A method for adjusting road feel in steer-by-wire based on a driving simulator and related equipment.

Abstract

This invention provides a method and related equipment for adjusting road feel in steer-by-wire based on a driving simulator, relating to the field of steer-by-wire systems. This technical solution uses a driving simulator to collect driver operation characteristics and vehicle driving scenario data in real time. It then uses a fuzzy C-means clustering algorithm to accurately identify the driver's driving style type and outputs a corresponding driving style coefficient. This coefficient is then substituted into a dynamic adjustment algorithm for road feel parameters, combined with driving scenario data, to calculate the target road feel torque in real time. Finally, this is converted into a motor control signal to drive the steering wheel feel simulator motor to output the corresponding feedback torque. This achieves adaptive response of the steer-by-wire system to the driver's personalized control preferences, effectively improving the accuracy, safety, and comfort of road feel feedback in complex driving scenarios.

Description

Technical Field

[0001] This invention relates to the field of steer-by-wire systems, and in particular to a steer-by-wire road feel adjustment method and related equipment based on a driving simulator. Background Technology

[0002] The application of steer-by-wire systems in the process of automotive intelligence is becoming increasingly widespread. This system eliminates the mechanical connection between the steering wheel and the wheels, instead generating road feel through algorithmic simulation. Current mainstream technologies typically use fixed lookup tables or polynomial models based on vehicle speed, lateral acceleration, and steering angle to calculate road feel. This traditional method has significant drawbacks: its road feel parameters are fixed and universal, failing to adapt to the subjective preferences of different driving styles, easily leading to aggressive drivers feeling the steering is too light and mild drivers feeling it is too heavy; at the same time, existing algorithms rely only on objective vehicle state parameters, failing to fully incorporate real-time driver operation characteristics, which can easily lead to road feel distortion in complex and extreme scenarios, affecting handling safety; furthermore, current driving simulators are mostly limited to the verification stage of road feel models, and their potential in algorithm development and adaptive adjustment has not yet been fully explored. Summary of the Invention

[0003] The purpose of this invention is to provide a steer-by-wire road feel adjustment method and related equipment based on a driving simulator, in order to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions that can dynamically generate personalized road feel feedback torque according to the real-time identified driver operating style and driving scenario, thereby taking into account the subjective preferences and comfort needs of different drivers while ensuring driving safety.

[0004] On the one hand, this application provides a method for adjusting the road feel of steer-by-wire based on a driving simulator, including the following steps: S100: Collects driver operation characteristic data and vehicle driving scene data in real time in the driving simulator; S200: The operation feature data is processed using the fuzzy C-means FCM clustering algorithm to identify the driver's driving style type and output the corresponding driving style coefficient. S300: Substitute the driving style coefficient into the road feel parameter dynamic adjustment algorithm, and combine it with the driving scenario data to calculate the target road feel torque in real time; S400: Convert the target road feel torque into a motor control signal and output it to the steering wheel feel simulator motor in the steer-by-wire system to drive the motor to output the corresponding feedback torque.

[0005] Furthermore, in S100, the driver's operational characteristic data includes: steering wheel angle, steering angular velocity, steering angular acceleration, driver applied torque, and steering operation frequency; The driving scenario data includes: vehicle speed, lateral acceleration, and road surface adhesion coefficient.

[0006] Furthermore, in step S200, the fuzzy C-means FCM clustering algorithm is used to identify the driver's driving style. Specific steps include: The collected driver operation feature data is normalized and an operation feature vector is constructed. Perform fuzzy C-means FCM clustering to cluster the driver's steering operation feature vector into multiple preset driving style categories, and output the continuous membership degree of each sample to each category; The driving style coefficient is calculated based on the continuous membership degree.

[0007] Furthermore, the calculation of the driving style coefficient based on the continuous membership degree specifically involves: The membership degree of each category is weighted and calculated with the preset style weights; The execution of fuzzy C-means FCM clustering specifically includes setting an objective function and iteratively updating the membership matrix and cluster centers until the objective function converges or reaches the preset maximum number of iterations.

[0008] Furthermore, in S300, the real-time calculation of the target road feel torque specifically involves: Construct a basic road feel model that includes vehicle speed gain, lateral acceleration damping, and road surface adhesion self-correction coefficient; Introduce a style adaptive correction coefficient based on the driving style coefficient; The base torque calculated from the basic road feel model is weighted and calculated with the style adaptive correction coefficient to obtain the final target road feel torque.

[0009] Furthermore, the logic for determining the style adaptive correction coefficient is as follows: Obtain a preset first style threshold and a second style threshold, wherein the first style threshold is greater than the second style threshold; When the identified driving style coefficient is greater than or equal to the first style threshold, it is determined that the driver's operation is in the first preset style range, and the style adaptive correction coefficient is increased according to the first gain ratio to enhance the output target road feel torque. When the identified driving style coefficient is less than or equal to the second style threshold, it is determined that the driver's operation is in the second preset style range, and the style adaptive correction coefficient is adjusted down according to the second attenuation ratio to reduce the output target road feel torque.

[0010] Furthermore, the method also includes: Subjective evaluation indicators of multiple groups of drivers with different driving styles were collected in various preset test scenarios. The current road feel parameters are evaluated based on the subjective evaluation index. If the evaluation result does not meet the design threshold, the optimization process of the basic road feel parameters and FCM cluster centers is triggered, and the process returns to S100 for closed-loop iteration until the design requirements are met.

[0011] On the other hand, this application provides a steer-by-wire road feel adjustment system based on a driving simulator, including the following hardware modules: The data acquisition unit, installed on the driving simulator, is used to collect driver operation characteristics and vehicle status signals; The central processing unit is used to execute the road feel adjustment method as described above, specifically including a style recognition unit and a torque calculation unit; The actuator includes a feel simulator motor mounted on the steering wheel assembly, used to output a target road feel torque according to instructions from the central processing unit.

[0012] On the other hand, this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned steer-by-wire road feel adjustment method based on a driving simulator.

[0013] On the other hand, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned steer-by-wire road feel adjustment method based on a driving simulator.

[0014] The beneficial effects of this invention are as follows: This application provides a steer-by-wire road feel adjustment method based on a driving simulator. This technical solution uses a driving simulator to collect driver operation characteristics and vehicle driving scenario data in real time. It then uses a fuzzy C-means clustering algorithm to accurately identify the driver's driving style type and outputs a corresponding driving style coefficient. This coefficient is then substituted into a road feel parameter dynamic adjustment algorithm, combined with driving scenario data, to calculate the target road feel torque in real time. Finally, this is converted into a motor control signal to drive the steering wheel feel simulator motor to output the corresponding feedback torque. This achieves adaptive response of the steer-by-wire system to the driver's personalized control preferences, effectively improving the accuracy, safety, and comfort of road feel feedback in complex driving scenarios. This application also provides related equipment for the above method. The beneficial effects of the related equipment are similar to those of the above method and will not be elaborated here.

