JP2009214752A - Vehicle travel state estimating device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle travel state estimating device capable of accurately estimating vehicle behavior when the vehicle behavior suddenly changes. <P>SOLUTION: This vehicle travel state estimating device is provided with a vehicle behavior means for arithmetically operating a tire slip angle based on a vehicle model, a cornering stiffness arithmetic operation means for arithmetically operating cornering stiffness from a lateral force corresponding physical quantity and the tire slip angle of a wheel, and a divergence preventive means for preventing divergence of an arithmetic operation value of the cornering stiffness. A vehicle behavior arithmetic operation means corrects the vehicle model by the cornering stiffness. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a vehicle travel state estimation device that estimates a vehicle state.

As this type of technology, the technology described in Patent Document 1 is disclosed. In this publication, the yaw rate and lateral acceleration are calculated based on the vehicle motion model in the linear region, and the cornering power of the vehicle motion model is determined according to the deviation between the calculated yaw rate and lateral acceleration and the yaw rate and lateral acceleration detected by the sensor. There is disclosed a method for estimating vehicle behavior by performing a correction for increasing or decreasing the vehicle speed.
Japanese Patent No. 3571379

  In the above prior art, the cornering power of the vehicle model is corrected according to the deviation between the yaw rate and lateral acceleration calculated based on the vehicle motion model and the yaw rate and lateral acceleration detected by the sensor. When the vehicle behavior suddenly changes, the cornering power correction cycle is not in time, and there is a problem that the vehicle behavior may not be estimated accurately.

  The present invention has been made paying attention to the above problems, and an object of the present invention is to provide a vehicle running state estimation device capable of accurately estimating the vehicle behavior even when the vehicle behavior suddenly changes.

  In order to achieve the above object, the vehicle running state estimating means of the present invention calculates cornering stiffness by preventing divergence from the physical force equivalent to the lateral force of the wheel and the tire slip angle, and corrects the vehicle model based on the cornering stiffness. I made it.

  Therefore, the vehicle behavior can be accurately estimated even when the vehicle behavior suddenly changes.

  Hereinafter, the best mode for realizing the vehicle running state estimation device of the present invention will be described based on the first embodiment.

[overall structure]
First, the configuration of a vehicle equipped with eight vehicle running state estimation devices according to the first embodiment will be described.
FIG. 1 is a configuration diagram showing a schematic configuration of a vehicle. The vehicle according to the first embodiment is a steering device including a steering wheel 9, a steering shaft 10, an electric power steering (hereinafter referred to as EPS) motor 7, a rack and pinion mechanism (pinion 12, rack 13), and a tie rod. 14 is provided.

  The control system includes a steering angle sensor 1, a yaw rate sensor 2, a lateral acceleration sensor 3, a longitudinal acceleration sensor 4, a wheel speed sensor 5, and an electric power steering electronic control unit (Electric Power Steering Electronic Control Unit: EPSECU) 6 and a vehicle running state estimation device 8 are provided.

  A steering force is input to the steering wheel 9 by the driver, and the steering force is transmitted to the steering shaft 10 that rotates together with the steering wheel 9. Further, the EPS motor 7 is driven by the drive signal of the EPSECU 6 to apply assist torque to the steering shaft 10. Steering force and assist torque are transmitted to the left and right wheels 11FL and 11FR via the rack and pinion mechanisms 12 and 13 and the tie rods 14, and the left and right wheels 11FL and 11FR are steered.

The steering angle sensor 1 detects the rotation angle (steering angle δ) of the steering shaft 10 and outputs the detected steering angle information to the vehicle running state estimation device 8.
The yaw rate sensor 2 detects the yaw rate γ of the vehicle, and outputs the detected yaw rate information to the vehicle running state estimation device 8.

The lateral acceleration sensor 3 detects the lateral acceleration Gy of the vehicle and outputs the detected lateral acceleration information to the vehicle running state estimation device 8.
The longitudinal acceleration sensor 4 detects the longitudinal acceleration Gx of the vehicle and outputs the detected longitudinal acceleration information to the vehicle running state estimation device 8.
The wheel speed sensor 5 detects the wheel speed Vw of each wheel 11FL, 11FR, 11RL, 11RR provided on the vehicle body, and outputs the detected wheel speed information to the vehicle running state estimation device 8.

