, which include driving behavior, road conditions, speed, and traffic conditions; performing parameter selection and analysis on each test dimension to determine the test parameter thresholds for each driving condition dimension; designing a test parameter table based on the multiple driving style feature clusters and the test parameter thresholds for each driving condition dimension; and conducting real-vehicle driving tests based on the driving condition test parameter table to record the multiple driving style-drive-by-wire chassis response datasets.

5. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 1, characterized in that, Establishing a multi-channel steerable chassis parameter control system includes: evaluating and filtering the response effects of multiple driving style-steerable chassis response datasets to obtain multiple usable driving style-steerable chassis response datasets; using a deep neural network to train and fit parameters to the multiple usable driving style-steerable chassis response datasets to obtain multiple driving style-steerable chassis parameter control sub-channels; and performing parallel integration and labeling of the multiple driving style-steerable chassis parameter control sub-channels to establish the multi-channel steerable chassis parameter control system.

6. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 1, characterized in that, Determining the target drive-by-wire chassis parameters includes: performing style matching analysis on the target driver behavior data based on the multiple driving style feature clusters to determine the target driving style; matching and calling multiple channels of the drive-by-wire chassis parameter control according to the target driving style to obtain the target drive-by-wire chassis parameter control channel; and performing parameter control analysis on the target driver behavior data based on the target drive-by-wire chassis parameter control channel to determine the target drive-by-wire chassis parameters.

7. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 1, characterized in that, Driving feedback optimization control based on the target drive-by-wire chassis parameters includes: controlling vehicle driving and monitoring the driving process in real time using the target drive-by-wire chassis parameters to obtain a vehicle driving state data stream; evaluating the effect of the vehicle driving state data stream to obtain vehicle driving effect parameters; optimizing the target drive-by-wire chassis parameters based on the obtained vehicle driving effect parameters to obtain drive-by-wire chassis optimization parameters, and performing driving optimization control using the drive-by-wire chassis optimization parameters.

8. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 7, characterized in that, Obtaining the optimized parameters of the drive-by-wire chassis includes: analyzing and adjusting the target drive-by-wire chassis parameters based on the obtained vehicle driving effect parameters to determine the drive-by-wire chassis parameter adjustment threshold; and performing parameter selection simulation and global effect optimization within the drive-by-wire chassis parameter adjustment threshold to obtain the optimized parameters of the drive-by-wire chassis.

9. The method for optimizing drive-by-wire chassis parameters based on driving behavior feedback as described in claim 8, characterized in that, The method further includes: collecting driver control feedback evaluation parameters through a driver feedback device installed in the vehicle; and performing adaptive optimization of the drive-by-wire chassis optimization parameters based on the driver control feedback evaluation parameters.

10. A drive-by-wire chassis parameter optimization system based on driving behavior feedback, characterized in that, The steps for implementing the drive-by-wire chassis parameter optimization method based on driving behavior feedback according to any one of claims 1 to 9, wherein the drive-by-wire chassis parameter optimization system based on driving behavior feedback comprises: a data acquisition module, used to acquire historical driving behavior datasets, perform feature extraction and style analysis on the historical driving behavior datasets to obtain multiple driving style feature clusters; a driving style feature extraction module, used to design a test schedule and conduct real-vehicle driving tests based on the multiple driving style feature clusters, and record multiple driving style-drive-by-wire chassis response datasets; a multi-channel parameter control module, used to perform parameter training and fitting based on the multiple driving style-drive-by-wire chassis response datasets to establish a multi-channel drive-by-wire chassis parameter control, wherein the multi-channel drive-by-wire chassis parameter control includes multiple driving style drive-by-wire chassis parameter control sub-channels; and a matching control analysis module, used to monitor target driver behavior data in real time, perform matching control analysis on the target driver behavior data based on the multi-channel drive-by-wire chassis parameter control, determine target drive-by-wire chassis parameters, and perform driving feedback optimization control through the target drive-by-wire chassis parameters.

Images