[0015] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0016] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0017] Figure 1 This is a flowchart of the steer-by-wire road feel adjustment method based on a driving simulator provided in this application; Figure 2 This is a schematic diagram of the road feel model obtained from the output provided in this application; Figure 3 This is a structural diagram of the steer-by-wire road feel adjustment system based on a driving simulator provided in this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] The present application will be further described below with reference to the accompanying drawings and specific embodiments. The described embodiments should not be considered as limitations on the present application, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present application.

[0020] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] With the rapid development of automotive intelligence and electrification, steer-by-wire systems have eliminated the mechanical connection between the steering wheel and the steering actuator. Instead, they rely on algorithms to construct a road feel simulation model and output feedback torque through a hand feel simulation motor to recreate a realistic driving feel. Although ideal road feel feedback should take into account the subjective preferences of different drivers and diverse road scenarios, and driving simulators have the advantages of being risk-free and repeatable in development, current technologies still have significant shortcomings in terms of personalized matching and scenario adaptability.

[0023] Currently, most mainstream technologies rely on fixed lookup table methods or polynomial mathematical models based on vehicle speed, lateral acceleration, and steering wheel angle. While this fixed parameter mapping strategy is simple to implement, it cannot be dynamically adjusted according to the driver's real-time operating characteristics. As a result, aggressive drivers feel that the steering is too light and the feedback is insufficient when driving aggressively, while mild-mannered drivers feel that the steering is too heavy and the operation is difficult in daily driving. It is difficult to achieve a balance between different user groups, which seriously affects driving comfort and satisfaction.

[0024] Furthermore, existing road sensing algorithms mainly rely on objective vehicle state parameters such as vehicle speed, lacking a deep perception of the driver's real-time operational intentions. This results in the system being unable to respond to the driver's transient operational needs in a timely manner in extreme scenarios such as sharp steering, emergency obstacle avoidance, or low-friction surfaces, which can easily lead to distorted road sensing feedback or even situations that contradict the driver's intentions, thereby endangering driving safety.

[0025] Meanwhile, in the existing technology system, the function of driving simulators is mostly limited to the verification stage after the algorithm development is completed. Their potential in data acquisition, model training and parameter iterative optimization has not been fully explored. Due to the lack of a mechanism to feed simulator data back to the algorithm model in real time for closed-loop iteration, the calibration process of road feel parameters relies on a large number of real vehicle road tests, resulting in a disconnect between the development and verification process. This not only prolongs the development cycle and increases costs, but also restricts the rapid improvement and iteration of the road feel quality of the steer-by-wire system.

[0026] To address the shortcomings of the existing technologies, this application proposes a steer-by-wire road feel adjustment method and related equipment based on a driving simulator, constructing a data-driven adaptive closed-loop control system. This method first utilizes a driving simulator to collect real-time micro-operational characteristic data of the driver and macro-driving scene data of the vehicle under a safe environment. Then, a fuzzy C-means clustering algorithm is introduced to deeply mine and process the massive amount of collected operation data, thereby accurately identifying the driver's current driving style type and outputting a quantified driving style coefficient.

[0027] Subsequently, the system uses the style coefficient as a key variable in the dynamic adjustment algorithm of road feel parameters. Combined with real-time driving scenario data, it generates a target road feel torque that perfectly matches the driver's intention and road conditions through multi-dimensional fusion calculation. Finally, it converts the torque into a motor control signal to drive the steering wheel feel simulator motor, realizing a fundamental shift from fixed road feel to personalized and dynamic road feel feedback.

[0028] First, the method for adjusting the road feel of steer-by-wire based on a driving simulator provided in this application will be described in detail below with reference to the accompanying drawings.

[0029] Reference Figure 1 The implementation process of the steer-by-wire road feel adjustment method based on a driving simulator provided in this application includes, but is not limited to, the following steps.

[0030] Step S100: Real-time acquisition of driver operation characteristic data and vehicle driving scene data in the driving simulator.

[0031] In step S100, a comprehensive, high-fidelity data acquisition environment is constructed to provide accurate data support for subsequent personalized road feel generation. This step fully leverages the advantages of driving simulators—risk-free, repeatable, and parameter-controllable—enabling real-time capture of the driver's most realistic micro-level operational behaviors in simulated complex traffic flows or extreme weather scenarios. The system not only focuses on the vehicle's macroscopic driving state, such as vehicle speed and lateral acceleration, but also emphasizes mining the driver's operational characteristic data, such as the transient rate of change of steering wheel angle, the fluctuation frequency of torque input, and the fineness of directional correction.

[0032] These multi-dimensional raw data form the basis of the system's perception, enabling the system to not only see how the car runs, but also understand how the driver drives, just like an experienced racing coach. This solves the problem of data deficiency in traditional methods that rely solely on a single vehicle state parameter and ignore the driver's subjective intentions, providing a detailed and timely input source for subsequent algorithm analysis.

[0033] Step S200: The fuzzy C-means FCM clustering algorithm is used to process the operation feature data, identify the driver's driving style type, and output the corresponding driving style coefficient.

[0034] In step S200, the advanced data mining technique of fuzzy C-means clustering is used to transform abstract driving behavior into quantifiable mathematical indicators. Traditional classification methods often have clear boundaries and are difficult to deal with the continuity and ambiguity of human driving behavior. However, fuzzy C-means clustering can calculate the membership degree of each sample point to different categories, thus more accurately handling the fuzzy concept of driving style, which is neither black nor white.

[0035] By iteratively calculating and recognizing the massive amount of operational feature data collected in step S100, the system can automatically classify drivers into different style categories such as mild, normal, or aggressive, and output a continuous driving style coefficient. This coefficient is not just a simple label; it represents a quantitative value of the driver's current level of aggressiveness. It can keenly capture the subtle psychological changes of the driver when switching from smooth driving to aggressive control, providing highly targeted and personalized parameters for subsequent torque adjustment, completely breaking the rigidity of traditional road feel control.

[0036] In step S300, the driving style coefficient is substituted into the dynamic adjustment algorithm of road feel parameters, and the target road feel torque is calculated in real time by combining driving scenario data.

[0037] In step S300, the dynamic calculation and adaptive matching of the road feel torque are implemented. In this step, the system no longer uses a fixed lookup table method, but instead uses the driving style coefficient output in step S200 as a key variable, which is then substituted into the dynamic adjustment algorithm for the road feel parameters. This algorithm can adjust the weights of the gain coefficient, damping ratio, or self-aligning torque in the road feel model in real time according to different driving style coefficients.

