Methods, devices, computer equipment, media, and products for simulating road feel in steer-by-wire
By receiving driving mode switching commands in the steer-by-wire system, acquiring vehicle parameters, and generating dynamic transition weight factors for interpolation, the problem of torque discontinuity during mode switching in the steer-by-wire system is solved, achieving a seamless driving mode switching experience and improved comfort.
Patent Information
- Application Number
- CN202611504618.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-24
- Publication Date
- 2026-11-17
AI Technical Summary
Existing steer-by-wire systems cause non-linear step changes in the steering wheel when switching driving modes because they ignore the continuity of underlying physical parameters, which affects driving comfort.
By receiving driving mode switching commands, the system obtains the vehicle's longitudinal speed and steering characteristic parameters, generates a dynamic transition weight factor, and interpolates the mode feature gain vector based on this factor to generate the execution mode feature gain vector. Finally, it calculates the target feedback torque to drive the road feel motor, ensuring the continuity and smoothness of the torque.
It achieves a seamless experience during driving mode switching, improves driving comfort, and eliminates the problem of sudden torque changes.
Smart Images

Figure CN123104768A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of automotive control technology, and in particular to a method, apparatus, computer device, computer-readable storage medium, and computer program product for simulating road feel during steer-by-wire. Background Technology
[0002] With the continuous improvement of automotive electronics, steer-by-wire systems have eliminated the mechanical connection between the steering wheel and the steering actuator. Road feel feedback relies entirely on the road feel simulation feedback motor in the upper steering column to generate the feedback. This allows the steering feel to be freely defined at the software level, providing a hardware foundation for achieving personalized, multi-mode driving feel.
[0003] In existing technologies, when switching modes, the "result-end torque synthesis" method is usually adopted. That is, two independent algorithms calculate the final target torque of the two modes, and then the two final target torque values are directly weighted and superimposed, for example, by using a fixed time linear gradient factor for smooth transition.
[0004] However, the aforementioned existing technologies ignore the dynamic continuity of the underlying physical parameters of the steering system (such as stiffness and damping). When the torque values corresponding to the two modes differ significantly, simple linear weighting cannot guarantee that the resultant torque at any instant during the transition process conforms to the natural physical feel change law. When switching modes, it is easy to cause nonlinear step changes in the steering wheel, resulting in an abrupt feeling when the driver operates the steering wheel and poor driving comfort. Summary of the Invention
[0005] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for steer-by-wire road feel simulation that can seamlessly switch driving modes and effectively improve driving comfort, in response to the aforementioned technical problems.
[0006] In a first aspect, this application provides a method for simulating road feel during steer-by-wire control, the method comprising:
[0007] Receive a driving mode switching command for the target vehicle; wherein the driving mode switching command is used to switch the target vehicle from a first driving mode to a second driving mode;
[0008] Obtain the longitudinal speed and steering characteristic parameters of the target vehicle at the current moment;
[0009] Generate a transition weight factor for the current moment; wherein the transition weight factor dynamically changes over time within a preset switching period starting from the current moment, and the value range of the transition weight factor is [0,1].
[0010] Obtain the first mode feature gain vector matching the first driving mode and the second mode feature gain vector matching the second driving mode, and perform interpolation processing on the first mode feature gain vector and the second mode feature gain vector based on the transition weight factor to obtain the execution mode feature gain vector at the current time.
[0011] The target feedback torque at the current moment is obtained based on the execution mode feature gain vector, the longitudinal vehicle speed, and the steering feature parameters;
[0012] The road sensor motor of the target vehicle is driven to operate based on the target feedback torque.
[0013] In one embodiment, before receiving the driving mode switching command for the target vehicle, the method further includes:
[0014] For various driving modes, a set of multi-dimensional mode feature gain vectors are pre-calibrated; wherein, for any driving mode, the mode feature gain vector includes multiple gain coefficients corresponding to each component torque module in a preset general reference mapping table.
[0015] In one embodiment, the step of pre-calibrating a set of multi-dimensional mode feature gain vectors for each of the multiple driving modes includes:
[0016] The static physical parameters of the target vehicle are obtained, and objective evaluation indicators for characterizing driving feel are set based on the static physical parameters; wherein, the objective evaluation indicators include the average steering stiffness in the center area, steering wheel hysteresis, and target damping ratio;
[0017] For at least two driving modes, set target values for objective evaluation indicators for each driving mode;
[0018] For any driving mode, the target value of its objective evaluation index is used as the offline calibration target, and a set of multi-dimensional mode feature gain vectors is calculated by inversion.
[0019] The mode feature gain vectors matched with each driving mode are stored in a fixed form to form a mode feature gain matrix.
[0020] In one embodiment, the component torque module includes a main torque module, an active self-aligning torque module, a damping compensation torque module, and a friction simulation torque module; the multiple gain coefficients in the mode feature gain vector include: the basic strength gain coefficient and the rotation angle nonlinearity gain coefficient corresponding to the main torque module, the proportional gain coefficient, integral gain coefficient, and differential gain coefficient corresponding to the active self-aligning torque module, the damping strength gain coefficient corresponding to the damping compensation torque module, and the friction amplitude gain coefficient corresponding to the friction simulation torque module.
[0021] In one embodiment, the transition weight factor is generated based on the steering feature parameters; generating the transition weight factor at the current moment includes:
[0022] The steering intensity safety penalty coefficient for the current moment is determined based on the steering characteristic parameters; wherein, when the absolute value of the steering characteristic parameter is less than or equal to the minimum steering characteristic threshold, the steering intensity safety penalty coefficient is 1; when the absolute value of the steering characteristic parameter is greater than the minimum steering characteristic threshold and less than the maximum steering characteristic threshold, the steering intensity safety penalty coefficient decreases linearly as the absolute value of the steering characteristic parameter increases, and the value range of the steering intensity safety penalty coefficient is (0,1); when the absolute value of the steering characteristic parameter is greater than or equal to the maximum steering characteristic threshold, the steering intensity safety penalty coefficient is 0.
[0023] Based on the steering strength safety penalty coefficient and the switching cycle, a smooth transition process is performed to obtain the transition weight factor at the current moment.
[0024] In one embodiment, obtaining the target feedback torque at the current moment based on the execution mode feature gain vector, the longitudinal vehicle speed, and the steering feature parameters includes:
[0025] Each gain coefficient in the execution mode feature gain vector is used as a scaling factor and applied to the corresponding component torque module in the preset general reference mapping table. The longitudinal vehicle speed and the steering feature parameters are then input into each component torque module.
[0026] The corresponding component torques are obtained from each component torque module, and the component torques are superimposed to obtain the target feedback torque at the current moment; wherein, the general reference mapping table represents the physical characteristic mapping relationship that does not change with the driving mode, and the component torque modules in the general reference mapping table include the main torque module, the active self-centering torque module, the damping compensation torque module, and the friction simulation torque module.
[0027] Secondly, this application also provides a steer-by-wire road feel simulation device, the device comprising:
[0028] A mode switching sensing unit is used to receive a driving mode switching command from a target vehicle; wherein the driving mode switching command is used to switch the target vehicle from a first driving mode to a second driving mode.
