JP5206490B2 - Vehicle ground contact surface friction state estimation apparatus and method - Google Patents

Description

  The present invention relates to a technique for estimating a friction state of a wheel contact surface or a road surface grip state of a wheel.

  As a conventional technique, there is one that estimates the road surface μ from the maximum value of the rotational angular acceleration of the drive wheel and performs optimal torque control so that the drive wheel does not slip.

Japanese Examined Patent Publication No. 6-78736

Incidentally, Patent Document 1 estimates the road surface μ from the maximum value of the rotational angular acceleration. On the other hand, it is also conceivable to estimate the road surface μ by comparing the detected value with a tire model (for example, a map).
However, when the tire used as the reference of the tire model is replaced with a tire having different tire characteristics (grip force or the like), there is a risk that the estimation accuracy of the road surface μ using the tire model is lowered.
An object of the present invention is to estimate the road surface μ with high accuracy based on a tire model that is not affected by tire characteristics.

In order to solve the above-mentioned problems, the present invention assumes tire characteristics representing a characteristic line established by the correlation between the tire force of the wheel obtained by the reference road surface of the reference road surface friction coefficient by the reference tire and the slip degree of the wheel. A tire characteristic correlation map modeled in this way is provided.
The present invention obtains a linear range detection value ratio that is a ratio of the tire force and the slip degree in a region where the correlation between the detected tire force and the detected slip degree is a linear relationship. Further, the present invention obtains a linear region reference value ratio between the tire force and the slip degree in a region where the correlation between the tire force and the slip degree of the assumed tire characteristic correlation map is a linear relationship. In the present invention, the ratio between the linear range detection value ratio and the linear range reference value ratio is used as a correction coefficient, and the tire characteristic correlation map is corrected based on the correction coefficient.

Here, when the tire characteristics are different, the correlation between the tire force and the slip degree may be different. At this time, the ratio between the tire force and the slip degree in a region where the correlation is linear is also different.
In such a case, there is a correlation between the change in the correlation between the tire force and the slip degree and the change in the ratio between the tire force and the slip degree in a region where the correlation is a linear relationship.

According to the present invention, based on the linear range detection value ratio and the linear range reference value ratio, the tire characteristic correlation map can be matched with that of the tire characteristic from which the linear range detection value ratio can be obtained.
Thereby, since the tire characteristic correlation map which comprises a tire model can be matched with the thing of an actual tire characteristic, road surface (micro | micron | mu) can be estimated with high precision based on the tire model which is not influenced by a tire characteristic.

FIG. 6 is a characteristic diagram showing a tire characteristic curve (Fx-λ characteristic curve) established between a wheel slip ratio λ and a wheel braking / driving force Fx, which is used for explaining a presupposed technology. It is a figure used in order to explain a premise technique, and is a characteristic view showing a tire characteristic curve (Fx-λ characteristic curve) and a friction circle of each road surface μ. It is the figure used in order to explain the premise technology, and shows the inclination of the tangent at the intersection of the tire characteristic curve (Fx-λ characteristic curve) of each road surface μ and the straight line passing through the origin of the tire characteristic curve FIG. It is the figure used in order to explain the premise technology, and shows the inclination of the tangent at the intersection of the tire characteristic curve (Fx-λ characteristic curve) of each road surface μ and the straight line passing through the origin of the tire characteristic curve It is another characteristic view. It is the figure used in order to explain the technology which is the premise, the ratio of braking / driving force Fx, the ratio of slip ratios λ, or the ratio of line lengths and road surface μ obtained for tire characteristic curves with different road surface μ It is a characteristic view which shows that the ratio of becomes equal. FIG. 5 is a characteristic diagram showing the relationship between braking / driving force Fx and slip ratio λ obtained on road surfaces with different road surface μ, which is used to explain the underlying technology. It is a figure used in order to explain the technology used as a premise, and is a characteristic view showing the relation between braking / driving force Fx and slip ratio λ obtained on road surfaces with different road surface μ for studless tires. It is the figure used in order to explain the premise technology, ratio (Fx / λ) of braking / driving force Fx and slip ratio λ of the arbitrary point of the tire characteristic curve (Fx-λ characteristic curve) and the arbitrary point FIG. 6 is a characteristic diagram composed of a set of plot points with a slope (μ gradient) of a tangent line of a tire characteristic curve in FIG. FIG. 9 is a characteristic diagram showing a characteristic curve (grip characteristic curve) obtained from the plotted points in FIG. 8, which is used to explain the underlying technology. FIG. 5 is a characteristic diagram showing a tire characteristic curve (Fy-βt characteristic curve) established between a wheel slip angle βt and a wheel lateral force Fy. It is a figure used in order to explain a premise technique, and is a characteristic figure showing a tire characteristic curve (Fy-βt characteristic curve) and a friction circle of each road surface μ. It is the figure used in order to explain the premise technology, and shows the inclination of the tangent at the intersection of the tire characteristic curve (Fy-βt characteristic curve) of each road surface μ and the straight line passing through the origin of the tire characteristic curve FIG. It is the figure used in order to explain the premise technology, and shows the inclination of the tangent at the intersection of the tire characteristic curve (Fy-βt characteristic curve) of each road surface μ and the straight line passing through the origin of the tire characteristic curve It is another characteristic view. It is the figure used in order to explain the technology which is the premise, ratio of the lateral forces Fy which are obtained about the tire characteristic curve where road surface μ differs, ratio of slip angle βt or ratio of line length and road surface μ It is a characteristic view which shows that ratio becomes equal. It is the figure used in order to explain the technology which is the premise, the ratio of tire force F which is obtained about the tire characteristic curve where road surface μ differs, ratio of slip degree S or ratio of line length and road surface μ It is a characteristic view which shows that ratio becomes equal. It is the figure used in order to explain a premise technique, and is a ratio (Fy / βt) of lateral force Fy and slip angle βt of an arbitrary point of a tire characteristic curve (Fy-βt characteristic curve), and the arbitrary point It is a characteristic view which shows the relationship (grip characteristic curve) with the inclination (micro gradient) of the tangent of the tire characteristic curve. It is a figure showing the schematic structure of the vehicles of a 1st embodiment. It is a block diagram which shows the structure of a vehicle running state estimation apparatus. It is a block diagram which shows the structure of a vehicle body slip angle estimation part. It is the figure used in order to explain the field force which acts on the body during turning. It is the figure used in order to explain the field force which acts on the body during turning. It is the characteristic view used in order to demonstrate the control map for setting a compensation gain. It is the figure used in order to explain the linear two-wheel model of vehicles. FIG. 5 is a flowchart showing a processing procedure of a road surface μ estimated value calculation unit, which calculates a road surface μ estimated value based on a straight line length connecting an actual measurement point and the origin of a tire characteristic curve. It is a characteristic view used in order to explain an EPS output adjustment map. It is a flowchart which shows the process sequence of a linear region Cp value estimation part. It is a flowchart which shows the process sequence of the determination process of a linear region Cp value estimation part. It is the figure used in order to demonstrate correction | amendment of a tire characteristic curve map. It is a characteristic view which shows two tire characteristic curves from which a tire characteristic differs. It is the figure used in order to explain correction of the characteristic map based on correction ratio R. It is a flowchart which shows the calculation procedure of the estimated value of road surface micro | micron | mu of the driving | running | working road surface performed using the characteristic map after correction | amendment. It is a characteristic view which shows two grip characteristic curves from which a tire characteristic differs. FIG. 6 is a characteristic diagram used for explaining correction of an actual measurement value based on a correction ratio R. It is a figure which shows schematic structure of the vehicle of 2nd Embodiment. It is a block diagram which shows the structure of a system control part. FIG. 5 is a flowchart showing a processing procedure of a road surface μ estimated value calculation unit, which calculates a road surface μ estimated value based on a straight line length connecting an actual measurement point and the origin of a tire characteristic curve. It is a characteristic view showing the relationship between the road surface μ (estimated value) and the gain Gain. It is a flowchart which shows the process sequence of a linear region Cp value estimation part. It is the figure used in order to demonstrate correction | amendment of a characteristic map.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(Technology that is the premise of the embodiment)
First, a technique that is a premise of the present embodiment will be described.
(1) Relationship between wheel slip ratio and wheel braking / driving force FIG. 1 shows a tire characteristic curve. This tire characteristic curve shows a general correlation established between the slip ratio λ of the drive wheel and the braking / driving force (or longitudinal force) Fx of the drive wheel. For example, a tire characteristic curve is obtained from a tire model such as MagicFormula. Here, the braking / driving force Fx is a force acting on the ground from the tire. The braking / driving force Fx corresponds to the wheel force acting on the wheel on the ground contact surface. The slip ratio λ of the wheel corresponds to the slip degree of the wheel.

  As shown in FIG. 1, in the tire characteristic curve, the relationship between the slip ratio λ and the braking / driving force Fx changes from linear (linear relationship) to non-linear (curve relationship) as the absolute value of the slip rate λ increases. That is, in the tire characteristic curve, when the slip ratio λ is within a predetermined range from zero, a linear relationship is established between the slip ratio λ and the braking / driving force Fx. In the tire characteristic curve, when the slip ratio λ (absolute value) increases to some extent (exceeding the predetermined range), the relationship between the slip ratio λ and the braking / driving force Fx becomes a non-linear relationship. Thus, the tire characteristic curve has a linear portion and a non-linear portion.

Such a transition between the slip ratio λ and the braking / driving force Fx or a transition from a linear relationship to a non-linear relationship is obvious when attention is paid to the slope of the tangent line of the tire characteristic curve. The slope of the tangent line of the tire characteristic curve here is a value indicated by a partial differential coefficient with respect to a ratio between a change amount of the slip ratio λ and a change amount of the braking / driving force Fx, that is, the slip ratio λ of the braking / driving force Fx. is there.
Here, as shown in FIG. 1, arbitrary straight lines a, b, c, d,... Passing through the origin of the tire characteristic curve are drawn. Then, the slope of the tangent line of the tire characteristic curve can be obtained at the intersections with arbitrary straight lines a, b, c, d,. The slope of the tangent line of the tire characteristic curve is different at each intersection. By paying attention to the inclination of the tangent line of the tire characteristic curve, it is possible to know the relationship between the slip ratio λ and the braking / driving force Fx or the state of transition from the linear relationship to the nonlinear relationship.

  Thereby, estimation of the friction state of the tire is also possible. For example, as shown in FIG. 1, if the tire characteristic curve is at a position x0 that is close to the linear region even in the non-linear region, it can be estimated that the tire friction state is in a stable state. And if the friction state of a tire is in a stable state, for example, it can be estimated that the tire is at a level at which its ability can be exhibited. Alternatively, it can be estimated that the vehicle is in a stable state.

  FIG. 2 shows tire characteristic curves and friction circles of various road surfaces μ. FIG. 2A shows tire characteristic curves of various road surfaces μ. 2 (b) to 2 (d) show the friction circle of each road surface μ. The road surface μ is, for example, 0.2, 0.5, or 1.0. As shown in FIG. 2 (a), the tire characteristic curve shows the same tendency qualitatively at each road surface μ. Further, as shown in FIGS. 2B to 2D, the friction circle becomes smaller as the road surface μ becomes smaller. That is, the braking / driving force that the tire can tolerate decreases as the road surface μ decreases. As described above, the tire characteristics are characteristics using the road surface friction coefficient (road surface μ) as a parameter. For this reason, as shown in FIG. 2, depending on the value of the road surface friction coefficient, a tire characteristic curve in the case of low friction, a tire characteristic curve in the case of medium friction, a tire characteristic curve in the case of high friction, and the like. Can be obtained.

  FIG. 3 shows the relationship between tire characteristic curves of various road surfaces μ and arbitrary straight lines b, c, d passing through the origin of the tire characteristic curve. As shown in FIG. 3, as in FIG. 1, tangent slopes are obtained at intersections with arbitrary straight lines b, c, d for tire characteristic curves of various road surfaces μ. That is, for the tire characteristic curves on various road surfaces μ, tangent slopes are obtained at the intersections with the straight line b. For tire characteristic curves on various road surfaces μ, tangential slopes are obtained at intersections with the straight line c. With respect to tire characteristic curves on various road surfaces μ, tangent slopes are obtained at intersections with the straight line d. As a result, a result can be obtained in which the tangent slopes of the tire characteristic curves of various road surfaces μ obtained at the intersections with the same straight line are the same.

  For example, FIG. 4 focuses on the straight line c shown in FIG. As shown in FIG. 4, the inclination of the tangent line at the intersection with the straight line c is the same in the tire characteristic curves of various road surfaces μ. That is, the ratio (Fx1 / λ1) between the braking / driving force Fx1 and the slip ratio λ1 indicating the intersection x1 with the tire characteristic curve with the road surface μ of μ = 0.2 is obtained. Further, a ratio (Fx2 / λ2) between the braking / driving force Fx2 and the slip ratio λ2 indicating the intersection x2 with the tire characteristic curve with the road surface μ of μ = 0.5 is obtained. Further, a ratio (Fx3 / λ3) between the braking / driving force Fx3 and the slip ratio λ3 indicating the intersection point x3 with the tire characteristic curve where the road surface μ is μ = 1.0 is obtained. Each value thus obtained is the same value. And the inclination of the tangent in each of these intersection x1, x2, x3 becomes the same value.

Thus, even if the road surface μ is different, the slope of the tangent line is the same for each tire characteristic curve at the value (λ, Fx) at which the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ is the same. Become.
The ratio of the braking / driving force Fx obtained between the different tire characteristic curves with respect to the values (λ, Fx) at which the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ is the same in each tire characteristic curve. Alternatively, the ratio between the slip ratios λ is equal to the ratio of the road surface μ.

With reference to FIG. 5, the relationship between the ratio between the braking / driving force Fx or the ratio between the slip ratios λ and the ratio between the road surface μ and each tire characteristic curve having a different road surface μ will be described. FIG. 5 shows tire characteristic curves obtained respectively on the road surface A (road surface μ = μ _A ) and the road surface B (road surface μ = μ _B ) having different road surfaces μ.
As shown in FIG. 5, the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ is the same (λ, Fx) (values indicated by ■ and ● in the same figure), respectively. What is the ratio (a2 / b2) between the braking / driving force a2 and the braking / driving force b2 and the ratio (μ _A / μ _B ) between the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B? It becomes the same value.