[0038] For example, when the algorithm detects an aggressive driving style, it automatically increases the centering torque and self-centering stiffness at high speeds to satisfy the driver's desire for precise control. Conversely, when the coefficient is moderate, the algorithm appropriately reduces the torque gradient to improve driving ease and comfort. Simultaneously, this step deeply integrates vehicle driving scenario data to ensure that, under special conditions such as sharp turns, emergency obstacle avoidance, or low-traction surfaces, the calculated target road feel torque satisfies the driver's personalized preferences while strictly adhering to the vehicle's dynamic safety boundaries. This ensures driving pleasure while building a robust safety barrier.

[0039] In step S400, the target road feel torque is converted into a motor control signal and output to the steering wheel feel simulator motor in the steer-by-wire system to drive the motor to output the corresponding feedback torque.

[0040] In step S400, the digitized calculation results are converted into realistic tactile feedback from the driver's hands, completing the closed loop of human-machine interaction. The steer-by-wire system eliminates mechanical connections, relying entirely on the motor to simulate road feel; therefore, the accuracy of this step directly determines the realism of the driving experience. The system converts the target road feel torque calculated in step S300 into a high-precision motor control signal, typically employing a composite control strategy combining current and speed loops to drive the steering wheel feel simulator motor to output the corresponding electromagnetic torque.

[0041] This process demands extremely high dynamic response speed and torque control precision to ensure that the feedback torque output by the motor can follow changes in the target value within milliseconds. Through this step, the driver's wrist can truly feel the feedback of road information and the torque characteristics tailored to its features by the system, perfectly integrating the virtual driving scenario with real physical feedback. This not only verifies the effectiveness of the algorithm but also fundamentally improves the adaptability, comfort, and safety of the steer-by-wire system in the hands of different drivers.

[0042] In some embodiments of this application, in step S100, the driver's operational characteristic data includes: steering wheel angle, steering angular velocity, steering angular acceleration, driver-applied torque, and steering operation frequency. Driving scenario data includes: vehicle speed, lateral acceleration, and road surface adhesion coefficient.

[0043] In some embodiments of this application, the steering wheel angle, as the most direct physical quantity reflecting the driver's steering intention, plays a crucial role in establishing the fundamental target of the vehicle's driving trajectory. In the driving simulator environment, the system uses high-precision sensors to capture the angle of the steering wheel's deviation from the center position in real time. This is not only the basis for calculating the vehicle's theoretical steering ratio but also the primary signal for determining whether the driver is currently maintaining a straight line, changing lanes to overtake, or making a large-angle turn. Through continuous monitoring of the steering angle data, the system can construct a time-series curve of the driver's operations, thus providing raw material for subsequent analysis of the smoothness and continuity of their operations, ensuring that the system can perceive the driver's macroscopic command to change the vehicle's posture in the first instance.

[0044] In some embodiments of this application, steering angular velocity and steering angular acceleration are key differential indicators for measuring the urgency and dynamic characteristics of driver operation. Steering angular velocity reflects the speed of steering wheel rotation and can directly reflect the driver's urgency in responding to the vehicle; while steering angular acceleration further reveals the explosive force and transient rate of change of the driver's hand movements.

[0045] These two parameters are crucial in identifying driving styles. For example, high-frequency and dramatic fluctuations in angular velocity and acceleration are typical characteristics of aggressive drivers during spirited driving or emergency obstacle avoidance, while gradually changing values ​​correspond to milder drivers. Introducing these two dynamic parameters allows the system to move beyond static position determination and gain deeper insights into the driver's transient intentions, thus providing highly sensitive feature vectors for the fuzzy C-means clustering algorithm to distinguish different driving styles.

[0046] In some embodiments of this application, the torque applied by the driver is the core data connecting the physical touchpoints of human-machine interaction. It directly reflects the physical exertion and subjective will of the driver in overcoming the return torque and achieving the steering goal. In the road feel simulation system, there is a subtle game and fusion relationship between the torque input by the driver's hand and the torque fed back by the system.

[0047] Collecting this data helps the system determine whether the driver's perception of "weight" in the current road conditions is appropriate and whether there is any difficulty in operation. Furthermore, the fluctuation characteristics of the torque signal often imply the driver's perceptual feedback on road information. By analyzing the magnitude and trend of the torque applied by the driver, the system can assess whether the current road feel model matches the driver's muscle memory and force expectations. This provides an ergonomic basis for subsequently dynamically adjusting the gain of the feedback torque, ensuring that the final generated torque conforms to both vehicle dynamics and the driver's physical load preferences.

[0048] In some embodiments of this application, steering operation frequency is a statistical characteristic parameter characterizing the number of times and the level of activity of the driver in correcting the steering wheel per unit time. Its unique role is to reveal the driver's ability to control the vehicle's driving stability and the level of psychological tension. At high speeds or on roads with low coefficient of friction, frequent and minor steering corrections often indicate that the driver lacks confidence in maintaining the vehicle's trajectory or that there is significant road interference, which is a high-frequency intervention driving behavior; while a lower operation frequency usually corresponds to a stable and confident driving state.

[0049] Including this parameter in the data collection range can effectively distinguish whether the driver is making large-scale route planning or dealing with high-frequency road disturbances. It helps the system identify those "restless" operations that are small in magnitude but extremely high in frequency, so that appropriate damping optimization or torque smoothing can be performed for such high-frequency noise when adjusting the road feel, thereby improving the driving stability.

[0050] In some embodiments of this application, vehicle speed is a global constraint variable that determines the basic characteristics of the steer-by-wire road feel. Its main function is to establish the transmission ratio and the magnitude of the basic self-centering torque of the steering system, so as to ensure the driving stability of the vehicle in different speed ranges. According to the principles of vehicle dynamics, a light and flexible steering feel is required at low speeds to facilitate parking and U-turns, while a stable and heavy steering torque is required at high speeds to prevent the vehicle from drifting and to ensure tracking.

[0051] Therefore, real-time vehicle speed data collection is intended to enable the road feel adjustment algorithm to automatically switch the basic control strategy according to speed changes, ensuring that no matter how aggressive the driving style, the torque output by the system always follows the safe physical principle of "lightweight at low speeds and stable at high speeds," providing a safe benchmark platform that dynamically changes with speed for personalized adjustments.

[0052] In some embodiments of this application, lateral acceleration is a key dynamic parameter reflecting the vehicle's extreme state and tire stress, providing a realistic "feeling of gravity" and limit feedback for road feel simulation. When a vehicle turns, lateral acceleration is directly related to the tire's slip angle and lateral force, serving as an important physical feedback source for the driver to perceive whether the vehicle is about to lose control. In road feel adjustment, introducing lateral acceleration data allows the algorithm to calculate the vehicle's current centrifugal force in real time and accordingly superimpose the corresponding self-centering torque and damping sensation.

