A highly adjustable road feel simulation device for a steer-by-wire system

Abstract

The application relates to a height-adjustable road feeling simulation device of a steer-by-wire system, which comprises a sensor unit, a hand force design unit, a hand force control unit and an interactive setting unit. The sensor unit is used to collect vehicle motion state information and hand force simulation mechanism angle torque information, and serves as the input of the hand force design unit and the hand force control unit. The hand force design unit comprises a rack force estimation module and a desired hand force calculation module, and is used to obtain a height-adjustable desired hand force. The hand force control unit is used to generate a corresponding road feeling motor control instruction, so as to realize tracking control of an actual hand force on the desired hand force. The interactive setting unit is used to provide a plurality of height-adjustable interfaces in succession, so as to meet the personalized steering feeling requirements of a driver. Compared with the prior art, the application has the advantages of low cost, reliable performance and height adjustability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobile drive-by-wire chassis, and particularly relates to a height-adjustable drive-by-wire steering system road feeling simulation device. BACKGROUND

[0002] As a representative technology of automobile intelligence, drive-by-wire steering technology has been highly concerned by colleges and enterprises. Since the drive-by-wire steering system cancels the mechanical connection and realizes the steering function through the drive-by-wire mode, the driver cannot perceive the vehicle motion state and road information through the steering hand feeling, which further leads to the decline of the vehicle handling and the driver's safety feeling. Therefore, the accurate and highly adjustable road feeling simulation function can not only make the driver have clear road feeling, but also can be highly adjustable in multiple dimensions, so as to fully exert the advantages of the force characteristic free design of the drive-by-wire steering, and the driver can customize his own steering hand feeling from multiple dimensions.

[0003] The existing road feeling simulation algorithm can be roughly divided into two kinds: ① the road feeling simulation method based on experience estimation of return torque; ② the road feeling simulation method based on measurement of real return torque. The former is to equivalent the hand force as a function of steering wheel angle and vehicle speed according to experience and based on a simple linear dynamics model, but the driver has steering feeling but lacks the perception of road conditions; the latter is to estimate the real return torque of the vehicle through the sensor installed on the steering actuator, so as to calculate the hand force and feedback to the driver in real time, so that the driver can have clear road feeling, which is a method more in line with actual demand. However, the existing road feeling simulation algorithm still has many problems to be solved.

[0004] In one aspect, due to the cancellation of mechanical connection, the steer-by-wire system needs to control the end of the steering wheel angle, that is, when the driver turns the steering wheel to the maximum value (end) specified by the system, the system needs to generate a large hand force to prevent the driver from continuing to turn the steering wheel. However, the existing end control method does not consider the contradiction between hand force smoothness and end control realizability. That is, to realize the end control function by closed-loop control, a large range torque sensor needs to be used at the hand force simulation mechanism column, resulting in low measurement resolution and poor hand force smoothness in normal operating conditions. To ensure hand force smoothness, the use of a small range high-precision torque sensor will limit the end control torque and result in poor limit effect, and the steering wheel angle is easy to exceed the end under the driver's operation, which reduces the system safety. If a high-precision large-range torque sensor is used, the actual application realizability is low, and the hardware cost is high. Therefore, considering that the hand force range is not large in normal operating conditions and the hand force precision is high, and the control precision is not high in the end control condition, only open-loop control is needed to provide sufficient torque, and a small range high-precision torque sensor should be selected on the hardware, and an open and closed loop smooth switching strategy should be combined on the software to solve the contradiction.

[0005] On the other hand, most of the existing road feeling simulation algorithms can only calibrate one kind of hand feeling, and cannot meet the different steering hand feeling needs caused by the difference between drivers. The remaining few adjustable road feeling simulation algorithms only provide simple adjustment in the level and gradient of the steering force, and do not provide other adjustable dimensions, and do not consider the SBW system stability uncertainty problem caused by system adjustment.

[0006] In summary, the two aspects of the existing steer-by-wire system road feeling simulation problems are that the existing hand force control method cannot simultaneously consider the hand force smoothness in normal operating conditions and the safety limit in the end condition, which relates to the sensor cost, steering feeling and system safety of the steer-by-wire system; and the other aspect is that most of the existing hand force design methods can only calibrate one kind of hand feeling, and cannot meet the different hand feeling needs of drivers, and the remaining adjustable hand force design is limited to the level and gradient of the steering force, and does not set the road feeling clarity adjustable, steering force VS speed gradient adjustable and road feeling bandwidth adjustable interface, the system adjustment is low, the coupling degree between multiple adjustable dimensions is too high, and the hand force stability change problem is not considered. SUMMARY

[0007] The purpose of the present application is to overcome the defects of the prior art and provide a highly adjustable steer-by-wire system road feeling simulation device and method.

[0008] The purpose of the present application can be realized by the following technical solutions:

[0009] The application provides a highly adjustable road feel simulation device for a steer-by-wire system, which comprises:

[0010] A sensor unit is used to collect vehicle motion state information and hand force simulation mechanism angular torque information, and serve as input for a hand force design unit and a hand force control unit;

[0011] The hand force design unit comprises a rack force estimation module and a desired hand force calculation module, and is used to obtain a highly adjustable desired hand force;

[0012] The hand force control unit is used to generate control instructions for a corresponding road feel motor, so as to realize tracking control of an actual hand force on a desired hand force;

[0013] An interaction setting unit is used to provide a plurality of continuously adjustable interfaces to meet the personalized control feeling requirements of a driver.

[0014] Preferably, the sensor unit comprises a torque sensor at a hand force simulation mechanism column, a steering wheel angular displacement sensor at the hand force simulation mechanism column, a steering execution motor torque sensor, a rack displacement sensor and a vehicle speed sensor;

[0015] The torque sensor is used to measure the torque of the column, i.e. the actual hand force, and the torque sensor is matched with an open-loop and closed-loop smooth switching algorithm to realize high-precision closed-loop control of the hand force in a normal control mode and large torque safety limiting after the hand force exceeds the range in an end control mode;

[0016] The angular displacement sensor is used to measure the angular displacement of the column, i.e. the steering wheel angular displacement, and the angular displacement of the column is used for calculation of the desired hand force in the hand force design unit and feedforward compensation of friction in the hand force control unit;

[0017] The steering execution motor torque sensor is used to measure the motor output torque, i.e. the steering execution motor torque, and the motor output torque is used for estimation of the rack force in the hand force design unit;

[0018] The rack displacement sensor is used to measure the rack displacement, and the rack displacement is used for estimation of the rack force in the hand force design unit;

[0019] The vehicle speed sensor is used to measure the vehicle speed, and the vehicle speed is used for calculation of the desired hand force in the hand force design unit.