Similarly, the ratio (a3 / b3) of the slip ratio a3 and the slip ratio b3, which are respectively obtained with values (λ, Fx) at which the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ is the same. The ratio (μ _A / μ _B ) between the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B is the same value.
For this reason, the line length a1 obtained by connecting the value (λ, Fx) and the origin (0, 0) where the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ is the same, and and the ratio of the line length b1 (a1 / b1), becomes the same value as the ratio of the road surface mu values mu _B of the road surface mu values mu _a and the road surface B of road surface _{a (μ a / μ B)} . This can be proved geometrically as follows.

The triangle that can be drawn using the tire characteristic curve of road surface A (triangle with sides a1, a2, and a3) and the triangle that can be drawn using the tire characteristic curve of road surface B (triangle with sides b1, b2, and b3) are similar. It becomes a triangle. From this, the ratio of a1 and b1, the ratio of a2 and b2, and the ratio of a3 and b3 are the same value (a1: b1 = a2: b2 = a3: b3). And the ratio (a2 / b2) of a2 and b2 regarding the braking / driving force Fx and the ratio (a3 / b3) of a3 and b3 regarding the slip ratio λ are the values of the road surface μ μ value μ _A and the road surface B. It becomes the same value as the ratio (μ _A / μ _B ) with the road surface μ value μ _B. Accordingly, the ratio (a1 / b1) between the line length a1 and the line length b1 is the same value as the ratio (μ _A / μ _B ) between the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B. The following conclusion can be obtained.
As described above, if the ratio between the braking / driving forces Fx, the ratio between the slip ratios λ, or the ratio between the line lengths can be known, the ratio of the road surface μ can be known.

  FIG. 6 shows the relationship between the braking / driving force Fx and the slip ratio λ obtained on road surfaces having different road surfaces μ. In FIG. 6, the vibration waveform indicates actual measurement values obtained on the Dry road, the Wet road, and the low μ road. A dotted line indicates a characteristic curve of a tire (normal tire) on each road surface. As shown in FIG. 6, the tire characteristic curve on each road surface with different road surface μ maintains the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ, but the braking / driving force Fx decreases as the road surface μ decreases. And the slip ratio λ becomes small.

FIG. 7 shows the relationship between the braking / driving force Fx and the slip ratio λ obtained on the road surface with different road surface μ for the studless tire. In FIG. 7, the vibration waveform indicates actual values obtained on the Dry road, the Wet road, and the low μ road. Moreover, a dotted line shows the tire characteristic curve in each road surface. A thick dotted line indicates a tire characteristic curve of a normal tire.
As shown in FIG. 7, the tire characteristic curve (thin dotted line) on each road surface with different road surface μ maintains the ratio (Fx / λ) of braking / driving force Fx and slip ratio λ, but the road surface μ is small. As it is, the braking / driving force Fx and the slip ratio λ are reduced. Furthermore, the ratio (Fx / λ) of the braking / driving force Fx and slip ratio λ of the tire characteristic curve (thick dotted line) of the normal tire, and the braking / driving force Fx and slip of the tire characteristic curve (thin dotted line) of the studless tire The ratio (Fx / λ) to the rate λ is the same value. That is, the tire characteristic curve of the normal tire and the tire characteristic curve of the studless tire have a similar shape. That is, even when the gripping force, the tire surface shape, and the like are different as in a studless tire, the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ of the tire characteristic curve of the normal tire is the same value.

  FIG. 8 shows the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ at an arbitrary point of the tire characteristic curve and the slope of the tangent line of the tire characteristic curve at that arbitrary point (the braking / driving force / ∂slip ratio). ). In FIG. 8, the values obtained at each road surface μ (for example, μ = 0.2, 0.5, 1.0) are plotted. As shown in FIG. 8, regardless of the road surface μ, the ratio of the braking / driving force Fx to the slip ratio λ (Fx / λ) and the slope of the tangent line of the tire characteristic curve show a constant relationship.

FIG. 9 shows a characteristic curve obtained based on the plotted points in FIG. As shown in FIG. 9, in this characteristic curve, the ratio of the braking / driving force Fx to the slip ratio λ (Fx / λ) and the slope of the tangent line of the tire characteristic curve are always constant regardless of the road surface μ. It will be shown.
That is, this characteristic curve is established even on a road surface having a different road surface μ, such as a dry asphalt road surface or a frozen road surface. Alternatively, it can be said that the characteristic curve includes a high friction tire characteristic curve for a high friction road surface having a high friction coefficient and a low friction tire characteristic curve for a low friction road surface having a low friction coefficient lower than the high friction coefficient. Thus, it can be said that the characteristic curve shown in FIG. 9 shows the tire characteristic curve, as in FIG. However, in distinction from FIG. 1, the characteristic curve of FIG. 9 can also be called, for example, a grip characteristic curve.

  As shown in FIG. 9, in the region where the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ is small (small ratio region), the tangent slope of the tire characteristic curve is a negative value. In this region, as the ratio (Fx / λ) increases, the tangential slope of the tire characteristic curve once decreases and then increases. Here, the slope of the tangent line of the tire characteristic curve being a negative value indicates that the partial differential coefficient relating to the slip ratio of the braking / driving force is a negative value.

  Further, as shown in FIG. 9, in a region where the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ is large (large ratio region), the slope of the tangent line of the grip characteristic curve becomes a positive value. In this region, as the ratio (Fx / λ) increases, the tangential slope of the tire characteristic curve increases. That is, in the region where the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ is large, the grip characteristic curve has a monotonically increasing function.

  Here, the slope of the tangent line of the tire characteristic curve being a positive value indicates that the partial differential coefficient relating to the slip ratio of the braking / driving force is a positive value. In addition, the maximum inclination of the tangent line of the tire characteristic curve indicates that the inclination of the tangent line is in the linear region of the tire characteristic curve. In the linear region, the slope of the tangent line of the tire characteristic curve always shows a constant value regardless of the ratio between the braking / driving force Fx and the slip ratio λ.

  The slope of the tangent to the tire characteristic curve that can be obtained in this way (μ slope, hereinafter also referred to as Cp value) is a grip characteristic parameter, a variable that represents the grip state of the tire, or saturation of the force that the tire can exert in the lateral direction. It becomes a parameter indicating the state. Specifically, when the slope of the tangent line of the tire characteristic curve is a positive value, it indicates that a larger braking / driving force Fx can be generated by increasing the slip ratio λ. When the slope of the tangent line of the tire characteristic curve is zero or a negative value, it indicates that the braking / driving force Fx does not increase even if the slip ratio λ is increased and may decrease. Thus, it can be known from the slope of the tangent to the tire characteristic curve that the grip force of the tire is in the limit region. Thereby, for example, even when the wheel grip force is in the limit region, it is possible to appropriately estimate the margin of the tire grip force with respect to the friction limit.

A grip characteristic curve (FIG. 9) can be obtained by performing partial differential calculation on the tire characteristic curve (FIG. 1) and drawing continuously.
As described above, the inventor of the present application has the same tangential slope at the intersection of any one straight line passing through the origin of the tire characteristic curve and the tire characteristic curve for the tire characteristic curve of each road surface μ. I found a spot. That is, with respect to the tire characteristic curve of each road surface μ, a point was found where the slope of the tangent line was the same at the value (λ, Fx) where the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ was the same.

Thereby, the inventor of the present application has a characteristic curve (grip characteristic curve) having a relationship between the ratio of the braking / driving force Fx and the slip ratio λ (Fx / λ) and the slope of the tangent line of the tire characteristic curve regardless of the road surface μ. The result which can be expressed as was obtained (FIG. 9). By using this result, if the braking / driving force Fx and the slip ratio λ are known, based on the characteristic curve (grip characteristic curve), the information on the frictional state of the tire can be obtained without requiring information on the road surface μ. Can be obtained.
Further, the inventor of the present application uses a tire characteristic curve with different road surface μ, and the braking / driving forces Fx between the braking / driving forces Fx at values (λ, Fx) at which the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ is the same. It has been found that the ratio, the ratio between the slip ratios λ or the ratio between the line lengths becomes equal to the ratio of the road surface μ.
Thereby, if the ratio between the braking / driving forces Fx, the ratio between the slip ratios λ, or the ratio between the line lengths is known, the ratio of the road surface μ can be known.

(2) Relationship between wheel slip angle and wheel lateral force FIG. 10 shows a tire characteristic curve. This tire characteristic curve shows a general correlation established between the wheel slip angle βt and the wheel lateral force Fy. For example, by tuning a tire model based on experimental data, an equivalent characteristic diagram (tire characteristic curve) for two wheels is obtained for each of the front and rear wheels. Here, for example, a tire model is constructed on the basis of a magic formula. The lateral force Fy is a value represented by a cornering force or a side force. Here, the lateral force Fy is a force acting on the ground from the tire. Further, the lateral force Fy corresponds to the wheel force acting on the wheel on the ground contact surface. The wheel slip angle βt corresponds to the slip degree of the wheel.

  As shown in FIG. 10, in the tire characteristic curve, the relationship between the slip angle βt and the lateral force Fy changes from linear to non-linear as the absolute value of the slip angle βt increases. That is, in the tire characteristic curve, when the slip angle βt is within a predetermined range from zero, a linear relationship is established between the slip angle βt and the lateral force Fy. In the tire characteristic curve, when the slip angle βt (absolute value) increases to some extent (exceeding the predetermined range), the relationship between the slip angle βt and the lateral force Fy becomes a non-linear relationship. Thus, the tire characteristic curve has a linear portion and a non-linear portion.

The transition from the relationship between the slip angle βt and the lateral force Fy or the linear relationship to the non-linear relationship is obvious when attention is paid to the slope (gradient) of the tangent line of the tire characteristic curve. The inclination of the tangent line of the tire characteristic curve here is a value represented by a ratio of a change amount of the slip angle βt and a change amount of the lateral force Fy, that is, a partial differential coefficient related to the slip angle βt of the lateral force Fy.
Here, as shown in FIG. 10, arbitrary straight lines a, b, c,... Passing through the origin of the tire characteristic curve are drawn. Then, the tangent slope of the tire characteristic curve can be obtained at the intersections (intersections indicated by circles in FIG. 10) with arbitrary straight lines a, b, c,... Intersecting the tire characteristic curve. The slope of the tangent line of the tire characteristic curve is different at each intersection. By paying attention to the inclination of the tangent line of the tire characteristic curve, it is possible to know the relationship between the slip angle βt and the lateral force Fy or the state of transition from the linear relationship to the non-linear relationship.

  Thereby, estimation of the friction state of the tire is also possible. For example, as shown in FIG. 10, if the tire characteristic curve is at a position x0 close to the linear region even in the non-linear region, it can be estimated that the tire friction state is in a stable state. And if the friction state of a tire is a stable state, it can be estimated that a tire is in the level which can exhibit the capability, for example. Alternatively, it can be estimated that the vehicle is in a stable state.

  FIG. 11 shows tire characteristic curves and friction circles of various road surfaces μ. FIG. 11A shows tire characteristic curves of various road surfaces μ. FIGS. 11B to 11D show the friction circle of each road surface μ. The road surface μ is, for example, 0.2, 0.5, or 1.0. As shown in FIG. 11 (a), the tire characteristic curve shows the same tendency qualitatively at each road surface μ. Further, as shown in FIGS. 11B to 11D, the friction circle becomes smaller as the road surface μ becomes smaller. That is, as the road surface μ decreases, the lateral force that the tire can tolerate decreases. As described above, the tire characteristics are characteristics using the road surface friction coefficient (road surface μ) as a parameter. Therefore, as shown in FIG. 11, according to the value of the road surface friction coefficient, a tire characteristic curve in the case of low friction, a tire characteristic curve in the case of medium friction, a tire characteristic curve in the case of high friction, and the like can be obtained. it can.

  FIG. 12 shows a relationship between tire characteristic curves of various road surfaces μ and arbitrary straight lines a, b, c passing through the origin. As shown in FIG. 12, as in FIG. 12, tangent slopes are obtained at intersections with arbitrary straight lines a, b, c for tire characteristic curves of various road surfaces μ. That is, for the tire characteristic curves on various road surfaces μ, tangent slopes are obtained at the intersections with the straight line a. For tire characteristic curves on various road surfaces μ, tangent slopes are obtained at intersections with the straight line b. For tire characteristic curves on various road surfaces μ, tangential slopes are obtained at intersections with the straight line c. As a result, a result can be obtained in which the tangent slopes of the tire characteristic curves of various road surfaces μ obtained at the intersections with the same straight line are the same.

  For example, in FIG. 13, attention is paid to the straight line c shown in FIG. As shown in FIG. 13, the slope of the tangent at the intersection with the straight line c is the same for the tire characteristic curves of various road surfaces μ. That is, the ratio (Fy1 / βt1) between the lateral force Fy1 and the slip angle βt1 indicating the intersection x1 with the tire characteristic curve with the road surface μ of μ = 0.2 is obtained. Further, a ratio (Fy2 / βt2) between the lateral force Fy2 and the slip angle βt2 indicating the intersection x2 with the tire characteristic curve with the road surface μ of μ = 0.5 is obtained. Further, a ratio (Fy3 / βt3) between the lateral force Fy3 and the slip angle βt3 indicating the intersection x3 with the tire characteristic curve with the road surface μ of μ = 1.0 is obtained. Each value thus obtained is the same value. And the inclination of the tangent line in each intersection x1, x2, x3 becomes the same value.

Thus, even if the road surface μ is different, the slope of the tangent line is the same for each tire characteristic curve at the value (βt, Fy) where the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is the same. .
Then, regarding the values (βt, Fy) at which the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is the same in each tire characteristic curve, the ratio or slip angle between the lateral forces Fy obtained with different tire characteristic curves. The ratio between βt is equal to the ratio of the road surface μ.

