JP6623771B2 - Control method and control device for variable damping force shock absorber - Google Patents

Description

  The present invention relates to a method and a device for controlling a variable damping force shock absorber.

  As a technique for controlling a suspension of a vehicle, a technique described in Patent Document 1 is disclosed. This publication discloses a technique for estimating a stroke speed of a suspension from a fluctuation component of a wheel speed with respect to a vehicle body speed and performing skyhook control.

JP 2009-241813 A

  However, when tires having different front and rear moving radii (hereinafter, referred to as front and rear different diameter tires) are mounted, the fluctuation component of the wheel speed calculated from the vehicle body speed and the wheel speed is offset in an increasing or decreasing direction, There was a problem that the stroke speed could not be calculated with high accuracy, and an appropriate damping force control amount could not be calculated.

  The present invention has been made in view of the above problem, and has as its object to accurately estimate a stroke speed and appropriately control a damping force even when front and rear tires are attached.

  In order to achieve the above object, according to the present invention, among the fluctuation components of the wheel speed calculated from the vehicle speed and the wheel speed, the offset value superimposed by mounting tires having different diameters on the front wheel and the rear wheel is reduced. Calculate the stroke speed of the damping force variable shock absorber based on the corrected wheel speed fluctuation component, and use the skyhook control based on the stroke speed to perform damping force control that suppresses changes in the sprung behavior of the vehicle. Do.

  Therefore, even when tires having different diameters are attached to the front wheel and the rear wheel, the stroke value is calculated after the offset value is reduced. The desired damping force control can be achieved.

FIG. 1 is a system schematic diagram illustrating a control device for a vehicle according to a first embodiment. FIG. 2 is a control block diagram illustrating a control configuration of a vehicle control device according to the first embodiment. FIG. 2 is a conceptual diagram illustrating a configuration of a wheel speed feedback control system according to the first embodiment. FIG. 3 is a control block diagram illustrating a configuration of a traveling state estimation unit according to the first embodiment. FIG. 4 is a control block diagram illustrating control contents in a stroke speed calculation unit according to the first embodiment. FIG. 3 is a block diagram illustrating a configuration of a reference wheel speed calculation unit according to the first embodiment. It is a schematic diagram showing a vehicle body vibration model. FIG. 4 is a control block diagram illustrating a configuration of roll rate suppression control according to the first embodiment. 6 is a time chart illustrating an envelope waveform forming process of roll rate suppression control according to the first embodiment. FIG. 3 is a block diagram illustrating a control configuration of unsprung vibration control according to the first embodiment. FIG. 4 is a control block diagram illustrating a control configuration of a damping force control unit according to the first embodiment. 5 is a time chart illustrating a limit value process in a GEO conversion unit. FIG. 13 is a control block diagram illustrating control contents in a stroke speed calculation unit according to the second embodiment. FIG. 10 is a block diagram illustrating a configuration of a tire moving radius correction unit according to a second embodiment.

[Example 1]
FIG. 1 is a system schematic diagram illustrating a control device of a vehicle according to a first embodiment. The vehicle includes an engine 1 as a power source and a brake 20 for generating braking torque by frictional force on each wheel (hereinafter, when displaying brakes corresponding to individual wheels, right front wheel brake: 20FR, left front wheel brake: 20FL) , Right rear wheel brake: 20 RR, left rear wheel brake: 20 RL) and a shock absorber 3 (hereinafter referred to as S / A) provided between each wheel and the vehicle body and capable of variably controlling damping force. When displaying the S / A corresponding to the individual wheels, the right front wheel S / A: 3FR, the left front wheel S / A: 3FL, the right rear wheel S / A: 3RR, and the left rear wheel S / A: 3RL are described. And).

  The engine 1 includes an engine controller (hereinafter, also referred to as an engine control unit, which corresponds to a power source control unit) 1a for controlling a torque output from the engine 1, and the engine controller 1a includes a throttle valve opening of the engine 1 and By controlling the fuel injection amount, ignition timing, and the like, a desired engine operating state (engine speed and engine output torque) is controlled. The brake 20 generates a braking torque based on the hydraulic pressure supplied from the brake control unit 2 that can control the brake hydraulic pressure of each wheel according to the running state. The brake control unit 2 includes a brake controller (hereinafter, also referred to as a brake control unit) 2a that controls a braking torque generated by the brake 20, and a master cylinder pressure generated by a driver's operation of a brake pedal or a built-in motor. The pump pressure generated by the driving pump is used as a hydraulic pressure source, and a desired hydraulic pressure is generated in the brakes 20 of the respective wheels by opening and closing a plurality of solenoid valves.

  S / A3 is a damping force generating device for damping the elastic motion of a coil spring provided between the unsprung part (axle, wheels, etc.) and the sprung part (vehicle body, etc.) of the vehicle. Is configured to be variable. The S / A 3 has a cylinder filled with a fluid, a piston that strokes in the cylinder, and an orifice that controls fluid movement between fluid chambers formed above and below the piston. Further, an orifice having a plurality of kinds of orifice diameters is formed on the piston, and when the S / A actuator is operated, an orifice according to a control command is selected from the plurality of kinds of orifices. Thereby, a damping force according to the orifice diameter can be generated. For example, if the diameter of the orifice is small, the movement of the piston is likely to be restricted, so that the damping force is high. If the diameter of the orifice is large, the movement of the piston is not easily restricted, and the damping force is small.