[0020] Preferably, the rack force estimation module adopts an inverse model plus low-pass filtering method for the process of rack force estimation, and the specific process is as follows:

[0021] A model based on a steering execution mechanism is established, and the expression is as follows:

[0022]

[0023] wherein mr and cr are the rack equivalent mass and damping coefficient of the steering actuator, respectively, x r is the rack displacement, F r , and F m are the rack force to be estimated, the rack equivalent friction of the steering actuator and the motor equivalent force acting on the rack, respectively, F m and x r are obtained by a steering actuator output torque sensor and a rack displacement sensor, respectively;

[0024] obtaining a rack force estimation nominal value, the rack force estimation nominal value being calculated by:

[0025]

[0026] wherein F r ′ is the rack force estimation nominal value;

[0027] the obtained rack force estimation nominal value F r ′ is filtered by a low-pass filter, the estimated rack force being obtained by:

[0028] F rack_est = H lowpass * F r ′

[0029] wherein F rack_est is the estimated rack force, H lowpass (s) is the transfer function of the low-pass filter, H lowpass is its inverse Laplace transform, the low-pass filter being a third-order low-pass filter, the three poles of the third-order low-pass filter being configured as [p1 p2 p3], wherein the pole p1 is used to adjust the bandwidth, the poles p2 and p3 are used to adjust the characteristics of the filter, the adjustable range of the feedback road bandwidth B w being set as [B wmin , B wmax ], the pole configuration method of the third-order low-pass filter being specifically:

[0030] when the bandwidth B w =B wmax , the cut-off frequency of the transfer function H lowpass (s) is set as a third-order Butterworth low-pass filter with B wmax , the configured poles being respectively:

[0031] p1 = -Bw

[0032]

[0033] transfer function H lowpass (s)(m r s 2 +c r s) has a maximum gain A gain (B wmax ) to represent the maximum gain of the rack force estimation value fluctuation caused by sensor noise;

[0034] When the bandwidth B w =B wmin , the pole p1 controls the bandwidth, the poles p2 and p3 form a second-order Butterworth filter with a bandwidth B w1 , and the configured poles are respectively:

[0035] p1=-B w

[0036]

[0037] Adjusting the bandwidth B w1 to make the transfer function H lowpass (s)(m r s 2 +c r s) has a maximum gain A gain (B wmin )≈A gain (B wmax ), and at the same time, the phase angle lag is close to the working condition when the bandwidth B w =B wmax ;

[0038] When the bandwidth B w ∈(B wmin , B wmax ), the poles p1, p2 and p3 are obtained by linear interpolation of the corresponding pole values when the bandwidth B w =B wmax and the bandwidth B w =B wmin .

[0039] Preferably, the expected hand force calculation module obtains the expected hand force based on the estimated rack force, vehicle speed and steering wheel angle, and the composition of the expected hand force includes six torque, respectively, the height-adjustable main steering torque T main , the height-adjustable stiffness torque T stiff , the mode-adaptive high-speed damping torque T damp , the simulated friction torque T fric , the low-speed active return torque T ar and the end control torque T limit .

[0040] The height-adjustable main control torque T main The expression is:

[0041] T main =ρ·f main (F rack_est )

[0042] Among them, T main The main control torque is ρ, where ρ is the road feel clarity adjustment factor, and T represents the main control torque. main Its relationship with stiffness moment T stiff The proportion of the two, F rack_est The estimated rack force, f, is obtained from the rack force estimation module. main Based on the rack force estimate F rack_est Generate main control force T main The function;

[0043] The height-adjustable stiffness moment T stiff The expression is:

[0044] T stiff =f stiff (F rack_virtual ,ρ,γ·η(V))

[0045] Among them, T stiff Let f be the stiffness torque, γ be the control force level and gradient adjustment factor, η(V) be the control force vs. vehicle speed gradient adjustment factor, and f be the stiffness torque. stiff Based on the virtual value F of rack force rack_virtual Generating stiffness moment T stiff The function, which is a two-part piecewise function, is used to independently generate the stiffness moments at intermediate and non-intermediate positions, F. rack_virtual (V, θ) sw (Based on rotation angle θ) sw The virtual rack force is calculated based on the vehicle speed V;

[0046] Virtual rack force F rack_virtual (V, θ) sw The expression for ) is:

[0047] F rack_virtual =K stiff (V)·θ sw

[0048] K stiff (V) represents the stiffness coefficient at different vehicle speeds, and θ sw For the corner;

[0049] Based on the virtual value of rack force F rack_virtual Generating stiffness moment T stiff function fstiff The expression of the virtual rack force threshold is:

[0050]

[0051] Wherein, f stiff1 (·) is the stiffness torque for generating the intermediate position, f stiff2 (·) is the stiffness torque for generating the non-intermediate position, F thes1 is the virtual rack force threshold for distinguishing the intermediate position and the non-intermediate position at different vehicle speeds, ρ is the road feel definition adjustment factor, γ is the steering force level and gradient adjustment factor, η(V) is the steering force VS vehicle speed gradient adjustment factor;

[0052] The mode-adaptive high-speed damping torque T damp The expression of the mode-adaptive high-speed damping torque is:

[0053]

[0054] Wherein, T damp is the mode-adaptive high-speed damping torque, c damp (V, ρ, γ·η(V)) is the mode-adaptive damping coefficient, f dt (θ sw ) is the steering wheel angle influence factor on the damping coefficient, which is used to weaken the damping in the small steering angle working area, and thus reduce the stickiness caused by the active damping in the small steering angle area, is the steering wheel speed;

[0055] The simulated friction torque T fric is used to simulate the mechanical system friction of the mechanical system or the EPS, and the expression of the simulated friction torque T fric is:

[0056]

[0057] Wherein, is the friction force model based on the tanh(·) function;

[0058] The low-speed active return torque T ar is used to ensure the return performance in the low-speed working condition, and the expression of the low-speed active return torque T ar is:

[0059]

[0060] Wherein, K ar represents the return strength, σ V (V) represents the vehicle speed factor, is the steering angle adjustment factor, is the speed adjustment factor, a positive direction judgment factor;

[0061] The terminal control torque T limit To ensure the safe limit of the steering wheel.

[0062] Preferably, the controlled object of the hand force control unit is a hand force simulation mechanism, and the dynamic characteristic equation of the hand force simulation mechanism is:

[0063]

[0064]

[0065] T s = K Ts (θ sw - θ m )

[0066] Wherein, J sw and J m represent the rotational inertia of the steering wheel and the road feel motor respectively, C sw and C m represent the linear damping of the steering wheel and the road feel motor respectively, K Ts represents the stiffness of the torque sensor, T h , T s , T m and T f represent the steering force of the driver, the torque sensor torque, the torque at the output end of the motor and the nonlinear friction respectively, θ sw is the steering wheel angle, and θ m is the angle of the road feel motor.

[0067] The control target of the hand force control unit is specifically to transmit the control instruction to the road feel motor to control the output torque T m of the road feel motor, so that the actual hand force tracks the expected hand force.

[0068] Preferably, the hand force control unit controls the hand force based on three control channels and generates the control instruction of the road feel motor, and the three control channels are respectively a closed-loop control channel, a feedforward control channel and a nonlinear friction feedforward compensation channel.

[0069] The closed-loop control channel adopts a lag-lead correction device designed torque closed-loop controller based on the expected frequency characteristic method, and then controls the hand force closed-loop tracking based on the deviation between the expected hand force and the actual hand force measured by the torque sensor.

[0070] The feedforward control channel is specifically an open-loop control, which is used to continue transmitting the control instruction to the road feel motor after the torque exceeds the range, so as to continue to provide the hand force, and the expression of the control instruction is:

[0071]

[0072] where u ffd is the torque, T cmd is the desired hand force, represents the active damping imposed on the open-loop control to reduce the hand force jitter caused by the feedforward control channel;

[0073] The nonlinear friction feedforward compensation channel is used to feedforward compensate the system friction.

[0074] Preferably, the disconnection or connection of the closed-loop control channel and the feedforward control channel is controlled by an open-closed-loop smooth switching strategy, and the input of the open-closed-loop smooth switching strategy is the sensor torque T s The feedforward control channel is gradually controlled, and the closed-loop control channel is switched, and the two channel switching quantities of the open-closed-loop switching strategy are respectively the feedforward opening degree β ffd and the closed-loop opening degree β fb , and the expressions are respectively:

[0075]

[0076] where β ffd is the feedforward opening degree, f ffd (T s ) is the feedforward switch steady-state value, s is the Laplace transform complex frequency, β fb is the closed-loop opening degree, τ ffd and τ fb are time constants, and f fb (T s ) is the closed-loop switch steady-state value.

