JP2016002844A - Vehicle spring-upper/spring-lower relative speed estimation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a relative speed estimation device based on a single-wheel two-freedom model which can exactly estimate a relative speed between a vehicle spring upper structure and a vehicle spring lower structure without the necessity for the detection or estimation of road face displacement and a device for the detection and estimation.SOLUTION: A vehicle spring-upper/spring-lower relative speed estimation device 10 has an upper acceleration detection device 12, and an estimated relative speed calculation device 14 including a plant model 16 and an observer 18 which are constituted on the basis of a state space of an equation of motion based on a single-wheel two freedom model. The plant model has a system 16X which drives road face displacement by only a variable attenuation force of a shock absorber without the necessity for an outside input, the system calculates a substitution value of a disturbance item based on the road face displacement on the basis of the variable attenuation force, the observer 18 sets spring-upper acceleration and the variable attenuation force as inputs without making the road face displacement as disturbance, and a gain of a Kalman filter of the observer is set while responding to the system 16X.

Description

  The present invention relates to a sprung-unsprung relative speed estimation device for estimating a relative speed between a sprung structure and an unsprung structure of a vehicle, and more specifically, a sprung based on a single-wheel two-degree-of-freedom model. -It relates to an unsprung relative speed estimation device.

  As a sprung-unsprung relative speed estimation apparatus, a sprung-unsprung relative speed estimation apparatus having a plant model and an observer is known. For example, in Patent Document 1 below, a sprung-unsprung relative speed estimation device having a plant model and an observer based on a state space based on a single-wheel two-degree-of-freedom model (“single-wheel two-degree-of-freedom model”). Based on "relative speed estimator"). According to this type of relative speed estimation device, a sprung-unsprung relative speed estimation device having a plant model and an observer based on a state space based on a single-wheel one-degree-of-freedom model (“single-wheel one-degree-of-freedom model”). The relative speed estimation accuracy in the high frequency region can be improved as compared with a relative speed estimation device based on

Japanese Patent Laid-Open No. 10-95214

[Problems to be Solved by the Invention]
In a plant model configured based on a state space based on a single-wheel two-degree-of-freedom model, the damping force (nonlinear term) of the shock absorber is set as process noise, and road surface displacement is set as an external input. Therefore, in order to estimate the relative speed between the sprung structure and the unsprung structure of the vehicle by this type of relative speed estimation device, it is necessary to detect or estimate the road surface displacement.

  However, not only a special device is required to detect or estimate the road surface displacement, but the road surface displacement cannot be accurately detected or estimated by the special device. For example, it is conceivable to detect the vertical acceleration of the unsprung structure and substitute the twice integrated value of the vertical acceleration of the unsprung structure as the road surface displacement. However, in this case, a sensor for detecting the vertical acceleration of the unsprung structure is necessary, and it is inevitable that errors will occur in the detection of the vertical acceleration of the unsprung structure and its integration twice. The road surface displacement cannot be accurately estimated even by the method.

  Therefore, depending on the conventional relative speed estimation device based on the single-wheel two-degree-of-freedom model, the relative speed between the sprung structure and the unsprung structure of the vehicle can be obtained without requiring detection or estimation of road surface displacement and a device for the detection or estimation. Cannot be estimated accurately.

  The present invention has been made in view of the above-described problems in the conventional sprung-unsprung relative speed estimation device based on the single-wheel two-degree-of-freedom model. The main problem of the present invention is that it is possible to accurately estimate the relative speed between the sprung structure and the unsprung structure of the vehicle without the need for detecting or estimating the road surface displacement and a device therefor. To provide a relative speed estimation device based on a two-degree-of-freedom model.