  In addition to the selection of the orifice diameter, for example, an electromagnetic control valve is arranged on a communication path connecting fluids formed above and below the piston, and the damping force is set by controlling the opening and closing amount of the electromagnetic control valve. There is no particular limitation. The S / A3 has an S / A controller 3a (corresponding to a damping force control means) for controlling the damping force of the S / A3, and controls the damping force by operating the orifice diameter by the S / A actuator.

  Further, a wheel speed sensor 5 for detecting the wheel speed of each wheel (hereinafter, when displaying wheel speeds corresponding to individual wheels, right front wheel speed: 5FR, left front wheel speed: 5FL, right rear wheel speed: 5RR) , Left rear wheel speed: 5RL), an integrated sensor 6 for detecting longitudinal acceleration, yaw rate and lateral acceleration acting on the center of gravity of the vehicle, and a steering angle which is a steering operation amount of the driver. Steering angle sensor 7, vehicle speed sensor 8 for detecting vehicle speed, engine torque sensor 9 for detecting engine torque, engine speed sensor 10 for detecting engine speed, and master pressure sensor 11 for detecting master cylinder pressure A brake switch 12 that outputs an ON state signal when a brake pedal is operated, and an accelerator opening sensor 13 that detects an accelerator pedal opening. The signals of these various sensors are input to the engine controller 1a, the brake controller 2a, and the S / A controller 3a as necessary. In the engine controller 1a and the brake controller 2a, vehicle behavior control (hereinafter referred to as VDC) for stabilizing the behavior of the vehicle at the time of braking slip, driving slip, and skidding is appropriately performed. The arrangement of the integrated sensor 6 may be at the position of the center of gravity of the vehicle, or may be at any other location as long as various values at the position of the center of gravity can be estimated, and are not particularly limited. Further, it is not necessary to be an integrated type, and the yaw rate, the longitudinal acceleration, and the lateral acceleration may be individually detected.

(Overall configuration of vehicle control device)
In the vehicle control device according to the first embodiment, the S / A3 is used to control a vibration state generated on a spring. At S / A3, skyhook control is performed. At this time, without using a stroke sensor or a sprung vertical acceleration sensor generally required for the skyhook control, the skyhook control can be performed with an inexpensive configuration using wheel speed sensors mounted on all vehicles. Realize. Hereinafter, specific contents for realizing this will be described.

FIG. 2 is a control block diagram illustrating a control configuration of the vehicle control device according to the first embodiment. In the first embodiment, the S / A controller 3a is provided as a controller, and forms a wheel speed feedback control system.
(Configuration of S / A controller)
The S / A controller 3a includes a driver input control unit 31 that performs driver input control for achieving a desired vehicle posture based on driver operations (such as steering operation, accelerator operation, and brake pedal operation), and detection values of various sensors. A first traveling state estimating unit 32 for estimating a traveling state based on (mainly the wheel speed sensor value of the wheel speed sensor 5), and a sprung mass damper for controlling a vibration state on a spring based on the estimated traveling state. A control unit 33, an unsprung vibration suppression control unit 34 for controlling the unsprung vibration state based on the estimated traveling state, a shock absorber attitude control amount output from the driver input control unit 31, and a sprung vibration suppression Based on the sprung mass damping control amount output from the control unit 33 and the unsprung mass damping control amount output from the unsprung mass damping control unit 34, a damping force to be set in S / A3 is determined, / And a damping force control unit 35 for performing the damping force control of the A.

(About the running state estimation unit)
First, a first traveling state estimating unit which is a common configuration provided in each feedback control system will be described. FIG. 4 is a control block diagram illustrating a configuration of the first traveling state estimation unit according to the first embodiment. In the first traveling state estimating unit 32 of the first embodiment, basically, based on the wheel speed detected by the wheel speed sensor 5, the stroke of each wheel used for the skyhook control of the sprung mass damping control unit 33 described below. Calculate speed, bounce rate, roll rate and pitch rate. First, the value of the wheel speed sensor 5 of each wheel is input to the stroke speed calculation unit 321, and the sprung speed is calculated from the stroke speed of each wheel calculated by the stroke speed calculation unit 321.

  FIG. 5 is a control block diagram illustrating control contents in the stroke speed calculation unit of the first embodiment. The stroke speed calculation unit 321 is provided individually for each wheel, and the control block diagram shown in FIG. 5 is a control block diagram focusing on a certain wheel. The stroke speed calculation unit 321 includes a value of the wheel speed sensor 5, a front wheel steering angle δf detected by the steering angle sensor 7, and a rear wheel steering angle δr (actual rear wheel steering when a rear wheel steering device is provided). And a reference wheel speed calculation unit 300 that calculates a reference wheel speed based on the vehicle body lateral speed and the actual yaw rate detected by the integrated sensor 6. A tire rotation vibration frequency calculation unit 321a for calculating a tire rotation vibration frequency based on the calculated reference wheel speed, and a deviation calculation unit 321b for calculating a deviation (wheel speed fluctuation) between the reference wheel speed and the wheel speed sensor value. A high-pass filter 321f for filtering the deviation calculated by the deviation calculator 321b and removing the DC component, and applying a predetermined upper limit value and lower limit value to the deviation from which the DC component has been removed. The GEO converter 321c converts the filtered deviation within the lower limit range into a suspension stroke amount, a stroke speed calibration unit 321d that calibrates the converted stroke amount to a stroke speed, and a stroke speed calibration unit 321d. A signal processing unit 321e that removes a tire rotation primary vibration component by applying a band elimination filter corresponding to the frequency calculated by the tire rotation vibration frequency calculation unit 321a to the calculated value and calculates a final stroke speed. . The high-pass filter 321f will be described later in detail.

