JP6089959B2 - Tire behavior estimation device - Google Patents

Description

  The present invention relates to a tire behavior estimation device that estimates the behavior of a vehicle tire.

  In the tire behavior estimation device, an experimental formula model such as Magic Formula Tire Model or a physical model such as Fiala Model is generally used as a tire model. Although the experimental model is excellent in estimation accuracy, a large amount of experimental data is required for model fitting. In contrast, physical models do not necessarily require experimental data, but have poor estimation accuracy.

  In addition, in the experimental model, the relationship between the tire representative characteristics and model parameters, which are noticed in the basic study of vehicle motion and the initial study of vehicle development, is complicated, and parameter studies cannot be performed easily. In contrast, for physical models, experimental data is not always available in basic studies of vehicle motion or initial studies of vehicle development, but the impact of the tire representative characteristics of interest on the vehicle motion and the target vehicle motion characteristics. It is necessary to grasp the tire representative characteristics to be satisfied with appropriate accuracy.

  For this reason, there has been a demand for the development of a tire model that can realize the estimation accuracy corresponding to the experimental model as well as the simplicity corresponding to the physical model.

  On the other hand, as described in, for example, Japanese Patent Application Laid-Open No. 2008-122253, a tire model in which both models are fused is also known. This fusion model is intended to realize a simulation that takes into account the temperature dependence of the tire while reducing labor and time.

JP 2008-122253 A JP 2005-335588 A

  However, the publication related to the fusion model sufficiently discloses specific means for realizing a tire behavior estimation device having estimation accuracy equivalent to an empirical model as well as simplicity corresponding to a physical model. I can't say that.

  Therefore, the present invention intends to provide a tire behavior estimation device that can easily and accurately estimate the tire behavior.

A tire behavior estimation device according to the present invention is a tire behavior estimation device that estimates the behavior of a tire of a vehicle, obtains a vertical shift using a camber angle change rate of a lateral friction coefficient of the tire, and uses a canvas stiffness to An estimation unit for _obtaining a shift and estimating a lateral force of the tire during a pure lateral slip using a horizontal shift and a vertical shift is provided, wherein the vertical shift is S _v , the camber angle change rate is _{dγ μy} , and the tire contact load the _{F z,} the camber angle of the tire gamma, the scaling factor of the friction coefficient of the tire when the mu _scale, the vertical shift _{S v} is the equation _{S v = dγ μy · F z} · γ · μ scale sought, the horizontal shift _{S h,} the canvas Chifu Ness CS, the cornering Chifu ness when the BCD _y, the horizontal shift _{S h} , Determined by the formula _{S h = (CS · γ-} S v) / BCD y, the estimation unit, the lateral force _{F y,} the peak factor _{D y,} the shape factor _{C y,} the steel Stiffness Factor _B y, the sum of the slip angle α and the horizontal shift _{S h} x, the Kaaba Chua factor is taken as _{E y,} wherein the lateral force _{F y = D y · sin [} C y · arctan {B y · x-E y ( B _y · x-arctan (B _y · x))}] + S _v .

  Here, in the conventional tire behavior estimation apparatus, the vertical shift and the horizontal shift are obtained using a large number of parameters determined by fitting a tire model to a lot of experimental data. On the other hand, according to the tire behavior estimation apparatus according to the present invention, the vertical shift is obtained as a function of the camber angle change rate of the tire lateral friction coefficient and the horizontal shift is obtained as a function of the canvas stiffness, thereby obtaining the tire camber angle. Can easily and accurately estimate the effect of the sway on lateral force during pure side slip. As a result, the tire behavior can be estimated easily and with high accuracy.

  ADVANTAGE OF THE INVENTION According to this invention, the tire behavior estimation apparatus which can estimate a tire behavior simply and with high precision can be provided.

It is a block diagram which shows the tire behavior estimation apparatus which concerns on embodiment of this invention. It is a flowchart which shows operation | movement of the tire behavior estimation apparatus which concerns on embodiment of this invention. It is a figure which shows the estimation result of a pure front-back slip characteristic (front-back force). It is a figure which shows the correction result of a lateral friction coefficient. It is a figure which shows the correction | amendment result of normalized cornering stiffness. It is a figure which shows the correction | amendment result of normalized canvas stiffness. It is a figure which shows the correction | amendment result of normalized self-aligning stiffness. It is a figure which shows the correction | amendment result of normalized camber torque stiffness. It is a figure which shows the estimation result of a pure side slip characteristic (lateral force). It is a figure which shows the estimation result of a pure side slip characteristic (self-aligning torque). It is a figure which shows the estimation result of a composite slip characteristic (lateral force). It is a figure which shows the estimation result of a composite slip characteristic (self-aligning torque).

  Hereinafter, a tire behavior estimating device according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

  First, a tire behavior estimation device 10 according to an embodiment of the present invention will be described with reference to FIG. The tire behavior estimation device 10 is a device that estimates tire behavior that affects vehicle motion and vehicle development. The tire behavior estimation device 10 is configured as a computer including a CPU, a ROM, a RAM, an input / output circuit, and the like.

  FIG. 1 is a block diagram showing a tire behavior estimation device 10 according to an embodiment of the present invention. As shown in FIG. 1, the tire behavior estimation device 10 includes a setting unit 11, an input unit 12, an estimation unit 13, an output unit 14, and a storage unit 15. The functions of the setting unit 11, the input unit 12, the estimation unit 13, the output unit 14, and the storage unit 15 are realized by the CPU reading and executing a program for executing a tire behavior estimation method described later from a ROM or the like. The

  The setting unit 11 sets the model coefficient of the tire model based on given tire data. As the model coefficient, tire data itself or a derived value from tire data is used. Details of the model coefficient will be described later.

The input unit 12 inputs tire model input parameters. As input parameters, a tire ground contact load F _z [N], a slip ratio κ [−], a slip angle α [rad], a camber angle γ [rad], and a scaling factor μ _scale [−] of a friction coefficient μ are input. ([] Indicates a unit. [-] Indicates no unit. The same shall apply hereinafter.) Input parameters are input from a storage device, an input device, or the like.