With reference to FIG. 14, it will be described that the ratio of the lateral forces Fy or the ratio of the slip angles βt and the ratio of the road surface μ are equal for each tire characteristic curve having a different road surface μ. FIG. 14 shows tire characteristic curves obtained on a road surface A (road surface μ = μ _A ) and a road surface B (road surface μ = μ _B ) having different road surfaces μ.
As shown in FIG. 14, the values (βt, Fy) in which the ratio (Fy / βt) between the lateral force Fy and the slip angle βt are the same (values indicated by ■ and ● in the same figure), respectively. The ratio of the lateral force a2 and the lateral force b2 (a2 / b2) and the ratio (μ _A / μ _B ) between the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B are the same value. Become.

Similarly, the ratio (a3 / b3) between the slip ratio a3 and the slip ratio b3, which are respectively obtained with values (βt, Fy) at which the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is the same, The ratio (μ _A / μ _B ) between the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B is the same value.
Therefore, the tire characteristic curve obtained on the road surface A and the tire characteristic curve obtained on the road surface B have the same ratio (Fy / βt) between the lateral force Fy and the slip angle βt (βt, Fy). ) And the origin (0, 0), respectively, the ratio (a1 / b1) of the line length a1 and the line length b1, and the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B The ratio (μ _A / μ _B ) is the same value. This can be proved geometrically as follows.

A triangle (triangle with sides a1, a2 and a3) obtained using the tire characteristic curve of road surface A and a triangle (triangle with sides b1, b2 and b3) obtained using the tire characteristic curve of road surface B, and Becomes a similar triangle. From this, the ratio of a1 and b1, the ratio of a2 and b2, and the ratio of a3 and b3 are the same value (a1: b1 = a2: b2 = a3: b3). The ratio of a2 and b2 for the lateral force Fy and the ratio of a3 and b3 for the slip angle βt are the ratio of the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B (μ _A / Μ _B ). Accordingly, the ratio (a1 / b1) between the line length a1 and the line length b1 is the same value as the ratio (μ _A / μ _B ) between the road surface μ value μ _A of the road surface _A and the road surface μ value μ _B of the road surface B. The following conclusion can be obtained.

In FIG. 15, the horizontal axis represents the slip degree S and the vertical axis represents the tire force F.
Here, the braking / driving force Fx and the lateral force Fy can be considered as the tire force F, and the slip ratio λ and the slip angle βt can be considered as the slip degree S. Further, for example, the resultant force of the braking / driving force Fx and the lateral force Fy can be considered as the tire force F.
Thus, since the braking / driving force Fx and the lateral force Fy can be considered as the tire force F, and the slip ratio λ and the slip angle βt can be considered as the slip degree S, the tire force F and the slip degree S are also shown in FIGS. 14 can be obtained.

Therefore, as shown in FIG. 15, for each tire characteristic curve with different road surface μ, the ratio between tire forces F (a2 / b2), the ratio between slip degrees S (a3 / b3), or the ratio of line lengths (a1 / and b1), the ratio of the road surface μ (μ _a / μ _B) and are equal.
FIG. 16 shows the relationship between the ratio (Fy / βt) between the lateral force Fy and the slip angle βt at an arbitrary point of the tire characteristic curve and the slope of the tangent line of the tire characteristic curve at that arbitrary point (∂Fy / ∂βt). Indicates. As shown in FIG. 16, the ratio (Fy / βt) between the lateral force Fy and the slip angle βt and the tangent line of the tire characteristic curve for each road surface μ (for example, μ = 0.2, 0.5, 1.0). Shows a certain relationship with the slope.

  That is, this characteristic curve is established even on a road surface having a different road surface μ, such as a dry asphalt road surface or a frozen road surface. Alternatively, it can be said that the characteristic curve includes a high friction tire characteristic curve for a high friction road surface having a high friction coefficient and a low friction tire characteristic curve for a low friction road surface having a low friction coefficient lower than the high friction coefficient. Here, the characteristic curve shown in FIG. 16 is a tire characteristic curve as in FIG. Differentiating from FIG. 10, the characteristic curve of FIG. 16 can also be called a grip characteristic curve, for example.

  As shown in FIG. 16, in the region where the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is small (small ratio region), the tangent slope of the tire characteristic curve becomes a negative value. In this region, as the ratio (Fy / βt) increases, the tangential slope of the tire characteristic curve once decreases and then increases. Here, the negative slope of the tangent of the tire characteristic curve indicates that the partial differential coefficient regarding the slip angle of the lateral force is a negative value.

  Further, as shown in FIG. 16, in the region where the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is large (large ratio region), the tangent slope of the tire characteristic curve becomes a positive value. In this region, as the ratio (Fy / βt) increases, the tangential slope of the tire characteristic curve increases. That is, in the region where the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is large, the grip characteristic curve has a monotonically increasing function.

  Here, the inclination of the tangent line of the tire characteristic curve being a positive value indicates that the partial differential coefficient regarding the slip angle of the lateral force is a positive value. In addition, the maximum inclination of the tangent line of the tire characteristic curve indicates that the inclination of the tangent line is in the linear region of the tire characteristic curve. In the linear region, the slope of the tangent line of the tire characteristic curve always shows a constant value regardless of the ratio between the lateral force Fy and the slip angle βt.

  The tangential slope (μ slope) of the tire characteristic curve that can be obtained in this way is a grip characteristic parameter, a variable that represents the grip state of the tire, or a parameter that represents a saturated state of force that the tire can exert in the lateral direction. Specifically, when the slope of the tangent line of the tire characteristic curve is a positive value, it indicates that a stronger lateral force Fy (cornering force or the like) can be generated by increasing the slip angle βt. If the slope of the tangent line of the tire characteristic curve is zero or a negative value, it indicates that even if the slip angle βt is increased, the lateral force Fy (cornering force or the like) does not increase and may decrease. Thus, it can be known from the slope of the tangent to the tire characteristic curve that the grip force of the tire is in the limit region. Thereby, for example, even when the wheel grip force is in the limit region, it is possible to appropriately estimate the margin of the tire grip force with respect to the friction limit.

Note that the grip characteristic curve (FIG. 16) can be obtained by performing partial differential calculation on the tire characteristic curve (FIG. 10) and drawing continuously.
As described above, the inventor of the present application has the same tangential slope at the intersection of any one straight line passing through the origin of the tire characteristic curve and the tire characteristic curve for the tire characteristic curve of each road surface μ. I found a spot. That is, the tire characteristic curve of each road surface μ was found to have the same tangent slope at a value (βt, Fy) at which the ratio (Fy / βt) of the lateral force Fy to the slip angle βt is the same.

Accordingly, the inventor of the present application uses a characteristic curve (grip characteristic curve) having a relationship between the ratio of the lateral force Fy and the slip angle βt (Fy / βt) and the slope of the tangent line of the tire characteristic curve regardless of the road surface μ. The result which can be expressed was obtained (FIG. 16). By using this result, if the lateral force Fy and the slip angle βt are known, information on the friction state of the tire is obtained based on the characteristic curve (grip characteristic curve) without requiring information on the road surface μ. be able to.
Further, the inventor of the present application is a tire characteristic curve with different road surface μ, and the ratio between the lateral forces Fy at the values (βt, Fy) where the ratio (Fy / βt) between the lateral force Fy and the slip angle βt is the same, It has been found that the ratio between the slip angles βt or the ratio between the line lengths becomes equal to the ratio of the road surface μ.
Thereby, if the ratio between the lateral forces Fy, the ratio between the slip angles βt, or the ratio between the line lengths is known, the ratio of the road surface μ can be known.

(Embodiment)
Next, an embodiment realized by adopting the above technique will be described.
(First embodiment)
(Constitution)
The first embodiment is a vehicle to which the present invention is applied. FIG. 17 shows a schematic configuration of the vehicle. As shown in FIG. 17, the vehicle includes a steering angle sensor 21, a yaw rate sensor 22, a lateral acceleration sensor 23, a longitudinal acceleration sensor 24, a wheel speed sensor 25, an EPSECU (Electric Power Steering Electronic Control Unit) 26, and an EPS (Electric Power Steering). A motor 27 and a vehicle running state estimation device 28 are included.

The steering angle sensor 21 detects the rotation angle of the steering shaft 30 that rotates integrally with the steering wheel 29. The steering angle sensor 21 outputs the detection result (steering angle) to the vehicle running state estimation device 28. The yaw rate sensor 22 detects the yaw rate of the vehicle. The yaw rate sensor 22 outputs the detection result to the vehicle running state estimation device 28. The lateral acceleration sensor 23 detects the lateral acceleration of the vehicle. The lateral acceleration sensor 23 outputs the detection result to the vehicle running state estimation device 28. The longitudinal acceleration sensor 24 detects the longitudinal acceleration of the vehicle. The longitudinal acceleration sensor 24 outputs the detection result to the vehicle running state estimation device 28. The wheel speed sensor 25 detects the wheel speed of each wheel 31 _{FL to} 31 _RR of the vehicle body. The wheel speed sensor 25 outputs the detection result to the vehicle running state estimation device 28.

The EPS ECU 26 outputs a steering assist command to the EPS motor 27 based on the steering angle detected by the steering angle sensor 21. The steering assist command here is a command signal for performing steering force assist. Further, the EPS ECU 26 outputs a steering assist command to the EPS motor 27 based on the estimated road surface μ value output from the vehicle travel state estimating device 28.
The EPS motor 27 applies rotational torque to the steering shaft 30 based on the steering assist command output from the EPS ECU 26. Accordingly, the EPS motor 27 steers the left and right front wheels 31 _FL and 31 _FR via the rack and pinion mechanism (pinion 32 and rack 33) connected to the steering shaft 30, the tie rod 14 and the knuckle arm 15. To assist.

The vehicle traveling state estimation device 28 estimates the road surface state (road surface μ) of the traveling road surface based on the detection results of the steering angle sensor 21, the yaw rate sensor 22, the lateral acceleration sensor 23, the longitudinal acceleration sensor 24, and the wheel speed sensor 25. . The vehicle running state estimation device 28 outputs the estimation result to the EPS ECU 26.
FIG. 18 shows a configuration of the vehicle running state estimation device 28. As shown in FIG. 18, the vehicle running state estimation device 28 includes a vehicle body speed calculation unit 41, a vehicle body slip angle estimation unit 42, a tire slip angle calculation unit 43, a tire lateral force calculation unit 44, a road surface μ estimated value calculation unit 45, A linear region Cp value estimating unit (linear region μ gradient estimating unit) 46 and a map correcting unit 47 are provided.

The vehicle body speed calculation unit 41 estimates the vehicle body speed based on the wheel speed detected by the wheel speed sensor 25 and the longitudinal acceleration detected by the longitudinal acceleration sensor 24. Specifically, the vehicle body speed calculation unit 41 calculates the average value of the wheel speeds of the driven wheels 31 _RL and 31 _RR or the average value of the wheel speeds of the wheels 31 _{FL to} 31 _RR , and uses the calculated value as the vehicle body. This is the basic value of speed. The vehicle body speed calculation unit 41 corrects the basic value by the longitudinal acceleration. Specifically, correction is made from the basic value so as to eliminate the influence of errors caused by tire slipping during sudden acceleration and tire lock during sudden braking. The vehicle body speed calculation unit 41 uses the corrected value as the estimation result of the vehicle body speed. The vehicle body speed calculation unit 41 outputs the estimation result to the vehicle body slip angle estimation unit 42 and the tire lateral force calculation unit 44.

The vehicle body slip angle estimation unit 42 includes a steering angle detected by the steering angle sensor 21, a yaw rate detected by the yaw rate sensor 22, a lateral acceleration detected by the lateral acceleration sensor 23, a longitudinal acceleration detected by the longitudinal acceleration sensor 24, and a vehicle body speed calculation unit. Based on the vehicle body speed calculated by 41, the side slip angle (slip angle) of the vehicle is estimated.
FIG. 19 shows a configuration example of the vehicle body slip angle estimation unit 42. As shown in FIG. 19, the vehicle body slip angle estimation unit 42 includes a linear two-input observer 51 that estimates a vehicle state quantity (a vehicle side slip angle β, a slip angle β). As a result, the vehicle body slip angle estimation unit 42 estimates the side slip angle (slip angle) β of the vehicle. Here, a linear two-input observer 51 is constructed based on a two-wheel model of the vehicle. The two-wheel model of the vehicle can be expressed by the following equation (1) from the balance between the lateral force and moment of the vehicle.

  Here, A, B, C, and D are matrices determined by the linear two-wheel model of the vehicle. When the tire rudder angle is input u, and the yaw rate and lateral acceleration are output y, the state equation (output equation) of the equation (1) is as shown in the following equation (2).

Here, m is the vehicle mass. I is the yaw moment of inertia. l _f is the distance between the vehicle center of gravity and the front axle. l _r is the distance between the vehicle center of gravity and the rear axle. The cp _f is the front wheel cornering power (right and left wheels total). Cp _r is the rear wheel cornering power (the left and right wheels total value). V is the vehicle speed. β is the side slip angle of the vehicle. γ is the yaw rate. G _y is the lateral acceleration. a ₁₁ , a ₁₂ , b ₁ are the elements of the matrices A and B.
Based on this state equation, the yaw rate and the lateral acceleration are input, and a linear two-input observer 51 is created as the observer gain K1. Here, the observer gain K1 is a value set so as to be less susceptible to modeling errors and perform stable estimation.

  The linear two-input observer 51 includes a β estimation compensator 53 that corrects the input of the integrator 52. Thereby, the linear two-input observer 51 can ensure estimation accuracy even in the limit region. In other words, by including the β estimation compensator 53, not only in the road surface condition assumed at the time of designing the two-wheel model of the vehicle and in the linear region where the tire side slip angle does not become a non-linear characteristic, but also when the road surface μ changes or when the vehicle travels marginally. Even if it exists, the side slip angle β can be estimated with high accuracy.