[Reference wheel speed calculator]
Here, the reference wheel speed calculation unit 300 will be described. FIG. 6 is a block diagram illustrating a configuration of the reference wheel speed calculation unit according to the first embodiment. The reference wheel speed refers to a value at which various disturbances are removed from each wheel speed. In other words, the reference wheel speed is a value that shows a strong correlation with the vehicle speed, and the difference between the wheel speed sensor value and the reference wheel speed is caused by the bouncing behavior, roll behavior, pitch behavior, or unsprung vertical vibration of the vehicle body. This value is related to a component that fluctuates according to the stroke. In the embodiment, the stroke speed is estimated based on this difference.

  The plane motion component extraction unit 301 calculates a first wheel speed V0 as a reference wheel speed of each wheel based on the vehicle body plan view model with the wheel speed sensor value as an input. Here, the wheel speed sensor value detected by the wheel speed sensor 5 is ω (rad / s), the front wheel actual steering angle detected by the steering angle sensor 7 is δf (rad), and the rear wheel actual steering angle is δr (rad). ), The vehicle body lateral speed is Vx, the yaw rate detected by the integrated sensor 6 is γ (rad / s), the vehicle body speed estimated from the calculated reference wheel speed ω0 is V (m / s), the reference to be calculated. The wheel speed is VFL, VFR, VRL, VRR, the front wheel tread is Tf, the rear wheel tread is Tr, the distance from the vehicle center of gravity to the front wheel is Lf, and the distance from the vehicle center of gravity to the rear wheel is Lr. Using the above, the vehicle body plan view model is represented as follows.

(Equation 1)
VFL = (V−Tf / 2 · γ) cosδf + (Vx + Lf · γ) sinδf
VFR = (V + Tf / 2 · γ) cosδf + (Vx + Lf · γ) sinδf
VRL = (V−Tr / 2 · γ) cosδr + (Vx−Lr · γ) sinδr
VRR = (V + Tr / 2 · γ) cosδr + (Vx−Lr · γ) sinδr
Assuming that the vehicle is running normally without side slip, the vehicle body lateral speed Vx may be input as zero. When this is rewritten into a value based on V in each equation, it is expressed as follows. In this rewriting, V is described as V0FL, V0FR, V0RL, V0RR (corresponding to the first wheel speed) as a value corresponding to each wheel.
(Equation 2)
V0FL = {VFL−Lf · γsinδf} / cosδf + Tf / 2 · γ
V0FR = {VFR−Lf · γsinδf} / cosδf−Tf / 2 · γ
V0RL = {VRL + Lr.γsinδr} / cosδr + Tr / 2 · γ
V0RR = {VRR + Lf.γsinδf} / cosδr−Tr / 2 · γ

The roll disturbance removing unit 302 receives the first wheel speed V0 as input and calculates second wheel speeds V0F and V0R as reference wheel speeds of front and rear wheels based on the vehicle body front view model. The vehicle body front view model removes the wheel speed difference caused by the roll motion generated around the roll rotation center on the vertical line passing through the vehicle center of gravity when the vehicle is viewed from the front. Is done.
V0F = (V0FL + V0FR) / 2
V0R = (V0RL + V0RR) / 2
As a result, the second wheel speeds V0F and V0R from which disturbance based on the roll has been removed are obtained.

The pitch disturbance removing unit 303 calculates the third wheel speeds VbFL, VbFR, VbRL, and VbRR, which are the reference wheel speeds of all the wheels, based on the vehicle body side view model by using the second wheel speeds V0F and V0R as inputs. Here, the vehicle body side view model is to remove a wheel speed difference caused by a pitch motion generated around a pitch rotation center on a vertical line passing through the vehicle center of gravity when the vehicle is viewed from the lateral direction. It is represented by the following equation.
(Equation 3)
VbFL = VbFR = VbRL = VbRR = {Lr / (Lf + Lr)} V0F + {Lf / (Lf + Lr)} V0R
The reference wheel speed redistribution unit 304 substitutes VbFL (= VbFR = VbRL = VbRR) into V of the vehicle body plan view model shown in (Equation 1), and finally obtains the reference wheel speeds VFL, VFR, and VRL of each wheel. , VRR and calculate the reference wheel speed ω0 by dividing each by the tire radius r0.

  When the reference wheel speed ω0 of each wheel is calculated by the above-described processing, a deviation between the reference wheel speed ω0 and the wheel speed sensor value is calculated, and since this deviation is a wheel speed fluctuation accompanying a suspension stroke, Converted to the stroke speed Vz_s. Basically, the suspension does not only stroke in the vertical direction when holding each wheel, but also the wheel rotation center moves back and forth with the stroke, and the axle itself equipped with the wheel speed sensor 5 also tilts. And causes a rotation angle difference with the wheel. Since the wheel speed changes with this forward / backward movement, the deviation between the reference wheel speed and the wheel speed sensor value can be extracted as a change accompanying this stroke. It should be noted that the degree of the variation may be appropriately set according to the suspension geometry.