[Means for Solving the Problems and Effects of the Invention]
The main problem described above is that according to the present invention, the sprung of a vehicle for estimating the relative speed between a sprung structure and an unsprung structure that are connected to each other via a damping force variable shock absorber and a spring. An unsprung relative speed estimation device configured based on a sprung acceleration detection device that detects a vertical acceleration of the sprung structure as a sprung acceleration and a state space of an equation of motion based on a single-wheel two-degree-of-freedom model; In a sprung-unsprung relative speed estimation device for a vehicle having a plant model and an estimated relative speed calculation device including an observer, the plant model is capable of changing the shock absorber as process noise without using road surface displacement as an external input. A system driven only by a damping force, which calculates an alternative value of a disturbance term based on a road surface displacement based on a variable damping force; The buzzer does not make the road surface displacement a disturbance, receives the sprung acceleration and variable damping force as input, includes the estimated relative speed between the sprung structure and the unsprung structure in the output, and the gain of the Kalman filter of the observer Is achieved by a vehicle sprung-unsprung relative speed estimation device characterized in that it is set in correspondence with the system.

  According to the above configuration, the plant model has a system that is driven only by the variable damping force of the shock absorber as process noise without using the road surface displacement as an external input, and the system converts the road surface displacement based on the variable damping force. Calculate the substitute value of the disturbance term based on it. The observer does not make the road surface displacement a disturbance, receives the sprung acceleration and variable damping force as inputs, includes the estimated relative speed between the sprung structure and the unsprung structure in the output, and the gain of the Kalman filter of the observer is It is set to correspond to the plant model system.

  Therefore, it is not necessary to detect or estimate the road surface displacement in order to calculate the disturbance term based on the road surface displacement, and an apparatus for that purpose is also unnecessary. Therefore, the relative speed between the sprung structure and the unsprung structure of the vehicle can be estimated with a simpler configuration than the conventional relative speed estimation device based on the single-wheel two-degree-of-freedom model.

  Moreover, according to said structure, the alternative value of the disturbance term based on a road surface displacement is calculated based on variable damping force. Therefore, it is possible to estimate the relative speed more accurately than the conventional relative speed estimation device based on the single-wheel one-degree-of-freedom model in which the disturbance due to the road surface displacement is not considered, and the conventional relative speed based on the single-wheel two-degree-of-freedom model. The relative speed can be estimated as accurately as the speed estimation device.

  In the above configuration, the system calculates an alternative value of the disturbance term as a product of a variable damping force of the shock absorber and a coefficient matrix having only the mass of the unsprung structure as a parameter, and substitutes the disturbance term. The value may be a value obtained by dividing the variable damping force by the mass of the unsprung structure.

  According to the above configuration, since the substitute value of the disturbance term is a value obtained by dividing the variable damping force by the mass of the unsprung structure, as described in detail later, as a value corresponding to the vertical acceleration of the road surface Calculated. Therefore, for example, the vertical speed of the unsprung structure can be obtained by integrating the substitute value of the disturbance term. Therefore, the relative velocity can be obtained as a deviation between the integrated value of the vertical acceleration of the sprung structure detected by the sprung acceleration detector and the integrated value of the substitute value of the disturbance term.

  In the above configuration, the observer may have the same-dimensional observer configuration with respect to the plant model.

  According to the above configuration, since the observer has the configuration of the same dimensional observer with respect to the plant model, the estimated value of the vertical acceleration of the sprung structure can be calculated corresponding to the plant model.