  The estimation unit 13 estimates tire slip characteristics based on model coefficients, input parameters, and the like. The slip characteristics include pure front / rear slip characteristics, pure lateral slip characteristics, and composite slip characteristics. A pure front / rear slip is a slip generated in the traveling direction (rolling direction) of the tire, a pure lateral slip is a slip generated in the lateral direction of the tire, and a composite slip is in the traveling direction and lateral direction of the tire. It is a slip that occurs at the same time.

The output unit 14 outputs an output parameter of the tire model. As output parameters, longitudinal force F _x [N] at the time of pure slip, lateral force F _y [N] at the time of pure lateral slip and self-aligning torque M _z [Nm], and longitudinal force F _xC at the time of compound slip [N], lateral force F _yC [N], and self-aligning torque M _zC [Nm] are output. The output parameter is output to a storage device, a display device, a printing device, or the like.

  The storage unit 15 stores model coefficient setting values, parameter input / output values, and the like.

  Next, a tire behavior estimation method by the tire behavior estimation apparatus 10 according to the embodiment of the present invention will be described with reference to FIGS.

  FIG. 2 is a flowchart showing the operation of the tire behavior estimation apparatus 10 according to the embodiment of the present invention. As shown in FIG. 2, the tire behavior estimation apparatus 10 first performs model coefficient setting processing prior to various slip characteristics estimation processing (S11). The setting unit 11 sets the following model coefficient based on given tire data.

μ _x0 : _Longitudinal friction coefficient μ _x (at standard load) [-]
dμ _x : Load change rate of the longitudinal friction coefficient μ _x [1 / N]
F _{z0 μx} : reference load [N] of the longitudinal friction coefficient μ _{x
dγ μy} : camber angle change rate of the lateral friction coefficient μ _y [1 / rad]
_{c μy0:} lateral friction coefficient mu _y approximation equation constant terms of the [-]
_c μy1: 1-order coefficient of the approximate expression of the lateral friction coefficient μ _y [1 / N]
F _zLI : Critical load [N]
w: Tire contact width [m]
C _Fx : Normalized braking stiffness [−]
cCSS ₀ : Constant term [1 / rad] of approximate expression of normalized cornering stiffness
cCSS ₁ : first order coefficient [1 / (radN)] of approximate expression of normalized cornering stiffness
cCSS ₂ : quadratic coefficient [1 / (radN ² )] of approximate expression of normalized cornering stiffness
cCSS ₃ : third-order coefficient [1 / (radN ³ )] of the approximate expression of normalized cornering stiffness
cCS ₀ : Constant term of the approximate expression of normalized canvas stiffness [1 / rad]
cCS ₁ : first order coefficient [1 / (radN)] of approximate expression of normalized canvas stiffness
cCS ₂ : quadratic coefficient [1 / (radN ² )] of approximate expression of normalized canvas stiffness
cSAS ₁ : First order coefficient [1 / (radN)] of approximate expression of normalized self-aligning stiffness
cSAS ₂ : quadratic coefficient [1 / (radN ² )] of the approximate expression of normalized self-aligning stiffness
cCTQ ₀ : Constant term [m / rad] of approximate expression of normalized camber torque stiffness
cCTQ ₁ : First order coefficient [m / (radN)] of approximate expression of normalized camber torque stiffness

The model coefficients μ _x0 , dμ _x , F _{z0 μx} , _{dγ μy} , F _zLI , w, and C _Fx are normally set as tire data itself. The model coefficients c _μy0 , c _μy1 , cCSS ₀ , cCSS ₁ , cCSS ₂ , cCSS ₃ , cCS ₀ , cCS ₁ , cCSS ₂ , cSAS ₁ , cSAS ₂ , cCTQ ₀ , cCTQ ₁ are usually derived values from tire data Set as

The longitudinal friction coefficient μ _x is a friction coefficient between the tire and the ground contact surface in the longitudinal direction of the tire, and the lateral friction coefficient μ _y is a friction coefficient between the tire and the ground contact surface in the lateral direction of the tire. is there. The camber angle change rate d [gamma] _Myuwai lateral friction coefficient mu _y, a rate of change of the lateral friction coefficient mu _y with respect to a change in camber angle gamma. The critical load F _zLI represents the upper limit standard of the use load of the tire, and is usually set to the value of the road index. The tire contact width is the width of the tire in contact with the contact surface in the lateral direction of the tire, and is usually set to a value of about 70% of the tire width.

The braking stiffness BCD _x is an initial change gradient of the longitudinal force F _x with respect to the change of the slip ratio κ. The cornering stiffness BCD _y is an initial change gradient of the lateral force F _y with respect to the slip angle α. The canvas Chifu Ness CS, is the initial change gradient of the lateral force F _y for camber angle gamma. The self-aligning stiffness SAS is an initial change gradient of the self-aligning torque M _z with respect to the change of the slip angle α, and the camber torque stiffness CTQ is an initial change gradient of the self-aligning torque M _z with respect to the change of the camber angle γ. It is.

Each of the braking stiffness BCD _x , cornering stiffness BCD _y , canvas stiffness CS, self-aligning stiffness SAS, and camber torque stiffness CTQ is normalized by dividing by the ground load F _z .

The lateral friction coefficient μ _y , normalized cornering stiffness, normalized canvas stiffness, normalized self-aligning stiffness, and normalized camber torque stiffness are estimated based on actual measurement values (or virtual values, the same applies hereinafter). These estimates are, for example, expressed as an approximate expression relating the vertical load F _z by fitting against measured values.