[0029] The vehicle monitoring unit is used to acquire the longitudinal speed and steering characteristic parameters of the target vehicle at the current moment.
[0030] A dynamic weight generation unit is used to generate a transition weight factor at the current moment; wherein the transition weight factor changes dynamically with time within a preset switching period starting from the current moment, and the value range of the transition weight factor is [0,1].
[0031] The gain interpolation unit is used to obtain a first mode feature gain vector matching the first driving mode and a second mode feature gain vector matching the second driving mode, and to perform interpolation processing on the first mode feature gain vector and the second mode feature gain vector based on the transition weight factor to obtain the execution mode feature gain vector at the current time.
[0032] The torque synthesis execution unit is used to obtain the target feedback torque at the current moment based on the execution mode feature gain vector, the longitudinal vehicle speed, and the steering feature parameters.
[0033] The vehicle control unit is used to drive the road sensor motor of the target vehicle to operate based on the target feedback torque.
[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the preceding claims.
[0035] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the preceding claims.
[0036] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the preceding claims.
[0037] The aforementioned steer-by-wire road feel simulation method, device, computer equipment, computer-readable storage medium, and computer program product, upon receiving a mode switching command, do not perform a weighted average of the torque calculated from the two modes. Instead, based on a dynamically changing transition weight factor, they interpolate the multi-dimensional mode feature gain vector representing the feel characteristics to generate an intermediate gain vector that fuses the characteristics of the two modes before and after the switch. Subsequently, the target feedback torque is synthesized from this continuously changing intermediate gain vector, and the road feel motor of the target vehicle is driven based on the synthesized target feedback torque. During this process, because the object of dynamic interpolation during mode switching is the underlying gain coefficient describing physical characteristics, the differential dynamic equation of the road feel control law is self-consistent and continuous within any small time slice, resulting in a physically smooth final synthesized torque. Based on this, this application can eliminate the problem of torque step jumps at the source of the physical mechanism, making the driving mode switching process completely imperceptible to the driver, effectively improving driving comfort. Attached Figure Description
[0038] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0039] Figure 1 This is an application environment diagram of the steer-by-wire road feel simulation method in one embodiment;
[0040] Figure 2 This is a flowchart illustrating a road feel simulation method for steer-by-wire in one embodiment;
[0041] Figure 3 This is a schematic diagram of the component torque module in one embodiment;
[0042] Figure 4 This is a structural block diagram of a steer-by-wire road feel simulation device in one embodiment;
[0043] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation
[0044] 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.
[0045] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.
[0046] The steer-by-wire road feel simulation method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, the road sensor motor controller 101 communicates with the electronic control unit (ECU) 102 via an in-vehicle communication bus (such as CAN (Controller Area Network, a serial communication protocol) bus, in-vehicle Ethernet, etc.). A data storage module can store the data that the ECU 102 needs to process. The data storage module can be integrated into the ECU 102 or placed on a cloud or server connected to the ECU 102. The ECU 102 is deployed inside the target vehicle. It is responsible for receiving driving mode switching commands from the target vehicle in real time to switch the target vehicle from a first driving mode to a second driving mode, acquiring the target vehicle's longitudinal speed and steering characteristic parameters, generating a transition weight factor, and interpolating the first mode feature gain vector matching the first driving mode and the second mode feature gain vector matching the second driving mode based on the transition weight factor to obtain the execution mode feature gain vector. Then, it calculates the target feedback torque by combining the longitudinal speed and steering characteristic parameters, so as to drive the road sensor motor through the road sensor motor controller 101. The mode feature gain vectors and universal reference mapping table for various driving modes can be pre-stored in the data storage module, or remotely calibrated, optimized, and updated via a server. The electronic control unit 102 can be, but is not limited to, various intelligent in-vehicle devices, autonomous driving controllers, in-vehicle infotainment terminals, and domain controllers, etc., and is not limited here. The electronic control unit 102 may include a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steer-by-wire road feel simulation method provided in this embodiment.
[0047] In one exemplary embodiment, such as Figure 2 As shown, a method for simulating road feel in steer-by-wire is provided, which is then applied to... Figure 1 Taking the electronic control unit as an example, the explanation includes the following steps 201 to 206. Wherein:
[0048] Step 201: Receive a driving mode switching command from the target vehicle; wherein the driving mode switching command is used to switch the target vehicle from the first driving mode to the second driving mode.
[0049] The target vehicle refers to a vehicle equipped with a steer-by-wire system. Driving mode refers to a preset operating mode with different steering feel characteristics. The first driving mode is the driving mode the vehicle is currently executing before the switch occurs, and the second driving mode is the target driving mode the vehicle will execute after the switch occurs. Driving modes can include, but are not limited to, Comfort mode, Standard mode, and Sport mode. Driving mode switching commands can be actively triggered by the driver through the in-vehicle human-machine interface (such as a mode switching button, touchscreen controls, etc.), or automatically triggered by the vehicle's domain controller or autonomous driving controller based on conditions such as vehicle speed, road conditions, and driving style recognition results.
[0050] For example, the electronic control unit continuously listens for mode switching request signals from the in-vehicle network. When a valid mode switching command is received, the command is parsed to determine the current first driving mode and the second driving mode to be switched to, and the subsequent smooth mode switching process is triggered.
[0051] Step 202: Obtain the longitudinal speed and steering characteristic parameters of the target vehicle at the current moment.
[0052] Among them, longitudinal speed refers to the speed at which the vehicle travels along the longitudinal axis of the vehicle body. The electronic control unit can obtain the current longitudinal speed by reading the signal collected by the vehicle speed sensor from the CAN bus.
[0053] Steering characteristic parameters refer to one or more physical quantities used to characterize the driver's current steering operation intensity or the vehicle's steering motion state. In one embodiment, steering characteristic parameters may include steering wheel angular velocity. In another embodiment, steering characteristic parameters may include vehicle yaw rate. In yet another embodiment, steering characteristic parameters may include vehicle lateral acceleration. In still another embodiment, steering characteristic parameters may include rack force change rate. Furthermore, steering characteristic parameters may also employ a weighted combination of the above parameters to more comprehensively reflect the instantaneous steering intensity of the vehicle. In this embodiment, the electronic control unit can acquire the above parameters through corresponding sensors (such as steering angle sensors, yaw rate sensors, acceleration sensors, etc.) or state observers, which are not limited here.
[0054] In this embodiment, steps 201 and 202 can be executed in parallel or sequentially, and there is no restriction here.
[0055] Step 203: Generate the transition weight factor for the current time; wherein, the transition weight factor changes dynamically with time within a preset switching period starting from the current time, and the value range of the transition weight factor is [0,1].
[0056] The transition weighting factor is a dimensionless scalar coefficient that changes continuously within the interval [0,1]. In this embodiment, it is denoted as α(t), where t represents the current time. The switching cycle is used to provide a basic time reference for switching between different driving modes, ensuring that the mode feature gain vector parameters gradually transition according to the preset time gradient during the mode switching process, avoiding problems such as abrupt changes in road feel and sudden steering wheel feel caused by parameter mutations.