1 is a block diagram showing one embodiment of a vehicle sprung-unsprung relative speed estimation device according to the present invention based on a two-degree-of-freedom model. FIG. It is a figure which shows the single-wheel 2 degrees-of-freedom model of a vehicle. It is a block diagram which shows the structure of the plant model shown by FIG. It is a block diagram which shows the structure of the observer shown by FIG. It is a block diagram which shows the conventional sprung-unsprung relative speed estimation apparatus based on a two-degree-of-freedom model. It is a block diagram which shows the structure of the plant model shown by FIG. FIG. 6 is a block diagram illustrating a configuration of an observer illustrated in FIG. 5. When the road surface displacement _xr is a sine waveform and the non-linear damping force fv of the shock absorber is not present, the gain fluctuation range (upper stage) and phase fluctuation range (lower stage) are represented by the sprung-unsprung relative speed estimation apparatus ( It is a graph shown about the conventional sprung-unsprung relative speed estimation apparatus (right-downward hatching) based on a 1-degree-of-freedom model. When the road surface displacement _xr is a random waveform and the non-linear damping force fv of the shock absorber is not present, the gain fluctuation range (upper stage) and phase fluctuation range (lower stage) are determined based on the sprung-unsprung relative speed estimation device of the embodiment ( It is a graph shown about the conventional sprung-unsprung relative speed estimation apparatus (right-downward hatching) based on a 1-degree-of-freedom model. When the road surface displacement _xr is a sine waveform and there is a nonlinear damping force fv of the shock absorber, the gain fluctuation range (upper stage) and the phase fluctuation range (lower stage) are determined based on the sprung-unsprung relative speed estimation apparatus ( It is a graph shown about the conventional sprung-unsprung relative speed estimation apparatus (right-downward hatching) based on a 1-degree-of-freedom model. When the road surface displacement _xr is a random waveform and there is a nonlinear damping force fv of the shock absorber, the fluctuation range (upper stage) and the fluctuation range (lower stage) of the phase are represented by the sprung-unsprung relative speed estimation device ( It is a graph shown about the conventional sprung-unsprung relative speed estimation apparatus (right-downward hatching) based on a 1-degree-of-freedom model.

  Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing a sprung-unsprung relative speed estimation device 10 for a vehicle according to an embodiment of the present invention. In FIG. 1, the sprung-unsprung relative speed estimation device 10 includes a sprung acceleration sensor 12 and an estimated relative speed calculation device 14, and the estimated relative speed calculation device 14 includes a plant model 16 and an observer 18. . As can be seen from FIG. 1, the sprung-unsprung relative speed estimation device 10 does not include a device for detecting or estimating the vertical displacement of the road surface.

  The sprung acceleration sensor 12 functions as a sprung acceleration detection device that detects vertical acceleration of a sprung structure of a vehicle not shown in FIG. 1 as sprung acceleration. As will be described in detail later, the estimated relative speed calculation device 14 calculates the estimated relative speed between the sprung structure and the unsprung structure without requiring information on the vertical displacement of the road surface. Further, as will be described in detail later, the plant model 16 and the observer 18 are configured based on a state space of an equation of motion based on the single-wheel two-degree-of-freedom model shown in FIG.

In FIG. 2, Mu is the mass of the sprung structure 102 of the vehicle 100 (hereinafter referred to as sprung mass), and Mw is the mass of the unsprung structure 104 (hereinafter simply referred to as unsprung mass). It is. The masses of the arm (not shown), the spring 106 and the shock absorber 108 constituting the suspension are distributed to the sprung structure and the unsprung structure at an appropriate ratio. Ks is a spring constant of the spring 106, and Kt is a spring constant of the tire 110. fc is a linear damping force (fixed damping force) of the shock absorber 108, and fv is a nonlinear damping force (variable damping force) of the shock absorber. Further, x _u represents the vertical displacement of the sprung structure 102, x _w represents the vertical displacement of the unsprung structure 104, and x _r represents the vertical displacement of the road surface.
Further, in the present application, “′” represents a time differential value of a variable to which this is attached, and “” ”represents a second-order time differential value of the variable to which this is attached. Thus, for example, x _u ′ represents the vertical speed of the sprung structure 102, x _u ″ represents the vertical acceleration of the sprung structure 102, and “ ^T ” represents the row example in which this is attached. This represents a transposed matrix, and “ _e ” represents that this is an estimated value of the attached variable.
The equations of motion for the sprung structure 102 and the unsprung structure 104 of the system shown in FIG. 2 are expressed by the following expressions (1) and (2), respectively. Co is a linear damping coefficient of the shock absorber 108.
_{Mu · x u "= Ks (} x w -x u) + Co (x w '-x u') + fv ... (1)
_{Mw · x w "= Ks (} x w -x u) -Co (x w '-x u')
_{+ Kt (X r -x w)} -fv ... (2)