  When the stroke speeds Vz_sFL, Vz_sFR, Vz_sRL, and Vz_sRR of each wheel are calculated by the above-described processing in the stroke speed calculation unit 321, the bounce rate, roll rate, and pitch rate for skyhook control are calculated in the sprung speed calculation unit 322. It is calculated.

(About the estimation model)
The skyhook control sets a damping force based on the relationship between the stroke speed of the S / A3 and the sprung speed, and achieves a flat running state by controlling the attitude on the sprung. Here, in order to achieve the sprung attitude control by the skyhook control, it is necessary to feed back the sprung speed. Now, the value that can be detected from the wheel speed sensor 5 is the stroke speed, and since there is no vertical acceleration sensor or the like on the spring, the sprung speed needs to be estimated using an estimation model. Hereinafter, the problem of the estimation model and the model configuration to be adopted will be described.

  FIG. 7 is a schematic diagram illustrating a vehicle body vibration model. FIG. 7A is a model of a vehicle having a constant damping force S / A (hereinafter referred to as a conveyor vehicle), and FIG. 7B is a model having a variable damping force S / A. This is a model for performing skyhook control. In FIG. 7, Ms represents the mass of the sprung mass, Mu represents the mass of the unsprung mass, Ks represents the elastic modulus of the coil spring, Cs represents the damping coefficient of S / A, and Ku represents the unsprung mass (tire). , Cu represents an unsprung (tire) damping coefficient, and Cv represents a variable damping coefficient. Further, z2 represents a position on the sprung, z1 represents a position on the unsprung, and z0 represents a road surface position.

When the conveyor vehicle model shown in FIG. 7A is used, the equation of motion with respect to the sprung mass is expressed as follows. The first derivative (ie, speed) of z1 is represented by dz1, and the second derivative (ie, acceleration) is represented by ddz1.
(Estimation formula 1)
Ms.ddz2 = -Ks (z2-z1) -Cs (dz2-dz1)
When this relational expression is Laplace transformed and arranged, it is expressed as follows.
(Estimation formula 2)
dz2 = − (1 / Ms) · (1 / s ² ) · (Cs · s + Ks) (dz2−dz1)
Here, since dz2-dz1 is the stroke speed (Vz_sFL, Vz_sFR, Vz_sRL, Vz_sRR), the sprung speed can be calculated from the stroke speed. However, when the damping force is changed by the skyhook control, the estimation accuracy is remarkably reduced, so that there is a problem that a large attitude control force (a change in the damping force) cannot be given to the conveyor vehicle model.

Therefore, it is conceivable to use a vehicle model based on skyhook control as shown in FIG. Changing the damping force basically means changing the force that limits the piston moving speed of the S / A3 with the suspension stroke. In order to use a semi-active S / A3 that cannot positively move the piston in the desired direction, a semi-active skyhook model is adopted, and the sprung velocity is expressed as follows.
(Estimation formula 3)
dz2 = − (1 / Ms) · (1 / s ² ) · {(Cs + Cv) · s + Ks} (dz2−dz1)
However,
When dz2 · (dz2-dz1) ≧ 0 Cv = Csky · {dz2 / (dz2-dz1)}
When dz2 · (dz2-dz1) <0 Cv = 0
That is, Cv is a discontinuous value.

  Now, when it is desired to estimate the sprung velocity using a simple filter, in the semi-active skyhook model, when this model is viewed as a filter, each variable corresponds to a filter coefficient, and a pseudo differential term { Since (Cs + Cv) · s + Ks} includes the discontinuous variable attenuation coefficient Cv, the filter response becomes unstable, and appropriate estimation accuracy cannot be obtained. In particular, when the filter response becomes unstable, the phase shifts. If the correspondence between the phase of the sprung speed and the sign is broken, skyhook control cannot be achieved. Therefore, even when the semi-active S / A3 is used, the active skyhook model that can directly use a stable Csky without depending on the sign relationship between the sprung speed and the stroke speed is used. The speed was estimated. When the active skyhook model is adopted and the sprung speed is obtained, it is expressed as follows.

(Estimation formula 4)
dz2 = − (1 / s) · {1 / (s + Csky / Ms)} · {(Cs / Ms) s + (Ks / Ms)} (dz2−dz1)
In this case, no discontinuity occurs in the pseudo-differential term {(Cs / Ms) s + (Ks / Ms)}, and the term {1 / (s + Csky / Ms)} can be constituted by a low-pass filter. Therefore, the filter response is stabilized, and appropriate estimation accuracy can be obtained. In this case, even if the active skyhook model is adopted, only the semi-active control can be actually performed, so that the controllable area is halved. Therefore, the magnitude of the estimated sprung speed is smaller than the actual value in the frequency band below the sprung resonance, but the most important in the skyhook control is the phase, and if the correspondence between the phase and the sign can be maintained, the skyhook speed can be maintained. Control is achieved and there is no problem since the magnitude of sprung speed can be adjusted by other factors or the like.