The lateral friction coefficient μ _y is expressed as, for example, a first-order approximate expression related to the ground load F _z . Normalized cornering steel Stiffness, for example, expressed as cubic approximate expression about the vertical load F _z, normalized canvas Chifu Ness, for example, it is represented as a quadratic approximate expression for the vertical load F _z. Normalized aligning steel Stiffness, for example, be represented as a quadratic approximate expression about the vertical load F _z, normalized camber Torx Chifu Ness, for example, be represented as a linear approximation expression for the vertical load F _z.

The approximate expression of the normalized self-aligning steel Stiffness, since normalization aligning steel Stiffness is 0 with decreasing vertical load F _z, a constant term becomes zero. The order of the approximate expression is not limited to the above example, but is determined according to the required approximation accuracy. n The fitting of the following approximate expression, n + 1 or more sets of measured values (e.g., the fitting of a third-order approximate expression of the normalized cornering steel Stiffness, 4 or more sets of vertical load F _z and the measured value of the normalized cornering steel Stiffness) Is needed.

The tire behavior estimation apparatus 10 performs a parameter input process following the setting process (S12). The input unit 12 includes, as input parameters, a tire ground contact load F _z [N], a slip ratio κ [−], a slip angle α [rad], a camber angle γ [rad], and a scaling factor μ _scale [−] of a friction coefficient μ. Enter.

The ground contact load F _z is a load that presses the tire against the ground contact surface. The slip ratio κ represents the difference between the rotational angular velocity ω of the tire and the rotational angular velocity ω ₀ of the tire during free rolling, and is usually defined as (ω−ω ₀ ) / ω ₀ . The slip angle α is an angle formed between the traveling direction of the tire ground contact center and the direction of the wheel center plane (the direction in which the tire faces). The camber angle γ is an angle formed by a rotation surface of the tire and a surface perpendicular to the ground contact surface. The scaling factor μ _{scale of} the friction coefficient μ is a variable for adjusting the friction coefficient μ (μ _x , μ _y ) according to the state of the ground contact surface.

The tire behavior estimation apparatus 10 performs pure front / rear slip characteristic estimation processing following the parameter input processing (S13). The pure front / rear slip is a slip generated in the front / rear direction of the tire when a braking / driving force is applied to the tire with a slip angle α of zero. The estimation unit 13 estimates the tire longitudinal force F _x [N] from the tire model equation (1) at the time of pure front / rear slip, based on the model coefficient and the input parameter. The longitudinal force F _x, a force generated in the front-rear direction of the tire between the tire and the ground surface when pure longitudinal slip.

_{F x = D x · sin [} C x · arctan {B x · κ-E x (B x · κ-arctan (B x · κ))}] ... (1)
here,
C _x = p ₁ (2)
_{D x = {μ x0 + dμ} x (F z -F z0μx)} F z · μ scale ... (3)
B _x = BCD _x / (C _x · D _x ) (4)
E _x = p ₂ (5)
BCD _x = C _Fx · F _z (6)

The coefficients C _x , D _x , B _x , and E _x are parameters that define the tire model at the time of pure front / rear slip, and are called shape factor C _x , peak factor D _x , stiffness factor B _x , and curvature factor E _x. Is done. As shown in Equation (4), the coefficient B _x corresponding to the stiffness factor at the time of pure front and rear slip is obtained by dividing the braking stiffness by the shape factor C _x and the peak factor D _x at the time of pure front and rear slip. The coefficient _Bx is also used in the composite slip characteristic estimation process described later. The coefficient BCD _x is the braking stiffness. Constant _{p 1} is in the range of 1.3 to 1.7 (e.g. 1.5), the constant _{p 2} is in the range of -2～1 (e.g. -0.5).

FIG. 3 is a diagram illustrating an estimation result of a pure front / rear slip characteristic (front / rear force F _x ). 3 shows, as a target the vertical load F _z of three patterns, the change trend of the pure longitudinal slip characteristic when the reduced again after increasing the slip ratio kappa, results of comparing estimated value and the measured value is shown Has been. According to FIG. 3, it can be seen that the pure front / rear slip characteristics of the tire are estimated with appropriate accuracy.

Returning to the description of FIG. 2, the tire behavior estimation device 10 performs a pure lateral slip characteristic estimation process following the pure front / rear slip characteristic estimation process (S <b> 14). Pure lateral slip is slip generated in the lateral direction of a tire when a braking angle and a driving force are set to zero and a slip angle is added. The estimation unit 13 estimates the tire lateral force F _y [N] from the tire model equation (7) at the time of a pure lateral slip, based on the model coefficient and the input parameter. The lateral force _Fy is a force generated in the lateral direction of the tire between the tire and the contact surface during a pure lateral slip.

Further, the estimation unit 13 estimates the tire self-aligning torque M _z [Nm] from the tire model equation (17) at the time of a pure lateral slip, based on the model coefficient, the input parameter, and the calculation result of the lateral force F _y. To do. The self-aligning torque _Mz is a moment generated around the vertical axis of the tire during a pure side slip.

_{F y = D y · sin [} C y · arctan {B y · x-E y (B y · x-arctan (B y · x))}] + S v ... (7)
here,
C _y = p ₃ (8)
D _y = (c _μy0 + C _μy1 · F _z ) F _z · μ _scale (9)
B _y = BCD _y / (C _y · D _y ) (10)
E _y = p ₄ {1 + p ₅ · γ · sign (x)} (11)
S _v = _{dγ μy} · F _z · γ · μ _scale (12)
S _h = (CS · γ−S _v ) / BCD _y (13)
x = α + S _h (14)
BCD _y = (cCSS ₀ + cCSS ₁ · F _z + cCSS ₂ · F _z ² + cCSS ₃ · F _z ³ ) F _z (15)
CS = (cCS ₀ + cCS ₁ · F _z + cCS ₂ · F _z ² ) F _z (16)

Coefficient _{C y, D y, B y} , E y, S v, S h is the parameter defining the tire model during pure lateral slip, shape factor _{C y,} peak factor _{D y,} steel Stiffness Factor _{B y,} Carver Chua factor _{E y,} the vertical shift _{S v,} a horizontal shift _{S h} referred. As shown in Expression (10), the coefficient B _y corresponding to the stiffness factor at the time of a pure side slip is obtained by dividing the cornering stiffness BCD _y by the shape factor C _y and the peak factor D _y at the time of a pure side slip. Factor B _y is also used for the estimation process of estimating and complex slip properties of the self-aligning torque M _z, which will be described later.