[0053] This means that when the driver drives aggressively and the lateral acceleration increases sharply, the system will automatically increase the resistance of the steering wheel to simulate the heavy feeling when the tire grip reaches its limit. This provides the driver with clear physical feedback, alerting him to the vehicle's dynamic boundaries and preventing the driver from misjudging the speed and cornering difficulty due to an overly light road feel.

[0054] In some embodiments of this application, the road surface adhesion coefficient is a core indicator among various environmental parameters characterizing the severity of road conditions and tire grip capability. It provides the steer-by-wire system with prior knowledge about the "road surface texture" to achieve adaptive road feel adjustment. Different road surface adhesion coefficients (such as dry asphalt, wet ski slopes, or icy roads) correspond to drastically different tire mechanical characteristics.

[0055] Collecting this data allows the system to proactively adjust its road feel simulation strategy when entering low-traction surfaces. For example, it can appropriately reduce the abrupt change rate of the self-centering torque to prevent the torque from impacting the driver's hand, or alter the torque gradient characteristics to simulate the "floating" feeling of icy or snowy roads. This step ensures that road feel adjustment not only considers the "human" factor but also the "road" factor, enabling the output feedback torque to accurately reflect the current road surface's friction limits. This helps drivers perceive risks through changes in feel when road grip is insufficient, allowing them to take more cautious driving actions.

[0056] In some embodiments of this application, step S200 involves using the fuzzy C-means FCM clustering algorithm to identify the driver's driving style. The specific steps include: Step S210: Normalize the collected driver operation feature data and construct an operation feature vector.

[0057] In step S210, the interference of different physical dimensions on the algorithm calculation is eliminated, and the discrete sensor signals are converted into a mathematical form that the algorithm can recognize. Because the raw data such as steering wheel angle, steering angular velocity, and driver applied torque collected have huge differences in numerical range, unit magnitude, and frequency of change, directly inputting them into the clustering algorithm will cause the features with larger values ​​to dominate the calculation results, thereby masking the feature information that has smaller values ​​but is equally important.

[0058] Through normalization, the system maps all operational feature data to a unified numerical range, ensuring that each feature dimension has equal weight in subsequent analysis. The operational feature vector constructed on this basis essentially abstracts the driver's micro-operational behavior over a period of time into geometric points in a high-dimensional space. This not only achieves standardized fusion of multi-source heterogeneous data but also provides a clearly structured and weighted mathematical input for the subsequent fuzzy C-means clustering algorithm, guaranteeing the convergence speed and computational accuracy of model training.

[0059] Step S220: Perform fuzzy C-means FCM clustering to cluster the driver steering operation feature vector into multiple preset driving style categories, and output the continuous membership degree of each sample to each category.

[0060] In step S220, the fuzzy C-means clustering algorithm overcomes the limitations of traditional hard clustering algorithms, which often present only one type of driving style, and achieves flexible classification of complex human driving behaviors. Traditional clustering methods often force drivers to be classified into a specific style, ignoring the transitional and mixed characteristics that driving behaviors may exhibit in different scenarios. This step, by performing fuzzy C-means clustering, allows an operational feature vector to belong to multiple preset driving style categories simultaneously, and outputs the continuous membership degree of each sample to each category.

[0061] This means the system can recognize that a driver may exhibit both gentle, stable driving and sudden bursts of aggressive driving at any given moment. This membership-based calculation method more realistically reflects the ambiguity and continuity of human driving behavior, avoiding abrupt changes in control strategies caused by rigid classification, and laying a solid logical foundation for generating smooth-transition road feel adjustment parameters.

[0062] Step S230: Calculate the driving style coefficient based on continuous membership degree.

[0063] In step S230, the multi-dimensional probability distribution is mapped to a scalarized driving style coefficient to facilitate direct invocation of the road feel adjustment algorithm. Based on the continuous membership degree output in step S220, the system uses weighted calculation or other mapping logic to fuse the probability values ​​of different driving styles, such as aggressive and mild, into a continuous numerical index. This driving style coefficient is no longer a simple classification label, but a scale that can accurately quantify the aggressiveness of the driver's current operation. It retains the subtle perception capability of fuzzy clustering for complex behaviors while simplifying it into a single variable that is easy for the steer-by-wire control system to handle.

[0064] Through this step, the system can output a style coefficient that dynamically changes according to the driver's operating habits in real time. This makes the subsequent road feel torque adjustment no longer a discrete gear shift, but can achieve a linear and smooth transition at the millisecond level based on the coefficient, truly achieving a seamless fit between road feel feedback and driver intention.

[0065] In some embodiments of this application, step S230, which calculates the driving style coefficient based on continuous membership, specifically includes the following steps: Step S231: Calculate the weighted average of the membership degree of each category with the preset style weights.

[0066] In step S231, a quantitative mapping bridge is constructed from the fuzzy probability space to specific physical control parameters. Through weighted calculation, the abstract clustering results are transformed into scalar indicators that the drive-by-wire system can directly call upon. Since the continuous membership degree output in step S220 only represents the probability or likelihood that the driver's operating characteristics belong to a specific style category, the system cannot directly use these probability values ​​to adjust the motor's output torque.

[0067] Therefore, this step introduces preset style weights. These weights are typically calibrated based on the demands of driving style on vehicle dynamics. For example, an aggressive style corresponds to a higher weight value to improve the sensitivity of road feel feedback, while a mild style corresponds to a lower weight value to enhance comfort. By weighted summing of the membership degrees of each category with their corresponding style weights, the system can calculate a driving style coefficient that lies between the extreme values ​​of different styles.

[0068] This coefficient not only accurately quantifies the driver's current operating tendency, but also ensures that when the driver's driving behavior smoothly transitions between different styles, the output control parameters can also change linearly, thereby avoiding the step change in road feel torque caused by abrupt changes in style judgment, and ensuring the continuity and naturalness of driving feel.

[0069] Step S232: Perform fuzzy C-means FCM clustering, which includes setting an objective function and iteratively updating the membership matrix and cluster centers until the objective function converges or the preset maximum number of iterations is reached.

[0070] In step S232, the cluster centers and membership distributions that best reflect the inherent structure of the data are obtained through iterative optimization, ensuring the accuracy and robustness of driving style identification. This step first sets an objective function, which is typically defined as the weighted sum of squared distances from all sample points to their corresponding cluster centers, aiming to measure the compactness of the clustering results.

[0071] Subsequently, the algorithm enters an iterative loop, alternately executing two key operations: first, with the cluster centers fixed, updating the membership matrix so that sample points are more likely to belong to the closer cluster centers; second, with the membership matrix fixed, updating the position of the cluster centers so that they move to the weighted center of that class of samples.