It is assumed that the state quantity x is as shown in the following equation (3), the external input u is the road surface vertical displacement (X _r ), and the process noise w is the nonlinear damping force (fv). The state space (state equation and output equation) of the above equations of motion (1) and (2) in the conventional relative velocity estimation apparatus is expressed by the following equations (4) and (5).
x = [x _u x _u 'x _w x _w '] ^T (3)
x ′ = A · x + B · w + G · u (4)
y = C · x + D · u (5)

The output y of the above equation (5) is an estimated value x _ue ″ of the vertical acceleration of the sprung structure 102 that can be detected by the sprung acceleration sensor 12, and in the above equations (4) and (5) The coefficient matrices A to G are as follows.

On the other hand, since the relative speed estimation device 10 of the embodiment does not use the road surface displacement _xr as an external input, the state space of the motion equations (1) and (2) in the relative speed estimation device of the embodiment is as follows. It represents with Formula (6) and (7). The coefficient matrices A to D in the following formulas (6) and (7) are the same as the coefficient matrices A to D in the above formulas (4) and (5). The coefficient matrix G2 is as follows, and only the mass Mw of the unsprung structure 104 is used as a parameter, and the spring constant Kt of the tire 110 is not included.
x ′ = A · x + B · G 2 · w (6)
y = C · x + D · w (7)

As understood from the comparison between the above formula (7) and formula (5), these formulas are the same. However, the formula (6), as seen from comparison with formula (4), only the non-linear damping force fv of the shock absorber 108 of the system 16X road surface displacement x _r of the plant model 16 as process noise without the external input It has been modified to become a system driven by

As shown in FIG. 1, the estimated relative speed calculation device 18 includes a data map 20. The data map 20 shows the relative velocity x _se ′ between the sprung structure 102 and the unsprung structure 104 and the damping coefficient kc of the shock absorber 108 controlled by a damping force control device not shown in the figure. The relationship with the nonlinear damping force fv of the shock absorber 108 is stored. The data map 20 calculates the nonlinear damping force fv of the shock absorber 108 as the product of the estimated relative velocity x _se ′ output from the observer 16 and the damping coefficient kc of the shock absorber 108.

  A signal indicating the nonlinear damping force fv is input to the system 16X of the plant model 16, and the system 16X calculates the time differential value x ′ of the state quantity x according to the above equation (6) and the state quantity x which is an integral value thereof. Is calculated. The nonlinear damping force fv and the state quantity x are multiplied by coefficient matrices C and D, respectively, and the sum of the values after multiplication is calculated by the adder 22, whereby the calculation of the above equation (7) is performed.

A signal indicating the vertical acceleration x _u ″ of the sprung structure 102 detected by the sprung acceleration sensor 12 and a signal indicating an estimated value x _ue ″ of the vertical acceleration of the sprung structure 102 calculated by the adder 22 Is input to the adder 24. The adder 24 calculates an estimated error ν (= x _u ″ −x _ue ″) of the vertical acceleration of the sprung structure 102, and a signal indicating the estimated error ν is input to the observer 16.

Observer 16 on the basis of the estimated value x _e and the estimated error ν state quantity x, according to the following equation for the L and gain of the Kalman filter (8) and (9), the estimated value of the state quantity x _e and its time derivative Calculate the value x _e '. Therefore, the observer 16 has the same-dimensional observer configuration for the system represented by the above equations (6) and (7). Since the time differential value x _e ′ includes estimated vertical speeds x _ue ′ and x _we ′ of the sprung structure 102 and the unsprung structure 104, the estimated relative speed x _se ′ input to the data map 20. Is calculated as the difference between the estimated speeds x _ue '-x _we '.
x _e ′ = A · x _e + L · ν (8)
y _e = C · x _e + D · w (9)

  3 and 4 show the configurations of the plant model 16 and the observer 18 shown in FIG. 1, respectively. 3 and 4, “s” indicates a Laplace operator, and this is the same for FIGS. 6 and 7 described later.