FIG. 20 shows a turning vehicle running at a vehicle body side slip angle β. As shown in FIG. 20, the field force acting on the vehicle body, that is, the centrifugal force acting outward from the turning center is also generated in the direction shifted by the side slip angle β from the vehicle width direction. Therefore, the β estimation compensator 53 calculates a field force deviation β ₂ according to the following equation (3). This deviation β ₂ becomes a reference value (target value) G for correcting the vehicle slip angle β estimated by the linear two-input observer 51.

Here, G _x is the longitudinal acceleration. Further, as shown in FIG. 21, the balance of force due to the speed change is also taken into consideration. Thereby, when only the thing by turning is extracted, the said (3) Formula can be represented as following (4) Formula.

Then, the β estimation compensator 53 subtracts the target value β ₂ from the sideslip angle β estimated by the linear two-input observer 51. Further, the β estimation compensator 53 multiplies the subtraction result by the compensation gain K2 set by the control map of FIG. Then, the β estimation compensator 53 uses the multiplication result as an input of the integrator 52.
In the control map of FIG. 22, when the absolute value (| G _y |) of the lateral acceleration G _y of the vehicle is equal to or smaller than the first threshold value, the compensation gain K2 is zero. Further, if the absolute value of lateral acceleration G _y of the vehicle is of the second or more threshold greater than the first threshold value, the compensation gain K2 is relatively large constant value. Further, when the absolute value of lateral acceleration G _y of the vehicle is between the first threshold and the second threshold value, the absolute value becomes larger in the lateral acceleration G _y, compensation gain K2 is increased.

Thus, in the control map of FIG. 22, when the absolute value of lateral acceleration G _y has a value close to zero or less the first threshold value, and the compensation gain K2 is zero. Thereby, since it is not necessary to correct | amend under the condition where the turning G does not generate | occur | produce like the time of straight running, it is trying not to correct by mistake. Further, in the control map of FIG. 22, when the absolute value of lateral acceleration G _y is greater than the first threshold value increases (e.g., becomes greater than 0.1 G), proportional to the absolute value of lateral acceleration G _y Thus, the feedback gain (compensation gain) K2 is increased. Further, in the control map of FIG. 22, when the absolute value of lateral acceleration G _y is equal to or greater than the second threshold value (for example, equal to or greater than 0.5G), it has a constant value to stabilize the control of the compensation gain K2. By doing so, the estimation accuracy of the side slip angle β is improved.

The tire slip angle calculator 43 estimates the steering angle (tire steering angle δ) detected by the steering angle sensor 21, the yaw rate γ detected by the yaw rate sensor 22, the vehicle speed V calculated by the vehicle speed calculator 41, and the vehicle slip angle estimation. Based on the vehicle side slip angle (vehicle slip angle) β calculated by the unit 42, front and rear wheel slip angles β _f and β _r (wheel slip angle βt) are calculated according to the following equation (5).

The tire slip angle calculation unit 43 outputs the calculated front and rear wheel slip angles β _f and β _r to the road surface μ estimated value calculation unit 45 and the linear region Cp value estimation unit 46.
Tire lateral force calculating section 44, calculated based on the lateral acceleration G _y of the yaw rate γ and lateral acceleration sensor 23 the yaw rate sensor 22 detects detects the lateral force Fy f of the front and rear wheels in accordance with the following equation _(6), the Fy _r To do.

Here, the yaw rate γ and the lateral acceleration G _y are values as shown in FIG. Tire lateral force calculating section 44 outputs the lateral forces Fy f of the calculated front and rear _wheels, the Fy _r to road surface μ estimated value calculating section 45 and the linear range Cp value estimating unit 46.
The road surface μ estimated value calculation unit 45 stores a characteristic map (tire characteristic curve map) including tire characteristic curves of the reference road surface in a memory or the like. For example, the tire characteristic curve of the road surface of the road surface mu values mu _A shown in FIG. 14, and the tire characteristic curve of the reference road surface. The road surface μ estimated value calculation unit 45 has a tire characteristic curve map including the tire characteristic curve of the reference road surface corresponding to the front and rear wheels. That is, the road surface μ estimated value calculation unit 45 has a tire characteristic curve map for two front wheels and a tire characteristic curve map for two rear wheels.

  For example, a tire characteristic curve map is obtained by performing a running experiment in advance. As a running experiment, an acceleration circle turning running experiment is performed. As a result, a tire characteristic curve map including the tire characteristic curve of the reference road surface is obtained from the relationship between the fluctuation of the slip angle and the fluctuation of the lateral force obtained by the acceleration circle turning test on the reference road surface. It is also possible to obtain a tire characteristic curve map made up of tire characteristic curves on the reference road surface by calculations based on simulations and the like instead of running experiments.

The road surface μ estimated value calculation unit 45 calculates the road surface μ of the actual traveling road surface as an estimated value based on the tire characteristic curve map.
FIG. 24 shows a processing procedure for calculating the road surface μ (estimated value) of the actual traveling road surface based on the line length. Here, the road surface μ of the actual traveling road surface is calculated based on the lateral force Fy _f of the front wheels and the slip angle βt _f .
As shown in FIG. 24, first in step S1, the road surface μ estimated value calculation unit 45 detects a lateral force Fy _f (referred to as Fy _fb ) of the front wheels (obtains an output from the tire lateral force calculation unit 44).
Subsequently, in step S2, the road surface μ estimated value calculation unit 45 detects the slip angle βt _f (referred to as βt _fb ) of the front wheels (obtains an output from the tire slip angle calculation unit 43).

Subsequently, in step S3, the road surface μ estimated value calculation unit 45 determines that a straight line passing through the origin (0, 0) of the tire characteristic curve (tire characteristic curve map) of the reference road surface and the actual measurement point intersects the tire characteristic curve. (Βt _fa , Fy _fa ) is specified. Here, the actual measurement point is a value (βt _fb , Fy _fb ) indicated by the lateral force Fy _fb and the slip angle βt _fb detected in the steps S1 and S2.

In this step S3, the road surface μ estimated value calculation unit 45 determines that the lateral force and the slip angle are zero in two-dimensional coordinates having the wheel lateral force (tire force) and the wheel slip angle (slip degree) as coordinate axes. An inclination calculating means for calculating the inclination of the detected point (βt _fb , Fy _fb ) indicated by the detected current lateral force Fy _fb and slip angle βt _fb with respect to the origin of a certain two-dimensional coordinate is realized. Further, the road surface μ estimated value calculation unit 45 calculates a reference point that is an intersection of a straight line (a straight line having a line length a1 described later) extending from the origin of the two-dimensional coordinates with the inclination calculated by the inclination calculating means and the tire characteristic curve map. A reference point calculating means is realized.

In step S4 Subsequently, road surface mu estimation value calculation unit 45 calculates the road surface mu values mu _B of the actual traveled road surface. That is, the road surface μ estimated value calculation unit 45 has a straight line length b1 (= √ (βt _fb ² + Fy _fb ² )) connecting the measured point (βt _fb , Fy _fb ) and the origin of the tire characteristic curve of the reference road surface. Get. Further, the road surface μ estimated value calculating unit 45 is a straight line length a1 (which connects the value (βt _fa , Fy _fa ) of the tire characteristic curve of the reference road surface specified in step S3 and the origin of the tire characteristic curve). = √ (βt _fa ² + Fy _fa ² )). Further, the road surface μ estimated value calculation unit 45 calculates a ratio (b1 / a1) between the line length b1 and the line length a1. The road surface μ calculator 3 multiplies the calculated ratio (b1 / b1) by the road surface μ value μ _A of the reference road surface from which the tire characteristic curve map is obtained, and the multiplied value is the road surface of the actual road surface. obtained as the estimated value mu _B of _{μ (μ B = μ a ·} b1 / a1).

The processing procedure described above, the road surface mu estimation value calculation unit 45 calculates the estimated value mu _B of the road surface mu actual traveled road surface.
Here, by the same procedure, on the basis of the lateral force Fy _r and slip angle [beta] t _r of the rear wheels, it is also possible to calculate the road surface μ of the actual traveled road surface. Further, by the same procedure, the lateral force _Fy f of the front and rear wheels, Fy _r and slip angle [beta] t _f, based on [beta] t _r, for example, an average value _{((Fy f + Fy r)} / 2, (βt f + βt r ) / 2), the road surface μ of the actual traveling road surface can be calculated.

Road mu estimation value calculation unit 45 outputs the estimated value mu _B of the road surface mu actual traveled road surface calculated in the EPSECU 26.
The EPS ECU 26 outputs a steering assist command to the EPS motor 27 based on the calculated road surface μ. Specifically, the steering assist command is a command signal that reduces the output of the EPS motor 27 as the road surface μ decreases. For example, the steering assist command is determined based on the EPS output adjustment map. FIG. 25 shows an example of the EPS output adjustment map. As shown in FIG. 25, the EPS output adjustment map is a map that reduces the output of the EPS motor 27 as the road surface μ decreases.
The linear region Cp value estimating unit 46 estimates the linear region Cp value, and calculates a correction value (correction ratio) for map correction based on the estimated linear region Cp value.

FIG. 26 shows a processing procedure in the linear region Cp value estimation unit 46. As shown in FIG. 26, first, in step S11, the linear region Cp value estimation unit 46 acquires various data. Specifically, the linear region Cp value estimation unit 46 calculates the vehicle body speed calculated by the vehicle body speed calculation unit 41, the front and rear wheel slip angles β _f and β _r calculated by the tire slip angle calculation unit 43, and the tire lateral force calculation. lateral force _Fy f of the front and rear wheels part 44 is calculated to obtain the Fy _r. Further, the linear region Cp value estimation unit 46 acquires the yaw rate γ detected by the yaw rate sensor 22 and the longitudinal acceleration G detected by the longitudinal acceleration sensor 24.

  Subsequently, in step S12, the linear region Cp value estimation unit 46 determines whether or not the vehicle body speed V is greater than a predetermined threshold value Vth. The predetermined threshold value Vth is an experimental value, an empirical value, a theoretical value, or the like. For example, the predetermined threshold value Vth is 15 (km / h). When the vehicle body speed V is greater than the predetermined threshold value Vth (V> Vth), the linear region Cp value estimation unit 46 proceeds to step S13. In addition, when the vehicle body speed V is equal to or lower than the predetermined threshold value Vth (V ≦ Vth), the linear region Cp value estimation unit 46 ends the process shown in FIG. 26 (the process starts again from step S11). .

  In step S13, the linear region Cp value estimating unit 46 determines whether or not the host vehicle is turning in the linear region of the tire characteristic curve. The linear region of the tire characteristic curve is a region where the correlation (tire characteristic curve) between the lateral force Fy and the slip angle βt is a linear relationship. When the host vehicle is turning in the linear region of the tire characteristic curve, the linear region Cp value estimating unit 46 proceeds to step S14. If not, for example, when the host vehicle is turning in a non-linear region (curve region) of the tire characteristic curve, the linear region Cp value estimating unit 46 ends the processing shown in FIG. 26 (from step S11). Start processing again).

  FIG. 27 shows a specific determination procedure of step S13. As shown in FIG. 27, first, in step S21, the linear region Cp value estimation unit 46 determines whether or not the yaw rate γ acquired in step S11 is within a predetermined range. Specifically, the linear region Cp value estimation unit 46 determines whether or not the absolute value | γ | of the yaw rate γ is greater than a predetermined lower limit value γth1 and less than a predetermined upper limit value γth2 (γth1 <γth2). . The predetermined lower limit value γth1 and upper limit value γth2 are experimental values, empirical values, theoretical values, or the like. For example, the predetermined lower limit value γth1 is 2 (deg / s). The predetermined upper limit value γth2 is 20 (deg / s). When the yaw rate γ is within the predetermined range (lower limit value γth1 <| γ | <γth2), the linear region Cp value estimating unit 46 proceeds to step S22. If not, the linear region Cp value estimation unit 46 proceeds to step S25 and determines that the host vehicle is not turning in the linear region of the tire characteristic curve.

  In step S22, the linear region Cp value estimation unit 46 determines whether the longitudinal acceleration G acquired in step S11 is within a predetermined range. Specifically, the linear region Cp value estimation unit 46 determines whether or not the absolute value | G | of the longitudinal acceleration G is less than a predetermined threshold Gth. The predetermined threshold Gth is an experimental value, an empirical value, a theoretical value, or the like. For example, the predetermined threshold Gth is 0.2 (G). When the longitudinal acceleration G is within the predetermined range (| G | <Gth), the linear region Cp value estimation unit 46 proceeds to step S23. If not, the linear region Cp value estimating unit 46 proceeds to step S25.

In step S23, the linear region Cp value estimation unit 46 determines whether or not the slip angle βt acquired in step S11 is within a predetermined range. Specifically, the linear region Cp value estimation unit 46 determines whether or not the absolute value | βt | of the slip angle βt is less than a predetermined threshold value βtth. The predetermined threshold value βtth is an experimental value, an empirical value, a theoretical value, or the like. For example, the predetermined threshold value βtth is 5 (deg). Further, the slip angle βt mentioned here is one of the front and rear wheel slip angles β _f and β _r . Alternatively, the slip angle βt is an average value ((β _f + β _r ) / 2) of the slip angles β _f and β _r of the front and rear wheels.
When the slip angle βt is within a predetermined range (| βt | <βtth), the linear region Cp value estimating unit 46 proceeds to step S24 and determines that the host vehicle is turning in the linear region of the tire characteristic curve. To do. If not, the linear region Cp value estimating unit 46 proceeds to step S25.
The determination is made according to the above processing procedure.

Subsequently, in step S14, the linear region Cp value estimation unit 46 determines the ratio (Fy / βt) between the lateral force Fy and the slip angle βt based on the lateral force Fy and the slip angle βt acquired in step S11, that is, Cp. Calculate the value or μ gradient. Lateral force Fy here is any value of the lateral angle _Fy f, Fy _r of the front and rear wheels. Or, the lateral force Fy is an average value of the lateral angle _Fy f, Fy _r of the front and rear wheels _{((Fy f + Fy r)} / 2). The slip angle βt is one of the front and rear wheel slip angles β _f and β _r . Alternatively, the slip angle βt is an average value ((β _f + β _r ) / 2) of the slip angles β _f and β _r of the front and rear wheels.