The term in parentheses in equation (9) corresponds to an approximate expression of the lateral friction coefficient μ _y , the term in parentheses in equation (15) corresponds to an approximate expression of normalized cornering stiffness, and The term in parentheses corresponds to an approximate expression of normalized canvas stiffness. Constant _{p 3} is in the range of 1 to 1.5 (e.g. 1.3), the constant _{p 4} is in the range of -2～1 (e.g. -1.3), the constant _{p 5} is in the range of -20 to 5 (e.g., -11 ).

Here, the lateral force F _y during net lateral slip of the formula (12), as shown in (13), in order to camber angle γ of the tire taking into account the effect on the lateral force F _y when pure lateral slip obtains the horizontal shift S _h with canvas Chifu Ness CS with obtaining the vertical shift S _v with the camber angle change rate d [gamma] _Myuwai lateral friction coefficient mu _y, is estimated using the vertical shift S _v and a horizontal shift S _h The

The horizontal shift S _h subtracts the vertical shift S _v from the value obtained by multiplying the input value of the camber angle γ canvas Chifu Ness CS, it obtained the subtraction result divided by the cornering steel Stiffness BCD _y.

The canvas stiffness CS, cornering stiffness BCD _y , and lateral friction coefficient μ _y are set approximately based on tire data.

_{M z = {t s · F} y0 + t c (F y -F y0)} cos (α) ... (17)
here,
_{F y0 = D y · sin [} C y · arctan (B y · α-E y (B y · α-arctan (B y · α)))] ... (18)
t _s = T (SAS / BCD _y , p ₆ , B _y , α) (19)
t _c = T (CTQ / CS, 0, B _y , α) (20)
nSAS = cSAS ₁ · F _z + cSAS ₂ · F _z ² (21)
SAS = nSAS · F _z (22)
CTQ = (cCTQ ₀ + cCTQ ₁ · F _z ) F _z (23)
_{T (D t, C t,} B y, α) = D t · cos [(1 + C t) · arctan {B t · α-E t (B t · α-arctan (B t · α))}] ... (24)
B _t = p ₇ · B _y ^p8 _μscale ^p9 (multiply by 0.25 when estimating t _c ) (25)
E _t = p ₁₀ (4 times when t _{c is} estimated.) (26)

The coefficient F _y0 is a contribution component due to the slip angle α in the lateral force F _y at the time of pure lateral slip. Coefficient t _s is the shift between the force application center in the longitudinal direction and the center of the lateral force F _y of the tire, referred to as the pneumatic trail. The coefficient t _c is a value obtained by dividing the self-aligning torque M _z generated by the camber angle γ by the lateral force F _y and is called a camber trail. The coefficient SAS is self-aligning stiffness, and the coefficient CTQ is camber torque stiffness.

The coefficient nSAS corresponds to an approximate expression for normalized self-aligning stiffness, and the term in parentheses in Expression (23) corresponds to an approximate expression for normalized camber torque stiffness. Constant _{p 6} is in the range of 0.05 to 0.15 (e.g. 0.1), the constant _{p 7} is a 2-3 range (e.g. 2.6), the constant _{p 8} is the range of 0.5 to 0.8 (e.g., 0.65). Constant _{p 9} is in the range of -0.5 to 0.1 (e.g. -0.35), the constant _{p 10} is in the range of -5 to-1 (e.g., -2.5).

Incidentally, when the vertical load F _z is used with respect to a vertical load F _z used in the model coefficient setting processing on the slip estimation processing increases, likely value of the approximate expression for the vertical load F _z is an abnormal value. Therefore, when the vertical load F _z for use in slip characteristic estimation processing is large, for example, the following correction is performed.

Horizontal friction coefficient mu _y is up to the vertical load F _z in accordance with an increase in the vertical load F _z is increased to about critical load F _ZLI decreases in proportion to the vertical load F _z, drop reduction slope at higher To do. Therefore, when the ground load F _z is equal to or greater than the critical load F _zLI , the equation (27) is applied instead of the approximate term of the lateral friction coefficient μ _y included in the equation (9). Constant _{p 11} is in the range of 0.5 to 0.9 (e.g., 0.7).

Here, the self-aligning torque M _z when pure lateral slip, as shown in equation (17), multiplied by the pneumatic trail t _s contribution component by the slip angle α of the lateral force F _y when pure lateral slip It is estimated by using a value obtained by adding a value obtained by multiplying the contribution component due to the camber angle γ to the camber trail t _c in the lateral force F _y at the time of a pure side slip.

Further, pneumatic trail t _s is set to approximate the self aligning steel Stiffness SAS and cornering steel Stiffness BCD _y using tire data, as shown in equation (19), the self-aligning steel Stiffness cornering the SAS steel Stiffness BCD _y And the stiffness factor B _y used to estimate the lateral force F _y at the time of a pure lateral slip.

Further, the camber trail t _c is set by approximating the canvas stiffness CS and the camber torque stiffness CTQ using tire data, and a value obtained by dividing the camber torque stiffness CTQ by the canvas stiffness CS as shown in the equation (20), And the stiffness factor B _y used to estimate the lateral force F _y at the time of a pure side slip.

Also, at least one approximate value of the tire lateral friction coefficient μ _y , canvas stiffness CS, cornering stiffness BCD _y , self-aligning stiffness SAS, and camber torque stiffness CTQ is the tire ground contact load F _z and critical load F _zLI. Is corrected according to the comparison result.