As shown in FIG. 3, the plant model 16, only the non-linear damping force fv as process noise w is input, the road surface displacement x _r as an external input u is not inputted. In the system 16X of the plant model 16, a value obtained by multiplying the nonlinear damping force fv by the coefficient matrix B and a value obtained by multiplying the nonlinear damping force fv by the coefficient matrix G2 are input to the adder 26. The output of the adder 26 is input to the integrator 28, and the output of the integrator 28 is input to the adder 26 after being multiplied by the coefficient matrix A. Therefore, the output of the adder 26, and hence the output of the system 16X, is the time differential value x ′ of the state quantity x, and the output of the integrator 28 is the time integral value of the time differential value x ′, that is, the state quantity x.

As shown in FIG. 4, an estimated error ν (= x _u ″ −x _ue ”) of the vertical acceleration of the sprung structure 102 is input from the adder 24 to the observer 18, and the external input u The road surface displacement _xr is not input. The estimation error ν is input to the adder 30 after being multiplied by the gain L of the Kalman filter. The output of the adder 30 is input to the integrator 32, and the output of the integrator 32 is input to the adder 30 after being multiplied by the coefficient matrix A. Therefore, the output of the adder 30 is a time differential value x _e ′ of the estimated value x _e of the state quantity x.

The output of the integrator 32 is input to the adder 34 after being multiplied by the coefficient matrix C. The adder 34 also receives the product of the nonlinear damping force fv as the process noise w and the coefficient matrix D. Therefore, the output of the adder 34, and hence the output of the observer 18, is an estimated value x _ue ″ of the vertical acceleration of the sprung structure 102.

Moreover, the difference between the by the coefficient matrix C2 is multiplied by the output of the adder 30, _'estimated velocity x _we in the vertical direction of the sprung structure _104' vertical estimated velocity x _ue of the sprung structure 102 _{x ue} '- x _we', ie the estimated relative velocity _{x se} 'is calculated. The estimated relative velocity x _se ′ is input to the data map 20 shown in FIG. The coefficient matrix C2 is as shown in the following equation (10).
C2 = [0 1 0 -1] (10)

  The estimated relative speed calculation device 14 may actually be a microcomputer having a known configuration including a CPU, a ROM, a RAM, an input / output port device, and the like, which are connected to each other via a bidirectional common bus. The functions of the functional blocks shown in FIGS. 1, 3, and 4 may be achieved by a control program stored in the ROM.

  Next, in order to further clarify the characteristics of the sprung-unsprung relative speed estimation device 10 according to the embodiment, a conventional sprung-unsprung relative based on a two-degree-of-freedom model with reference to FIGS. The speed estimation device will be described. 5 to 7, the same components as those shown in FIGS. 1, 3, and 4 are denoted by the same reference numerals as those in FIGS. 1, 3, and 4. Yes.

FIG. 5 is a block diagram similar to FIG. 1 showing a conventional sprung-unsprung relative speed estimation device 50 based on a two-degree-of-freedom model. The relative speed estimation device 50 includes a plant model 56 and an observer 58 corresponding to the plant model 16 and the observer 18 of the embodiment, respectively. The relative speed estimation unit 50, in addition to the data map 20, and a device 60 for detecting or estimating a road surface displacement x _r.

System 56X plant model 56, as seen from the above equation (4), the road surface displacement x _r as an external input, a system for the non-linear damping force fv of the shock absorber 108 and process noise. Therefore, the system 56X, together with the non-linear damping force fv from the data map 20 is input, the road surface displacement x _r is input from the device 60. The system 56X calculates the time differential value x ′ of the state quantity x according to the above equation (4), and calculates the state quantity x that is an integral value thereof. The calculation of the above formula (5) is performed in the same manner as the calculation of the above formula (7).