Subsequently, in step S15, the linear region Cp value estimation unit 46 stores the ratio (Fy / βt) between the lateral force Fy and the slip angle βt calculated in step S14 as history information.
Subsequently, in step S16, the linear region Cp value estimation unit 46 statistically calculates a ratio (Fy / βt) between the plurality of lateral forces Fy and the slip angle βt stored as history information in step S15. Specifically, the linear region Cp value estimation unit 46 calculates the average value of the ratio (Fy / βt) between the plurality of lateral forces Fy and the slip angle βt that are the history information (((Fy / βt) ₁ + (Fy / βt) ₂ +... + (Fy / βt) _N ) / N).

  Subsequently, in step S17, the linear region Cp value estimation unit 46 determines whether or not the history data number (N) stored in step S15 is larger than a predetermined threshold value Nth. When the history data number (N) is larger than the predetermined threshold value Nth (N> Nth), the linear region Cp value estimation unit 46 proceeds to step S18. If not (N ≦ Nth), the linear region Cp value estimation unit 46 ends the processing shown in FIG. 26 (starts the processing again from step S11).

In step S18, the linear region Cp value estimation unit 46 calculates a correction ratio (correction value). Specifically, the linear region Cp value estimating unit 46 corrects the correction ratio R (= Cp0 /) as a ratio between the reference Cp value Cp0 (linear region reference value ratio) and the measured Cp value Cp0 ′ (linear region detected value ratio). Cp0 ′) is calculated.
Here, the measured Cp value Cp0 ′ is the average value obtained in step S16. This measured Cp value Cp0 ′ is composed of the measured Cp value, and is a value obtained when it is determined that the host vehicle is turning in the linear region of the tire characteristic curve as described above. The measured Cp value Cp0 ′ corresponds to the Cp value in the linear region of the tire characteristic curve of the tire mounted on the vehicle. The measured Cp value Cp0 ′ is also the ratio (Fy / βt) between the lateral force Fy and the slip angle βt in the linear region of the tire characteristic curve of the mounted tire. The reference Cp value Cp0 corresponds to the Cp value in the linear region of the tire characteristic curve on the reference road surface. The reference Cp value Cp0 is a ratio (Fy / βt) between the lateral force Fy and the slip angle βt in a region where a linear relationship is established between the lateral force Fy and the slip angle βt in the tire characteristic curve on the reference road surface. It is also.

The linear region Cp value estimation unit 46 finally calculates the correction ratio R by the processing procedure as described above. The linear region Cp value estimation unit 46 outputs the calculated correction ratio R to the map correction unit 47.
The correction ratio R is smaller than 1 when the measured Cp value Cp0 ′ is larger than the reference Cp value Cp0. That is, the correction ratio R is smaller than 1 when the μ gradient in the linear region of the tire characteristic curve obtained with an actual tire is standing. On the contrary, the correction ratio R is larger than 1 when the measured Cp value Cp0 ′ is smaller than the reference Cp value Cp0. That is, the correction ratio R is greater than 1 when the μ gradient in the linear region of the tire characteristic curve obtained with an actual tire is lying.

The map correction unit 47 corrects the tire characteristic curve map of the road surface μ estimated value calculation unit 45 with the correction ratio R. Specifically, the map correction unit 47 multiplies the slip angle βt ₀ of the tire characteristic curve map by the correction ratio R.
FIG. 28A shows an example of a tire characteristic curve map 45a. The tire characteristic curve map 45a has coordinate values (βt ₀₀ , Fy ₀₀ ), (βt ₀₁ , Fy ₀₁ ),..., (Βt _0n , Fy _0n ) that draw the tire characteristic curve of the reference road surface in two-dimensional coordinates. (Where n is an arbitrary integer).
The map correction unit 47 multiplies the slip angles βt ₀₀ , βt ₀₁ ,..., Βt _0n of the tire characteristic curve map 45a by the correction ratio R to perform correction as shown in FIG. A later tire characteristic curve map 45a is created.

(Operation and action)
(Vehicle control based on road surface μ estimation)
While the vehicle is traveling, the vehicle body slip angle estimation unit 42 is configured to detect the steering angle detected by the steering angle sensor 21, the yaw rate detected by the yaw rate sensor 22, the lateral acceleration detected by the lateral acceleration sensor 23, the longitudinal acceleration detected by the longitudinal acceleration sensor 24, and Based on the vehicle body speed calculated by the vehicle body speed calculation unit 41, the side slip angle (slip angle) of the vehicle is estimated.
On the other hand, the tire lateral force calculation unit 44 calculates the lateral force Fy based on the yaw rate γ detected by the yaw rate sensor 22 and the lateral acceleration G _y detected by the lateral acceleration sensor 23.
Then, the road surface μ estimated value calculation unit 45 calculates the road surface μ of the actual traveling road surface based on the lateral force Fy, the slip angle βt, and the tire characteristic curve map (for example, FIG. 28A).
The EPS ECU 26 controls the EPS motor 27 by a steering assist command based on the calculated road surface μ. Specifically, control is performed to reduce the output of the EPS motor 27 as the road surface μ becomes smaller.

(Map correction)
When the vehicle body speed V is greater than the predetermined threshold value Vth and it can be determined that the host vehicle is turning in the linear region of the tire characteristic curve, the linear region Cp value estimating unit 46 calculates (detects) at that time. The ratio (Fy / βt) of the lateral force Fy and slip angle βt to be stored is stored as history information. Further, the linear region Cp value estimating unit 49 calculates an average value (steps S11 to S16). Then, the linear area Cp value estimating unit 46 calculates the correction ratio R (= Cp0 / Cp0 ′) when the number of history data (N) becomes larger than the predetermined threshold value Nth (steps S17 to S17). Step S18). The map correction unit 47 corrects the tire characteristic curve map of the road surface μ estimated value calculation unit 45 based on the calculated correction ratio R (FIG. 28).

Here, in a general production vehicle, it is easily assumed that after the sale, the user replaces the tire with one different from that at the time of shipment. In this case, the tire characteristic of the vehicle also changes, and there is a possibility that the tire characteristic curve map held in the system may not be established. This is for the following reason.
For example, 15-inch tires and 18-inch tires having the same outer circumference and different flatness ratios are worn, and respective tire characteristic curves (Fy-βt characteristic curves) are obtained with the same vehicle. At this time, each tire characteristic curve (Fy-βt characteristic curve) is obtained as the same road surface μ value (for example, road surface μ value = 1.0). FIG. 29 shows tire characteristic curves (Fy-βt characteristic curves) of a 15-inch tire and an 18-inch tire, respectively.

  In the result shown in FIG. 29, the tire characteristic curve of the 15-inch tire and the tire characteristic curve of the 18-inch tire are inconsistent. Specifically, the tire characteristic curve of the 15-inch tire is shifted in a direction in which the slip angle βt is increased as a whole with respect to the tire characteristic curve of the 18-inch tire. That is, the tire characteristic curve of the 15-inch tire is expanded in the slip angle βt axis direction. Specifically, in the region where the slip angle βt is small (the region where the correlation is linear), the slope of the tangent line (μ gradient) of the tire characteristic curve is smaller in the 15-inch tire. On the other hand, in a region where the slip angle βt is large (for example, a region where the μ gradient is zero or less), the lateral force Fy shows the same or almost the same value regardless of the 15-inch tire and the 18-inch tire.

Here, for example, a case is considered where the road surface μ estimated value calculation unit 45 has a tire characteristic curve obtained from an 18-inch tire as a tire characteristic curve map (characteristic map). In such a case, it is assumed that a 15-inch tire is actually put on the vehicle. In this case, the road surface μ estimated value calculation unit 45 refers to the tire characteristic curve map obtained based on the 18-inch tire (reference tire), and the line length b of the actual measurement point and the extension line of the line length of the actual measurement point. Based on the ratio (b / a18) to the straight line length a18 that intersects the tire characteristic curve of the reference road surface, the road surface μ of the actual traveling road surface is calculated. That is, the ratio (b / a18) is multiplied by the reference road surface μ value μ ₁₈ obtained from the tire characteristic curve map, and the multiplied value is calculated as an estimated value μ _b of the actual road surface μ. (Μ _b = μ ₁₈ · b / a ₁₈ ).

However, as described above, there is an error between the tire characteristic curve of the 15-inch tire and the tire characteristic curve of the 18-inch tire. As a result, the estimation accuracy of the road surface μ of the actual traveling road surface is also lowered. That is, as the line length obtained for the tire characteristic curve of the reference road surface, the line length a15 obtained for the tire characteristic curve of the 15-inch tire to be mounted was used for the tire characteristic curve of the 18-inch tire. The line length a18 is used. Therefore, it will include an estimate mu _b of the road surface mu actual traveled road surface error corresponding to the difference between the line length a15 and line length a18. For example, the estimated value is smaller than the actual road surface μ value.

On the other hand, in this embodiment, the map correction unit 47 corrects the tire characteristic curve map of the road surface μ estimated value calculation unit 45 with the correction ratio R. Specifically, the map correction unit 47 corrects the tire characteristic curve map by multiplying the slip angle βt ₀ of the tire characteristic curve map by the correction ratio R.
Here, the correction ratio R is a ratio between the reference Cp value Cp0 and the measured Cp value Cp0 ′. When viewed in the direction of the slip angle βt axis (assuming that the lateral force Fy is the same value), this correction ratio R is determined by the slip angle βt in the linear region of the tire characteristic curve on the reference road surface and the tire characteristics of the tire mounted on the vehicle. This corresponds to the ratio to the slip angle βt in the linear region of the curve.

The map correction unit 47 multiplies the correction ratio R thus defined by the slip angle βt ₀ of the tire characteristic curve map. As a result, the tire characteristic curve map changes by a correction ratio R times in the slip angle βt axis direction. That is, the shape of the tire characteristic curve map is enlarged or reduced by a correction ratio R times in the slip angle βt axis direction with the origin fixed. As a result, the tire characteristic curve map shows the tire characteristic curve of the tire to be mounted from the one showing the tire characteristic curve of the reference road surface.

When applied to the above-described example, the reference Cp value Cp0 is a Cp value in the linear region of the tire characteristic curve of the 18-inch tire. The measured Cp value Cp0 ′ is a Cp value in the linear region of the tire characteristic curve of the 15-inch tire mounted on the vehicle.
Map correcting unit 47 multiplies the correction ratio R is the ratio of such a reference Cp value Cp0 and the measured Cp values Cp0 'to slip angle [beta] t ₀ of the tire characteristic curve map (tire characteristic curve 18-inch tire). As a result, the tire characteristic curve map changes R times in the slip angle βt axis direction. That is, as changed from (a) to (b) of FIG. 30, the shape of the tire characteristic curve map is expanded by the correction ratio R times in the slip angle βt axis direction with the origin fixed. As a result, the tire characteristic curve map shows the tire characteristic curve of the 15-inch tire mounted from the one showing the tire characteristic curve of the 18-inch tire.

  That is, when changing to a tire having a large Cp value in the linear region of the tire characteristic curve, the map correction unit 47 reduces (compresses) the shape of the tire characteristic curve map in the slip angle βt axis direction. On the other hand, the map correction unit 47 expands the shape of the tire characteristic curve map in the slip angle βt axis direction when the tire is changed to a tire having a small Cp value in the linear region of the tire characteristic curve. As a result, the tire characteristic curve map shows the tire characteristic curve of the changed tire.

Then, the road surface μ estimated value calculation unit 45 calculates the road surface μ of the actual traveling road surface based on the corrected tire characteristic curve map. When applied to the above-described example, the road surface μ estimated value calculation unit 45 calculates the road surface μ of the actual traveling road surface based on the tire characteristic curve map corrected to the tire characteristic curve of the 15-inch tire. That is, as shown in FIG. 31, the road surface μ estimated value calculation unit 45 calculates the ratio between the line length b of the actual measurement point and the line length a15 up to the intersection of the tire characteristic curve of the reference road surface (corrected tire characteristic curve map). (B / a15) is obtained. Then, the road surface μ estimated value calculation unit 45 multiplies the ratio (b / a15) by the road surface μ value μ ₁₅ of the reference road surface from which the tire characteristic curve map is obtained, and multiplies the multiplied value by the road surface μ of the actual traveling road surface. to calculated as an estimate _{μ b (μ b = μ 15} · b / a15).

Thereby, the road surface μ estimated value calculation unit 45 can estimate the estimated value μ _b of the road surface μ of the actual traveling road surface with high accuracy. The EPS ECU 26 controls the EPS motor 27 by a steering assist command based on the estimated value of the road surface μ.
FIG. 32 shows the result of comparison on the grip characteristic curve. As shown in FIG. 32, there is an error between the 15-inch tire and the 18-inch tire even in the grip characteristic curve. Specifically, the maximum value of the μ gradient of the grip characteristic curve moves on a straight line passing through the origin of the grip characteristic curve (a virtual line described only for explanation). And if it becomes a 15-inch tire, the maximum value of (mu) gradient will become small.

  Furthermore, as shown in FIG. 32, the grip characteristic curves are different in size while maintaining their shapes. That is, the grip characteristic curves are similar and have different sizes. In this example, when a 15-inch tire is used, the shape of the grip characteristic curve is reduced. This is equivalent to an increase in the shape of the tire characteristic curve in the slip angle βt axis direction for a 15-inch tire.

(Modification of the first embodiment)
(1) In the first embodiment, the tire characteristic curve map 45a is corrected based on the correction ratio R. On the other hand, the actual measurement value can be corrected based on the correction ratio R. Specifically, the slip angle βt of the actually measured value (βt, Fy) is multiplied by the reciprocal (1 / R) of the correction ratio R (βt · 1 / R, Fy). Accordingly, the road surface μ of the actual traveling road surface is calculated based on the values (βt · 1 / R, Fy) and the ratio (Fy / (βt / R)) converted to be comparable with the reference road surface characteristic map 45a. (See FIG. 24). In this way, a correction process substantially similar to that when correcting the tire characteristic curve map 45a based on the correction ratio R is realized, and the road surface μ of the actual traveling road surface can be estimated with high accuracy.