[0057] During the switching cycle starting from the moment t0 when the switching command is received, the transition weight factor α(t) changes dynamically with time. Specifically, α(t) gradually transitions from 0 to 1. When α=0, it indicates that the system is still fully executing the first driving mode. When α=1, it indicates that the system has fully transitioned to the second driving mode. Values between 0 and α<1 indicate the transition state between the two driving modes.
[0058] The method for generating the transition weight factor can be found in the detailed description of the following embodiments, in each preset control period T. s Inside, the electronic control unit calculates and updates the value of α(t) based on the vehicle's real-time status and preset switching strategies.
[0059] Step 204: Obtain the first mode feature gain vector matching the first driving mode and the second mode feature gain vector matching the second driving mode, and perform interpolation processing on the first mode feature gain vector and the second mode feature gain vector based on the transition weight factor to obtain the execution mode feature gain vector at the current time.
[0060] The mode feature gain vector is a set of independent, multi-dimensional gain coefficients acting on each component torque module at the bottom layer of the road feel simulation control. It is used to determine the value of each component torque, and thus determine the final target feedback torque. In this embodiment, each driving mode is pre-calibrated with a specific set of mode feature gain vectors, which contain the gain coefficients of each component torque module in that mode.
[0061] For example, the first mode feature gain vector is denoted as M. gain,A The second mode feature gain vector is denoted as M. gain,B They are all column vectors or row vectors containing multiple elements, such as M gain =[G1,G2,G3,...,G n ] T These gain coefficients G1, G2, ..., G nThese correspond to different control modules or characteristic dimensions in the road sense synthesis algorithm.
[0062] In this embodiment of the application, interpolation refers to the process of interpolating M in the vector space. gain,A and M gain,B Perform linear or nonlinear mixing to obtain the instantaneous execution mode feature gain vector M in the current transition state. gain (t), which is the execution mode feature gain vector at the current moment.
[0063] In one embodiment, the execution mode feature gain vector at the current moment is obtained using linear interpolation:
[0064] M gain (t)=[1-α(t)]*M gain,A +α(t)*M gain,B
[0065] Based on this, the interpolation operation is applied to the underlying parameters rather than the final torque value, which is fundamentally different from the existing "result-end torque synthesis" scheme.
[0066] Step 205: Based on the execution mode feature gain vector, longitudinal vehicle speed, and steering feature parameters, obtain the target feedback torque at the current moment.
[0067] Here, the target feedback torque refers to the total feedback torque that the road feel motor is expected to output to the driver's steering wheel. In this embodiment, the electronic control unit uses the execution mode feature gain vector M obtained in step 204 at the current moment. gain (t), along with longitudinal vehicle speed and steering characteristic parameters (such as steering wheel angular velocity, vehicle yaw rate, vehicle lateral acceleration, rack force change rate, etc.) are used as input parameters. Combined with a pre-set universal reference mapping table that does not change with the driving mode, torque synthesis is achieved. During torque synthesis, the target feedback torque T at the current moment is calculated comprehensively. total (t).
[0068] Step 206: Drive the road sensor motor of the target vehicle to run according to the target feedback torque.
[0069] Specifically, the electronic control unit will calculate the target feedback torque T total (t) After performing necessary amplitude limits, rate of change limits and signal validity judgments, it is converted into a target torque request command (such as a torque request signal) that the road sensor motor can recognize. It is then sent to the road sensor motor controller through the drive circuit or directly through the bus interface, thereby driving the road sensor motor to output the corresponding physical torque, and finally presenting the desired steering feel to the driver on the steering wheel.
[0070] In the aforementioned steer-by-wire road feel simulation method, upon receiving a mode switching command, instead of weighted averaging the torque calculated from the two modes, a dynamically changing transition weight factor is used to interpolate the multi-dimensional mode feature gain vector representing the feel characteristics. This generates an intermediate gain vector that fuses the characteristics of the two modes before and after the switch. Subsequently, this continuously changing intermediate gain vector is used to synthesize the target feedback torque, and the synthesized target feedback torque drives the target vehicle's road feel motor. During this process, because the object of dynamic interpolation during mode switching is the underlying gain coefficient describing physical characteristics (such as stiffness and damping), the differential dynamic equation of the road feel control law is self-consistent and continuous within any small time slice, resulting in a physically smooth final synthesized torque. Based on this, this application can eliminate the problem of torque step jumps at the source of the physical mechanism, making the driving mode switching process completely imperceptible to the driver and effectively improving driving comfort.
[0071] In an exemplary embodiment, before receiving the driving mode switching command for the target vehicle, the method further includes:
[0072] Step 200: For multiple driving modes, a set of multi-dimensional mode feature gain vectors are pre-calibrated; wherein, for any driving mode, the mode feature gain vector includes multiple gain coefficients corresponding to each component torque module in the preset general reference mapping table.
[0073] The various driving modes may include, but are not limited to, comfort mode, standard mode, and sport mode. In this embodiment, during the offline development and debugging phase of the vehicle system, the optimal parameter combination for each driving mode is determined through simulation or real-vehicle testing, thereby achieving pre-calibration of the mode feature gain vector.
[0074] A universal reference mapping table refers to a set of physical characteristic mapping relationships that are pre-obtained through extensive real-vehicle or simulation calibration, used to describe the basic physical and dynamic characteristics of the steering system and do not change with driving mode switching. It can also be called a multi-dimensional lookup table or a set of functions. In this embodiment, the universal reference mapping table defines the mapping relationships between physical quantities such as torque, vehicle speed, steering angle, and angular velocity that are common to all driving modes, such as basic road feel curves, basic damping characteristic curves, and basic friction characteristic curves. In this embodiment, after successful verification, the universal reference mapping table and the mode characteristic gain vector are jointly stored in the non-volatile memory of the electronic control unit for real-time retrieval during vehicle operation.
[0075] The component torque module is an independent algorithm unit obtained by decomposing the total target road feel torque according to its physical source and mechanism of action. In one embodiment, each component torque module includes at least a main torque module, an active self-centering torque module (referred to as the active self-centering module), a damping compensation torque module (referred to as the damping module), and a friction simulation torque module (referred to as the friction simulation module). During implementation, each component torque module can independently calculate a component torque, and all components are superimposed to form the total target feedback torque.
[0076] The gain coefficient is a specific element in the mode feature gain vector. Each element corresponds to a specific control term in a torque module. The gain coefficient is used to scale the torque components output by the torque module in the general reference mapping table, thereby flexibly adjusting the feel characteristics of each mode without changing the basic mapping relationship.
[0077] For example, a general benchmark mapping table can be used with the function Map base It is represented by the torque component, and its output is a component torque.
[0078] In this embodiment, the adjustment of multi-mode feel is transformed into adjusting only a set of low-dimensional gain coefficients, while the universal reference mapping table remains unchanged. This allows the final feedback torque to be changed by adjusting the mode feature gain vector when switching driving modes, ensuring the consistency of the most basic vehicle physical characteristics between different modes and avoiding a sense of disconnect between different modes.