  From the above relationship, it can be understood that the sprung speed can be estimated if the stroke speed of each wheel is known. Next, since the actual vehicle is four wheels instead of one wheel, consider estimating the sprung state using the stroke speed of each of these wheels by modal decomposition into roll rate, pitch rate, and bounce rate. I do. Now, when the above three components are calculated from the stroke speeds of the four wheels, one corresponding component is missing, and the solution becomes indefinite. Therefore, a war plate representing the movement of the diagonal wheel is introduced. When the bounce term of the stroke amount is xsB, the roll term is xsR, the pitch term is xsP, the warp term is xsW, and the stroke amounts corresponding to Vz_sFL, Vz_sFR, Vz_sRL, and Vz_sRR are z_sFL, z_sFR, z_sRL, and z_sRR, Holds.

以上の関係式から、xsB、xsR、xsP、xsWの微分ｄxsB等は以下の式で表される。
ｄxsB＝1/4（Vz_sFL＋Vz_sFR＋Vz_sRL＋Vz_sRR）
ｄxsR＝1/4（Vz_sFL−Vz_sFR＋Vz_sRL−Vz_sRR）
ｄxsP＝1/4（−Vz_sFL−Vz_sFR＋Vz_sRL＋Vz_sRR）
ｄxsW＝1/4（−Vz_sFL＋Vz_sFR＋Vz_sRL−Vz_sRR） (Equation 4)

From the above relational expression, the derivative dxsB of xsB, xsR, xsP, xsW, etc. is represented by the following expression.
dxsB = 1/4 (Vz_sFL + Vz_sFR + Vz_sRL + Vz_sRR)
dxsR = 1/4 (Vz_sFL−Vz_sFR + Vz_sRL−Vz_sRR)
dxsP = 1/4 (-Vz_sFL-Vz_sFR + Vz_sRL + Vz_sRR)
dxsW = 1/4 (-Vz_sFL + Vz_sFR + Vz_sRL-Vz_sRR)

Here, since the relationship between the sprung speed and the stroke speed is obtained from the above-mentioned estimation formula 4, in the estimation formula 4, − (1 / s) · {1 / (s + Csky / Ms)} · {(Cs / Ms) s + (Ks / Ms)} portion is described as G, and modal parameters (CskyB, CskyR, CskyP, CsB, CsR, CsP, If the values considering KsB, KsR, and KsP) are GB, GR, and GP, and each bounce rate is dB, the roll rate is dR, and the pitch rate is dP, dB, dR, and dP can be calculated as the following values.
dB = GB · dxsB
dR = GR · dxsR
dP = GP ・ dxsP
From the above, it is possible to estimate the sprung state of the actual vehicle based on the stroke speed of each wheel.

(Spring damping control unit)
Next, the skyhook control configuration executed in the sprung mass damping control unit 101a, the skyhook damping control unit 201, and the sprung mass damping control unit 33 will be described. In the skyhook control, the sprung state estimated based on the wheel speed as described above is controlled to be the target sprung state. In other words, the wheel speed change changes in accordance with the sprung state. When the sprung state such as bounce, roll, and pitch is controlled to the target sprung state, the detected change in the wheel speed is the target sprung state. The control is performed so that the wheel speed changes according to the state.

[Configuration of Skyhook control unit]
The skyhook controller 33a in the S / A controller 3a controls three objects, a bounce rate, a roll rate, and a pitch rate.

Skyhook control amount in the bounce direction is
FB = CskyB · dB
Skyhook control amount in the roll direction is
FR = CskyR · dR
The skyhook control amount in the pitch direction is
FP = CskyP ・ dP
It becomes.
(Skyhook control amount FB in bounce direction)
The skyhook control amount FB in the bounce direction is calculated as a part of the S / A attitude control amount in the skyhook control unit 33a.
(Sky hook control amount FR in roll direction)
The sky hook control amount FR in the roll direction is calculated by the sky hook control unit 33a as a part of the S / A attitude control amount.
(Sky hook control amount FP in pitch direction)
The skyhook control amount FP in the pitch direction is calculated by the skyhook control unit 33a as a part of the S / A attitude control amount.

(About S / A side driver input control unit)
Next, the S / A side driver input control unit will be described. The S / A-side driver input control unit 31 calculates a driver input damping force control amount corresponding to the vehicle behavior desired by the driver based on signals from the steering angle sensor 7 and the vehicle speed sensor 8, and the damping force control unit 35. Output to For example, if the nose side of the vehicle rises while the driver is turning, the driver's view tends to be off the road. In this case, the damping force of the four wheels is reduced by the driver's input damping force so as to prevent the nose from rising. Output as control amount. In addition, a driver input damping force control amount for suppressing roll generated during turning is output.

(About roll control by S / A side driver input control)
Here, the roll suppression control performed by the S / A side driver input control will be described. FIG. 8 is a control block diagram illustrating a configuration of the roll rate suppression control according to the first embodiment. The lateral acceleration estimating unit 31b1 estimates the lateral acceleration Yg based on the front wheel steering angle δf detected by the steering angle sensor 7 and the vehicle speed VSP detected by the vehicle speed sensor 8. The lateral acceleration Yg is calculated from the following equation based on the vehicle body plan view model.
Yg = (VSP ² / (1 + A · VSP ² )) · δf
Here, A is a predetermined value.