_{μ y = a · exp {c} μy1 (F z -F zLI) / a} + μ y∞ ... (27)
here,
_{a = μ yLI -μ y∞ ... (} 28)
_{μ y∞ = p 11 · μ yLI} ... (29)
_μyLI = c _μy0 + c _μy1 · _FzLI (30)

FIG. 4 is a diagram illustrating a correction result of the lateral friction coefficient μ _y . FIG. 4 shows a comparison between the value of the approximate expression of the lateral friction coefficient μ _y obtained based on the tire data of various tires and the value of the correction expression (27). As shown in FIG. 4, the approximate value is appropriate until the contact load F _z is about the critical load F _zLI , but it is too small beyond that. On the other hand, by applying the correction formula, an appropriate value is obtained even when the ground load _Fz is large.

Also, the normalized cornering stiffness is _likely to be set excessively by the approximate expression in parentheses in the expression (15) when the ground load F _z exceeds the critical load F _zLI . Therefore, when the ground load F _z is equal to or greater than the critical load F _zLI , the correction formula (31) is applied instead of the approximate formula. In addition, the approximate expression is applied until the contact load F _z is about 70% of the critical load F _zLI , and the correction expression is defined so that the approximate expression and the correction expression are smoothly connected in a section of about 70 to 100%. May be.

BCD _{y
= D y 'sin [2arctan {} F z / B y' -E y '(F z / B y' -arctan (F z / B y '))}] ... (31)
_{CS 300 '= p 12 · CS} 300 + p 13 ... (32)

Here, the coefficients B _y ′, D _y ′, and E _y ′ in Expression (31) are, for example, values of normalized cornering stiffness corresponding to loads corresponding to 70%, 100%, and 300% of the critical load F _zLI. Determined by fitting to. The cornering stiffness CS ₃₀₀ ′ corresponding to a load corresponding to 300% of the critical load F _zLI is obtained from the extrapolated value CS ₃₀₀ using, for example, Expression (32). However, a linear expression is used for extrapolation when obtaining CS ₃₀₀ . Constant _{p 12} is in the range of 0.2 to 0.35 (e.g. 0.27), the constant _{p 13} is in the range of 0.3 to 0.6 (e.g., 0.45).

FIG. 5 is a diagram illustrating a result of correcting the normalized cornering stiffness. FIG. 5 shows a comparison between the value of the approximate expression of the normalized cornering stiffness and the value of the correction expression together with the experimental data of the left and right tires. As tire data, an average value of experimental data of left and right tires is used in consideration of the influence of tire asymmetry. As shown in FIG. 5, the approximate value is appropriate until the ground load F _z is about the critical load F _zLI , but it is excessive beyond that. On the other hand, by applying the correction formula, an appropriate value is obtained even when the ground load _Fz is large.

Normalized canvas Chifu Ness, tends to vertical load F _z converges to constant are more than critical load F _ZLI. On the other hand, the normalized _canvas stiffness is _likely to be set excessively when the ground contact load F _z is about the critical load F _zLI or more. Therefore, the critical load _{F ZLI} greater vertical load _{F z,} in place of the value of the approximate expression in the parentheses of formula (16), a value corresponding to the critical load _{F ZLI} is applied.

FIG. 6 is a diagram illustrating a correction result of normalized canvas stiffness. FIG. 6 shows a comparison between an approximate expression and a correction expression for normalized canvas stiffness, along with experimental data for left and right tires. As shown in FIG. 6, the value of the approximate expression is appropriate until the ground load F _z is about the critical load F _zLI , but it is excessive beyond that. On the other hand, by applying the correction formula, an appropriate value is obtained even when the ground load _Fz is large.

Normalized aligning steel Stiffness increases with an increase in the vertical load F _z, tends to be constant after exceeding the maximum value. In contrast, the normalized self-aligning steel Stiffness is the vertical load F _z is greater than a vertical load F _z corresponding to the maximum value, likely to be under-set by approximate expression (21). Therefore, the maximum value is applied instead of the value of the approximate expression at the ground load F _z larger than the ground load F _z corresponding to the maximum value.

FIG. 7 is a diagram showing a result of correcting the normalized self-aligning stiffness. FIG. 7 shows a comparison between the value of the approximate expression of the normalized self-aligning stiffness and the value of the correction expression together with the experimental data of the left and right tires. As shown in FIG. 7, the value of the approximate expression is up vertical load F _z which corresponds to the maximum value is appropriate, has become excessively large at higher. On the other hand, by applying the correction formula, an appropriate value is obtained even when the ground load _Fz is large.

Camber Torx Chifu Ness, increases with an increase in the vertical load F _z, tends to be constant after exceeding the maximum value. On the other hand, the camber torque stiffness is likely to be set too small at a ground contact load F _z greater than the ground load F _z corresponding to the maximum value. Therefore, the maximum value is applied when the contact load F _{z is} larger than the contact load F _z corresponding to the maximum value.

FIG. 8 is a diagram illustrating a correction result of the normalized camber torque stiffness. In FIG. 8, the values of the approximate expression in parentheses of the normalized camber torque stiffness expression (23) and the correction expression are shown in contrast with the experimental data of the left and right tires. As shown in FIG. 8, the value of the approximate expression is up vertical load F _z which corresponds to the maximum value is appropriate, has become too small at higher. FIG. 8 shows the normalized value of the camber torque stiffness CS, and the normalized value is a value corresponding to the maximum value of the camber torque stiffness CS above the ground load F _z corresponding to the critical load F _zLI. Is shown. On the other hand, by applying the correction formula, an appropriate value is obtained even when the ground load _Fz is large.

FIG. 9 is a diagram illustrating an estimation result of the pure lateral slip characteristic (lateral force F _y ). FIG. 10 is a diagram illustrating an estimation result of the pure lateral slip characteristic (self-aligning torque M _z ). FIG 9 and FIG 10, the target contact load F _z of three patterns, the change trend of the pure lateral slip characteristic when the reduced again after increasing the slip angle alpha, of comparing estimated value and measured values Results are shown. According to FIGS. 9 and 10, it can be seen that the pure lateral slip characteristic of the tire is estimated with appropriate accuracy.