[0079] In an exemplary embodiment, for multiple driving modes, a set of multi-dimensional mode feature gain vectors are pre-calibrated, including:
[0080] a1. Obtain the static physical parameters of the target vehicle, and set objective evaluation indicators to characterize the driving feel based on the static physical parameters; among them, the objective evaluation indicators include the average steering stiffness in the center area (referred to as center area stiffness), steering wheel hysteresis (referred to as hysteresis ring width), and target damping ratio;
[0081] The static physical parameters of the target vehicle refer to the basic chassis parameters determined during the design and manufacturing phases, whose values do not change with the vehicle's motion. For example, static physical parameters may include, but are not limited to, curb weight (m), wheelbase (L), and distance from the center of gravity to the front axle (l). f ), distance from center of mass to rear axle (l) r ) and steering system angular ratio (i s These parameters can be obtained by reading the vehicle configuration file or by manual input by the developers. By obtaining static physical parameters, subsequent objective evaluation indicators can be linked to the specific physical characteristics of the vehicle, ensuring the realism and accuracy of the road feel simulation.
[0082] Objective evaluation indicators transform the driver's subjective feelings of "light / heavy", "sensitive / sluggish", "stable / drifting" into objective engineering parameters that can be physically measured and mathematically calculated.
[0083] For example, the embodiments of this application employ at least the following three objective evaluation indicators: average steering stiffness of the target center area, target steering wheel hysteresis (hysteresis loop width), and target damping ratio.
[0084] Among them, the average steering stiffness of the target center area is the average slope of the steering wheel torque as the steering angle changes when the steering wheel angle is within the small angle center area (e.g., ±5° to ±10°). It directly determines the feel of the steering wheel when the driver is driving straight or making slight corrections to the direction.
[0085] The target steering wheel hysteresis is the characteristic width of the hysteresis loop formed by the difference in steering angle under the same torque, or the difference in torque under the same steering angle, when slowly and repeatedly turning the steering wheel. It primarily simulates the internal friction characteristics of the steering system, reflecting the system's ability to filter high-frequency road interference and the feel of mechanical friction, thus affecting the clarity of road feel and the filtering effect on high-frequency vibrations. In actual physical systems, the conversion from static friction to kinetic friction causes the hysteresis loop to be wider at low speeds and narrower at high speeds. Therefore, the hysteresis loop can be narrowed or widened depending on the driving mode.
[0086] The target damping ratio is a dimensionless parameter used to describe how quickly the oscillations of the steering wheel decay after it has returned to center or been disturbed. It directly affects the smoothness of the return-to-center process and the stability of the center area at high speeds.
[0087] a2. For at least two driving modes, set target values for objective evaluation indicators under each driving mode;
[0088] In this embodiment of the application, the developers set specific target values for the aforementioned objective evaluation indicators based on the design goals of each mode. For example:
[0089] Sport mode: Set a higher target value for average steering stiffness in the center area (stable feel), a narrower target value for steering wheel angle lag (clear road feel), and a higher target value for damping ratio (quick return to center and strong high-speed stability).
[0090] Comfort mode: Set a lower target value for average steering stiffness in the center area (lighter feel), a wider target value for steering wheel lag (filters out minor bumps), and a moderate target value for damping ratio (smooth return to center).
[0091] Based on this, subjective driving experience requirements can be transformed into actionable technical specifications.
[0092] For example, a target stiffness reference model is established directly based on the vehicle's steady-state dynamics, wherein the average steering stiffness of the target center area is... The mathematical model is as follows:
[0093]
[0094] In the formula, This is the stiffness characteristic coefficient for the driving mode, which can be set to different values depending on the driving mode. The target torque-lateral acceleration gradient is preset based on industry objective evaluation standards; v is the vehicle's longitudinal speed. The steady-state lateral acceleration gain of the vehicle at longitudinal speed v is obtained through a two-degree-of-freedom vehicle dynamics model:
[0095]
[0096] In the formula, L is the wheelbase of the vehicle; K is the inherent understeer gradient of the vehicle, which is determined by the front and rear axle lateral stiffness and axle load distribution.
[0097] During implementation, different stiffness characteristic coefficients are assigned to different driving modes to achieve personalization:
[0098] Sports Mode: Settings This amplifies the torque gradient, making the steering wheel feel heavier and providing more direct road feedback under the same lateral acceleration.
[0099] Comfort Mode: Settings This reduces the torque gradient, resulting in a lighter cruising feel;
[0100] Standard Mode: Settings Maintain basic calibration.
[0101] For example, an exponential decay model based on vehicle speed is used to quantify the target steering wheel angle hysteresis; correspondingly, the target steering wheel angle hysteresis... The mathematical model is as follows:
[0102]
[0103] In the formula, The maximum base hysteresis is set under stationary or extremely low-speed conditions. The minimum basic residual hysteresis at high speed; This is the time constant factor that decays with vehicle speed; This refers to the hysteresis characteristic coefficient for different driving modes. Different hysteresis characteristic coefficients are assigned to different driving modes to achieve personalization.
[0104] Sports Mode: Settings (e.g., 0.7), forcibly narrowing the hysteresis loop width reduces the masking of road feel signals caused by friction simulation, ensuring that the driver can perceive subtle changes in road surface adhesion;
[0105] Comfort Mode: Settings (As in 1.3), widening the hysteresis loop width utilizes the increased physical hysteresis characteristics to passively filter out unnecessary high-frequency micro-vibrations on the road surface (such as steering wheel vibrations caused by gravel roads), thereby improving driving smoothness.
[0106] Target damping ratio The mathematical model is as follows:
[0107]
[0108] In the formula, This represents the vehicle speed-dependent basic damping ratio function without considering differences in driving modes. It describes the target damping level that gradually increases with increasing vehicle speed. This function can be obtained from a one-dimensional lookup table or expressed as a continuous expression:
[0109]
[0110]
[0111] in, and These are the base damping ratios for low speed and high speed, respectively; As a speed transition scale; To determine the lower and upper limits of the permissible damping ratio, during implementation, the target damping ratio under each mode should be constrained to not be lower than the permissible lower limit. This reduces the risk of oscillation or instability caused by insufficient damping; sat() is the amplitude limiting function, also known as the saturation function.
[0112] Different driving modes are assigned different damping characteristic coefficients to achieve personalization:
[0113] Standard mode usually makes Comfort mode and Sport mode can be selected within the specified range according to the requirements of smooth self-centering and high-speed stability.
[0114] a3. For any driving mode, the target value of its objective evaluation index is used as the offline calibration target, and a set of multi-dimensional mode feature gain vectors is calculated by inversion.
[0115] Inverse calculation refers to the process of reversing the objective evaluation index to find the control parameters (i.e., the gain coefficients in the mode feature gain vector) that can achieve the desired performance. For example, iterative optimization is carried out in a vehicle dynamics simulation environment (such as CarSim, CarMaker and Simulink joint simulation platform) or a real vehicle test bench to achieve inverse calculation.
[0116] During implementation, for each driving mode, developers continuously fine-tune the gain coefficients, driving virtual or real vehicles under standard test conditions (such as center-area steering tests and free-centering tests), extracting the actual objective evaluation index values from the simulation or test results, and comparing them with the target values set in step a2 until the error converges to the allowable range. The mode feature gain vector calculated by inversion can meet the target values of each objective evaluation index. Based on this, the blindness of traditional blind tuning methods and the over-reliance on developer experience can be avoided. At the same time, a mapping relationship between subjective feelings and underlying parameters is established, allowing the differences in feel between multiple modes to be quantitatively designed.