[0072] This process is repeated until the objective function converges to a minimum or the preset maximum number of iterations is reached. Through this rigorous mathematical iterative process, the system can automatically extract the most representative driving mode prototypes from massive amounts of driver operation data without prior labels, providing accurate membership data support for step S231. This ensures that the entire driving style recognition system can maintain a high degree of adaptability and classification accuracy when facing different drivers and complex road conditions.

[0073] In some embodiments of this application, step S300, specifically calculating the target road feel torque in real time, is as follows: Step S310: Construct a basic road feel model that includes vehicle speed gain, lateral acceleration damping, and road surface adhesion self-correction coefficient.

[0074] In step S310, a basic torque model conforming to the physical laws of vehicle dynamics and driving safety benchmarks is established, providing a solid underlying physical architecture for road feel simulation. This step aims to simulate the steering feedback characteristics of a real vehicle under ideal conditions by constructing a composite model that includes vehicle speed gain, lateral acceleration damping, and road adhesion self-correction coefficient.

[0075] The speed gain function is used to make the steering feel change with speed, ensuring that the steering wheel is light and flexible when the vehicle is driving at low speeds, such as parking, while the steering wheel is stable and heavy when driving at high speeds, preventing the vehicle from feeling unstable due to too light road feel; the introduction of lateral acceleration damping is to simulate the resistance torque generated by the tire lateral deflection when the vehicle is turning, which increases the steering resistance as the centrifugal force increases, giving the driver clear limit feedback; and the road adhesion self-centering coefficient adjusts the gradient of the self-centering torque according to the road conditions (such as dry or wet), allowing the driver to perceive the change in road grip through feel.

[0076] This step ensures that regardless of the driver's style, the system's output torque always follows the laws of physics, guaranteeing the vehicle's tracking performance and safety.

[0077] Step S320: Introduce a style adaptive correction coefficient based on the driving style coefficient.

[0078] In step S320, a personalized adjustment factor is introduced, breaking away from the rigid, one-size-fits-all approach of traditional road feel control and achieving precise adaptation to the operating habits of different drivers. In this step, the system converts the driving style coefficient identified in step S200 into a style adaptive correction coefficient. This correction coefficient acts as a regulating valve connecting the driver's subjective intentions with objective physical feedback.

[0079] For aggressive drivers, the correction factor tends to enhance road feel sensitivity and feedback, retaining more road surface details and tire dynamics to meet their high demands for handling precision. For milder drivers, the correction factor tends to smooth torque fluctuations, reduce high-frequency interference, and provide a more comfortable and relaxed steering feel. By introducing this correction factor based on driving style, the system can dynamically reshape the output characteristics of the basic road feel model, making road feel feedback no longer a cold, physical simulation, but an adaptive interaction full of human-centered care.

[0080] Step S330: The basic torque calculated from the basic road feel model is weighted and calculated with the style adaptive correction coefficient to obtain the final target road feel torque.

[0081] In step S330, the basic physical torque is organically integrated with personalized psychological needs to output a final target road feel torque that is both realistic and considerate. This step involves a weighted calculation of the basic torque calculated from the basic road feel model and the style adaptive correction coefficient. This process is not a simple numerical addition, but rather a personalized gain or attenuation based on a safety benchmark. Through this weighted calculation, the system can retain the necessary physical feedback for vehicle driving (such as self-centering torque and damping feel) while fine-tuning the magnitude, gradient, and sensitivity of the torque according to the driver's real-time style.

[0082] For example, while maintaining stability at high speeds, the road feel and weight are appropriately increased during aggressive driving. This step ensures that the target road feel torque ultimately output to the steering wheel feel simulator conforms to the objective laws of vehicle dynamics and perfectly matches the driver's current subjective driving intentions, achieving the best balance in human-machine co-driving.

[0083] In some embodiments of this application, the logic for determining the style adaptive correction coefficient in step S320 is as follows: Step S321: Obtain a preset first style threshold and a second style threshold, wherein the first style threshold is greater than the second style threshold.

[0084] In step S321, clear physical boundaries and judgment benchmarks are established for the quantitative evaluation of driving style, thereby transforming ambiguous driving behavior into interval divisions with clear engineering guidance significance. By acquiring and setting preset first style thresholds and second style thresholds, and strictly ensuring that the first style threshold is greater than the second style threshold, the system constructs a stable reference coordinate system on a continuous driving style coefficient axis.

[0085] These two thresholds act as a watershed in the algorithm's logic. They not only define the triggering conditions for extreme driving behaviors but also reserve a buffer or smooth transition range between the two thresholds. This rigorous threshold setting mechanism effectively avoids frequent jumps in the control strategy near the critical point due to minor data fluctuations, ensuring that the subsequent road feel adjustment logic has extremely high robustness and anti-interference capabilities.

[0086] Step S322: When the identified driving style coefficient is greater than or equal to the first style threshold, it is determined that the driver's operation is in the first preset style range, and the style adaptive correction coefficient is increased according to the first gain ratio to enhance the output target road feel torque.

[0087] In step S322, the system accurately identifies and responds to the driver's high-intensity control demands, and improves the vehicle's dynamic feedback performance under extreme conditions through a positive gain mechanism. When the system monitors in real time that the identified driving style coefficient is greater than or equal to the set first style threshold, it decisively determines that the driver is currently in the first preset style range (usually corresponding to intense, aggressive, or high-frequency intervention driving states).

[0088] Under this condition, the algorithm will increase the style adaptive correction coefficient according to the preset first gain ratio. This increase directly affects the torque calculation model, resulting in a significant enhancement of the final output target road feel torque. The enhanced torque provides the driver with a more stable steering wheel feel, clearer road information, and higher steering damping, thereby meeting the driver's stringent requirements for vehicle tracking and handling confidence during fast cornering or emergency obstacle avoidance.

[0089] Step S323: When the identified driving style coefficient is less than or equal to the second style threshold, it is determined that the driver's operation is in the second preset style range, and the style adaptive correction coefficient is adjusted down according to the second attenuation ratio to reduce the output target road feel torque.

[0090] In step S323, the smoothness and effortless experience of human-computer interaction are optimized through a negative attenuation mechanism. When the system detects that the identified driving style coefficient is less than or equal to the set second style threshold, it accurately determines that the driver's operation is in the second preset style range (usually corresponding to a mild, smooth, or low-frequency correction driving state).

[0091] At this point, the algorithm automatically adjusts the style adaptive correction coefficient according to the second attenuation ratio. The introduction of this attenuation ratio effectively reduces unnecessary drag and high-frequency vibration feedback in the basic road feel model, thereby reducing the output target road feel torque. This lightweight torque output significantly reduces the burden on the driver's hand muscles, making steering wheel operation lighter and smoother, perfectly matching the pursuit of driving comfort and relaxation in scenarios such as daily commuting or long-distance cruising.