Returning to the description of FIG. 2, the tire behavior estimation device 10 performs a composite slip characteristic estimation process following the pure lateral slip characteristic estimation process (S <b> 15). The composite slip is a slip that occurs simultaneously in the front-rear direction and the lateral direction of the tire, and corresponds to a slip that combines a pure front-rear slip and a pure lateral slip. Based on the model coefficient, the input parameters, and the calculation results of the longitudinal force F _x and the lateral force F _y , the estimation unit 13 calculates the tire longitudinal force F _xC [from the tire model equations (33) and (34) at the time of composite slip. N] and lateral force F _yC [N] are estimated. The longitudinal force F _xC and the lateral force F _yC are forces generated in the longitudinal direction and lateral direction of the tire between the tire and the contact surface, respectively, at the time of composite slip.

Further, the estimation unit 13 calculates the tire self-aligning torque M _zC [Nm from the tire model equation (53) at the time of the composite slip based on the calculation result of the model coefficient, the input parameter, the longitudinal force F _xC and the lateral force F _yC. ] Is estimated. The self-aligning torque M _zC is a moment generated around the vertical axis of the tire during compound slip.

F _xC = F _x · G _x (33)
F _yC = F _y · G _y + S _vyx (34)
here,
G _x = cos [arctan {q ₃ · cos (arctan (q ₄ · κ)) α}] (35)
_{G y = cos [C yx ·} arctan {B yx (κ + S hyx)}] / cos {C yx · arctan (B yx · S hyx)} ... (36)
B _yx = q ₇ · cos {arctan (q ₈ · α)} (37)
C _yx = p ₁₄ (38)
S _hyx = S _hyx0 · B _xRef / B _x (39)
_{S vyx = w (c μy0 +} C μy1 · F z) × F z {1-F z '/ (b · F zLI)} · sin {arctan (-10γ)} × cos [arctan {B y · α-E _{g y} (B _y · α-arctan (B _y · α))}] × sin [arctan {B _x · κ-E _gx (B _x · κ -arctan (B _x · κ))}] G _y · μ _scale ... (40)
E _gx = p ₁₅ (41)
E _gy = p ₁₆ (42)
q ₃ = p ₁₇ · B _y / B _yRef (43)
q ₄ = p ₁₈ · B _x / B _xRef (44)
q ₇ = p ₁₉ · B _x / B _xRef · q _s (45)
q ₈ = p ₂₀ · B _y / B _yRef · q _s (46)
B _xRef = p ₂₁ / (p ₂₂ · C _x ) (47)
_{B yRef = p 23 / (p} 24 · C y) ... (48)
_{q s = p 25 + p 26} · tanh {(F z / F zLI -p 27) / p 28} ... (49)
S _hyx0 = p ₂₉ {1-min (F _z , F _zLI ) / F _zLI } (50)
F _z ′ = min (F _z , b · F _{z zLI} ) (51)
b = p ₃₀ (52)

The coefficients G _x and G _y are weighting coefficients for representing the influence of the composite slip on the pure front / rear slip characteristics and the pure lateral slip characteristics. The coefficients B _yx , C _yx , E _gx , E _gy , S _hyx , and S _vyx are parameters that define the tire model at the time of the composite slip, and include the shape factor C _yx , the stiffness factor B _yx , the curvature factor E _gx , E _This corresponds to _gy , vertical shift S _vyx , and horizontal shift S _hyx .

Constant _{p 14} is in the range of 0.9 to 1.1 (e.g. 1.03), the constant _{p 15} is in the range of -20 to 0 (e.g. -5), the constant _{p 16} is in the range of -20 to 0 (e.g., -10) And The constant p ₁₇ is in the range of 20 to 40 (eg 30.9), the constant p ₁₈ is in the range of 15 to 25 (eg 27.8), and the constant p ₁₉ is in the range of 10 to 20 (eg 14.1). The constant p ₂₀ is in the range of 7 to 15 (eg 11.9), the constant p ₂₁ is in the range of 20 to 30 (eg 24.5), and the constant p ₂₂ is in the range of 1.0 to 1.2 (eg 1.09). Constant _{p 23} is in the range of 15 to 30 (eg 20.2), the constant _{p 24} is in the range of 1.0 to 1.2 (e.g. 1.06), the constant _{p 25} is in the range of 0.5 to 1.0 (e.g., 0.7). Constant _{p 26} is in the range of -0.15～-0.05 (e.g. -0.1), the constant _{p 27} is in the range of -1.0 to-0.5 (e.g. -0.7), the constant _{p 28} is in the range of 0.05 to 0.15 (e.g. 0.1). The constant p ₂₉ is in the range of −0.01 to 0 (for example, −0.005), and the constant p ₃₀ is in the range of 1 to 2 (for example, 1.5).

_{M zC = {t s · F} y0 + t c (F y -F y0)} G y · cos (α) + G fy · F yC · F xC + G γ · F xC ... (53)
here,
t _s = T (SAS / BCD _y , p ₃₁ , B _y , x) (54)
t _c = T (CTQ / CS, 0, B _y , x) (55)
G _fy = p ₃₂ · w (1−F _z1 ′ / F _zLI ) (56)
_{G γ = p 33 · w ·} sin [arctan {p 34 · w (1-F z2 '/ (b · F zLI)) γ}] ... (57)
x = arctan {sqrt (x ₁ ² + x ₂ ² )} (58)
x ₁ = tan (α) (59)
x ₂ = B _x / B _y · κ (60)
F _z1 ′ = min (F _z , F _zLI ) (61)
F _z2 ′ = min (F _z , b · F _{z zLI} ) (62)
b = p ₃₅ (63)

The coefficients t _s and t _c correspond to a pneumatic trail and a camber trail, respectively, at the time of composite slip. The coefficients G _fy and G _γ correspond to the lateral movement coefficient of the longitudinal force application point due to the lateral force and the lateral movement coefficient of the longitudinal force application point due to the camber angle γ, respectively. Constant _{p 31} is in the range of 0.05 to 0.15 (e.g. 0.1), the constant _{p 32} is in the range of ^{-5 × 10 -5 ~-1 ×} 10 -5 ( e.g. -3 × 10 ^-5), the constant _{p 33} Is in the range of -0.6 to -0.3 (for example, -0.45). The constant p ₃₄ is in the range of 20 to 60 (for example, 40), and the constant p ₃₅ is in the range of 1 to 2 (for example, 1.5).