[0117] a4. Store the mode feature gain vectors that match each driving mode in a fixed format to form a mode feature gain matrix.
[0118] In this embodiment, the multiple sets of mode feature gain vectors, which are verified in step a3 and correspond one-to-one with each driving mode, are integrated in a certain order to form a multi-dimensional matrix, i.e., the mode feature gain matrix. This matrix can be used as a standard parameter library and can be burned into or written into the non-volatile memory of the electronic control unit of the steer-by-wire system for real-time retrieval during vehicle operation.
[0119] This embodiment discloses a calibration scheme for mode feature gain vectors. By acquiring the static physical parameters of the target vehicle and setting objective evaluation indicators to characterize driving feel based on these parameters, target values for these objective evaluation indicators are set for each driving mode. Subsequently, the target values of the objective evaluation indicators for each driving mode are used as offline calibration targets to inversely calculate a set of multi-dimensional mode feature gain vectors. Finally, the obtained mode feature gain vectors are solidified into a mode feature gain matrix. Based on this, calibration results from different vehicle models and developers can achieve high consistency and comparability, improving calibration efficiency and accuracy. Simultaneously, it allows enterprises to accumulate a reusable feel calibration database, possessing industrial application value.
[0120] In one exemplary embodiment, such as Figure 3 As shown, the torque module includes a main torque module, an active return torque module, a damping compensation torque module, and a friction simulation torque module, which can output the main torque, active return torque, damping compensation torque, and friction simulation torque respectively.
[0121] The main torque module is used to simulate the return torque transmitted from the lateral force of the tires to the steering wheel via the kingpin during steady-state steering.
[0122] The active self-centering torque module is used to simulate the torque characteristics of the steering wheel automatically returning to center after a vehicle exits a curve. This module is typically implemented using a proportional-integral-derivative (PID) controller, which calculates the auxiliary self-centering torque to be applied based on the deviation of the current steering wheel angle from the center position (0°) to improve the speed and accuracy of self-centering and suppress self-centering overshoot.
[0123] The damping compensation torque module is used to simulate the viscous damping characteristics of the steering system, generating a torque opposite to the direction of the steering wheel angular velocity to suppress excessive movement and oscillation during steering and increase the sense of stability in the center area at high speeds.
[0124] The friction simulation torque module is used to simulate the Coulomb friction and static friction characteristics of mechanical components such as steering column and steering gear. It generates a resistance torque that is opposite to the direction of steering wheel movement but exhibits nonlinear characteristics when changing direction at low speed or at rest. It is the main source of steering lag and filtering out some high-frequency vibrations of the road surface.
[0125] Correspondingly, the multiple gain coefficients in the mode feature gain vector include: the basic strength gain coefficient and the rotation nonlinearity gain coefficient corresponding to the main torque module; the proportional gain coefficient, integral gain coefficient and differential gain coefficient corresponding to the active homing torque module; the damping strength gain coefficient corresponding to the damping compensation torque module; and the friction amplitude gain coefficient corresponding to the friction simulation torque module.
[0126] For example, the pattern feature gain vector is represented as:
[0127]
[0128] in, These represent the basic strength gain coefficient, rotational nonlinearity gain coefficient, proportional gain coefficient, integral gain coefficient, differential gain coefficient, friction amplitude gain coefficient, and damping strength gain coefficient, respectively.
[0129] Among them, the basic strength gain coefficient is used to globally adjust the overall magnitude of the main torque; the steering angle nonlinear gain coefficient is used to adjust the torque gradient as the steering wheel angle increases.
[0130] The proportional gain coefficient, integral gain coefficient, and differential gain coefficient are used to adjust the proportional response strength of the aligning torque to the angular deviation, the integral elimination strength of the cumulative deviation, and the damping suppression strength of the angular change rate, respectively.
[0131] The damping strength gain coefficient is used to adjust the magnitude of the viscous damping torque as a whole.
[0132] The friction amplitude gain coefficient is used to adjust the width of the friction hysteresis loop and the amplitude of the friction torque.
[0133] In this embodiment, by adjusting these independent gain coefficients, precise control can be achieved over different dimensions such as steering feel (lightness / heaviness), return-to-center speed / smoothness, viscous feel, and clarity / filtering feel.
[0134] For example, the main moment module is defined through a multiplicative topology architecture, and its underlying mathematical model is defined as follows:
[0135]
[0136] In the formula, As the main moment; The basic strength gain coefficient; This represents the basic road feel mapping value that can be obtained from a preset basic road feel map, which determines the linearity of the feel as the rack force increases. F rack This represents the rack force, equivalent rack load, or its estimate for the vehicle, with the sign indicating the direction of force, and v representing the longitudinal vehicle speed. The basic road feel map can be implemented using F... rack The two-dimensional lookup table with v as input can also be composed of one-dimensional rack force-basic hand torque curves under several vehicle speed nodes; This refers to the nonlinear gain coefficient for the rotation angle. This represents the angle influence coefficient, which can be obtained from a two-dimensional lookup table using the absolute value of the steering wheel angle and the longitudinal vehicle speed. It is used to enhance road feel in large-angle areas. Steering wheel angle; This indicates the preset base correction parameters.
[0137] In Sport mode, the two gain coefficients in the main torque module are increased simultaneously to enhance the steering centering feel and improve the gradient of torque as the angle changes, resulting in a stable feel and direct road feedback; in Comfort mode, the gain is reduced to achieve a light steering experience.
[0138] The active self-aligning torque module is implemented using an adaptive PID architecture, and its underlying mathematical model is defined as follows:
[0139]
[0140] In the formula, This is the active restoring torque; The driver intervention inhibition factor, ranging from 0 to 1, is close to 1 when the intervention level is low and decreases smoothly as the intervention level increases. This prevents excessive self-centering force from being superimposed when the driver actively controls the steering wheel. This value can be achieved using one-dimensional or two-dimensional lookup tables, piecewise linear functions, or continuous saturation functions. A smooth representation of the degree of driver intervention can be obtained by sliding average of the steering wheel hand off / hand on state, or by mean filtering of the absolute value of the driver's hand torque or steering column torque. This represents the vehicle speed enable and low-speed transition factor, used to avoid a sudden increase in active return torque when stopping, starting, or at extremely low speeds. It can be obtained from a one-dimensional vehicle speed table: it takes a smaller value when stationary or in the extremely low speed zone, and smoothly increases and remains constant as the vehicle speed enters the normal return range. , and The basic P, I, and D component moments are defined as follows: , , ,in, Target angular velocity Compared with the actual steering wheel angular velocity difference, , , This indicates that the target return-to-center angular velocity is obtained based on the steering wheel angle and vehicle speed. Indicates the sign function, the negative sign and To keep the target velocity always pointing towards the middle position. This represents a differential operation function with low-pass filtering and dead time, where the integral term is set to limit or resist saturation, and the proportional gain is... Integral gain The differential gain can be obtained from the angle-speed two-dimensional table. It can be obtained from a one-dimensional vehicle speedometer; , and These are the proportional gain coefficient, integral gain coefficient, and differential gain coefficient, used to scale the P, I, and D components respectively in different modes. In implementations where it is not necessary to distinguish the active homing characteristics, the three can also take the same value in each mode.