[0077] Preferably, the properties of the two channel switching quantities of the open-closed-loop switching strategy are respectively:

[0078] The value range of the feedforward opening degree β ffd is [0, 1], and the feedforward opening degree β ffd linearly changes between the threshold T s_th1 and the threshold T s_th2 and the continuous change quantity of the series first-order inertia link during the intervention or withdrawal process gradually controls the feedforward control quantity, and the intervention or withdrawal speed of the feedforward opening degree β ffd is determined by the time constant τ ffd and the difference between the two thresholds (T s_th2 -T s_th1 );

[0079] The closed-loop opening degree βfb The value range of β is [0, 1]. fb Specifically, during the intervention or withdrawal process, a switching quantity of a first-order inertial element is connected in series to control the tracking error input of the closed-loop controller. The closed-loop switching coefficient β is mentioned above. fb The intervention or withdrawal speed is determined by the time constant τ fb Decide.

[0080] Preferably, the open-loop / closed-loop smooth switching strategy specifically includes:

[0081] When the sensor torque T s Less than the threshold At this time, the closed-loop switch is always 1 and the open-loop switch is always 0, that is, the closed-loop control channel and the friction force feedforward compensation channel are in working state to realize torque closed-loop tracking control in normal operating conditions.

[0082] When the sensor torque T s Greater than or equal to threshold T s_th1 Less than threshold T s_th2 At this time, the closed-loop switch is always 1, the open-loop switch is half closed, and the feedforward channel is not fully engaged. At this time, the system is controlled by the closed-loop control channel, the friction feedforward compensation channel, and the incomplete feedforward control channel.

[0083] When the sensor torque T s Greater than or equal to threshold T s_th2 When the closed-loop switch is quickly set to 0, the open-loop switch is fully closed, and the system is then controlled by the friction force feedforward compensation channel and the feedforward channel.

[0084] Preferably, the interactive setting unit includes four independent interfaces corresponding to four adjustable dimensions of hand force, allowing the driver to personalize settings through the human-computer interaction interface. The four dimensions can be freely combined and continuously adjusted according to the driver's needs, and a sport mode and a comfort mode can be calibrated. The interfaces are an adjustable road feel feedback bandwidth interface, a road feel clarity interface, an adjustable control force level and gradient interface, and an adjustable control force VS vehicle speed gradient interface. The adjustable road feel feedback bandwidth interface is used to set the road feel bandwidth of the rack force estimation module. The road feel clarity interface, the adjustable control force level and gradient interface, and the adjustable control force VS vehicle speed gradient interface are used to set the hand force characteristics of the desired hand force calculation module.

[0085] The adjustable dimensions of hand force height include adjustable road feel bandwidth, adjustable road feel clarity, adjustable steering force level and gradient, and adjustable steering force vs. vehicle speed gradient. The adjustable factors corresponding to these adjustable dimensions are the road feel bandwidth adjustable factor B. w, road sense clarity adjustment factor p, steering force level and gradient adjustable factor γ, and steering force VS speed gradient factor η(V);

[0086] The feedback road sense bandwidth is highly adjustable, and the method is pole self-adaptive configuration of a rack force filter. When switched to a comfortable mode, the bandwidth of the feedback road sense is set to a small value, so that the smoothness of the hand force is good. When switched to a sports mode, the bandwidth of the feedback road sense is set to a large value, so that the driver obtains more vehicle state and road information through the hand force.

[0087] The road sense clarity is highly adjustable, and the proportion of the main steering torque and the stiffness torque in the expected hand force at the middle position and the non-middle position is adjusted. When switched to a sports mode, the proportion of the main steering torque is large, the proportion of the stiffness torque is small, the peak-to-peak value of the hand force fluctuation caused by the same road roughness is large, the hand force on the low adhesion road is obvious, and the road sense clarity is increased. When switched to a comfortable mode, the proportion of the main steering torque is small, and the proportion of the stiffness torque is large, so that the hand force is smooth and stable.

[0088] The steering force level and gradient are highly adjustable, and the steering force level and gradient are adjusted. In the same vehicle speed working condition, when switched to a sports mode, the steering force level and gradient are large, and when switched to a comfortable mode, the steering force level and gradient are small.

[0089] The gradient of the steering force VS speed is highly adjustable, and the gradient of the steering force changing with the speed is adjusted. When switched to a sports mode, the gradient is enhanced with the speed, the perceptibility of the driver to the speed is enhanced, when switched to a normal mode, the gradient remains unchanged with the speed, and when switched to a comfortable mode, the gradient is weakened with the speed, so as to reduce the driving burden caused by the speed increase.

[0090] Compared with the prior art, the present application has the following beneficial effects:

[0091] 1. The present application realizes the technical effects of saving cost, reliable performance and high system safety by designing an open-closed loop smooth switching strategy, simultaneously ensuring high-precision closed-loop control of the hand force in normal steering working conditions and safety limiting in terminal working conditions, without using a large-range high-precision torque sensor with high cost.

[0092] 2、The application provides an interface for the driver to independently adjust the hand feeling through the setting of the interaction setting unit, so that the driver can continuously adjust the hand force from multiple dimensions, the road feeling bandwidth, the road feeling definition, the control force level and gradient and the gradient of the control force VS the vehicle speed can be adjusted, the system stability margin does not change with the mode switching, the decoupling between the adjustable hand force and stability, the decoupling of multiple adjustable dimensions, the personalized hand feeling demand of the driver and the highly adjustable technical effect are realized. BRIEF DESCRIPTION OF DRAWINGS

[0093] Figure 1 It is the principle diagram of the application.

[0094] Figure 2 It is the principle diagram of the hand force control unit of the application.

[0095] Figure 3 It is the on-off quantity steady-state value diagram in the open-closed loop switching strategy of the application.

[0096] Figure 4 It is the feedback road feeling bandwidth adjustable schematic diagram in the hand force design unit of the application.

[0097] Figure 5 It is the road feeling definition adjustable schematic diagram in the hand force design unit of the application.

[0098] Figure 6 It is the control force level and gradient adjustable, control force VS vehicle speed gradient adjustable schematic diagram in the hand force design unit of the application.

[0099] Figure 7 It is the influence of feedback road feeling bandwidth adjustable on the hand force ring stability without stability keeping measures.

[0100] Figure 8 It is the influence of road feeling definition adjustable and control force level and gradient adjustable on the hand force ring stability without stability keeping measures.

[0101] Figure 9 It is the Map diagram of the low-pass filter pole p2 and p3 adaptive configuration with bandwidth of the application.

[0102] Figure 10 It is the high-speed adaptive damping coefficient Map diagram when the road feeling definition adjustment factor of the application is set to ρ=1.0.

[0103] Figure 11 It is the control force characteristic diagram of the terminal control of the application.

[0104] Figure 12 It is the hand force time domain diagram in the open-closed loop switching process of the application.