  When applied to the above-described example, as shown in FIG. 33, the road surface μ estimated value calculation unit 45 or the map correction unit 47 performs the actual measurement (βt, Fy) obtained with the mounted 15-inch tire on the slip angle βt. Multiply the reciprocal of the correction ratio R (1 / R). As a result, the road surface μ estimated value calculation unit 45 can calculate the road surface μ of the actual traveling road surface based on the tire characteristic curve of the reference road surface that is not corrected (the tire characteristic curve of the 18-inch tire). Thereby, the road surface μ of the actual traveling road surface can be estimated in the same manner as when the tire characteristic curve map is corrected.

(2) In the first embodiment, when it can be determined that the host vehicle is turning in the linear region of the tire characteristic curve, the lateral force Fy and the slip angle βt during the turning are stored and averaged. The measured Cp value Cp0 ′ is calculated. However, the first embodiment is not limited to this. That is, if there is a margin in the memory and processing capacity of the system, the lateral force Fy and the slip angle βt are measured during traveling (traveling including straight traveling in addition to turning) regardless of whether or not the vehicle is turning, and an approximate curve As a tire characteristic curve is obtained. The measured Cp value Cp0 'can also be calculated from the vicinity of the origin of the tire characteristic curve obtained as an approximate curve, that is, from the linear region of the tire characteristic curve.

(3) The tire characteristic correlation map may be expressed as a mathematical expression (function expression) using the tire force and the slip degree obtained on the reference road surface of the reference road surface friction coefficient as variables. In this case, the tire characteristic correlation map is corrected by changing the slip degree as a variable based on the correction coefficient (correction ratio R). For example, the variable slip degree is changed by multiplying or dividing the variable slip degree by a correction coefficient.
(4) In the first embodiment, the average value is calculated by statistical calculation (step S16). On the other hand, the value can also be calculated by other statistical calculation, for example, the mode value.

(5) The detected value ratio during traveling calculated based on the tire force and the slip degree detected during traveling of the host vehicle is a region where the correlation between the tire force and the slip degree obtained on the dry road surface is a linear relationship. When the tire force is less than or equal to the slip ratio, the detected value during running can be used as a linear range detection value ratio to correct the tire characteristic correlation map. That is, the linear range detection value ratio can be selected based on the dry road surface.

(6) Based on the tire force and the slip degree (see FIG. 15), it is possible to calculate the road surface μ of the traveling road surface and correct the tire characteristic curve map as described above. In this case, a tire characteristic curve map is obtained based on the relationship between the tire force and the slip degree. For example, in the correction of the tire characteristic curve map, the slip is set so that the Cp value (Δ tire force / Δ slip degree) in the linear region of the tire characteristic curve matches the tire characteristic curve of the mounted tire. Multiply the degree by the correction ratio R.

(7) The estimated value of the road surface μ of the traveling road surface as described above can be calculated and the tire characteristic curve of the reference road surface can be corrected individually for each of the front and rear wheels. Thereby, even if the user wears tires having different characteristics between the front and rear wheels, the road surface μ of the traveling road surface can be estimated with the same accuracy as described above.
(8) In the first embodiment, the steering assist torque of the vehicle is controlled as the driving behavior control of the vehicle based on the estimated road surface μ. On the other hand, other control amounts (for example, braking / driving torque) for vehicle travel control can be controlled based on the estimated road surface μ.

In the first embodiment, the tire lateral force calculation unit 44 implements tire force detection means for detecting the tire force of the wheels.
Further, the tire slip angle calculation unit 43 realizes a slip degree detecting means for detecting the slip degree of the wheel.
The tire characteristic curve map (characteristic map) 45a of the road surface μ estimated value calculation unit 45 is established by the correlation between the wheel tire force obtained on the reference road surface of the reference road surface friction coefficient by the reference tire and the slip degree of the wheel. A tire characteristic correlation map modeled on the assumption of tire characteristics representing characteristic lines is realized.

  Here, the characteristic line established by the correlation between the tire force of the wheel and the slip degree of the wheel is a line including a curve that can be expressed as existing on the secondary plane. In the two-dimensional map, one of the vertical axis and the horizontal axis is one of the tire force and the slip, and the other of the vertical axis and the horizontal axis is the other of the tire force and the slip. In this embodiment, for example, as shown in FIG. 14, the characteristic line is a Fy-βt characteristic curve on coordinates where the horizontal axis is the slip angle βt and the vertical axis is the lateral force Fy.

  In addition, the process of step S3 of the road surface μ estimated value calculation unit 45 is based on the two-dimensional coordinates in which the tire force and the slip degree are zero in the two-dimensional coordinates having the wheel tire force and the wheel slip degree as coordinate axes. Inclination calculating means for calculating the current tire force detected by the tire force detecting means relative to the origin and the inclination of the detection point indicated by the current slip degree detected by the slip degree detecting means is realized.

Further, the process of step S3 of the road surface μ estimated value calculation unit 45 calculates a reference point that is an intersection of the straight line extending with the inclination calculated by the inclination calculating means from the origin of the two-dimensional coordinates and the tire characteristic correlation map. A reference point calculating means is realized.
Further, the process of step S4 of the road surface μ estimated value calculation unit 45 calculates the road surface friction coefficient of the current road surface based on the detection point, the reference point calculated by the reference point calculation means, and the reference road surface friction coefficient. A road surface friction coefficient calculating means is realized.

  Further, the process of step S4 of the road surface μ estimated value calculation unit 45 is a detection point distance calculation means for calculating a detection point distance that is a distance between the origin and the detection point (measurement point, actual measurement value) in two-dimensional coordinates. And reference point distance calculating means for calculating a reference point distance that is a distance between the origin and the reference point in the two-dimensional coordinates. Then, the process of step S4 of the road surface μ estimated value calculation unit 45 is performed based on the detection point distance calculated by the detection point distance calculation unit and the reference point distance calculated by the reference point distance calculation unit. It is possible to calculate the friction coefficient.

  Further, the linear range Cp value estimation unit 46 and the map correction unit 47 have a linear relationship (first-order linearity) between the tire force detected by the tire force detection unit and the slip degree detected by the slip degree detection unit. The linear range detection value ratio that is the ratio between the tire force and the slip degree in a region and the correlation between the tire force and the slip degree in the assumed tire characteristic correlation map is a linear relationship (primary linear). A correction unit that corrects the tire characteristic correlation map based on the correction coefficient is realized by using a ratio of the linear region reference value ratio that is a ratio of the tire force and the slip degree in a region as a correction coefficient.

  Further, in the first embodiment, in the vehicle contact surface friction state estimation method for estimating the contact surface grip characteristics of the vehicle wheel, the wheel tire force obtained on the reference road surface of the reference road surface friction coefficient by the reference tire In a region where the correlation between the detected tire force and the detected slip degree is a linear relationship, using a tire characteristic correlation map modeled on the assumption of the tire characteristic representing the characteristic line established by the correlation with the wheel slip degree A linear range detection value ratio that is a ratio between the detected tire force and the detected slip degree, and the tire force and the tire force in a region where the correlation between the tire force and the slip degree in the assumed tire characteristic correlation map is linear. A correction step for correcting the tire characteristic correlation map based on the correction coefficient, using a ratio with a linear range reference value ratio, which is a ratio to the slip degree, as a correction coefficient; In a two-dimensional coordinate having the yaw force and the slip degree as coordinate axes, the inclination of the detection point indicated by the current detected tire force and the detected slip degree with respect to the origin of the two-dimensional coordinate where the tire force and the slip degree are zero is calculated. An inclination calculating step; a reference point calculating step for calculating a reference point that is an intersection of a straight line extending from the origin of the two-dimensional coordinates with the inclination calculated in the inclination calculating step and the tire characteristic correlation map; And a road surface friction coefficient calculating step of calculating a road surface friction coefficient of a current road surface based on the reference point calculated in the reference point calculating step and the reference road surface friction coefficient. .

(Effects of the first embodiment)
(1) The road surface μ estimated value calculation unit 45 models a tire characteristic representing a characteristic line established by a correlation between a wheel tire force obtained from a reference road surface with a reference road surface friction coefficient and a slip degree of the wheel by the reference tire. A tire characteristic curve map which is a tire characteristic correlation map is provided.
Then, the correction means having the linear region Cp value estimation unit 46 and the map correction unit 47 has a ratio between the tire force and the slip degree in a region where the correlation between the detected tire force and the detected slip degree is a linear relationship. A linear range detection value ratio (actually measured Cp value Cp0) is obtained. Further, the correction means obtains a linear region reference value ratio (reference Cp value Cp0) between the tire force and the slip degree in a region where the correlation between the tire force and the slip degree of the tire characteristic correlation map is a linear relationship. . Then, the correction means uses the ratio between the linear area detection value ratio and the linear area reference value ratio as a correction coefficient, and corrects the tire characteristic correlation map based on the correction coefficient.

Here, when the tire characteristics are different, the correlation between the tire force and the slip degree may be different. At this time, the ratio between the tire force and the slip degree in a region where the correlation is linear is also different.
In such a case, there is a correlation between the change in the correlation between the tire force and the slip degree and the change in the ratio between the tire force and the slip degree in a region where the correlation is a linear relationship.
As a result, the correction means can match the tire characteristic correlation map with the tire characteristic with which the linear area detection value ratio can be obtained based on the linear area detection value ratio and the linear area reference value ratio.
As a result, the road surface μ estimated value calculation unit 45 can match the tire characteristic correlation map with that of the actual tire characteristics, and therefore, based on the tire characteristic correlation map that is a tire model that is not affected by the tire characteristics. The road surface μ can be estimated with high accuracy.

(2) The road surface μ estimated value calculation unit 45 includes detection point distance calculation means for calculating a detection point distance that is a distance between the origin and the detection point (actual measurement value) in two-dimensional coordinates, A reference point distance calculating means for calculating a reference point distance which is a distance between the origin and the reference point (step S4).
Then, the road surface μ estimated value calculation unit 45 calculates the road surface friction coefficient of the current road surface based on the detection point distance calculated by the detection point distance calculation unit and the reference point distance calculated by the reference point distance calculation unit.
The road surface μ estimated value calculation unit 45 uses the correlation between the relationship between the detection point distance and the reference point distance and the relationship between the current road surface friction coefficient and the reference road friction coefficient, A road surface friction coefficient can be calculated.

(3) If the ratio of the tire force and the slip degree on the reference road surface and the ratio of the tire force and the slip degree on the road surface of an arbitrary road surface friction coefficient are the same, the tire characteristic correlation map The ratio between the tire force and the tire force on the road surface with an arbitrary road surface friction coefficient, or the ratio between the slip degree on the reference road surface and the slip degree on the road surface with an arbitrary road surface friction coefficient is the reference road surface friction coefficient and an arbitrary road surface. It has a characteristic indicating a ratio to the coefficient of friction.
Then, the road surface μ estimated value calculation unit 45 calculates the current road surface based on the ratio between the detection point distance calculated by the detection point distance calculation unit and the reference point distance calculated by the reference point distance calculation unit, and the reference road surface friction coefficient. The road surface friction coefficient is calculated.

Here, when viewed in two-dimensional coordinates with the tire force and the slip degree as coordinate axes, the ratio of the tire force and the slip degree on the reference road surface and the ratio of the tire force and the slip degree on the road surface of an arbitrary road surface friction coefficient is as follows. When the same, the ratio of the tire force on the reference road surface to the tire force on the road surface of any road friction coefficient, or the ratio of the slip degree on the reference road surface to the slip degree on the road surface of any road friction coefficient The ratio between the detection point distance and the reference point distance is geometrically identical.
Therefore, the road surface μ estimated value calculation unit 45 can calculate the road surface friction coefficient of the current road surface from the ratio between the detection point distance and the reference point distance and the reference road surface friction coefficient.

(4) The tire characteristic correlation map is expressed as a two-dimensional curve composed of continuous characteristic lines as existing in two-dimensional coordinates.
In this case, the correction means corrects the tire characteristic correlation map by enlarging or reducing the tire characteristic correlation map in the coordinate axis direction of the slip degree of the two-dimensional coordinates based on the correction coefficient.
Thereby, it is possible to appropriately correct the tire characteristic correlation map expressed as a two-dimensional curve composed of continuous characteristic lines.

(5) The tire characteristic correlation map is expressed as a two-dimensional curve composed of continuous characteristic lines as existing in two-dimensional coordinates.
In this case, the correction means corrects the tire characteristic correlation map by multiplying a correction coefficient obtained by dividing the linear area reference value ratio by the linear area detection value ratio by the slip degree of the tire characteristic correlation map.
This makes it possible to utilize the correlation between the error that occurs in the tire characteristic correlation map when the tire is changed and the value obtained by dividing the linear range detection value ratio by the linear range reference value ratio, and the tire characteristic correlation. The map can be corrected appropriately.

(6) The tire characteristic correlation map is expressed numerically with the tire force and the slip degree obtained on the reference road surface as variables.
In this case, the correction means corrects the tire characteristic correlation map by changing the slip degree as a variable based on the correction coefficient.
Thereby, appropriate correction | amendment can be performed with respect to the tire characteristic correlation map expressed numerically.
(7) The correction means performs statistical calculation using a plurality of detected value ratios during running, which is a ratio between the tire force and the slip degree detected during traveling of the host vehicle, and the statistical calculation results are displayed in the tire characteristic correlation map. Used for correction.
Thereby, the tire characteristic correlation map can be corrected with an appropriate linear range detection value ratio.

(8) The correction means has a correlation between a tire tire force detected on a dry road surface and a wheel slip degree, in which the detected value ratio during running, which is a ratio between the tire force and the slip degree detected during traveling of the host vehicle. When the (tire characteristic curve) is equal to or less than the ratio between the tire force and the slip degree in a region having a linear relationship, the detected value ratio during running is used as a linear region detected value ratio for correction of the tire characteristic correlation map.
As a result, the linear range detection value ratio can be selected on the basis of the dry road surface, and as a result, the tire characteristic correlation map can be corrected by an appropriate linear range detection value ratio.