[0141] Among them, the P component torque As the primary force for active steering wheel return, it determines the absolute strength and dynamic response speed of the steering wheel's return; I component torque Specifically designed to solve the problem of excessive residual steering wheel return angle, it eliminates steady-state error through integration, ensuring precise zeroing of the steering wheel; D component torque. This is used to improve the smoothness of the return-to-center process at medium to high vehicle speeds, suppressing shimmy and overshoot during the return-to-center process through damping. In motion mode, the proportional gain coefficient is increased. To enhance centering feel and self-centering speed; in Comfort mode, the differential gain coefficient is increased. It is combined with a first-order delay to ensure a smooth and natural return-to-center process.
[0142] The damping compensation torque module is mainly used in medium- and high-speed operating conditions to suppress hand force fluctuations. Its underlying mathematical model is defined as follows:
[0143]
[0144] In the formula, For damping compensation torque; This is the damping strength gain coefficient; The non-negative basic damping amplitude can be obtained from a two-dimensional lookup table of the absolute value of the steering wheel angular velocity versus the vehicle speed. It employs bilinear interpolation and boundary limiting. This parameter does not need to be limited to a strictly linear viscous relationship, and can maintain a delicate damping in the low angular velocity region, while gradually increasing the damping in the medium-high angular velocity or medium-high vehicle speed region.
[0145] Damping strength gain coefficient It is used to adjust the overall strength of the damping compensation torque. The damping compensation torque acts in the opposite direction of the steering wheel angular velocity to suppress excessively fast return to center, swaying and fluctuations in hand force at medium and high speeds, and to create a stable feeling that changes with speed.
[0146] In high-speed operation during motion mode, increase the damping strength gain coefficient. To increase the responsiveness of the controls and boost confidence at high speeds; in comfort mode, the gain is reduced to ensure agility during maneuvers such as low-speed U-turns.
[0147] The friction simulation torque module is used to simulate the mechanical characteristics of a system. It directly defines the hysteresis width of the torque-rotation angle curve, and its underlying mathematical model is defined as follows:
[0148]
[0149] In the formula, To simulate frictional torque; This is the friction amplitude gain coefficient; A smoothed sign function for the steering wheel angular velocity can be used. Alternatively, a piecewise linear function with a zero-speed transition zone can be used to ensure that the friction torque is always opposite to the direction of the steering wheel movement and to avoid sign fluctuations near zero speed. The non-negative fundamental friction amplitude can be obtained using a multi-dimensional lookup table of angular velocity, rotation angle, and vehicle speed, or it can be described by a continuous function composed of static friction, Coulomb friction, and velocity decay terms. For example:
[0150]
[0151] In the formula, and These represent the low-speed static friction amplitude and the kinetic friction amplitude, respectively, as they vary with vehicle speed. This is the transitional scale between static and dynamic friction. This is used to describe necessary angle corrections. The above continuum is just one implementation of the publicly available mapping rules; in engineering practice, experimentally calibrated maps can be used directly.
[0152] Hysteresis loop width as a function of friction amplitude gain coefficient And the amplitude of the basic friction increases; smoothing sign function The frictional torque direction and frictional amplitude gain coefficient are determined by the frictional gain coefficient. Its overall amplitude can be adjusted, and this parameter mainly maps to the steering wheel angle hysteresis in objective indicators.
[0153] In motion mode, reduce the friction amplitude gain coefficient. The hysteresis loop is narrowed to reduce the road feel masking effect and improve the feel of the car. In comfort mode, the coefficient is increased to use nonlinear friction characteristics to physically filter high-frequency vibrations of the road surface, such as the feel of gravel, and improve the ride smoothness.
[0154] In this embodiment, by dividing the calculation module of the total target road feel torque into four component torque modules and extracting seven independent gain coefficients corresponding to them, it is possible to achieve refined and personalized vehicle feel customization. This overcomes the shortcomings of existing technologies where only the total torque is multiplied by a single comprehensive gain coefficient, resulting in the inability to independently adjust various feel characteristic parameters. Simultaneously, developers can fine-tune each gain coefficient independently, much like operating a "mixing console." For example, they can increase only the damping gain to enhance high-speed stability without affecting low-speed lightness, or narrow only the friction amplitude gain to improve road feel clarity without altering the center zone torque level, thereby meeting the objective indicator requirements of diverse driving modes.
[0155] Existing technologies often cause non-linear, abrupt changes in the steering wheel during mode switching. In high-dynamic conditions such as emergency obstacle avoidance, if a forced mode switch is performed, this sudden torque change may interfere with the driver's precise control, posing a significant safety risk. To further address the safety risks arising in high-dynamic conditions, this application makes the following improvements:
[0156] In an exemplary embodiment, the transition weight factor is generated based on the steering characteristic parameters; correspondingly, in step 203, generating the transition weight factor at the current time includes:
[0157] Step 2031: Determine the steering intensity safety penalty coefficient at the current moment based on the steering characteristic parameters; wherein, when the absolute value of the steering characteristic parameter is less than or equal to the minimum steering characteristic threshold, the steering intensity safety penalty coefficient is 1; when the absolute value of the steering characteristic parameter is greater than the minimum steering characteristic threshold but less than the maximum steering characteristic threshold, the steering intensity safety penalty coefficient decreases linearly with the increase of the absolute value of the steering characteristic parameter, and the value range of the steering intensity safety penalty coefficient is (0,1); when the absolute value of the steering characteristic parameter is greater than or equal to the maximum steering characteristic threshold, the steering intensity safety penalty coefficient is 0.
[0158] Among them, the steering intensity safety penalty coefficient (denoted as γ) is a coefficient with a value range of [0,1] determined by the absolute value of the steering characteristic parameter. It is used to characterize the intensity of the current driving operation and to impose a constraint on the speed of mode switching accordingly.
[0159] In one example embodiment, the steering characteristic parameter is the steering wheel angular velocity dθ / dt. In this embodiment, the absolute value of the steering characteristic parameter is simplified to... During implementation, the electronic control unit acquires the steering wheel angular velocity sensor signal in real time and performs segmented scheduling logic judgment based on a preset steering characteristic threshold (angular velocity threshold in this example):
[0160] steady state range ( ): When the absolute value of the steering characteristic parameter (| |) Less than or equal to the minimum steering characteristic threshold When the speed is 30° / s, the target vehicle is determined to be in a steady state of straight-line driving or minor adjustments. At this time, the steering strength safety penalty coefficient γ is set to 1, indicating that no penalty is imposed on the switching rate, allowing the transition weight factor α(t) to be updated normally at the preset rate.