  The 90 ° phase lead component creation unit 31b2 differentiates the estimated lateral acceleration Yg and outputs a lateral acceleration differential value dYg. The first adder 31b4 adds the lateral acceleration Yg and the lateral acceleration differential value dYg. The 90 ° phase delay component creation unit 31b3 outputs a component F (Yg) obtained by delaying the phase of the estimated lateral acceleration Yg by 90 °. The second adder 31b5 adds F (Yg) to the value added in the first adder 31b4. The Hilbert transform unit 31b6 calculates a scalar amount based on the envelope waveform of the added value. The gain multiplying unit 31b7 multiplies the scalar amount based on the envelope waveform by a gain, calculates a driver input attitude control amount for roll rate suppression control, and outputs the result to the damping force control unit 35.

  FIG. 9 is a time chart illustrating an envelope waveform forming process of the roll rate suppression control according to the first embodiment. At time t1, when the driver starts steering, a roll rate starts to gradually occur. At this time, by adding the 90 ° phase lead component to form an envelope waveform, and calculating the driver input attitude control amount based on the scalar amount based on the envelope waveform, it is possible to suppress the occurrence of the roll rate at the beginning of steering. it can. Next, at time t2, when the driver enters the steering holding state, the 90-degree phase lead component disappears, and the phase delay component F (Yg) is added this time. At this time, even if the roll rate itself does not change much in the steady turning state, a roll rate resonance component corresponding to the roll backlash is generated after rolling once. If the phase delay component F (Yg) is not added, the damping force from time t2 to time t3 is set to a small value, which may cause instability of the vehicle behavior due to the roll rate resonance component. In order to suppress the roll rate resonance component, a 90 ° phase delay component F (Yg) is added.

  At time t3, when the driver shifts from the steering holding state to the straight running state, the lateral acceleration Yg decreases, and the roll rate converges to a small value. Also in this case, since the damping force is securely secured by the action of the 90 ° phase delay component F (Yg), instability due to the roll rate resonance component can be avoided.

(Unsprung vibration control unit)
Next, the configuration of the unsprung mass damping control unit will be described. As described in the conveyor vehicle of FIG. 7A, since the tire also has an elastic coefficient and a damping coefficient, there is a resonance frequency band. However, the mass of the tire is smaller than the mass of the sprung mass and has a high elastic modulus, and therefore exists on the higher frequency side than the sprung mass resonance. Due to this unsprung resonance component, the tire flaps under the unsprung state, and there is a possibility that the contact property may be deteriorated. Also, the unsprung fluttering may cause discomfort to the occupant. Therefore, in order to suppress flutter due to unsprung resonance, a damping force corresponding to the unsprung resonance component is set.

  FIG. 10 is a block diagram illustrating a control configuration of the unsprung mass damping control according to the first embodiment. The unsprung resonance component extraction unit 341 extracts a unsprung resonance component by applying a band-pass filter to the wheel speed fluctuation output from the deviation calculation unit 321b in the traveling state estimation unit 32. The unsprung resonance component is extracted from a region of approximately 10 to 20 Hz among the wheel speed frequency components. The envelope waveform shaping unit 342 converts the extracted unsprung resonance component into a scalar, and shapes the envelope waveform using the EnvelopeFilter. The gain multiplying unit 343 multiplies the scalar unsprung resonance component by a gain, calculates an unsprung damping force control amount, and outputs the amount to the damping force control unit 35. In the first embodiment, the unsprung resonance component is extracted by applying a band-pass filter to the wheel speed fluctuation output from the deviation calculating unit 321b in the running state estimating unit 32. Alternatively, the unsprung resonance component may be extracted by applying a band-pass filter to, or the unsprung speed may be extracted by estimating the unsprung speed in accordance with the sprung speed in the traveling state estimation unit 32. Good.

(About the configuration of the damping force control unit)
Next, the configuration of the damping force control unit 35 will be described. FIG. 4 is a control block diagram illustrating a control configuration of a damping force control unit 35. In the saturation conversion unit 35a, the driver input damping force control amount output from the driver input control unit 31, the S / A attitude control amount output from the skyhook control unit 33a, and the output from the unsprung vibration suppression control unit 34 The controlled unsprung damping force control amount and the stroke speed calculated by the traveling state estimation unit 32 are input, and these values are converted into an equivalent viscous damping coefficient. Then, based on the stroke speed, the equivalent viscous damping coefficient Ce, and the damping coefficient maximum value Cemax and the minimum value Cemin at this stroke speed, the saturation DDS (%) is calculated by the following equation.
DDS = ((Ce−Cemin) / (Cemax−Cemin)) × 100

  In the saturation arbitration unit 35b, the saturations converted in the saturation conversion unit 35a (hereinafter, the saturations are referred to as driver input saturation k1, S / A attitude saturation k2, and unsprung vibration suppression saturation k4, respectively). ), The arbitration is performed based on which saturation is to be controlled, the arbitrated saturation is limited by a preset saturation limit map based on the stroke speed, and the limited saturation is finally determined. Output a high degree of saturation. The control signal converter 35c converts the signal into an S / A3 control signal (command current value) corresponding to the degree of saturation, and outputs the signal to the S / A3.