Here, the longitudinal force F _xC , lateral force F _yC , and self-aligning torque M _zC of the tire at the time of composite slip are _expressed by the equations (39), (40), (43) to (47), (54), ( 55) and (60), the stiffness factor B _x used for estimating the longitudinal force F _x at the time of pure front / rear slip, and the stiffness factor B _y used for estimating the lateral force F _y at the time of the pure side slip are calculated. Estimated.

FIG. 11 is a diagram illustrating an estimation result of the composite slip characteristic (lateral force F _yC ). FIG. 12 is a diagram illustrating an estimation result of the composite slip characteristic (self-aligning torque M _zC ). FIGS. 11 and 12 show composite slip characteristics (lateral force F _yC and self-aligning torque M _zC ) when the front / rear force F _xC is increased and then decreased again for three patterns of the ground load F _z. The result of comparing the measured value with the estimated value is shown. 11 and 12, it can be seen that the composite slip characteristic of the tire is estimated with appropriate accuracy.

  Returning to the description of FIG. 2, the tire behavior estimation device 10 performs a parameter output process following the composite slip characteristic estimation process (S <b> 16). The output unit 14 outputs the output parameters estimated by the estimation unit 13 and stored in the storage unit 15, that is, the estimated values of the pure front / rear slip characteristics, the pure lateral slip characteristics, and the composite slip characteristics.

As described above, according to the tire behavior estimating device 10 according to the embodiment of the present invention, the canvas Chifu Ness CS with obtaining the vertical shift S _v with the camber angle change rate d [gamma] _Myuwai lateral friction coefficient mu _y of the tire seek horizontal shift S _h using estimates the lateral force F _y of the tire at the time of net lateral slip with vertical shift S _v and a horizontal shift S _h. Here, in the conventional tire behavior estimation apparatus, the vertical shift _Sv and the horizontal shift _Sh are obtained using a large number of parameters determined by fitting a tire model to a lot of experimental data. In contrast, according to the tire behavior estimating device 10, by determining the horizontal shift S _h as a function of the canvas Chifu Ness CS with obtaining the vertical shift S _v as a function of camber angle change rate d [gamma] _Myuwai lateral friction coefficient mu _y, the effect of camber angle γ of the tire on the lateral force F _y during net lateral slip can be estimated easily and accurately. As a result, the tire behavior can be estimated easily and with high accuracy.

Further, by subtracting the vertical shift S _v from the value obtained by multiplying the input value of the camber angle γ canvas Chifu Ness CS, since obtaining a value obtained by dividing the subtraction result by the cornering steel Stiffness BCD _y as a horizontal shift S _h, camber angle γ is the effect on the lateral force F _y during net lateral slip can be estimated easily and accurately.

Further, since the canvas stiffness CS, cornering stiffness BCD _y , and lateral friction coefficient μ _y are approximated and set based on the tire data, parameters that define the tire model at the time of a pure lateral slip can be easily obtained.

Moreover, in the estimation process of the self-aligning torque M _z when pure lateral slip, as shown in equation (17), pneumatic trail t _s contribution component by the slip angle α of the lateral force F _y when pure lateral slip to the value obtained by multiplying, by adding the value obtained by multiplying the camber trail t _c on the contribution component by camber angle γ of the lateral force F _y when pure lateral slip, by using the addition result, the tire during net lateral slip The self-aligning torque _Mz is estimated. As a result, it is possible to easily and accurately estimate the influence of the slip angle α and the camber angle γ on the self-aligning torque _Mz during a pure side slip. By using the result of the estimation processing of the lateral force F _y when pure lateral slip, without requiring model data other than the model coefficients for the normalized aligning steel Stiffness normalized camber Torx Chifu Ness, self-aligning torque M _z Can be estimated easily and with high accuracy.

Further, the self-aligning stiffness SAS and the cornering stiffness BCD _y are set by approximation using tire data, and the value obtained by dividing the self-aligning stiffness SAS by the cornering stiffness BCD _y as shown in the equation (19) using steel Stiffness factor B _y used for estimation of the lateral force F _y at the transverse slip, it obtains the pneumatic trail t _s. As a result, the influence of the slip angle α on the self-aligning torque _Mz can be estimated easily and with high accuracy.

Further, the canvas stiffness CS and the camber torque stiffness CTQ are approximated and set using tire data, and the value obtained by dividing the camber torque stiffness CTQ by the canvas stiffness CS as shown in the equation (20) The camber trail t _c is obtained using the stiffness factor B _y used for estimating the lateral force F _y . As a result, the influence of the camber angle γ on the self-aligning torque M _z can be estimated easily and with high accuracy.

Further, at least one approximate value of the tire lateral friction coefficient μ _y , canvas stiffness CS, cornering stiffness BCD _y , self-aligning stiffness SAS, and camber torque stiffness CTQ is set as the tire ground contact load F _z and critical load F _zLI . It corrects according to the comparison result. Accordingly, even when the contact load F _z used in the estimation process with respect to the vertical load F _z used in the model coefficient setting process is large, it is possible to suppress the value of the approximate expression for the vertical load F _z becomes an abnormal value.

In the composite slip characteristic estimation processing, as shown in equations (39), (40), (43) to (47), (54), (55), and (60), the longitudinal force F _x at the time of pure front / rear slip using the steel Stiffness factor B _x used in the _estimation, and the steel Stiffness factor B _y used for estimation of the lateral force F _y during Jun'yoko slip, longitudinal force F _xC of the tire during the combined _slip, the lateral force F _yC, and At least one of the self-aligning torques M _zC is estimated. Thereby, the composite slip characteristics can be estimated easily and with high accuracy using the parameters used for estimating the pure front / rear slip characteristics and the pure lateral slip characteristics.