The observer 58 receives not only the estimated acceleration error ν (= x _u "-x _ue ") of the sprung structure 102 from the adder 24, but also the nonlinearity from the data map 20 and the device 60, respectively. damping force fv and the road surface displacement x _r is input. The observer 58 uses the following equations (11) and (9) corresponding to the equations (8) and (9) based on the estimated value x _e of the state quantity x, the nonlinear damping force f v, the road surface displacement x _r and the estimated error ν. According to 12), an estimated value x _e of the state quantity is calculated, and its time differential value x _e ′ is calculated.
_xe '= A * _xe + B * w + G * u + L * v (11)
y _e = C · x _e + D · w (12)

As shown in FIG. 6, in the system 56X of the plant model 56, in addition to the product of the coefficient matrix A and the state quantity x, the value input to the adder 26 is the nonlinear damping force fv and the coefficient matrix B. a multiplication value and the value coefficient matrix G is multiplied by the road surface displacement x _r. Other points of the plant model 56 are the same as the plant model 16 of the embodiment.

As shown in FIG. 7, in the observer 58, in addition to the product of the coefficient matrix A and the estimated value _xe and the product of the estimated error ν and the Kalman filter gain L, the nonlinear damping force fv and the coefficient matrix B product (Bw) and the product of the road surface displacement _{x r} and the coefficient matrix G (Gu) is input to the adder 30 with. The other points of the observer 58 are the same as the observer 18 of the embodiment.

As described above, the conventional spring based on two degrees of freedom model - in unsprung relative speed estimation unit 50, a road surface displacement x _r is treated as a disturbance term. Therefore, in order to obtain the gain L of the Kalman filter, it is necessary that “the road surface displacement _xr is a process noise to which no bias is applied”. Therefore, it can be considered that the disturbance term, that is, the product of the road surface displacement _xr and the coefficient matrix G is replaced with another parameter.

The units of the road surface displacement x _r and mm, the unit of the spring constant Kt of the tire 110 and N / mm, when the unit of the unsprung mass Mw and kg, units of the disturbance term, i.e. the road surface displacement x _r and the coefficient matrix G The unit of product of Kt / Mw is the unit of acceleration as follows.
mm ・ (N / mm) / kg = N / kg = (kg ・ m / sec ² ) / kg = m / sec ²

  Therefore, since the disturbance term may be considered as the vertical acceleration of the road surface, the value obtained by dividing the nonlinear damping force fv of the shock absorber 108 as a force by the unsprung mass Mw is a substitute parameter for the disturbance term based on the road surface displacement. Can be considered.

  According to the embodiment, a value obtained by multiplying the nonlinear damping force fv by the coefficient matrix G2 is calculated as a substitute parameter for the disturbance term based on the road surface displacement. The parameter of the coefficient matrix G is Kt / Mw, while the parameter of the coefficient matrix G2 is 1 / Mw. Therefore, in the embodiment, a value obtained by dividing the nonlinear damping force fv by the unsprung mass Mw is calculated as an alternative value for the disturbance term based on the road surface displacement.

Therefore, according to the embodiment, the sprung structure 102 and the unsprung structure are compared with the conventional sprung-unsprung relative speed estimation device based on the one-degree-of-freedom model that does not include the disturbance term based on the road surface displacement. The relative velocity x _se ′ with the body 104 can be accurately estimated.

In particular, as seen from comparison between comparative and each view 3,4 and 6 and 7 in FIGS. 1 and 5, according to the embodiment, information of the road surface displacement x _r it is unnecessary. Accordingly, without requiring the detection or estimation and apparatus therefor 60 of the road surface displacement x _r, the conventional spring based on one degree of freedom model - accurate than unsprung relative speed estimation device, the sprung structure 102 The relative velocity x _se ′ with the unsprung structure 104 can be estimated.

For example, 8 to 11, the variation range of the gain taking the frequency of the road surface displacement x _r in the horizontal axis (top) and the variation range of the phase (lower), hatched in the relative speed estimation unit (left edge of embodiments ) And a conventional relative speed estimation device (downward hatching) based on a one-degree-of-freedom model.