[0105] Figure 13Fig. 4 is a force-time diagram of the steering force characteristic of the steering process of the present application when different road feel bandwidths B w Fig. 5 is a force-time diagram of the steering force characteristic of the steering process of the present application when different road feel bandwidths B

[0106] Figure 14 Fig. 6 is a force-time diagram of the steering force characteristic of the steering process of the present application when different road feel bandwidths B

[0107] Figure 15 Fig. 7 is a force-time diagram of the steering force characteristic of the steering process of the present application when different road feel bandwidths B w Fig. 8 is a steering angle-time diagram of the hands-off return process of the present application when different road feel bandwidths B

[0108] Figure 16 Fig. 9 is a force-time diagram of the steering force characteristic of the steering process of the present application when different steering force levels and gradient adjustment factors γ

[0109] Figure 17 Fig. 10 is a steering angle-time diagram of the hands-off return process of the present application when different steering force levels and gradient adjustment factors γ

[0110] Figure 18 Fig. 11 is a force-time diagram of the steering force characteristic of the steering process of the present application when different road feel definition adjustment factors ρ

[0111] Figure 19 Fig. 12 is a force-time diagram of the steering force characteristic of the steering process of the present application when the road feel definition adjustment factor ρ = 0.5

[0112] Figure 20 Fig. 13 is a force-time diagram of the steering force characteristic of the steering process of the present application when the road feel definition adjustment factor ρ = 1.0

[0113] Figure 21 Fig. 14 is a steering angle-time diagram of the hands-off return process of the present application when different road feel definition adjustment factors ρ DETAILED DESCRIPTION

[0114] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments. The present embodiment is implemented on the basis of the technical solution of the present application, and detailed implementation modes and specific operation processes are given, but the protection scope of the present application is not limited to the following embodiments.

[0115] As shown in Figure 1 Fig. 1, the present application proposes a highly adjustable steer-by-wire system road feel simulation device, which comprises a sensor unit, a hand force design unit, a hand force control unit and an interactive setting unit:

[0116] The sensor unit is used to collect the vehicle motion state and the angle torque information of the hand force simulation mechanism. The sensors for collecting the vehicle motion state include a rack displacement sensor, a steering execution motor torque sensor, a vehicle speed sensor, a torque sensor at the hand force simulation mechanism column, and a steering wheel angle sensor at the hand force simulation mechanism column. The rack displacement sensor, the steering execution motor torque sensor, and the vehicle speed sensor are used to collect the vehicle motion state to estimate the vehicle return torque and as the input of the hand force design unit. The column angle sensor and the small-range high-precision torque sensor are used to collect the angle torque information of the hand force simulation mechanism to serve as the input of the hand force design unit and the hand force control unit.

[0117] The small-range high-precision torque sensor is selected to have a range of [-10 Nm, 10 Nm] and a resolution of 0.05 Nm, and the open-loop and closed-loop smooth switching strategy can simultaneously consider the hand force closed-loop high-precision control in the normal steering (hand force less than 10 Nm) condition and the safety limiting in the end control (hand force greater than 10 Nm) condition.

[0118] The hand force control unit is used to generate the control instruction of the corresponding road feel motor, which is fed back to the driver to track the expected hand force (expected feedback hand force) of the hand force design unit. The controlled object of the hand force control unit is the hand force simulation mechanism, and the dynamic characteristic equation of the hand force simulation mechanism is as follows:

[0119]

[0120]

[0121] T s =K Ts (θ sw -θ m )

[0122] Wherein, J sw and J m represent the rotational inertia of the steering wheel and the road feel motor respectively, C sw and C m represent the linear damping of the steering wheel and the road feel motor respectively, K Ts represents the stiffness of the torque sensor, T h , T s , T m and T f represent the steering force of the driver, the torque sensor torque, the motor output torque, and the nonlinear friction respectively, θ sw is the steering wheel angle, and θ m is the angle of the road feel motor.

[0123] The control target of the hand force control unit is to transmit appropriate control instructions to the road feel motor to control the output torque T of the road feel motor m , so that the actual torque sensor torque T s follows the expected hand force T cmd .

[0124] The hand force control unit controls the hand force in three control channels in total, as shown in Figure 2 , channel 1 is a closed-loop control channel, channel 2 is a feedforward control channel, and channel 3 is a friction feedforward compensation channel. The opening and closing of channels 1 and 2 are controlled by an open-closed loop smooth switching strategy.

[0125] Channel 1: A lag-lead correction device based on expected frequency characteristic method is used to design a torque closed-loop controller, that is, a series correction link (torque closed-loop controller G cc (s)) is designed based on the expected frequency domain characteristics through the lag-lead correction method, so that the closed-loop system performance meets the expected frequency domain characteristic requirements. The expected frequency domain characteristics include: a, the low-frequency segment gain is large, which ensures the high tracking accuracy of the system; b, the mid-frequency segment crossover frequency is large, which ensures the fast response of the closed loop and the transmission of the road feel; c, the high-frequency segment gain decays quickly, which ensures the noise suppression ability; d, the amplitude margin and phase angle margin are large, which ensures the stability of the system; the expected time domain characteristics include: the step response does not appear overshoot or the overshoot amount is small, that is, the over-damped characteristic of the system, which ensures the smoothness of the hand force when the closed loop is connected and disconnected;

[0126] Channel 2: It belongs to open-loop control, which can continue to transmit torque control instructions to the road feel motor after the torque exceeds the range, thereby continuing to provide hand force. The expression of the torque control instructions transmitted to the road feel motor is:

[0127]

[0128] Where, u ffd is the control torque, T cmd is the expected hand force, represents the active damping applied to the open-loop control, so that the originally under-damped system becomes an over-damped system under the action of the active damping, reducing the hand force jitter caused by the feedforward channel;

[0129] Channel 3: Its principle is to establish a non-linear friction force model based on the measurement of the rotation angle and rotation speed and the friction force properties of the system, and to incompletely cancel the friction force of the system itself, so that the sudden value of the hand force at the commutation instant is less than the human perception threshold, to ensure the smoothness of the hand force. Its expression is:

[0130]

[0131] Where, ufc represents the control instruction of channel 3 to the road feel motor, used to compensate the nonlinear friction of the system, f model represents the nonlinear friction model, in this embodiment, the nonlinear friction model adopts the gradual friction model.

[0132] The input of the open-closed loop smooth switching strategy is the column torque sensor torque T s , the gradual control is directly performed on channel 2, the open-closed loop control is switched on channel 1, and the feedforward open-closed loop coefficient of channel 2 is:

[0133]

[0134] wherein β ffd is the feedforward open-closed loop coefficient, τ ffd is the time constant, and f ffd (T s ) is the feedforward open-closed loop steady-state value, and s is the Laplace transform complex frequency.

[0135] The closed loop open-closed loop coefficient of channel 1 is:

[0136]

[0137] wherein β fb is the closed loop open-closed loop coefficient, τ fb is the time constant, and f fb (T s ) is the closed loop open-closed loop steady-state value.

[0138] The feedforward open-closed loop steady-state value f ffd (T s ) and the closed loop open-closed loop steady-state value f fb (T s ) change with the absolute value of the torque sensor measurement value in the range of the range as shown in the following formula (1) and (2): Figure 3 In the switching process, the feedforward open-closed loop steady-state value f ffd (T s ) is gradual, and the closed loop open-closed loop steady-state value f fb (T s ) is a sudden change, and the specific strategy is:

[0139] When the absolute value of the torque sensor torque is less than the threshold T s_th1 , the values of the feedforward open-closed loop steady-state value f ffd (T s ) and the closed loop open-closed loop steady-state value f fb (T s ) are respectively:

[0140] f ffd (T s ) = 0

[0141] f fb (T s )=1;

[0142] The closed-loop switch is always 1, and the open-loop switch is always 0, that is, the closed-loop control channel and the friction force feedforward compensation channel are in working state to realize torque closed-loop tracking control in normal operating conditions.