(Details of stroke speed calculator)
Next, details of the stroke speed calculation unit will be described. FIG. 12 is a time chart illustrating the limit value processing in the GEO conversion unit. When the deviation between the reference wheel speed and the wheel speed sensor value is calculated in the deviation calculating unit 321b, a limit value process is performed on the deviation, a value exceeding the upper limit is cut by the upper limit, and a value below the lower limit is calculated. It is output as a value cut at the lower limit. This is to avoid calculating the current supplied to the S / A3 based on the excessive deviation. Note that this limit value processing may be performed by a unit other than the GEO conversion unit 321c as long as the block has been subjected to the calculation of the deviation, and is not particularly limited.

  Here, there is no problem when the cut is made at the upper limit value or the lower limit value after an appropriate calculation result, but when the front wheel and the rear wheel have different tire radii, other than the actual fluctuation component due to the difference in the tire radius. Will be added. Then, a value different from the stroke that actually occurs in each wheel, in other words, a value offset upward or downward, is calculated, and the value is cut at the upper limit or the lower limit, thereby underestimating the variation of the deviation. This causes a problem that an appropriate damping force cannot be applied.

Therefore, in the first embodiment, a high-pass filter 321f is provided between the deviation calculation unit 321b and the GEO conversion unit 321c that performs a limit value process on the deviation, and the deviation calculated by the deviation calculation unit 321b is filtered. By removing the DC component, and correcting the value as shown by the thick solid line in FIG. 12, an appropriate damping force can be applied. The high-pass filter 321f is given by, for example, Expression 5 below.
(Equation 5)
^{H (z) = (z n} + a1 * z n-1 + ··· an) / (1 + a1 * z + a2 * z 2 + ··· + an * z n)
Here, in order to prevent the detection band signal of the sprung behavior from disappearing due to the filter itself, the cutoff frequency fc is set with respect to the sprung resonance frequency fb in a relationship of fc ≦ 1 / fb ^1/2 . Based on the cutoff frequency fc, the filter coefficient a of the above equation 5 is implemented.

  Based on the above-mentioned configuration, an experiment was conducted using an actual vehicle. As a result, the stroke speed when the front and rear tires were attached was restored by 6 dB or more. (In other words, information on the fluctuation component of 6 dB or more was lost due to the offset value.) However, the vibration control performance by skyhook control has been greatly improved. For example, when looking at the vertical acceleration amplitude as an index of the vibration damping performance, the amplitude after the high-pass filter 321f was attached was reduced to half the amplitude before the high-pass filter 321f was installed.

As described above, the first embodiment has the following functions and effects.
(1) A control method for S / A3 (variable damping force shock absorber), wherein a difference between a reference wheel speed (vehicle speed) and a deviation calculated from the wheel speed (variation component of wheel speed) is determined by a front wheel and a rear wheel. Correction to reduce the superimposed offset value by mounting tires having different diameters is performed, the S / A3 stroke speed is detected based on the corrected wheel speed fluctuation component, and skyhook control is performed based on the stroke speed. Calculates the damping force control amount that suppresses the change in the sprung behavior of the vehicle calculated by the above, and controls S / A3 based on the damping force control amount.
Therefore, even when tires having different diameters are attached to the front wheel and the rear wheel, the stroke value is calculated after the offset value is reduced. Since it can be calculated, desired damping force control can be achieved.

  (2) The correction for reducing the offset value is performed by a filter that removes the DC current value. Therefore, the offset can be easily corrected only by setting a filter such as a high-pass filter.

[Example 2]
Next, a second embodiment will be described. Since the basic configuration is the same as that of the first embodiment, only different points will be described. FIG. 13 is a control block diagram illustrating control contents in a stroke speed calculation unit according to the second embodiment. In the first embodiment, the high-pass filter 321f is mounted after the deviation calculation unit 321b. On the other hand, in the second embodiment, when calculating the reference wheel speed, the tire moving radius correcting unit 300a corrects the tire moving radius to remove the offset value when the front and rear tires with different diameters are mounted. Specifically, when calculating the reference wheel speeds, the final reference wheel speeds VFL, VFR, VRL, and VRR of each wheel are calculated, and each is divided by the tire radius r0 to calculate the reference wheel speed ω0. At this time, the tire radius r0 is set as the corrected tire dynamic radius rx, and the reference wheel speed ω0 is calculated using this value.

FIG. 14 is a block diagram illustrating a configuration of a tire moving radius correction unit according to the second embodiment. The operation threshold determination unit 51 receives information on the steering angle δf, the yaw rate γ, the wheel speeds ω_n of four wheels (n = FL, FR, RL, RR), the drive torque T, the brake fluid pressure P, and the VDC operation flag Fvdc. You. When all of the following ON conditions (a) to (f) of the operation permission threshold flag are satisfied, a signal for turning ON the operation permission threshold flag is output. The ON condition is
(A) The vehicle speed is within a predetermined range. (B) The yaw rate is within a predetermined range. (C) The steering angle is within a predetermined range. (D) The drive torque change is within a predetermined range. Value)
(F) VDC operation flag Fvdc is OFF
It is.