Each graph is a simulation value when the sprung mass Mu, the unsprung mass Mw, and the spring constant Ks of the spring 106 are the following values, respectively, and the fixed damping coefficient Co of the shock absorber 108 is 1000 Nmm / sec. is there. Furthermore, the gain and phase are 0 when an accurate estimation is made, so the gain and phase criteria are 0.
Sprung mass Mu: Median value: 524.8 kg, minimum value: 340 kg, maximum value: 662 kg
Unsprung mass Mw: Median value: 55.7 kg, Minimum value: 33 kg, Maximum value: 92 kg
Spring constant Ks of the spring 106: Median value: 26656 × 10 ³ N / mm,
Minimum value: 22400 × 10 ³ N / mm,
Maximum value: 63741 × 10 ³ N / mm

<Figure 8>
FIG. 8 shows the fluctuation range when the road surface displacement _xr is a sine waveform and there is no nonlinear damping force fv of the shock absorber. In the case of the conventional relative speed estimation device, the gain shift increases when the road surface displacement frequency is 100 Hz or more. However, in the case of the embodiment, even in the region where the road surface displacement frequency is 100 Hz or more, the gain is increased. The deviation is small. Further, in the case of the conventional relative speed estimation device, when the road surface displacement frequency is 2 Hz or higher, the phase shift gradually increases as the frequency increases, and the phase shift occurs in the region where the road surface displacement frequency is 10 Hz or higher. Become prominent. On the other hand, in the case of the embodiment, the phase shift is small even in the region where the road surface displacement frequency is 10 Hz or more.

<Fig. 9>
FIG. 9 shows the fluctuation range when the road surface displacement _xr has a random waveform and there is no nonlinear damping force fv of the shock absorber. In the case of the conventional relative speed estimation device, when the frequency of the road surface displacement becomes 10 Hz or more, the gain deviation becomes larger than that in the case of FIG. On the other hand, in the case of the embodiment, the gain deviation is small even in the region where the road surface displacement frequency is 10 Hz or more. In the case of the conventional relative speed estimation apparatus, when the frequency of the road surface displacement becomes 2 Hz or more, the phase shift gradually increases as the frequency increases, and the phase shift is not as great as in the case of FIG. In the region where the frequency is 10 Hz or more. On the other hand, in the case of the embodiment, the phase shift is small even in the region where the road surface displacement frequency is 10 Hz or more.

<Fig. 10>
FIG. 10 shows the fluctuation range when the road surface displacement _xr is a sine waveform and there is a nonlinear damping force fv of the shock absorber. In both cases of the conventional relative speed estimation device and the embodiment, a gain shift occurs when the frequency of the road surface displacement becomes 1 Hz or more. However, the magnitude of the gain shift in the embodiment is the same as the conventional relative speed estimation device. Smaller than the case. Further, in the case of the conventional relative speed estimation device, when the road displacement frequency becomes 0.5 Hz or more, the phase shift gradually increases as the frequency increases, and the phase shift causes the road displacement frequency to be 10 Hz or more. It becomes prominent in the area. On the other hand, in the case of the embodiment, when the frequency of the road surface displacement becomes 0.8 Hz or more, the phase shift gradually increases as the frequency increases, but the magnitude of the phase shift is the conventional relative speed estimation device. Smaller than the case.

<Fig. 11>
FIG. 11 shows the fluctuation range when the road surface displacement _xr has a random waveform and there is a nonlinear damping force fv of the shock absorber. In both cases of the conventional relative speed estimation device and the embodiment, a gain shift occurs when the frequency of the road surface displacement is 1 Hz or more. However, the magnitude of the gain shift in the embodiment is the same as that in the high frequency region. It is smaller than the case of the relative speed estimation device. In the case of the conventional relative speed estimation device, when the road surface displacement frequency is 0.4 Hz or higher, the phase shift gradually increases as the frequency increases, and the phase shift causes the road surface displacement frequency to be 11 Hz or higher. It becomes prominent in the area. On the other hand, in the case of the embodiment, when the frequency of the road surface displacement becomes 0.6 Hz or more, the phase shift gradually increases as the frequency increases, but the magnitude of the phase shift is the conventional relative speed estimation device. Smaller than the case.