[0161] Dynamic transition range ( When the absolute value of the steering characteristic parameter is greater than the minimum steering characteristic threshold: And less than the highest steering feature threshold At a speed of 200° / s, it is determined that the driver is performing a routine lane change or steering maneuver. At this time, γ follows | | decreases linearly as it increases, that is The value of γ ranges from (0,1). This achieves the effect of linearly slowing down the switching rate as the steering intensity increases, making the changes in feel more subtle and concealed.
[0162] Limit constraint interval ( When the absolute value of the steering characteristic parameter is greater than or equal to the highest steering characteristic threshold: When the vehicle is in a high-dynamic extreme condition such as severe steering or emergency obstacle avoidance, the steering intensity safety penalty coefficient γ is forcibly set to 0, which means freezing any switching progress and stopping the update of the transition weight factor α(t). At this time, the road feel related parameters are locked at the current instantaneous value until the steering intensity falls back to the safe range. During the freeze period, the steering characteristic parameters and longitudinal vehicle speed still participate in the calculation of the component torque normally.
[0163] Optionally, the aforementioned safety penalty coefficient setting applies not only to steering wheel angular velocity, but also equally to other physical quantities that characterize the dynamic strength of a vehicle, such as vehicle yaw rate, lateral acceleration, or rack force change rate, or any combination of these physical quantities.
[0164] It should be understood that the transition weight factor can also be generated based on the longitudinal vehicle speed. In this case, the minimum longitudinal vehicle speed and the maximum longitudinal vehicle speed are set accordingly. Of course, the transition weight factor can also be generated based on the combination of steering characteristic parameters and longitudinal vehicle speed, which is not restricted here.
[0165] Step 2032: Based on the steering strength safety penalty coefficient and the switching cycle, perform smooth transition processing to obtain the transition weight factor at the current moment.
[0166] For example, the preset switching period is defined as T. base (For example, 1 second). In each control cycle T s Within this context, the update calculation of the transition weight factor α(t) follows an integral or cumulative process. As an example, the update logic of the transition weight factor α(t) can be defined in the following integral form:
[0167]
[0168] In the formula, t0 is the time when the switching command is received. Let τ be the angular velocity of the steering wheel. Let x be the longitudinal vehicle speed at time τ, and let sat() be the limiting function. sat(x,0,1) means that the variable x is restricted to the interval [0,1].
[0169] In a discrete controller, the update logic for the transition weight factor can be defined as follows:
[0170]
[0171] In the formula, α(k+1) is the (k+1)th transition weight factor within the switching period, and α(k) is the kth transition weight factor within the switching period. The k-th steering wheel angular velocity within the switching cycle. This represents the k-th longitudinal speed within the switching cycle.
[0172] It should be understood that the transition weight factor can also be calculated using nonlinear transition algorithms such as smooth S-curves (Sigmoid function), cosine functions, or Bézier curves, and this is not restricted here. In contrast, using a limiting function to calculate the transition weight factor directly constrains the linearly accumulated weights through limiting operations, without the need for complex floating-point exponentiation or trigonometric operations. This can significantly reduce the computing power overhead of the onboard processor while ensuring a smooth weight transition.
[0173] When γ=1, the transition weighting factor is 1 / T base The transition weight factor increases at a constant rate. When γ is between 0 and 1, the growth rate of the transition weight factor slows down proportionally, thus lengthening the entire transition process. When γ=0, the transition weight factor stops updating (freezes), and the mode parameters are locked at the current instantaneous value. Once the target vehicle recovers from the extreme condition to the steady state or transition range, γ returns to a non-zero value, and the transition weight factor continues to update smoothly from the current frozen value until it reaches 1, at which point the driving mode switch is completed.
[0174] In this embodiment, by introducing a steering intensity safety penalty coefficient, the update rate of the transition weight factor is dynamically adjusted or even frozen. Based on this, the smoothness of mode switching and the safety of vehicle driving can be guaranteed. This allows the target vehicle to achieve smooth switching not only in steady state and normal operating conditions, but also to intelligently freeze the parameter switching process when the driver is detected to be driving aggressively, locking the road feel characteristics in a safe state. This avoids interference with the driver's prediction and precise control of vehicle dynamics due to sudden changes in road feel parameters at the moment of extreme control, thereby eliminating potential safety hazards of driving mode switching in high-risk scenarios such as emergency obstacle avoidance, and improving the safety of the target vehicle under all operating conditions.
[0175] In an exemplary embodiment, step 205, obtaining the target feedback torque at the current moment based on the execution mode feature gain vector, longitudinal vehicle speed, and steering feature parameters, includes:
[0176] Step 2051: Use the gain coefficients in the execution mode feature gain vector as scaling factors and apply them to the corresponding component torque modules in the preset general reference mapping table, and input the longitudinal vehicle speed and steering feature parameters into each component torque module.
[0177] In this embodiment, the component torque modules in the general reference mapping table include the main torque module, the active return torque module, the damping compensation torque module, and the friction simulation torque module. The component torques obtained from the above component torque modules include the main torque, the active return torque, the damping compensation torque, and the friction simulation torque.
[0178] For each component torque module, a basic torque value is first obtained by looking up a table or calculating a function based on the input longitudinal vehicle speed and steering characteristic parameters. Then, the basic value is multiplied by the corresponding gain coefficient in the execution mode feature gain vector to obtain the corresponding component torque.
[0179] Step 2052: Obtain the corresponding component torques according to each component torque module, and superimpose the component torques to obtain the target feedback torque at the current moment; wherein, the general reference mapping table represents the physical characteristic mapping relationship that does not change with the driving mode, and the component torque modules in the general reference mapping table include the main torque module, the active self-centering torque module, the damping compensation torque module, and the friction simulation torque module.
[0180] For example, the electronic control unit sums up the individual component torques calculated in step 2051 to obtain the total target feedback torque T. total (t):
[0181] T total (t)=T main (t)+T return (t)+T damping (t)+T friction (t)
[0182] In the formula, T main (t), T return (t), T damping (t) and T friction (t) represents the main torque, active return torque, damping compensation torque, and friction simulation torque output by the main torque module, active return torque module, damping compensation torque module, and friction simulation torque module, respectively.
[0183] In this embodiment, the target feedback torque T is obtained. total (t) after which T is also discussed total (t) Perform target torque amplitude limitation, rate of change limitation, and signal validity judgment to obtain the final target torque request T. req (t) to ensure the physical rationality of the output torque and the safety of the system.
[0184] By combining the dynamic gain vector with a general reference mapping table to obtain the target feedback torque, the method in this application can quickly calculate the target torque with smooth transition without modifying or switching any underlying two-dimensional lookup tables when running online. It only requires interpolation of a small number of gain coefficients.
[0185] In this embodiment, the target feedback torque is calculated based on the dynamic gain vector and the universal reference mapping table, making the online computing load almost the same as that of conventional control without mode switching, with only a small number of multiplication operations added. This places very low demands on the computing power of the vehicle processor. At the same time, since the operation of all driving modes is based on the universal reference mapping table, internal state conflicts that may arise from different driving modes using different algorithms can be avoided. This approach is suitable for drive-by-wire steering road feel simulation scenarios where resources are limited and reliability requirements are strict.
[0186] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.