  The wheel speed offset calculation unit 52 calculates the average value of the four wheel speeds, and calculates the average wheel speed ωref. In addition, in order to prevent erroneous estimation of the stroke speed due to a sharp change in the moving radius at the time of correcting the moving radius, the wheel speed ωref_ave corrected by the FIR filter that calculates the moving average so that the wheel speed while the operation permission flag is ON becomes the final value. Is output. The reason for using the FIR filter is to calculate strictly 0 when there is no offset amount (when tires of the same diameter are used instead of tires of different diameters). When the operation permission flag is OFF, the last value calculated up to the previous time is held.

The wheel speed offset amount calculation unit 52 receives the wheel speed ωref_ave and the wheel speed of each wheel, and calculates a reference vehicle speed Vref from a preset tire moving radius reference value Rref.
Vref = Rref * ωref_ave
Next, the estimated tire moving radius Re_n (n = FL, FR, RL, RR) is calculated from the following equation. This calculation is performed for each wheel.
Re_n = Vref * (1 / ω_n) = Rref * ωref_ave * (1 / ω_n)
Then, from the difference between the tire moving radius reference value Rref and the estimated tire moving radius Re_n, a radius offset amount ΔR_n (n = FL, FR, RL, RR) of each wheel is calculated by the following equation.
ΔR_n = Re_n-Rref
The calculated radius offset amount is output to the adding unit 55.

When the ON of the operation permission threshold value flag has continued for a predetermined time, the straight traveling stability time determination unit 54 sets the straight traveling determination flag to ON and outputs it to the adding unit 55.
The adder 55 outputs a value obtained by adding the radius offset amount ΔR_n to the current tire moving radius rx when the straight traveling determination flag is ON, and outputs the current tire moving radius rx as it is when the straight traveling determination flag is OFF. . In order to prevent the sprung resonance range from deteriorating due to the flag hunting, the predetermined time of the straight-running stabilization time determination unit 54 is set to at least 2 seconds or more (to prevent the effect on the phenomenon of 0.5 Hz or more).

As described above, in the second embodiment, the following operation and effect are obtained in addition to the operation and effect of the first embodiment.
(3) The correction for reducing the offset value is to calculate the reference wheel speed ωref_ave from each wheel speed ω_n when the vehicle detects the straight running stable state, and to calculate the reference wheel speed ω_n and the reference wheel speed ωref_ave based on each wheel speed ω_n. Calculate the radius offset amount ΔR_n (moving radius correction amount) of each wheel, correct the tire moving radius rx based on the radius offset amount ΔR_n, and deviate based on the corrected tire moving radius rx (variation component of wheel speed). Is calculated to reduce the offset value.
Therefore, it is possible to use a highly accurate tire dynamic radius rx at the stage of calculating the reference wheel speed, and when performing the skyhook control, an appropriate damping force control amount can be calculated. Control can be achieved.

1 Engine 1a Engine controller (engine control unit)
2 Brake control unit 2a Brake controller (brake controller)
3 S / A (variable damping force shock absorber)
3a S / A controller 5 wheel speed sensor 6 integrated sensor 7 steering angle sensor 8 vehicle speed sensor 20 brake 31 driver input control unit 32 running state estimation unit 33 sprung mass damping control unit 33a skyhook control unit 34 unsprung mass damping control Unit 35 damping force control unit 300a tire moving radius correction unit 321f high-pass filter 335 first target attitude control amount calculation unit 336 shock absorber attitude control amount calculation unit

Claims (4)

  A method of controlling a damping force variable shock absorber, wherein, among the fluctuation components of the wheel speed calculated from the vehicle speed and the wheel speed, an offset value superimposed by mounting a tire having a different diameter between the front wheel and the rear wheel is used. A reduction is performed, a stroke speed of the damping force variable shock absorber is detected based on the corrected wheel speed fluctuation component, and a change in vehicle sprung behavior calculated by skyhook control based on the stroke speed is suppressed. A damping force variable shock absorber control method that calculates a damping force control amount to be controlled and controls the damping force variable shock absorber based on the damping force control amount. The control method for a damping force variable shock absorber according to claim 1,
A method of controlling a variable damping force shock absorber, wherein the correction for reducing the offset value is performed by a filter that removes the offset value. The control method for a damping force variable shock absorber according to claim 1 or 2,
The correction for reducing the offset value calculates a reference wheel speed from each wheel speed when the vehicle detects a straight running stable state, and corrects a moving radius correction of each wheel based on the wheel speed of each wheel and the reference wheel speed. to calculate the amount, the dynamic radius correction amount to correct the tire dynamic radius based on the damping force variable to reduce said offset value by calculating a variation component of the wheel speed on the basis of the corrected tire dynamic radius How to control the shock absorber. Variable damping force shock absorber,
A controller for controlling the damping force variable shock absorber,
Has,
The controller performs a correction to reduce an offset value superimposed by mounting tires having different diameters on the front wheel and the rear wheel, among the fluctuation components of the wheel speed calculated from the vehicle body speed and the wheel speed, and correcting the corrected wheel. Detecting the stroke speed of the damping force variable shock absorber based on the speed fluctuation component, and calculating a damping force control amount for suppressing a change in the sprung behavior of the vehicle calculated by the skyhook control based on the stroke speed; A control device for a variable damping force shock absorber, wherein the variable damping force shock absorber is controlled based on the damping force control amount.

Images