[0092] In some embodiments of this application, the logic for determining the style adaptive correction coefficient in step S320 is as follows: when the identified driving style coefficient indicates that the driver's operation tends to be aggressive, the correction coefficient is increased to increase the road feel torque. When the identified driving style coefficient indicates that the driver's operation tends to be mild, the correction coefficient is decreased to reduce the road feel torque.

[0093] Specifically, when the identified driving style coefficient indicates that the driver's operation is becoming more aggressive, the system executes the logic of increasing the correction coefficient. Its core function is to meet the stringent demands of aggressive drivers for extreme vehicle control and road information perception by increasing the magnitude of the road feel torque and the response gradient. Aggressive driving style is usually accompanied by frequent steering wheel corrections, rapid steering angles, and large lateral accelerations. Under such conditions, drivers urgently need to obtain clear tire grip feedback and vehicle dynamic boundary information through the steering wheel.

[0094] Therefore, increasing the correction coefficient is equivalent to increasing the force feedback gain of the steering system, resulting in a heavier and more solid feel from the road feel simulator. This increased torque not only effectively suppresses vehicle posture oscillations caused by large-scale, high-frequency driver operations, improving directional stability during high-speed driving or cornering, but also simulates the physical characteristics of tires under extreme conditions through enhanced damping, providing the driver with clear warnings and confidence. This prevents misjudgments of vehicle speed and steering angle due to insufficient road feel, thus ensuring the accuracy and safety of human-machine interaction during aggressive driving.

[0095] When the identified driving style coefficient indicates that the driver's operation is becoming more moderate, the system executes a logic to reduce the correction coefficient. The key significance of this is to create a relaxed, comfortable, and low-load driving environment for moderate drivers by reducing the resistance and sensitivity of the road feel torque. Moderate driving style is usually characterized by smooth steering operation and low correction frequency. The driver pays more attention to the smoothness and effortlessness of driving rather than extreme handling feedback.

[0096] At this point, reducing the correction factor is equivalent to introducing power assist gain or torque smoothing filtering. The system actively reduces unnecessary high-frequency road surface interference feedback, making the steering wheel feel lighter and smoother. This adjustment strategy effectively reduces driver hand muscle fatigue during long drives or in congested urban traffic, avoiding tension caused by excessive road feel or abrupt feedback. At the same time, reducing torque does not mean sacrificing safety; rather, by optimizing the torque curve, it better matches the expected operating force of a mild-mannered driver, ensuring easy one-handed control during low-speed maneuvering or fine-tuning the steering, thus achieving a human-centered, comfortable driving experience.

[0097] In some embodiments of this application, the method further includes step S500, a closed-loop evaluation feedback step, specifically including: Step S510: Collect subjective evaluation indicators from multiple groups of drivers with different driving styles in multiple preset test scenarios.

[0098] In step S510, an objective evaluation benchmark based on real human perception is established, transforming the abstract driving experience into quantifiable data support, thereby verifying the effectiveness of the algorithm in a real physical environment. Traditional engineering development often relies excessively on physical model parameters, neglecting the psychological expectations and subjective feelings of the driver as the end user.

[0099] This step involves collecting subjective evaluation indicators from multiple groups of drivers with different driving styles across various pre-set test scenarios (such as high-speed straight driving, emergency lane changing, and low-speed parking), constructing a comprehensive evaluation database covering different groups and driving conditions. These subjective evaluation indicators typically include dimensions such as the appropriateness of steering feel, the richness of road information, the smoothness of return to center, and overall driving confidence.

[0100] By incorporating subjective human evaluation, the system can identify subtle experiential differences that cannot be captured by vehicle dynamics formulas alone, such as whether the nonlinearity of torque establishment is natural and whether there is a sense of play in the center zone. This step not only provides a "gold standard" for subsequent algorithm optimization but also ensures that the final road feel control strategy not only conforms to physical laws but also to the physiological and psychological habits of human drivers, which is a prerequisite for achieving harmonious and unified human-machine co-driving.

[0101] Step S520: Evaluate the current road feel parameters based on subjective evaluation indicators. If the evaluation results do not meet the design threshold, trigger the optimization process of basic road feel parameters and FCM cluster centers, and return to S100 for closed-loop iteration until the design requirements are met.

[0102] In step S520, a reverse optimization closed loop from subjective evaluation to objective parameter correction is constructed to ensure that the road feel simulation system has the ability to continuously iterate. This step first comprehensively evaluates the current road feel parameters based on the subjective evaluation indicators collected in step S510 to determine whether they meet the preset design thresholds. If the evaluation results show that drivers generally report that the feel is too heavy, too light, or vague, the system will determine that the current parameters do not meet the standards, thereby triggering the optimization process of basic road feel parameters and fuzzy C-means clustering centers.

[0103] The optimization here has a dual significance: on the one hand, it fine-tunes the physical parameters such as gain and damping in the basic road feel model to correct the baseline characteristics of torque output; on the other hand, it recalibrates the cluster centers of the fuzzy C-means clustering algorithm, which means that the system will correct the definition boundaries of "aggressive" or "mild" driving styles based on new feedback data, making style identification more accurate.

[0104] Subsequently, the system returns to step S100 for closed-loop iteration, re-collecting data, identifying the style, and calculating torque. Through this continuous iterative process, the system gradually converges to the optimal solution, eliminating the deviation between the model and the real-vehicle experience, and ultimately achieving precise and personalized road feel feedback. This ensures that what is delivered to the user is a highly mature and thoroughly tested intelligent steering control system.

[0105] In some embodiments of this application, reference is made to Figure 2 This flowchart illustrates a method for adjusting and optimizing steer-by-wire road feel based on a driving simulator. Its core logic is to achieve accurate calibration of the road feel model through closed-loop iteration. The process begins with initializing basic road feel parameters, followed by data acquisition to obtain driver operation and vehicle status signals, and then calculating driving style coefficients and adaptive road feel. Based on this, the system organizes multiple drivers to conduct subjective evaluations across multiple scenarios to verify the actual experience effect of the current road feel model. If the evaluation results meet the design requirements, the final road feel model is directly output; otherwise, the step of optimizing road feel parameters is triggered, and the corrected parameters are fed back to the data acquisition stage for recalculation and verification until an optimal road feel control strategy that meets expectations is formed.

[0106] In some embodiments of this application, a complete flow of a steer-by-wire adaptive road feel dynamic adjustment algorithm based on a driving simulator is provided. This algorithm dynamically matches driver operating characteristics with driving scenarios to improve the handling and comfort of steer-by-wire. The specific technical solution is as follows: First, data acquisition is based on the driving simulator. This step aims to build a comprehensive data perception system, covering three dimensions: driver operation characteristic data, driving scenario data, and steer-by-wire state data. Among these, driver operation characteristic data includes steering wheel angle. angular velocity of steering angular acceleration The driver applies torque. and steering operation frequency Driving scenario data includes vehicle speed. Lateral acceleration and road surface adhesion coefficient This coefficient is calculated in real time based on the simulator's preset scenario and the tire model; the steer-by-wire state data involves the output torque of the steering motor. rack displacement and steering response delay , here Specifically, it refers to the time difference between the driver's operation and the rack response, which needs to be calculated and fed back in real time to ensure the system's synchronization.