[0143] When the absolute value of the torque sensor torque exceeds the threshold T s_th1 , but does not exceed the threshold T s_th2 , the values of the feedforward switch steady-state value f ffd (T s ) and the closed-loop switch steady-state value f fb (T s ) are respectively:

[0144]

[0145] f fb (T s )=1;

[0146] The closed-loop switch is always 1, and the open-loop switch is semi-closed, and the feedforward channel is not completely involved, at this time the system is controlled by the closed-loop control channel, the friction force feedforward compensation channel and the incomplete feedforward control channel.

[0147] When the absolute value of the torque sensor torque exceeds the threshold T s_th2 , the values of the feedforward switch steady-state value f ffd (T s ) and the closed-loop switch steady-state value f fb (T s ) are respectively:

[0148] f ffd (T s )=1

[0149] f fb (T s )=0;

[0150] The closed-loop switch is quickly set to 0, and the open-loop switch is completely closed, at this time the system is controlled by the friction force feedforward compensation channel and the feedforward channel.

[0151] The speed of feedforward intervention or exit is determined by the difference between the two thresholds (T s_th2 -T s_th1 ) and the time constant τ ffd , the gradual intervention or exit and the faster response speed of the closed-loop system performance ensure that the intervention or exit of the feedforward channel will not cause a sudden change in torque, and the intervention or exit of the closed-loop is only determined by the time constant τ fbThe decision, although it will cause the mutation of closed-loop controller input, but due to the over-damping characteristics of the closed-loop system performance, ensure that the closed-loop channel intervention or exit will not cause the torque mutation.

[0152] In summary, the role of open and closed loop smooth switching strategy is: in the normal operating conditions, that is, the torque sensor torque within the threshold T s_th1 , the hand force control unit only channel 1 and channel 3 work, to ensure the high precision closed-loop tracking control of hand force; in the end condition, that is, the torque sensor torque exceeds the range, the hand force can still be provided by channel 2, to ensure the safety limit; when the open and closed loop switching, the hand force is smooth without mutation.

[0153] Hand force design unit: including rack force estimation module and expected hand force calculation module, the rack force estimation module estimates the load at the rack (rack force) by using the steering actuator torque sensor and rack displacement sensor, and then calculates the vehicle's return torque and adjustable road feel feedback, the rack force estimation module is based on the model of steering actuator, the expression based on the model of steering actuator is:

[0154]

[0155] Where, m r and c r are the mass and damping coefficient of the steering actuator equivalent to the rack, x r is the rack displacement, F r , f r and F m are the rack force to be estimated, the friction of the steering actuator equivalent to the rack and the force of the motor equivalent to the rack (motor force) respectively.

[0156] The goal of the rack force estimation module is: high steady-state accuracy; the bandwidth of the estimation module can be adjusted to control the frequency components of the estimation results; ensure that the fluctuations of the estimation results caused by sensor noise are small enough at different bandwidths; the phase lag corresponding to different bandwidth filters is basically consistent.

[0157] The rack force estimation method of the rack force estimation module is to use inverse model + low-pass filter to estimate the rack force. First, the nominal value F′ r of the rack force estimation is obtained, the calculation formula of the nominal value F r ′ of the rack force estimation is:

[0158]

[0159] Where, F m and x r are obtained through the steering actuator output torque sensor and rack displacement sensor respectively, friction force calculated by using the friction force model, i.e. the friction force compensation term;

[0160] In order to adjust the bandwidth of the rack force estimation module and suppress the fluctuation of sensor noise in the estimation result, the nominal value of the obtained rack force estimation F r The estimated rack force is obtained by low-pass filtering, and the expression of the estimated rack force is:

[0161] F rack_est = H lowpass * F r ′

[0162] Wherein, F rack_est is the estimated rack force, H lowpass (s) is the transfer function of the low-pass filter, in the present embodiment, the low-pass filter is a third-order low-pass filter, p1, p2 and p3 in the third-order low-pass filter are three poles required to be configured by the filter, the pole p1 is used to adjust the bandwidth, the poles p2 and p3 are used to adjust the phase lag of the filter, according to the bandwidth B w of the feedback road feel in the present embodiment, the adjustable range is [10Hz, 25Hz], and the pole configuration method is:

[0163] When the bandwidth B w = 25Hz, the transfer function H lowpass (s) is set as a third-order Butterworth low-pass filter with a cutoff frequency of 25Hz, and the three configured poles are respectively:

[0164] p1 = -B w ,

[0165] Wherein, the maximum amplitude gain of the transfer function H lowpass (s) (m r s 2 + c r s) is A gain (25Hz), which represents the maximum gain of the fluctuation of the rack force estimation value caused by the sensor noise;

[0166] When the bandwidth B w = 10Hz, p1 controls the bandwidth, and p2 and p3 constitute a second-order butterworth filter with a bandwidth of B w1 , and the three configured poles are respectively:

[0167] p1 = -B w ,

[0168] Adjust B w1 so that the transfer function H lowpass (s) (mr s 2 +c r The maximum amplitude gain of s) is A gain (10Hz)≈A gain (25Hz), while simultaneously lags the phase angle with B w =Approximately 25Hz operating conditions;

[0169] In bandwidth B w When ∈ (10Hz, 25Hz), linear interpolation is performed on poles p1, p2, and p3, and the transfer function H is achieved by appropriately adjusting the positions of poles p2 and p3. lowpass (s)(m r s 2 +c r The maximum amplitude gain and phase lag of s) are basically consistent.

[0170] Desired Torque Calculation Module: Used to calculate the desired feedback hand force T cmd (Expected hand force), expected feedback hand force T cmd It consists of six torques, namely the main control torque T main Stiffness moment T stiff Adaptive high-speed damping torque T damp Low-speed return torque T ar End-limiting torque T limit and simulated frictional torque T fric Expected feedback force T cmd The expression is:

[0171] T cmd =T main +T stiff +T damp +T limit +T ar +T fric ;

[0172] Main control torque T main It is one of the main components of the expected hand force, the main operating torque T main The expression is:

[0173] T main =ρ·f main (F rack_est )

[0174] Where ρ is an adjustable interface, representing the clarity of road feel, and F... rack_est f represents the rack force estimate obtained from the rack force estimation module. main (·) is a function that generates the main control force based on the rack force estimate. It can be highly designed, and its curve shape is as follows: Figure 5As shown in (a), in the middle position, the driver pays more attention to the rack force and requires a good sense of the middle position, therefore f main The gradient of (·) is relatively large; in non-middle positions, the driver's attention to rack force decreases, requiring a lower driving load, therefore f main The gradient of (·) is small.

[0175] Stiffness moment T stiff Another major component of the expected hand force is the stiffness moment T. stiff The expression is:

[0176] T stiff =f stiff (F rack_virtual ,ρ,γ·η(V))

[0177] Among them, T stiff ρ is the stiffness torque, ρ is the adjustable interface, representing the road feel clarity, and its main control torque T is... main In expected hand strength T cmd The components, such as Figure 5 As shown, when ρ is small, it provides a comfortable feel, the proportion of the main control torque decreases, and the proportion of the stiffness torque increases. In the intermediate and non-intermediate positions, the absolute value of the stiffness torque |T stiff |All depend on the absolute value of the steering wheel angle|θ sw The torque increases with increasing torque, but the torque gradient is greater in the middle position. The driver's handling feel mainly comes from the stiffness torque T. stiff It exhibits good smoothness and stability in terms of force application; when ρ is large, it provides a sporty feel, with a desired torque T. cmd In the middle, the main control torque T main The proportion increases, stiffness moment T stiff As the proportion decreases, the absolute value of the stiffness moment |T is in the middle position. stiff |Absolute value of steering wheel angle|θ sw |Increases and increases, assisting in improving the hand force gradient in the middle position, enhancing the driver's sense of the middle position, while in the non-middle position, the absolute value of the stiffness torque|T stiff |Absolute value of steering wheel angle|θ sw | Increases and decreases to highlight the road feel when the vehicle skids on low-friction surfaces, alerting the driver when lateral acceleration is large. γ is an adjustable interface representing the level of steering force and gradient adjustment factor, such as... Figure 6 As shown in (a), a smaller γ results in a comfortable feel with less driving burden, while a larger γ results in a sporty feel with a clearer grip. η(V) is an adjustable interface representing the gradient adjustment factor between hand force and vehicle speed, allowing adjustment of hand force at different vehicle speeds. Figure 6As shown in (b), the gradient of comfort-type hand force decreases with increasing vehicle speed, resulting in less driving burden; while the gradient of sport-type hand force increases with increasing vehicle speed, providing a clearer perception of vehicle speed for the driver. V represents vehicle speed, and F... rack_virtual (V, θ) sw The virtual rack force is calculated based on the rotation angle and vehicle speed, and its expression is:

[0178] F rack_virtual =K stiff (V)·θ sw

[0179] Among them, K stiff (V) represents the stiffness coefficient at different vehicle speeds V. stiff是 The function that generates stiffness moment based on the virtual value of rack force can be highly designed, and its curve shape is as follows: Figure 5 As shown in (b), it is a two-part piecewise function used to independently generate stiffness moments at intermediate and non-intermediate positions. The function f generates stiffness moments based on the virtual value of the rack force. stiff The expression is:

[0180]

[0181] Among them, f stiff1 (·) is used to generate the stiffness moment at the intermediate position, f stiff2 (·) is used to generate stiffness moments at non-intermediate positions, F thes1 To differentiate the virtual rack force threshold between the intermediate and non-intermediate positions at different vehicle speeds;

[0182] Mode-adaptive high-speed damping torque T damp The expression is:

[0183]

[0184] Where, θ sw For steering wheel angle, c is the steering wheel speed. damp Let be the damping coefficient, which is a function of vehicle speed V, the adjustable factor ρ for road feel clarity, and the product γ·η(V) of the other two adjustable factors. For simplicity, in this embodiment, the adjustable factor ρ for road feel clarity is discretized into two levels (ρ = 0.5, ρ = 1.0), where ρ = 0.5 is the comfort mode (vague road feel) and ρ = 1.0 is the sport mode (clear road feel). The other adjustable factors γ and η(V) are continuously adjustable. Therefore, c damp The settings can be equivalent to two Map diagrams, corresponding to the two gears ρ=1.0 and ρ=0.5 respectively. The setting for ρ=1.0 is as follows: Figure 10As shown, the two plane coordinates of the Map are vehicle speed V and the product of two adjustable factors γ·η(V), and the Map is obtained by frequency domain analysis and time domain analysis of key operating conditions (grid points in the Map) selected at equal intervals, to ensure that the stability of the hand force ring remains basically unchanged after mode switching (the frequency domain performance is basically unchanged, and the time domain performance is basically unchanged) (γ·η=1.0), and then linear interpolation is performed in two dimensions;f dt (θ sw ) is an adjustment factor of the damping coefficient, considering that the driver pays high attention to the feeling of the intermediate position, the stickiness of the damping control to the intermediate position should be reduced as much as possible, and the adjustment factor f dt (θ sw ) of the damping coefficient is set as follows:

[0185]

[0186] wherein p middle represents the weakening degree of the damping coefficient of the intermediate position, and the value range is [0.5, 1.0], θ sw_dpthes is the intermediate position boundary threshold of the damping coefficient adjustment.

[0187] The low-speed return torque T ar is used to solve the problem of insufficient hand release return in low-speed operating conditions, and also increases the intermediate position feeling in the low-speed operating condition control process, and the design method of the low-speed return torque T ar uses extended P control, and the expression of the low-speed return torque T ar is as follows:

[0188]

[0189] wherein K ar ·θ sw represents the conventional P control, K ar represents the return strength, σ V (V) represents the vehicle speed factor, which is used to make the low-speed return torque intervene at low speed and shield at high speed, is a rotation angle adjustment factor, which makes the intermediate position feeling stronger and the influence of the torque smaller at non-intermediate positions, is a rotation speed adjustment factor, which forces the active return torque to be 0 at the place where the rotation speed is 0, to avoid the impact of the hand force at the moment of reversing, represents a return judgment factor, which is used to ensure that the active return torque only intervenes in the return process and is shielded in the positive control process.

[0190] The end limit torque T limitSpecifically, its function is as follows: when the driver turns the steering wheel beyond the maximum permissible steering wheel angle, it outputs a large reaction force to prevent the driver from continuing to turn the steering wheel beyond the specified value. To ensure a good limiting effect, in this embodiment, the end-limiting torque T during the working condition is... limit The torque is approximately 20 Nm, which exceeds the range of the torque sensor [-10 Nm, 10 Nm]. In conjunction with the open-loop and closed-loop smooth switching strategy in this invention, the feedforward channel can still provide limiting force and safety limit even after the sensor's range is exceeded or the closed loop fails.

[0191] Simulated frictional torque T fric Specifically, its function is to simulate the friction in existing steering systems and electric power steering systems within the online steering system, thereby improving the feel and simulating the friction torque T. fric The expression is:

[0192]

[0193] in, This is a friction force model based on the tanh(·) function.

[0194] The interactive settings unit integrates four independent, continuously adjustable interfaces for driver settings, ensuring highly adjustable hand force. These four interfaces correspond to four dimensions of adjustable hand force: adjustable road feel bandwidth, road feel clarity, adjustable steering force level and gradient, and adjustable steering force vs. vehicle speed gradient. The corresponding adjustable factors are: road feel bandwidth adjustable factor B. w , Road feel clarity adjustment factor ρ, Handling force level and gradient adjustable factor γ, and Handling force vs. vehicle speed gradient factor η(V):

[0195] Road sense bandwidth adjustable factor B w Road sense bandwidth adjustable factor B w A larger value corresponds to a sporty feel, allowing the driver to obtain more high-frequency road surface information through steering effort; the road feel bandwidth is adjustable by factor B. w When the size is smaller, it corresponds to a more comfortable feel, and the driver's hand force is smoother;

[0196] Road feel clarity adjustment factor ρ: When the road feel clarity factor ρ is larger, it corresponds to a sporty feel. The feel of the vehicle skidding on uneven or low-friction surfaces is more obvious, and the road feel is clearer. The driver can obtain more specific road information through steering force. When the road feel clarity factor ρ is smaller, it corresponds to a comfortable feel. The road information is less reflected in the steering force, and the driver's steering force is smoother.

[0197] The steering force level and gradient adjustable factor γ: when the steering force level and gradient adjustable factor γ is large, the corresponding motion type hand feeling, the hand force is clear, when the steering force level and gradient adjustable factor γ is small, the corresponding comfortable type hand feeling, the driving burden is smaller;

[0198] Steering force VS vehicle speed gradient factor η(V): comfortable hand feeling with increasing gradient of vehicle speed, less driving burden, motion type hand feeling with increasing gradient of vehicle speed, clearer vehicle speed perception of the driver.

[0199] The four dimensions can be set to a connected adjustable interface, or calibrated to a plurality of discrete gears.

[0200] The algorithm verification of the present application is carried out:

[0201] The simulation effect of the device is shown in Figures 11-21 , Figure 11 and Figure 12 The torque sensor range is 10Nm, and in the open-loop and closed-loop switching process, the hand force does not appear above the feeling threshold (0.2Nm), and when the hand force is outside the torque sensor range, the feedforward channel can still provide more than 20Nm hand force at the end to ensure the safety limiting of the end.