Even when the sprung mass Mu, the unsprung mass Mw, the spring constant Ks of the spring 106, and the fixed damping coefficient Co of the shock absorber 108 are different from the above values, the results are the same as those shown in FIGS. The result which shows this tendency is obtained. That is, according to the embodiment, it has been confirmed that the relative speed x _se ′ can be estimated more accurately than in the case of the conventional sprung-unsprung relative speed estimation device based on the one-degree-of-freedom model.

Although not shown in the figure, according to the embodiment, a conventional sprung-unsprung relative speed estimation device based on a two-degree-of-freedom model in which road surface displacement _xr is estimated based on unsprung acceleration, It has been confirmed that the relative velocity x _se ′ can be estimated as accurately as possible. However, in the region where the frequency of the road surface displacement is very high, the gain shift and the phase shift become larger than in the case of the conventional sprung-unsprung relative speed estimation device based on the two-degree-of-freedom model.

  Although the present invention has been described in detail with respect to specific embodiments, the present invention is not limited to the above-described embodiments, and various other embodiments are possible within the scope of the present invention. This will be apparent to those skilled in the art.

  For example, in the above description of the embodiment, the type of the vehicle 100 is not specified, but the sprung-unsprung relative speed estimation device of the present invention is a sedan, a wagon car, a one-box car, an SUV, a compact car, or the like. It may be applied to various vehicles.

  Further, as described above, the estimated relative speed calculation device 14 may be a device other than a microcomputer, and the functions of the functional blocks shown in FIGS. 1, 3, and 4 are controlled by a control device having an arbitrary configuration. May be achieved.

DESCRIPTION OF SYMBOLS 10 ... Sprung-unsprung relative speed estimation apparatus, 12 ... Sprung acceleration sensor, 14 ... Estimated relative speed calculation apparatus, 16 ... Plant model, 18 ... Observer, 20 ... Data map, 100 ... Vehicle, 102 ... Sprung structure Body 104: Unsprung structure 106 Spring 106, 108 Shock absorber 110 Tire

Claims (3)

A sprung-unsprung relative speed estimation device for a vehicle for estimating a relative speed between a sprung structure and an unsprung structure that are connected to each other via a damping force variable shock absorber and a spring, wherein the spring A sprung acceleration detection device that detects the vertical acceleration of the upper structure as a sprung acceleration, and an estimated relative velocity calculation device that includes a plant model and an observer configured based on a state space of an equation of motion based on a two-degree-of-freedom model of a single wheel In a sprung-unsprung relative speed estimation device for a vehicle having
The plant model has a system that is driven only by the variable damping force of the shock absorber as process noise without using the road surface displacement as an external input, and the system has a disturbance term based on the road surface displacement based on the variable damping force. Calculate the substitute value,
The observer does not make the road surface displacement a disturbance, receives the sprung acceleration and variable damping force as input, and includes the estimated relative speed between the sprung structure and the unsprung structure in the output, and the Kalman filter of the observer A sprung-unsprung relative speed estimation device for a vehicle, wherein the gain is set corresponding to the system.   2. The vehicle sprung-unsprung relative speed estimation device according to claim 1, wherein the system is a product of a variable damping force of the shock absorber and a coefficient matrix having only the mass of the unsprung structure as a parameter. An alternative value of the disturbance term is calculated, and the alternative value of the disturbance term is a value obtained by dividing the variable damping force by the mass of the unsprung structure, and estimating an unsprung-unsprung relative speed of the vehicle. apparatus. The sprung-unsprung relative speed estimation device for a vehicle according to claim 1, wherein the observer has a configuration of a one-dimensional observer with respect to the plant model. .

Images