[0107] Secondly, a real-time driving style identification algorithm is executed. This algorithm achieves accurate quantification of driving style through five steps. The first step is feature normalization, which involves normalizing the statistical period... The samples within are processed, and the calculation formula is as follows: , , .

[0108] The second step is to construct the operational feature vector. The third step uses the fuzzy C-means clustering algorithm, setting the number of clusters. and fuzzy coefficient Its objective function is ,in Representing the The sample belongs to the first The probability of class style, and the sum of the membership degrees of a single sample to all classes is 1, cluster center. ( These correspond to the mild, normal, and aggressive types, respectively.

[0109] The fourth step is iterative optimization, updating the membership degree. and cluster center When the convergence condition is met. The iteration may terminate when it reaches 50 iterations.

[0110] The final step is to calculate the driving style coefficient. ,in , , They are respectively classified as mild, normal, and aggressive membership degrees, and satisfy the following conditions: .

[0111] Furthermore, the dynamic adjustment algorithm for road feel parameters is executed. This module is responsible for translating the identified style into specific torque feedback. The basic road feel model is defined as follows: In the formula For speed-related gain, For lateral acceleration-related damping, This is the road surface adhesion-related self-correction coefficient. The final output is the road feel torque. Adjustments are made by introducing a style adaptive correction coefficient, the formula being: The calculation logic for the style adaptive correction coefficient is as follows: , And another coefficient is also set to .

[0112] In some embodiments of this application, a specific process for adjusting the road feel of steer-by-wire based on a driving simulator is provided, which includes the following five key steps: Step 1: Initialize Parameters. The system first starts the driving simulator and sets the initial road sense parameters based on multi-driver sample data. These parameters mainly include the gain coefficient related to vehicle speed. Damping coefficient related to lateral acceleration The symmetry coefficient related to the road surface adhesion coefficient And the cluster centers required by the fuzzy C-means clustering algorithm, laying the foundation for subsequent real-time calculations.

[0113] Step two involves data acquisition and preprocessing. In the virtual scenario, the driver performs real-time driving operations. The system simultaneously acquires the driver's operational characteristics, current driving scenario data, and the status data of the steer-by-wire system. The acquired raw signals immediately enter the processing stage for filtering, noise reduction, and synchronization alignment to eliminate the impact of sensor noise and time delay on the accuracy of subsequent algorithms.

[0114] Step 3: Real-time driving style identification. The system is set to calculate the operation feature vector F every 50 milliseconds. This vector is then used to perform a fuzzy C-means clustering algorithm to calculate the driving style coefficient representing the current driver's operating habits. To ensure the stability of the control output, the system also uses a first-order low-pass filter to smooth the coefficient, preventing high-frequency fluctuations in the road feel torque caused by sudden data changes.

[0115] Step four: Calculate road feel torque in real time. Then, apply the smoothed driving style coefficient... Substitute the values ​​into the preset road feel adjustment model. The system, combined with the current vehicle state parameters, calculates the final target road feel torque in real time. It also uses a force feedback motor to act on the steering wheel, allowing the driver to instantly feel the steering feel that matches their driving style.

[0116] Step 5: Conduct subjective evaluation and closed-loop iteration. In various typical test scenarios, including highways, low-friction surfaces, and emergency obstacle avoidance, organize at least 30 drivers with different driving styles to conduct subjective scoring tests. Scoring is based on a 1-10 scale, and evaluation dimensions include steering weight, damping feel, self-centering performance, and overall comfort. If the scoring results meet the design requirements, the final road feel model is output; if not, the basic road feel parameters are optimized and adjusted, and the process returns to the data acquisition step to restart iterative verification until the optimal control strategy is obtained.

[0117] Secondly, refer to Figure 3 This application provides a steer-by-wire road feel adjustment system based on a driving simulator, including the following hardware modules: a data acquisition unit, a central processing unit, and an actuator.

[0118] In some embodiments of this application, the data acquisition unit is mounted on a driving simulator to collect driver operation characteristics and vehicle status signals. This unit uses a high-precision sensor array to monitor the driver's torque input to the steering wheel, the rate of change of steering angle, and the frequency of pedal operation in real time, while simultaneously acquiring key vehicle dynamics signals such as vehicle speed, lateral acceleration, and yaw rate.

[0119] These raw data are not only the foundation for subsequent algorithm analysis, but also the only basis for the system to understand the current driving scenario and the driver's intentions. The real-time and accuracy of their collection directly determine the response speed and recognition accuracy of the entire road feel adjustment system, ensuring that the system can perceive the changing trend of the driver's style at the first moment.

[0120] In some embodiments of this application, the central processing unit is used to execute the steer-by-wire road feel adjustment method as described above, specifically including a style identification unit and a torque calculation unit. The style identification unit utilizes a fuzzy C-means clustering algorithm to perform deep mining on the massive amount of collected feature data, quantifying abstract driving behaviors into specific driving style coefficients.

[0121] The torque calculation unit dynamically adjusts the basic road feel model based on this coefficient, integrating vehicle speed gain and road feedback information to calculate the most suitable target torque value under the current operating conditions. This unit achieves a leap from standardized to personalized road feel feedback, ensuring that the output control commands conform to both the vehicle's dynamic limits and the driver's subjective operating habits.

[0122] In some embodiments of this application, the actuator includes a feel simulator motor disposed on the steering wheel assembly for outputting a target road feel torque according to instructions from the central processing unit.

[0123] The actuator no longer relies on the mechanical connection between the wheels and the ground in traditional steering systems to provide feedback. Instead, it uses high-precision current control of the electric motor to reproduce the road surface's frictional resistance, tire lateral slip characteristics, and the vehicle's self-centering tendency in real time. When the central processing unit determines that the driver is in an aggressive driving style, the motor rapidly increases damping torque to provide a stable handling feel; conversely, it outputs a gentler torque to enhance comfort. The dynamic response characteristics of the actuator directly determine the realism and subtlety of the road feel, and are a key hardware guarantee for realizing the human-machine interaction experience of the steer-by-wire system.

[0124] Furthermore, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the aforementioned method for adjusting the steer-by-wire road feel based on a driving simulator.

[0125] Furthermore, embodiments of this application provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned steer-by-wire road feel adjustment method based on a driving simulator.