[0202] Figure 13 The different road feel bandwidth B w is set through the interactive setting unit Figure 14 The hand force time domain image is locally enlarged, and it can be seen that under the same road impact, the hand force fluctuation amplitude of the comfortable mode (B w =10Hz) is small, and the smoothness is good, the hand force fluctuation amplitude of the motion mode (B w =25Hz) is large, and the road feel is clearer, and the hand force smoothness and road feel clarity of the ordinary mode (B w =15Hz) is between the above two. Figure 15 When switching between different modes, the hand force stability margin is basically unchanged, the hand release return overshoot is small, and is basically consistent.

[0203] Figure 16 The different steering force characteristics obtained by setting different steering force level and gradient adjustment factors γ through the interactive setting unit are shown in Figure 17For setting different steering force levels and gradient adjustment factors Y of the hand-off return process by the interactive setting unit, it can be seen that the steering force level and the gradient can be continuously adjusted according to the individual needs of the driver, and the hand force stability does not change with the value of the steering force level and the gradient adjustment factor γ, and the hand-off return process basically does not appear overshoot.

[0204] Figure 18 For setting different road feel clarity adjustment factors ρ by the interactive setting unit, the steering force characteristic diagram of the vehicle driving on the same road with large unevenness, Figure 19 For ρ=0.5, the steering force characteristic diagram of the vehicle driving on the road with different adhesion coefficients (μ=0.3, μ=0.85), Figure 20 For ρ=1.0, the steering force characteristic diagram of the vehicle driving on the road with different adhesion coefficients (μ=0.3, μ=0.85), Figure 21 For setting different road feel clarity adjustment factors ρ of the hand-off return process by the interactive setting unit, it can be seen that for the comfortable hand feeling (ρ=0.5), the unevenness of the road and the change of the adhesion coefficient are not obvious in the hand force, and the hand force is smooth and stable; for the sports type hand feeling (ρ=1.0), the unevenness of the road and the change of the adhesion coefficient are obvious in the hand force, and the road feel is clear. And the hand force stability does not change with the value of the road feel clarity adjustment factor ρ, and the hand-off return process basically does not appear overshoot.

[0205] The above describes the preferred embodiments of the application in detail. It should be understood that those skilled in the art can make many modifications and changes without creative labor according to the concept of the application. Therefore, any technical solution obtained by logical analysis, reasoning or limited experiment on the basis of the prior art according to the concept of the application should be within the protection scope determined by the claims.

Claims

1. A road feel simulation device for a height-adjustable steer-by-wire system, characterized in that, The device includes: Sensor unit: Used to collect vehicle motion state information and angular torque information of the hand force simulation mechanism, and to serve as input for the hand force design unit and hand force control unit; Hand force design unit: includes rack force estimation module and desired hand force calculation module, used to obtain highly adjustable desired hand force; Hand force control unit: used to generate corresponding control commands for the road sensor motor, thereby realizing the tracking control of the actual hand force to the desired hand force; Interactive settings unit: Provides multiple continuously height-adjustable interfaces to meet the driver's personalized handling needs; The rack force estimation module uses an inverse-mode plus low-pass filtering method to estimate the rack force. The specific process is as follows: A model based on the steering actuator is established, and its expression is: in, and These are the equivalent mass and damping coefficient of the steering actuator to the rack, respectively. For rack displacement, , and These are the rack force that needs to be estimated, the frictional force equivalent to the rack from the steering actuator, and the force equivalent to the motor acting on the rack. and The values ​​are obtained respectively through the torque sensor at the output end of the steering actuator motor and the rack displacement sensor; The nominal value of the rack force estimate is obtained. The formula for calculating the nominal value of the rack force estimate is: in, Estimate the nominal value of the rack force; The obtained nominal value of rack force estimation The rack force is estimated by using a low-pass filter. The expression for the estimated rack force is as follows: in, For the estimated rack force, Let be the transfer function of the low-pass filter. Its inverse Laplace transform gives the low-pass filter a third-order low-pass filter, with the three poles configured as follows: Among them, the poles Used to adjust bandwidth, poles and extreme points Used to adjust the characteristics of the filter, setting the feedback loop bandwidth. The adjustable range is [ , The specific method for pole placement of a third-order low-pass filter is as follows: When bandwidth When setting the transfer function The cutoff frequency is The third-order Butterworth low-pass filter has the following poles configured: transfer function The maximum amplitude gain is , used to characterize the maximum gain of rack force estimation fluctuation caused by sensor noise; When bandwidth When setting the pole Control bandwidth, poles and extreme points The bandwidth is The second-order Butterworth filter has the following poles configured: Adjust bandwidth Make the transfer function The maximum gain of the amplitude is Simultaneously, the phase angle lags and the bandwidth... The working conditions at that time were similar; When bandwidth At that time, the extreme point ,pole and extreme points By bandwidth Operating conditions and bandwidth at the time The corresponding pole values ​​are obtained by linear interpolation under the operating conditions at that time.

2. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 1, characterized in that, The sensor unit includes a torque sensor at the manual force simulation mechanism column, a steering wheel angle sensor at the manual force simulation mechanism column, a steering actuator motor torque sensor, a rack displacement sensor, and a vehicle speed sensor. The torque sensor is used to measure the torque of the tubing, i.e. the actual hand force. The torque sensor works in conjunction with the open-loop and closed-loop smooth switching algorithm to achieve high-precision closed-loop control of hand force under normal operating conditions and high torque safety limit when hand force exceeds the range under end control conditions. The aforementioned angle sensor is used to measure the rotation angle of the column, i.e. the steering wheel angle. The rotation angle of the column is used for the calculation of the desired hand force in the hand force design unit and the feedforward compensation of the friction force in the hand force control unit. The steering actuator motor torque sensor is used to measure the torque at the motor output end, i.e., the steering actuator motor torque, which is used to estimate the rack force in the manual force design unit. The rack displacement sensor is used to measure rack displacement, and the rack displacement is used to estimate rack force in the manual force design unit; The vehicle speed sensor is used to measure vehicle speed, and the vehicle speed is used to calculate the desired hand force in the hand force design unit.

3. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 1, characterized in that, The desired hand force calculation module obtains the desired hand force based on the estimated rack force, vehicle speed, and steering wheel angle. The desired hand force consists of six torques, including a height-adjustable main control torque. Height-adjustable stiffness torque Mode-adaptive high-speed damping torque Simulated friction torque Low-speed active self-aligning torque and end control torque ; The height-adjustable main control torque The expression is: in, Main control torque, This is the road feel clarity adjustment factor, representing the main control force torque. In its relationship with stiffness moment The proportion of the two, The estimated rack force obtained by the rack force estimation module. For rack force estimation Generate main control force The function; The height-adjustable stiffness torque The expression is: in, For stiffness moment, For the control force level and gradient adjustment factor, For the control force vs. vehicle speed gradient adjustment factor, For virtual values ​​based on rack force Generating stiffness moment The function is a two-part piecewise function used to independently generate stiffness moments at intermediate and non-intermediate positions. For angle-based and vehicle speed The calculated virtual rack force; Virtual rack force The expression is: The stiffness coefficients at different vehicle speeds For the corner; Based on rack force virtual value Generating stiffness moment function The expression is: in, Used to generate stiffness moment at intermediate position, Used to generate stiffness moments at non-intermediate positions. To differentiate the virtual rack force thresholds at the intermediate and non-intermediate positions at different vehicle speeds, This is a factor for adjusting road feel clarity. For the control force level and gradient adjustment factor, The adjustment factor is the ratio of control force to vehicle speed gradient. The aforementioned mode-adaptive high-speed damping torque The expression is: in, For mode-adaptive high-speed damping torque, For mode-adaptive damping coefficient, This is a factor influencing the damping coefficient based on the steering wheel angle, used to reduce damping in the small steering angle working area, thereby reducing the stickiness caused by active damping in the small steering angle area. Steering wheel speed; The simulated friction torque Used to simulate the frictional force of mechanical systems or EPS systems, simulating frictional torque. The expression is: in, Based on Frictional force model of a function; The low-speed active self-aligning torque Used to ensure self-alignment performance under low-speed conditions, low-speed active self-alignment torque The expression is: in, Indicates the strength of the positive return. Represents the vehicle speed factor. This is the rotation angle adjustment factor. For speed adjustment factor, Indicates the positive return judgment factor; The end control torque Used to ensure the safe limit of the steering wheel.

4. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 1, characterized in that, The controlled object of the aforementioned hand force control unit is a hand force simulation mechanism, and the dynamic characteristic equation of the hand force simulation mechanism is: in, and These represent the moments of inertia of the steering wheel and the road sensor motor, respectively. and These represent the linear damping of the steering wheel and the road sensor motor, respectively. This indicates the stiffness of the torque sensor. , , and These represent the driver's control force, the torque sensor torque, the torque at the motor output, and the nonlinear friction force, respectively. For steering wheel angle, The rotation angle of the road sensor motor; The specific control objective of the hand force control unit is to transmit control commands to the road sensor motor to control the output torque of the road sensor motor. This allows the actual hand force to track the desired hand force.

5. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 4, characterized in that, The hand force control unit controls the hand force based on three control channels and generates control commands for the road feel motor. The three control channels are a closed-loop control channel, a feedforward control channel, and a nonlinear friction force feedforward compensation channel. The closed-loop control channel adopts a torque closed-loop controller designed with a lag-lead compensation device based on the desired frequency characteristic method, and then performs closed-loop tracking control of the system based on the deviation between the desired hand force and the actual hand force measured by the torque sensor. The aforementioned feedforward control channel is specifically an open-loop control system, used to continue transmitting control commands to the road sensor motor after the torque exceeds the range, thereby continuing to provide manual force. The expression for the control command is: in, For torque, To achieve the desired hand strength, This indicates the active damping applied to the open-loop control to reduce hand jitter caused by the feedforward control channel; The aforementioned nonlinear friction force feedforward compensation channel is used to perform feedforward compensation on the system friction force.

6. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 5, characterized in that, The opening or closing of the closed-loop control channel and the feedforward control channel is controlled by an open-loop / closed-loop smooth switching strategy, the input of which is the sensor torque. The feedforward control channel is subjected to gradual control, and the closed-loop control channel is subjected to switching control. The switching quantities of the two channels in the open-loop switching strategy are respectively the feedforward switching coefficients. and closed-loop switching coefficient The expressions are as follows: in, For feedforward switching coefficients, This is the steady-state value of the feedforward switch. For Laplace transform complex frequencies, This represents the closed-loop switching coefficient. and All are time constants. This represents the steady-state value of the closed-loop switch.

7. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 6, characterized in that, The properties of the two channel switching quantities in the described open-loop switching strategy are as follows: The feedforward switching coefficient The range of values ​​is The feedforward switching coefficient Specifically, the threshold during intervention or exit. and threshold The continuous change in the coefficient of force between the sensor and the feedforward control variable, which varies linearly with the sensor torque and is connected in series with a first-order inertial element, is used to gradually control the feedforward control variable. The feedforward switching coefficient is... The speed of intervention or withdrawal is determined by the time constant. and the difference between the two thresholds ( )Decide; The closed-loop switching coefficient The range of values ​​is The closed-loop switching coefficient mentioned above Specifically, during the intervention or withdrawal process, a switching quantity of a first-order inertial element is connected in series to control the tracking error input of the closed-loop controller. The closed-loop switching coefficient... The speed of intervention or withdrawal is determined by the time constant. Decide.

8. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 7, characterized in that, The aforementioned open-loop / closed-loop smooth switching strategy is specifically as follows: When the sensor torque Less than the threshold At this time, the closed-loop switch is always 1 and the open-loop switch is always 0, that is, the closed-loop control channel and the friction force feedforward compensation channel are in working state to realize torque closed-loop tracking control in normal operating conditions. When the sensor torque Greater than or equal to the threshold Less than the threshold At this time, the closed-loop switch is always 1, the open-loop switch is half closed, and the feedforward channel is not fully engaged. At this time, the system is controlled by the closed-loop control channel, the friction feedforward compensation channel, and the incomplete feedforward control channel. When the sensor torque Greater than or equal to the threshold When the closed-loop switch is quickly set to 0, the open-loop switch is fully closed, and the system is controlled by the friction force feedforward compensation channel and the feedforward channel.

9. The road feel simulation device for a height-adjustable steer-by-wire system according to claim 1, characterized in that, The interactive setting unit includes four independent interfaces corresponding to four adjustable dimensions of hand force, allowing drivers to personalize settings through the human-machine interface. The four dimensions can be freely combined and continuously adjusted according to the driver's needs, and sport and comfort modes can be calibrated. These interfaces are: adjustable road feel feedback bandwidth interface, road feel clarity interface, adjustable steering force level and gradient interface, and adjustable steering force vs. vehicle speed gradient interface. The adjustable road feel feedback bandwidth interface is used to set the road feel bandwidth of the rack force estimation module. The road feel clarity interface, adjustable steering force level and gradient interface, and adjustable steering force vs. vehicle speed gradient interface are used to set the hand force characteristics of the desired hand force calculation module. The adjustable dimensions of hand force include adjustable road feel bandwidth, adjustable road feel clarity, adjustable steering force level and gradient, and adjustable steering force vs. vehicle speed gradient. The adjustable factors corresponding to these adjustable dimensions are respectively the road feel bandwidth adjustable factors. Road feel clarity adjustment factor Maneuvering level and gradient adjustable factor and handling force vs. vehicle speed gradient factor ; The adjustable feedback road feel bandwidth specifically refers to adjusting the feedback road feel. The method used is the adaptive configuration of the poles of the rack force filter. When switching to comfort mode, the bandwidth of the feedback road feel is set to a small value to ensure smoothness of hand force. When switching to sport mode, the bandwidth of the feedback road feel is set to a large value to allow the driver to obtain more vehicle status and road information through hand force. The adjustable road feel clarity specifically refers to adjusting the proportion of the main control torque and stiffness torque in the desired hand force at the middle and non-middle positions, respectively. When switching to the sports mode, the proportion of the main control torque is large and the proportion of the stiffness torque is small. The peak-to-peak value of the hand force fluctuation caused by the same road surface unevenness is large, and the hand force on low-friction road surfaces is more obvious, thus increasing the road feel clarity. When switching to comfort mode, the proportion of main control torque is small, the proportion of stiffness torque is large, and the hand force is smooth and stable. The adjustable handling force level and gradient height specifically refers to adjusting the handling force level and gradient. Under the same vehicle speed conditions, when switching to sport mode, the handling force level and gradient are large, and when switching to comfort mode, the handling force level and gradient are small. The adjustable gradient height between control force and vehicle speed specifically refers to adjusting the gradient of control force as vehicle speed changes. When switching to Sport mode, the gradient increases with vehicle speed, enhancing the driver's perception of speed. When switching to Normal mode, the gradient remains constant with vehicle speed. When switching to Comfort mode, the gradient decreases with vehicle speed to reduce the driving burden caused by increased vehicle speed.

Images