[0187] Based on the same inventive concept, this application also provides a steerable steering road feel simulation device for implementing the above-mentioned steerable steering road feel simulation method. The solution provided by this device is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more steerable steering road feel simulation device embodiments provided below can be found in the limitations of the steerable steering road feel simulation method above, and will not be repeated here.
[0188] In one exemplary embodiment, such as Figure 4 As shown, a steer-by-wire road feel simulation device is provided, comprising:
[0189] The mode switching sensing unit is used to receive the driving mode switching command of the target vehicle; wherein, the driving mode switching command is used to switch the target vehicle from the first driving mode to the second driving mode.
[0190] The vehicle monitoring unit is used to acquire the longitudinal speed and steering characteristic parameters of the target vehicle at the current moment.
[0191] The dynamic weight generation unit is used to generate the transition weight factor at the current moment; wherein, the transition weight factor changes dynamically with time within a preset switching period starting from the current moment, and the value range of the transition weight factor is [0,1].
[0192] The gain interpolation unit is used to obtain the first mode feature gain vector matching the first driving mode and the second mode feature gain vector matching the second driving mode, and to interpolate the first mode feature gain vector and the second mode feature gain vector based on the transition weight factor to obtain the execution mode feature gain vector at the current time.
[0193] The torque synthesis execution unit is used to obtain the target feedback torque at the current moment based on the execution mode feature gain vector, longitudinal vehicle speed and steering feature parameters;
[0194] The vehicle control unit is used to drive the road sensor motor of the target vehicle according to the target feedback torque.
[0195] The modules in the aforementioned steer-by-wire road feel simulation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.
[0196] In one exemplary embodiment, a computer device is provided, which may be an electronic control unit, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data related to steer-by-wire road feel simulation. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a steer-by-wire road feel simulation method.
[0197] Those skilled in the art will understand that Figure 5The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0198] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0199] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.
[0200] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.
[0201] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.
[0202] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0203] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method for simulating road feel during steer-by-wire, characterized in that, The method includes: Receive a driving mode switching command for the target vehicle; wherein the driving mode switching command is used to switch the target vehicle from a first driving mode to a second driving mode; Obtain the longitudinal speed and steering characteristic parameters of the target vehicle at the current moment; Generate a transition weight factor for the current moment; wherein the transition weight factor dynamically changes over time within a preset switching period starting from the current moment, and the value range of the transition weight factor is [0,1]. Obtain the first mode feature gain vector matching the first driving mode and the second mode feature gain vector matching the second driving mode, and perform interpolation processing on the first mode feature gain vector and the second mode feature gain vector based on the transition weight factor to obtain the execution mode feature gain vector at the current time. The target feedback torque at the current moment is obtained based on the execution mode feature gain vector, the longitudinal vehicle speed, and the steering feature parameters; The road sensor motor of the target vehicle is driven to operate based on the target feedback torque.
2. The method according to claim 1, characterized in that, Before receiving the driving mode switching command from the target vehicle, the method further includes: For various driving modes, a set of multi-dimensional mode feature gain vectors are pre-calibrated; wherein, for any driving mode, the mode feature gain vector includes multiple gain coefficients corresponding to each component torque module in a preset general reference mapping table.
3. The method according to claim 2, characterized in that, For each of the various driving modes, a set of multi-dimensional mode feature gain vectors are pre-calibrated, including: The static physical parameters of the target vehicle are obtained, and objective evaluation indicators for characterizing driving feel are set based on the static physical parameters; wherein, the objective evaluation indicators include the average steering stiffness in the center area, steering wheel hysteresis, and target damping ratio; For at least two driving modes, set target values for objective evaluation indicators for each driving mode; For any driving mode, the target value of its objective evaluation index is used as the offline calibration target, and a set of multi-dimensional mode feature gain vectors is calculated by inversion. The mode feature gain vectors matched with each driving mode are stored in a fixed form to form a mode feature gain matrix.
4. The method according to claim 2, characterized in that, The component torque module includes a main torque module, an active return torque module, a damping compensation torque module, and a friction simulation torque module; The multiple gain coefficients in the mode feature gain vector include: the basic strength gain coefficient and the rotation nonlinearity gain coefficient corresponding to the main torque module; the proportional gain coefficient, integral gain coefficient and differential gain coefficient corresponding to the active self-correcting torque module; the damping strength gain coefficient corresponding to the damping compensation torque module; and the friction amplitude gain coefficient corresponding to the friction simulation torque module.
5. The method according to claim 1, characterized in that, The transition weight factor is generated based on the steering feature parameters; generating the transition weight factor at the current moment includes: The steering intensity safety penalty coefficient for the current moment is determined based on the steering characteristic parameters; wherein, when the absolute value of the steering characteristic parameter is less than or equal to the minimum steering characteristic threshold, the steering intensity safety penalty coefficient is 1; when the absolute value of the steering characteristic parameter is greater than the minimum steering characteristic threshold and less than the maximum steering characteristic threshold, the steering intensity safety penalty coefficient decreases linearly as the absolute value of the steering characteristic parameter increases, and the value range of the steering intensity safety penalty coefficient is (0,1); when the absolute value of the steering characteristic parameter is greater than or equal to the maximum steering characteristic threshold, the steering intensity safety penalty coefficient is 0. Based on the steering strength safety penalty coefficient and the switching cycle, a smooth transition process is performed to obtain the transition weight factor at the current moment.
6. The method according to claim 1, characterized in that, The step of obtaining the target feedback torque at the current moment based on the execution mode feature gain vector, the longitudinal vehicle speed, and the steering feature parameters includes: Each gain coefficient in the execution mode feature gain vector is used as a scaling factor and applied to the corresponding component torque module in the preset general reference mapping table. The longitudinal vehicle speed and the steering feature parameters are then input into each component torque module. The corresponding component torques are obtained from each component torque module, and the component torques are superimposed to obtain the target feedback torque at the current moment; wherein, the general reference mapping table represents the physical characteristic mapping relationship that does not change with the driving mode, and the component torque modules in the general reference mapping table include the main torque module, the active self-centering torque module, the damping compensation torque module, and the friction simulation torque module.
7. A steer-by-wire road feel simulation device, characterized in that, The device includes: A mode switching sensing unit is used to receive a driving mode switching command from a target vehicle; wherein the driving mode switching command is used to switch the target vehicle from a first driving mode to a second driving mode. The vehicle monitoring unit is used to acquire the longitudinal speed and steering characteristic parameters of the target vehicle at the current moment. A dynamic weight generation unit is used to generate a transition weight factor at the current moment; wherein the transition weight factor changes dynamically with time within a preset switching period starting from the current moment, and the value range of the transition weight factor is [0,1]. The gain interpolation unit is used to obtain a first mode feature gain vector matching the first driving mode and a second mode feature gain vector matching the second driving mode, and to perform interpolation processing on the first mode feature gain vector and the second mode feature gain vector based on the transition weight factor to obtain the execution mode feature gain vector at the current time. The torque synthesis execution unit is used to obtain the target feedback torque at the current moment based on the execution mode feature gain vector, the longitudinal vehicle speed, and the steering feature parameters. The vehicle control unit is used to drive the road sensor motor of the target vehicle to operate based on the target feedback torque.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.