[0126] In summary, the steer-by-wire road feel adjustment method and related equipment based on a driving simulator provided in this application have the following technical effects.

[0127] First, this application overcomes the limitations of traditional steer-by-wire systems, which rely on fixed road feel parameters, resulting in a monotonous feel and an inability to cater to the diverse needs of different drivers. By introducing a driving style recognition algorithm based on fuzzy C-means clustering, the system can quantify the driver's operating habits in real time and dynamically adjust the weight and gain of the road feel torque accordingly. For aggressive driving, the system automatically enhances steering damping and road feedback, providing a clear sense of limits; for mild driving, the system smooths out torque fluctuations, creating a light and comfortable driving atmosphere. This personalized adjustment strategy ensures that the steer-by-wire system provides optimal steering feel under different driving styles, enhancing driver confidence and driving pleasure.

[0128] Secondly, this application innovatively places the development and verification of the road feel adjustment algorithm in a driving simulator environment, using multi-dimensional subjective evaluation indicators to reverse-correct the basic road feel parameters and cluster centers. This development model, combining virtual simulation and subjective evaluation, not only avoids the reliance on expensive real-vehicle resources for early algorithm verification and eliminates potential safety risks from real-vehicle testing, but also allows engineers to safely explore the boundaries of road feel under extreme conditions. Through a rapid iterative closed-loop feedback mechanism, the system can quickly converge to the optimal parameter solution that conforms to human perception, providing an efficient and low-cost engineering path for the rapid implementation of steer-by-wire technology.

[0129] Third, this application successfully reconstructs the force communication link between the driver and the road surface by constructing a basic road feel model that includes vehicle speed gain, lateral acceleration damping, and road surface adhesion self-centering coefficient, and by using a hand-feel simulator motor for high-precision torque output, without eliminating mechanical connections. The system not only simulates the self-centering characteristics and friction feel of a traditional steering system, but also compensates for the torque discontinuity problem unique to steer-by-wire systems through algorithms, achieving a smooth transition from light and easy handling at low speeds to stable handling at high speeds. This highly realistic road feel simulation technology retains the advantages of steer-by-wire in terms of layout flexibility and autonomous driving cooperation, while eliminating the psychological panic experienced by the driver due to a lack of road information, achieving a perfect balance between safety and driving feel.

[0130] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.

[0131] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0132] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of ordinary skill of an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary skill. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0133] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0135] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or, if necessary, processing in a suitable manner, and then stored in computer memory.

[0136] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0137] In the foregoing description of this specification, the reference to terms such as "one embodiment / implementation," "another embodiment / implementation," or "certain embodiments / implementations," etc., indicates that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in an embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0138] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0139] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for adjusting the road feel of steer-by-wire based on a driving simulator, characterized in that, Includes the following steps: S100: Collects driver operation characteristic data and vehicle driving scene data in real time in the driving simulator; S200: The operation feature data is processed using the fuzzy C-means FCM clustering algorithm to identify the driver's driving style type and output the corresponding driving style coefficient. S300: Substitute the driving style coefficient into the road feel parameter dynamic adjustment algorithm, and combine it with the driving scenario data to calculate the target road feel torque in real time; S400: Convert the target road feel torque into a motor control signal and output it to the steering wheel feel simulator motor in the steer-by-wire system to drive the motor to output the corresponding feedback torque.

2. The steer-by-wire road feel adjustment method based on a driving simulator according to claim 1, characterized in that, In S100, the driver's operation characteristic data includes: steering wheel angle, steering angular velocity, steering angular acceleration, driver applied torque, and steering operation frequency; The driving scenario data includes: vehicle speed, lateral acceleration, and road surface adhesion coefficient.

3. The steer-by-wire road feel adjustment method based on a driving simulator according to claim 1, characterized in that, In step S200, the driver's driving style is identified using the fuzzy C-means FCM clustering algorithm. Specific steps include: The collected driver operation feature data is normalized and an operation feature vector is constructed. Perform fuzzy C-means FCM clustering to cluster the driver's steering operation feature vector into multiple preset driving style categories, and output the continuous membership degree of each sample to each category; The driving style coefficient is calculated based on the continuous membership degree.

4. The steer-by-wire road feel adjustment method based on a driving simulator according to claim 3, characterized in that, The calculation of the driving style coefficient based on the continuous membership degree is specifically as follows: The membership degree of each category is weighted and calculated with the preset style weights; The execution of fuzzy C-means FCM clustering specifically includes setting an objective function and iteratively updating the membership matrix and cluster centers until the objective function converges or reaches the preset maximum number of iterations.

5. The steer-by-wire road feel adjustment method based on a driving simulator according to claim 1, characterized in that, In S300, the real-time calculation of the target road feel torque is specifically as follows: Construct a basic road feel model that includes vehicle speed gain, lateral acceleration damping, and road surface adhesion self-correction coefficient; Introduce a style adaptive correction coefficient based on the driving style coefficient; The base torque calculated from the basic road feel model is weighted and calculated with the style adaptive correction coefficient to obtain the final target road feel torque.

6. The steer-by-wire road feel adjustment method based on a driving simulator according to claim 5, characterized in that, The logic for determining the style adaptive correction coefficient is as follows: Obtain a preset first style threshold and a second style threshold, wherein the first style threshold is greater than the second style threshold; When the identified driving style coefficient is greater than or equal to the first style threshold, it is determined that the driver's operation is in the first preset style range, and the style adaptive correction coefficient is increased according to the first gain ratio to enhance the output target road feel torque. When the identified driving style coefficient is less than or equal to the second style threshold, it is determined that the driver's operation is in the second preset style range, and the style adaptive correction coefficient is adjusted down according to the second attenuation ratio to reduce the output target road feel torque.

7. The method for adjusting steer-by-wire road feel based on a driving simulator according to claim 1, characterized in that, Also includes: Subjective evaluation indicators of multiple groups of drivers with different driving styles were collected in various preset test scenarios. The current road feel parameters are evaluated based on the subjective evaluation index. If the evaluation result does not meet the design threshold, the optimization process of the basic road feel parameters and FCM cluster centers is triggered, and the process returns to S100 for closed-loop iteration until the design requirements are met.

8. A steer-by-wire road feel adjustment system based on a driving simulator, characterized in that, Includes the following hardware modules: The data acquisition unit, installed on the driving simulator, is used to collect driver operation characteristics and vehicle status signals; The central processing unit is used to execute the steer-by-wire road feel adjustment method as described in any one of claims 1 to 7, specifically including a style recognition unit and a torque calculation unit; The actuator includes a feel simulator motor mounted on the steering wheel assembly, used to output a target road feel torque according to instructions from the central processing unit.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steer-by-wire road feel adjustment method based on any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the steer-by-wire road feel adjustment method based on a driving simulator as described in any one of claims 1 to 7.

Images