  In addition, embodiment mentioned above demonstrates best embodiment of the tire behavior estimation apparatus 10 which concerns on this invention, and the tire behavior estimation apparatus 10 which concerns on this invention is limited to what was described in this embodiment. It is not something. The tire behavior estimation device 10 according to the present invention is a modification of the tire behavior estimation device 10 according to the present embodiment or is applied to other devices without departing from the gist of the invention described in each claim. Also good.

  For example, in the above embodiment, the case where the principle of the present invention is applied to the tire behavior estimating device 10 has been described as an embodiment of the present invention. However, the principle of the present invention can be similarly applied to the tire behavior estimation method, the program for executing the tire behavior estimation method, and the recording medium on which the program is recorded.

In the above-described embodiment, the case where the pure front / rear slip characteristics, the pure lateral slip characteristics, and the composite slip characteristics are estimated by a series of estimation processes has been described. However, for example, the tire behavior estimation device 10 may be configured to estimate only the pure lateral slip characteristics or only the composite slip characteristics. In this case, for example, the coefficient B _y used for estimating the pure lateral slip characteristics or the coefficients B _x , B _y , F _x , F _y used for estimating the composite slip characteristics may be input as input parameters.

For example, in the above-described embodiment, an example of the tire model formula of slip characteristics has been given, but other terms may be added to the model formula or some terms may be added within a range not departing from the scope of the present invention. Modifications such as deleting or may be added. Although to an example of a range of constants _p 1 _{~p 35,} the range of the constant _p 1 _{~p 35} is not limited to the above examples.

  DESCRIPTION OF SYMBOLS 10 ... Tire behavior estimation apparatus, 11 ... Setting part, 12 ... Input part, 13 ... Estimation part, 14 ... Output part, 15 ... Storage part.

Claims (11)

A tire behavior estimation device for estimating the behavior of a vehicle tire,
The vertical shift is obtained using the camber angle change rate of the lateral friction coefficient of the tire and the horizontal shift is obtained using the canvas stiffness, and the lateral force of the tire during a pure lateral slip is estimated using the vertical shift and the horizontal shift. Including an estimation unit
The vertical shift S _v, the camber angle change rate _{dγ μy,} the ground contact load of the tire F _z, the camber angle of the tire gamma, the scaling factor of the friction coefficient of the tire when the mu _scale, the The vertical shift is determined by the equation S _v = _{dγ μy} · F _z · γ · μ _scale ,
When the horizontal shift is S _h , the canvas stiffness is CS, and the cornering stiffness is BCD _y , the horizontal shift is obtained by the formula S _h = (CS · γ−S _v ) / BCD _y ,
The estimating unit, the lateral force F _y, the peak factor D _y, the shape factor C _y, the steel Stiffness Factor B _y, the sum of the slip angle α and the horizontal shift S _h x, the Kaaba Chua factor and E _y Then, the lateral force is calculated using the formula F _y = D _y · sin [C _y · arctan {B _y · x−E _y (B _y · x−arctan (B _y · x))}] + S _v To estimate ,
Tire behavior estimation device.   2. The estimation unit according to claim 1, wherein the estimation unit subtracts the vertical shift from a value obtained by multiplying the canvas stiffness by an input value of the camber angle, and obtains a value obtained by dividing the subtraction result by the cornering stiffness as the horizontal shift. Tire behavior estimation device. The tire behavior estimation device according to claim 2 , further comprising a setting unit that approximates and sets the canvas stiffness and the cornering stiffness based on tire data. The tire behavior estimation device according to claim 3 , wherein the setting unit approximates and sets the lateral friction coefficient based on tire data. The estimation unit adds a camber trail to the contribution component due to the camber angle out of the lateral force during the pure lateral slip to the value obtained by multiplying the contribution component due to the slip angle out of the lateral force during the pure lateral slip. The tire behavior estimation device according to claim 3 or 4 , wherein the multiplied values are added, and the self-aligning torque of the tire at the time of a pure side slip is estimated using the addition result. The setting unit approximates and sets cornering stiffness and self-aligning stiffness using tire data,
The estimation unit obtains the pneumatic trail using a value obtained by dividing the cornering stiffness by the self-aligning stiffness and a stiffness factor used for estimating the lateral force at the time of a pure lateral slip. The tire behavior estimation device according to the description. The setting unit approximates and sets the canvas stiffness and camber torque stiffness using tire data,
The estimation unit obtains the camber trail by using the value obtained by dividing the camber torque stiffness by the canvas stiffness and the stiffness factor used for the estimation of the lateral force at the time of a pure lateral slip. The tire behavior estimation device according to the description.   The estimation unit uses the stiffness factor used for estimation of the longitudinal force at the time of pure front and rear slip, and the stiffness factor used for estimation of the lateral force at the time of pure lateral slip, and the longitudinal force of the tire at the time of compound slip, The tire behavior estimation device according to any one of claims 1 to 7, wherein at least one of lateral force and self-aligning torque is estimated.   The estimation unit corrects at least one approximate value of the lateral friction coefficient, the canvas stiffness, the cornering stiffness, the self-aligning stiffness, and the camber torque stiffness according to a comparison result between the ground contact load and the critical load of the tire. The tire behavior estimation device according to any one of claims 1 to 8. The tire behavior estimation device according to claim 6 , wherein the estimation unit divides the braking stiffness by a shape factor and a peak factor at the time of pure front and rear slip to obtain the stiffness factor at the time of pure front and rear slip.   The tire behavior estimation device according to claim 8, wherein the estimation unit obtains the stiffness factor during a pure side slip by dividing the cornering stiffness by a shape factor and a peak factor during a pure side slip.