(9) A determination unit is provided for determining whether or not the correlation (tire characteristic curve) between the tire force and the slip degree detected during traveling of the host vehicle is a linear relationship (step S13).
In this case, when the determination unit determines that the correlation (tire characteristic curve) is a linear relationship, the correction unit corrects the tire characteristic correlation map.
Thereby, the linear range detection value ratio when the correlation (tire characteristic curve) becomes a linear relationship is appropriately detected, and the tire characteristic correlation map can be corrected using the detected linear range detection value ratio.

(10) The determining means determines whether or not the correlation (tire characteristic curve) is a linear relationship based on the vehicle running state (FIG. 27).
Since the correlation (tire characteristic curve) changes between the linear range and the non-linear range according to the vehicle running state, the region where the correlation (tire characteristic curve) is linear is appropriate by referring to the vehicle running state. Can be detected.

(11) The determination unit determines that the correlation (tire characteristic curve) is a linear relationship when at least one of the vehicle acceleration, yaw rate, and slip degree of the host vehicle is within a predetermined value range including zero. (FIG. 27).
Since the correlation (tire characteristic curve) changes in the linear range and nonlinear range according to the vehicle acceleration, yaw rate, and slip degree of the host vehicle, the correlation is obtained by referring to the vehicle acceleration, yaw rate, and slip degree of the host vehicle. It is possible to appropriately detect a region where the relationship (tire characteristic curve) is a linear relationship.
(12) The tire force is a lateral force of the tire, and the slip degree is a tire slip angle.
Accordingly, the road surface friction coefficient of the current road surface can be estimated based on the tire characteristic correlation map indicating the correlation between the lateral force and the slip angle of the tire, and the tire characteristic correlation map can be corrected.

(Second Embodiment)
(Constitution)
The second embodiment is an electrically driven vehicle to which the present invention is applied.
FIG. 34 shows a schematic configuration of an electrically driven vehicle (two-wheel drive). As shown in FIG. 34, the electric drive vehicle, the accelerator pedal operation amount detection unit 71, the brake pedal operation amount detecting section 72, wheel speed detecting section ₇₃ FL _{to 73 RR,} the acceleration sensor 74, the drive motor ₇₅ FL, _{75 FR,} A system control unit 76, drive wheels 77 _FL and 77 _FR, and a battery 78 are included.

The accelerator pedal operation amount detector 71 detects the amount of operation of the accelerator pedal (accelerator opening) by the driver. The accelerator pedal operation amount detector 71 outputs the detection result (accelerator opening) to the system controller 76. The brake pedal operation amount detector 72 detects the operation amount of the brake pedal by the driver. The brake pedal operation amount detection unit 72 outputs the detection result to the system control unit 76. Wheel speed detecting section ₇₃ FL _{to 73 RR} detect the wheel speeds _v FL _{to v RR} of each wheel ₂₇ FL _{~ 27 RR} provided on the vehicle body. Wheel speed detection units 73 _{FL to} 73 _RR output the detection results to system control unit 76. The acceleration sensor 74 detects the longitudinal acceleration and the lateral acceleration of the vehicle. The acceleration sensor 74 outputs the detection result (front / rear G / lateral G) to the system control unit 76. The drive motors 75 _FL and 75 _FR generate a drive torque corresponding to the drive torque command value Tout output from the system control unit 76 and rotationally drive the drive wheels 77 _FL and 77 _FR . The drive torque command value Tout is a current supplied from the battery 78 in order to control the drive motors 75 _FL and 75 _FR .

FIG. 35 shows the configuration of the system control unit 76. As shown in FIG. 35, the system control unit 76 includes a vehicle body speed calculation unit 91, a drive torque command value calculation unit 92, a drive torque command value correction unit 93, a slip ratio calculation unit 94, a braking / driving force calculation unit 95, and a road surface μ. An estimated value calculation unit 96, a linear region μ gradient estimation unit 97, and a map correction unit 98 are included.
The vehicle body speed calculation unit 91 calculates the vehicle body speed based on the wheel speeds detected by the wheel speed detection units 73 _{FL to} 73 _RR . Specifically, the vehicle body speed calculation unit 91 calculates the vehicle body speed based on the left and right average values of the two driven wheels. Note that the vehicle body speed can also be estimated in consideration of the detection value of the longitudinal acceleration sensor. In this case, the vehicle body speed calculation unit 91 corrects the vehicle body speed from the vehicle body speed calculated based on the wheel speed so as to eliminate the influence of errors caused by tire slipping during sudden acceleration or tire lock during sudden braking. The vehicle body speed can also be detected using a GPS (Global Positioning System), an optical ground speed measuring device, or the like. The vehicle body speed calculation unit 91 outputs the calculated vehicle body speed to the drive torque command value calculation unit 92 and the slip ratio calculation unit 94.

The drive torque command value calculation unit 92 calculates a drive torque command value (drive torque basic command value) T based on the accelerator opening detected by the accelerator pedal operation amount detection unit 71 and the vehicle body speed detected by the vehicle body speed calculation unit 91. calculate. The drive torque command value (drive torque basic command value) T is a command value corresponding to the driver's accelerator operation, and is a current value for controlling the drive motors 75 _FL and 75 _FR . The drive torque command value calculation unit 92 outputs the calculated drive torque command value (drive torque basic command value) T to the drive torque command value correction unit 93.

The slip ratio calculation unit 94 calculates the slip ratio λ based on the wheel speed detected by the wheel speed detection units 73 _{FL to} 73 _RR and the vehicle body speed calculated by the vehicle body speed calculation unit 91. Specifically, the slip ratio calculation unit 94 calculates the slip ratio according to the difference between the wheel speed and the vehicle body speed. Further, as shown in the following equations (7) and (8), the slip ratio λ is calculated by switching between acceleration (driving) and deceleration (braking).
During acceleration (driving): λ = (V−w) / w (7)
During deceleration (during braking): λ = (V−w) / V (8)

Here, V is a vehicle body speed. w is the wheel speed. The slip ratio calculation unit 94 outputs the calculated slip ratio λ to the road surface μ estimated value calculation unit 96 and the linear region μ gradient estimation unit 97.
The braking / driving force calculation unit 95 calculates the braking / driving force Fx based on the motor current value for driving the drive motors 75 _FL and 75 _FR . Specifically, the braking / driving force calculation unit 95 calculates the braking / driving force Fx based on the motor current value and the wheel angular acceleration. The braking / driving force calculation unit 95 outputs the calculated braking / driving force Fx to the road surface μ estimated value calculation unit 96 and the linear region μ gradient estimation unit 97.

The road surface μ estimated value calculation unit 96 stores a characteristic map (tire characteristic curve map) composed of tire characteristic curves of the reference road surface in a memory or the like. For example, the tire characteristic curve of the road surface of the road surface mu values mu _A shown in FIG. 5, and a tire characteristic curve of the reference road surface.
For example, a tire characteristic curve map is obtained by performing a running experiment in advance. As a running experiment, a linear acceleration running test is performed. From a linear acceleration / acceleration running experiment on the reference road surface, a tire characteristic curve map including a tire characteristic curve on the reference road surface is obtained from the relationship between the change in slip ratio and the change in driving force or braking force obtained at that time. It is also possible to obtain a tire characteristic curve map composed of tire characteristic curves on the reference road surface, not by running experiments but by calculations such as simulations.
The road surface μ estimated value calculation unit 96 calculates the road surface μ of the actual traveling road surface as an estimated value based on the tire characteristic curve map.

FIG. 36 shows a processing procedure for calculating the road surface μ (estimated value) of the actual traveling road surface based on the line length.
As shown in FIG. 36, first, in step S31, road surface μ estimated value calculating section 96, (obtain an output from the braking-driving force calculating unit 95) for detecting the longitudinal force Fx _b.
Then at step S32, road surface μ estimated value calculating section 96, (obtain an output from the slip ratio calculating unit 94) for detecting a slip ratio lambda _b.
Subsequently, in step S33, the road surface μ estimated value calculation unit 96 determines that a straight line passing through the origin (0, 0) of the tire characteristic curve (tire characteristic curve map) of the reference road surface and the actual measurement point intersects the tire characteristic curve. (Λ _a , Fx _a ) is specified. Here, the actual measurement point is a value (λ _b , Fx _b ) indicated by the lateral force Fx _b and the slip angle λ _b detected in steps S31 and S32.

Subsequently, in Step S34, road surface mu estimation value calculation unit 96 calculates the road surface mu values mu _B of the actual traveled road surface. That is, the road surface μ estimated value calculation unit 96 has a straight line length b1 (= √ (λ _b ² + Fx _b ² )) connecting the measured point (λ _b , Fx _b ) and the origin of the tire characteristic curve of the reference road surface. Get. Further, the road surface μ estimated value calculation unit 96 has a straight line length a1 (which connects the intersection point value (λ _a , Fx _a ) of the tire characteristic curve of the reference road surface specified in step S33 and the origin of the tire characteristic curve. = √ (λ _a ² + λ _a ² )). Further, the road surface μ estimated value calculation unit 96 calculates a ratio (b1 / a1) between the line length b1 and the line length a1. Then, the road surface mu estimation value calculation unit 96, and the calculated ratio (b2 / b1), multiplying the road surface mu values mu _A of the reference road surface obtained from the tire characteristic curve map, the multiplied value actual traveled road surface Is obtained as an estimated value μ _B of the road surface μ (μ _B = μ _A · b1 / a1).

The processing procedure described above, the road surface mu estimation value calculation unit 96 calculates the estimated value mu _B of the road surface mu actual traveled road surface. Road mu estimation value calculation unit 96 outputs the estimated value mu _B of the road surface mu actual traveled road surface calculated in the driving torque command value correcting section 93.
The drive torque command value correcting unit 93 is based on the road surface μ (estimated value) calculated by the road surface μ estimated value calculating unit 96, and is calculated by the drive torque command value calculating unit 92 (drive torque basic command value). T is corrected. Specifically, the drive torque command value correction unit 93 corrects the drive torque command value (drive torque basic command value) T to be smaller as the road surface μ is smaller (becomes smaller from 1). For example, the drive torque command value (drive torque basic command value) T is corrected with a gain corresponding to the road surface μ.

FIG. 37 shows the relationship between the road surface μ (estimated value) and the gain Gain. As shown in FIG. 37, the gain Gain becomes smaller as the road surface μ becomes smaller (becomes smaller from 1). Using the gain Gain indicating such a relationship, a corrected drive torque command value T (left side) is calculated by the following equation (9).
T = T-Gain · L (9)
Here, L is a gain (> 0) for immediately stopping idling. According to the equation (9), as the estimated value of the road surface μ becomes smaller (becomes smaller from 1), the drive torque command value T becomes smaller.
The linear region μ gradient estimation unit 97 estimates the μ gradient in the linear region, and calculates the correction ratio based on the estimated μ gradient in the linear region.

FIG. 38 shows a processing procedure in the linear region μ gradient estimation unit 97. As shown in FIG. 38, first, in step S41, the linear region μ gradient estimation unit 97 acquires various data. Specifically, the linear region μ gradient estimating unit 97 acquires the slip ratio λ calculated by the slip ratio calculating unit 94 and the braking / driving force Fx calculated by the braking / driving force calculating unit 95.
In step S42, the linear region μ gradient estimating unit 97 determines whether or not the host vehicle is traveling (straight traveling) in the linear region of the tire characteristic curve. When the host vehicle is traveling in the linear region of the tire characteristic curve, the linear region μ gradient estimating unit 97 proceeds to step S43. If not, for example, if the host vehicle is traveling in a nonlinear region (curve region) of μ gradient, the linear region μ gradient estimation unit 97 ends the processing shown in FIG. 38 (from step S41 again). To start).

Subsequently, in step S43, the linear region μ gradient estimation unit 97, based on the braking / driving force Fx and the slip ratio λ acquired in step S41, the ratio (Fx / λ) of the braking / driving force Fx and the slip ratio λ, That is, the μ gradient is calculated.
Subsequently, in step S44, the linear region μ gradient estimation unit 97 stores the ratio (Fx / λ) between the braking / driving force Fx and the slip ratio λ calculated in step S43 as history information.

Subsequently, in step S45, the linear region μ gradient estimation unit 97 calculates an average value of the ratios (Fx / λ) of the plurality of braking / driving forces Fx and the slip ratio λ stored as history information in step S44. (((Fx / λ) ₁ + (Fx / λ) ₂ +... + (Fx / λ) _N ) / N).
Subsequently, in step S46, the linear region μ gradient estimation unit 97 determines whether or not the history data number (N) stored in step S44 is larger than a predetermined threshold value Nth. When the history data number (N) is larger than the predetermined threshold value Nth (N> Nth), the linear region μ gradient estimation unit 97 proceeds to step S47. If not (N ≦ Nth), the linear region μ gradient estimation unit 97 ends the process shown in FIG. 38 (starts the process again from step S41).

In step S47, the linear region μ gradient estimation unit 97 calculates a correction ratio. Specifically, the linear region μ gradient estimation unit 97 sets the ratio of the reference μ gradient Cp0 and the actually measured μ gradient Cp0 ′ as a correction ratio R (= Cp0 / Cp0 ′).
The linear region μ gradient estimation unit 97 finally calculates the correction ratio R by the processing procedure as described above. The linear region μ gradient estimation unit 97 outputs the calculated correction ratio R to the map correction unit 98.
The map correction unit 98 corrects the tire characteristic curve map of the road surface μ estimated value calculation unit 96 with the correction ratio R. Specifically, the map correction unit 98 multiplies the slip ratio λ ₀ of the tire characteristic curve map by the correction ratio R.

FIG. 39A shows an example of a tire characteristic curve map 96a. The tire characteristic curve map 96a has coordinate values (λ ₀₀ , Fx ₀₀ ), (λ ₀₁ , Fx ₀₁ ),..., (Λ _0n , Fx _0n ) that draw the tire characteristic curve of the reference road surface in two-dimensional coordinates. (Where n is an arbitrary integer).
The map correction unit 98 multiplies the slip ratios λ ₀₀ , λ ₀₁ ,..., Λ _0n of the tire characteristic curve map 96a by the correction ratio R to correct as shown in FIG. A later tire characteristic curve map 96a is created.

(Operation and action)
(Vehicle control based on road surface μ estimation)
While the vehicle is traveling, the accelerator pedal operation amount detection unit 71 detects the accelerator opening that the driver has operated, and the vehicle body speed calculation unit 91 calculates the vehicle body speed. The drive torque command value calculation unit 92 calculates a drive torque command value (drive torque basic command value) T based on the accelerator opening and the vehicle body speed. On the other hand, the wheel speed detectors 73 _{FL to} 73 _RR detect the wheel speed. Then, the slip ratio calculation unit 94 calculates the slip ratio based on the wheel speed and the vehicle body speed. Further, the braking / driving force calculation unit 95 calculates the braking / driving force based on the motor current value. The road surface μ estimated value calculation unit 96 calculates the road surface μ of the actual traveling road surface based on the braking / driving force, the slip ratio, and the characteristic map. Then, the drive torque command value correction unit 93 corrects the drive torque command value (drive torque basic command value) T based on the calculated road surface μ. Specifically, the drive torque command value correction unit 93 corrects the drive torque command value (drive torque basic command value) T to be smaller as the road surface μ is smaller (becomes smaller from 1).

(Map correction)
When the linear region μ gradient estimating unit 97 can determine that the host vehicle is traveling in the linear region of the tire characteristic curve, the ratio of the braking / driving force Fx and the slip ratio λ calculated (detected) at that time ( Fx / λ) is stored as history information. Further, the linear region μ gradient estimation unit 97 calculates an average value (steps S41 to S45). Then, the linear region μ gradient estimation unit 97 calculates the correction ratio R (= Cp0 / Cp0 ′) when the history data number (N) becomes larger than the predetermined threshold value Nth (steps S46 to S46). Step S47).

The map correction unit 98 corrects the tire characteristic curve map of the road surface μ estimated value calculation unit 96 with the correction ratio R (FIG. 39). Specifically, the map correction unit 98 multiplies the correction ratio R by the slip ratio λ ₀ of the tire characteristic curve map. As a result, the shape of the tire characteristic curve map changes by a correction ratio R times in the slip ratio λ-axis direction. In other words, the shape of the tire characteristic curve map is enlarged or reduced by a correction ratio R times in the slip ratio λ axis direction with the origin fixed. Thereby, the tire characteristic curve map shows the tire characteristic curve of the tire to be mounted from the one showing the tire characteristic curve of the reference road surface.

(Modification of the second embodiment)
(1) The maximum drive torque command value can be limited according to the road surface μ. For example, the smaller the road surface μ is, the smaller the maximum drive torque command value is. As a result, as the road surface μ becomes smaller (becomes smaller from 1), the drive torque command value T can be made smaller.
(2) In the second embodiment, the drive torque is corrected. On the other hand, the braking torque can be corrected. Also in this case, correction is made to reduce the braking torque as the road surface μ becomes smaller (becomes smaller from 1).

(3) In the second embodiment, the case of an electrically driven vehicle driven by a motor has been described. On the other hand, the present invention can be applied to a vehicle using another drive source as long as the vehicle can detect the braking / driving force and the slip ratio, or the physical quantities corresponding to them. For example, the present invention can be applied to a gasoline engine driven vehicle. In this case, an acceleration sensor can be mounted on a gasoline engine driven vehicle, and the road surface μ can be estimated by replacing the vehicle body acceleration detected by the acceleration sensor with the braking / driving force of the wheels.

(4) In the second embodiment, the braking / driving torque of the vehicle is controlled as the running behavior control of the vehicle based on the estimated road surface μ. On the other hand, based on the estimated road surface μ, it is possible to control another control amount (for example, steering assist torque) for vehicle travel control.
In the second embodiment, the braking / driving force calculation unit 95 realizes tire force detection means for detecting the tire force of the wheels.
Further, the slip ratio calculating unit 94 realizes a slip degree detecting means for detecting the slip degree of the wheel.

In addition, the characteristic map 96a of the road surface μ estimated value calculation unit 96 is a tire characteristic representing a characteristic line established by the correlation between the wheel tire force obtained by the reference tire on the reference road surface of the reference road surface friction coefficient and the slip degree of the wheel. A tire characteristic correlation map is modeled.
Further, the process of step S33 of the road surface μ estimated value calculation unit 96 is performed in the two-dimensional coordinates in which the tire force and the slip degree are zero in the two-dimensional coordinates having the wheel tire force and the wheel slip degree as coordinate axes. Inclination calculating means for calculating the current tire force detected by the tire force detecting means relative to the origin and the inclination of the detection point indicated by the current slip degree detected by the slip degree detecting means is realized.

Further, the process in step S33 of the road surface μ estimated value calculation unit 96 calculates a reference point that is an intersection of the straight line extending with the inclination calculated by the inclination calculating means from the origin of the two-dimensional coordinates and the tire characteristic correlation map. A reference point calculating means is realized.
Further, the process in step S34 of the road surface μ estimated value calculation unit 96 calculates the road surface friction coefficient of the current road surface based on the detection point, the reference point calculated by the reference point calculation means, and the reference road surface friction coefficient. A road surface friction coefficient calculating means is realized.

  Further, the linear region μ gradient estimation unit 97 and the map correction unit 98 are configured so that the tire in the region where the correlation between the tire force detected by the tire force detection unit and the slip degree detected by the slip degree detection unit is a linear relationship. The ratio between the tire force and the slip degree in a region where the correlation between the tire force and the slip degree in the tire characteristic correlation map is a linear relationship, and the linear range detection value ratio that is the ratio between the force and the slip degree A correction means for correcting the tire characteristic correlation map based on the correction coefficient is realized by using the ratio to the linear region reference value ratio as a correction coefficient.

(Effect of 2nd Embodiment)
(1) The tire force is the braking / driving force of the tire, and the slip degree is the tire slip rate.
Accordingly, the road surface friction coefficient of the current road surface can be estimated based on the tire characteristic correlation map indicating the correlation between the braking / driving force and the slip ratio of the tire, and the tire characteristic correlation map can be corrected.

  43 tire slip angle calculation unit, 44 tire lateral force calculation unit, 45, 96 road surface μ estimated value calculation unit, 45a, 96a tire characteristic curve map, 46 linear region Cp value estimation unit, 47, 98 map correction unit, 94 slip ratio Calculation unit, 95 braking / driving force calculation unit, 97 linear region μ gradient estimation unit

Claims (14)

Tire force detection means for detecting the tire force of the wheels;
Slip degree detecting means for detecting the slip degree of the wheel;
A tire characteristic correlation map modeled on the assumption that the tire characteristic represents a characteristic line established by the correlation between the wheel tire force obtained on the reference road surface of the reference road surface friction coefficient by the reference tire and the slip degree of the wheel;
In a two-dimensional coordinate having the tire force of the wheel and the slip degree of the wheel as coordinate axes, the current tire force detected by the tire force detecting means with respect to the origin of the two-dimensional coordinate where the tire force and the slip degree are zero, and the An inclination calculating means for calculating an inclination of a detection point indicated by the current slip degree detected by the slip degree detecting means;
Reference point calculating means for calculating a reference point that is an intersection of a straight line extending with the inclination calculated by the inclination calculating means from the origin of the two-dimensional coordinates and the tire characteristic correlation map;
Road surface friction coefficient calculation means for calculating a road surface friction coefficient of the current road surface based on the detection point, the reference point calculated by the reference point calculation means, and the reference road surface friction coefficient;
A linear range detection value ratio which is a ratio of the tire force and the slip degree in a region where the correlation between the tire force detected by the tire force detection means and the slip degree detected by the slip degree detection means is a linear relationship; The ratio of the linear region reference value ratio, which is the ratio between the tire force and the slip degree, in a region where the correlation between the tire force and the slip degree in the assumed tire characteristic correlation map is a linear relationship is used as a correction coefficient. Correcting means for correcting the tire characteristic correlation map based on the correction coefficient;
A vehicle ground contact surface friction state estimation device comprising:   The road surface friction coefficient calculation means includes detection point distance calculation means for calculating a detection point distance that is a distance between the origin and the detection point in the two-dimensional coordinates, and the origin and the points in the two-dimensional coordinates. A reference point distance calculating unit that calculates a reference point distance that is a distance between the reference point and the detection point distance calculated by the detection point distance calculating unit, and the reference point distance calculated by the reference point distance calculating unit 2. The vehicle ground contact surface friction state estimation device according to claim 1, wherein a road surface friction coefficient of a current road surface is calculated based on the reference road surface friction coefficient. If the ratio of the tire force and the slip degree on the reference road surface and the ratio of the tire force and the slip degree on the road surface of an arbitrary road surface friction coefficient are the same, the tire characteristic correlation map The ratio of the tire force and the tire force on the road surface of an arbitrary road surface friction coefficient, or the ratio of the slip degree on the reference road surface and the slip degree on the road surface of an arbitrary road surface friction coefficient is the reference road surface friction coefficient and the It has the characteristic to show the ratio with any road surface friction coefficient,
The road surface friction coefficient calculating means is configured to determine a current road surface based on a ratio between the detection point distance calculated by the detection point distance calculation means and the reference point distance calculated by the reference point distance calculation means, and the reference road surface friction coefficient. The road surface friction coefficient estimating device according to claim 2, wherein a road surface friction coefficient is calculated. The tire characteristic correlation map is expressed as a two-dimensional curve composed of continuous characteristic lines as existing in the two-dimensional coordinates,
The correction means corrects the tire characteristic correlation map by enlarging or reducing the tire characteristic correlation map in the coordinate axis direction of the slip degree of the two-dimensional coordinates based on the correction coefficient. The vehicle ground contact surface friction state estimation apparatus according to any one of claims 1 to 3. The tire characteristic correlation map is expressed as a two-dimensional curve composed of continuous characteristic lines as existing in the two-dimensional coordinates,
The correction means corrects the tire characteristic correlation map by multiplying a correction coefficient obtained by dividing the linear area reference value ratio by the linear area detection value ratio by the slip degree of the tire characteristic correlation map. The vehicle ground contact surface friction state estimation device according to any one of claims 1 to 4, wherein: The tire characteristic correlation map is expressed numerically with the tire force and the slip degree obtained on the reference road surface as variables,
4. The correction unit according to claim 1, wherein the correction unit corrects the tire characteristic correlation map by changing a slip degree as the variable based on the correction coefficient. 5. Vehicle ground contact surface friction state estimation device.   The correction means performs a statistical calculation using a plurality of detected value ratios during traveling, which is a ratio of the tire force detected by the tire force detecting means and the slip degree detected by the slip degree detecting means during traveling of the host vehicle. The vehicle contact surface friction state estimation device according to any one of claims 1 to 6, wherein the statistical calculation result is used for correction of the tire characteristic correlation map.   The correction means is a wheel on which a detected value ratio during running, which is a ratio of the tire force detected by the tire force detecting means and the slip degree detected by the slip degree detecting means during traveling of the host vehicle, is obtained on a dry road surface. When the correlation between the tire force and the slip degree of the wheel is equal to or less than the ratio of the tire force and the slip degree in a region where the relationship is a linear relationship, the detected value ratio during traveling is used as the linear region detected value ratio. The vehicle ground contact surface friction state estimation device according to any one of claims 1 to 7, wherein the vehicle ground contact surface friction state estimation device is used for correction of a characteristic correlation map. Determining means for determining whether or not the correlation between the tire force detected by the tire force detecting means and the slip degree detected by the slip degree detecting means during traveling of the host vehicle is a linear relationship;
The vehicle according to any one of claims 1 to 8, wherein the correction unit corrects the tire characteristic correlation map when the determination unit determines that the correlation is a linear relationship. Ground surface friction estimation device.   The vehicle contact surface friction state estimation device according to claim 9, wherein the determination unit determines whether or not the correlation is a linear relationship based on a vehicle running state.   The determination unit determines that the correlation is a linear relationship when at least one of the vehicle acceleration, yaw rate, and slip degree of the host vehicle is within a predetermined value range including zero. The vehicle ground contact surface friction state estimation apparatus according to claim 9 or 10.   The vehicle ground contact surface friction state estimation device according to any one of claims 1 to 11, wherein the tire force is a lateral force of a tire, and the slip degree is a tire slip angle.   The vehicle ground contact surface friction state estimation device according to any one of claims 1 to 11, wherein the tire force is a braking / driving force of a tire, and the slip degree is a tire slip ratio. In the vehicle ground contact surface friction state estimation method for estimating the ground contact surface grip characteristics of the vehicle wheel,
Using the tire characteristic correlation map modeled assuming tire characteristics representing the characteristic line established by the correlation between the tire force of the wheel obtained on the reference road surface of the reference road surface friction coefficient by the reference tire and the slip degree of the wheel,
A linear range detection value ratio that is a ratio between the detected tire force and the detected slip degree in a region where the correlation between the detected tire force and the detected slip degree is a linear relationship, and the tire force of the assumed tire characteristic correlation map The tire characteristic correlation map is based on the correction coefficient based on the ratio of the linear region reference value ratio, which is the ratio of the tire force and the slip degree, in a region where the correlation with the slip degree is a linear relation. A correction step for correcting
In a two-dimensional coordinate having the tire force and the slip degree as coordinate axes, the inclination of the detection point indicated by the current detected tire force and the detected slip degree with respect to the origin of the two-dimensional coordinate where the tire force and the slip degree are zero is calculated. A slope calculating step;
A reference point calculating step for calculating a reference point that is an intersection of the straight line extending from the origin of the two-dimensional coordinates with the inclination calculated in the inclination calculating step and the tire characteristic correlation map;
Based on the detection point, the reference point calculated in the reference point calculation step, and the reference road surface friction coefficient, a road surface friction coefficient calculation step for calculating a road surface friction coefficient of the current road surface;
A vehicle ground contact surface friction state estimation method comprising:

Images