The vehicle running state estimation device 8 includes a steering angle δ detected by the steering angle sensor 1, a yaw rate γ detected by the yaw rate sensor 2, a lateral acceleration G _y detected by the lateral acceleration sensor 3, and a longitudinal acceleration G detected by the longitudinal acceleration sensor 4. Unstable behavior suppression assist that estimates the running state of the vehicle based on _x and the wheel speed V _w detected by the wheel speed sensor 5 and controls the EPS motor 7 to suppress the unstable behavior of the vehicle based on the estimation result. The command is output to EPSECU6.

  The EPSECU 6 calculates a steering assist force based on the steering angle δ detected by the steering angle sensor 1, the wheel speed Vw detected by the wheel speed sensor 5, and outputs a steering assist command to the EPS motor 7.

  Further, the EPSECU 6 calculates a steering assist force corrected so as to suppress the unstable behavior of the vehicle based on the unstable behavior suppression assist command output from the vehicle running state estimation device 8, and uses the corrected steering assist command as the EPS. Output to the motor 7.

[Vehicle running state estimation device]
Next, the vehicle running state estimation device 8 will be described in detail. FIG. 2 is a control block diagram of the vehicle running state estimation device 8. The vehicle running state estimation device 8 includes a vehicle body speed calculation unit 16, a vehicle body slip angle estimation unit 17, a tire slip angle calculation unit 18, a tire lateral force calculation unit 19, a cornering stiffness calculation unit 20, a stability factor calculation unit 21, and a vehicle behavior. An estimator 22 and an unstable behavior suppression assist command value calculator 23 are provided.

(Car body speed calculator)
The vehicle body speed calculation unit 16 calculates the vehicle body speed V _t based on the wheel speed V _w detected by the wheel speed sensor 5 and the longitudinal acceleration G _x detected by the longitudinal acceleration sensor 4 and estimates the vehicle body speed information. To the unit 17 and the tire lateral force calculation unit 19.

Specifically, vehicle speed calculating section 16, the basic driven wheels 11RL, the average value of the wheel speed V _w of 11RR, or vehicle speed V _t by calculating the average value of the wheel speed V _w of each wheel 11FL~11RR The vehicle speed V _t is estimated by correcting the basic value based on the longitudinal acceleration G _x so that the influence of errors due to tire slipping during sudden acceleration and tire lock during sudden braking is removed from the basic value. As a result.

(Car body slip angle estimation part)
The vehicle body slip angle estimation unit 17 includes a steering angle δ detected by the steering angle sensor 1, a yaw rate γ detected by the yaw rate sensor, a lateral acceleration G _y detected by the lateral acceleration sensor 3, and a longitudinal acceleration G _x detected by the longitudinal acceleration sensor 4. The vehicle body slip angle estimated value β _t ^* is estimated based on the vehicle body speed V _t calculated by the vehicle body speed calculating unit 16, and the vehicle body slip angle estimated value is output to the tire slip angle calculating unit 18.

  FIG. 3 is a control block diagram of the vehicle body slip angle estimation unit 17. The vehicle body slip angle estimation unit 17 includes a direct integration method calculation unit 17a, a linear observer calculation unit 17b, and an Atan method calculation unit 17c.

  The vehicle body slip angle estimator 17 corrects the vehicle body slip angle estimated by the direct integration method based on the vehicle body slip angle estimated by the linear observer and the Atan method, and estimates the vehicle body slip angle with high accuracy.

なお、V_tは車体速度、β_tは車体スリップ角の時間微分値、γはヨーレイト、G_yは横加速度を示す。 <Direct integration method>
The direct integration method computing unit 17a obtains the vehicle body slip angle β _t based on the balance equation of the motion acting in the lateral direction (vehicle width direction) of the vehicle, and outputs it as β _t ^* _Integral .
The expression of the balance of the movement acting in the lateral direction (vehicle width direction) of the vehicle when turning can be expressed by the following expression (1).

Incidentally, V _t is the vehicle speed, beta _t is time derivative of the vehicle body slip angle, gamma is the yaw rate, G _y represents the lateral acceleration.

Solving Equation (1) for the time derivative β _t of the vehicle body slip angle yields Equation (2) below.

When the equation (2) is integrated over time, the following equation (3) is obtained, and the vehicle body slip angle β _t can be obtained.

<Linear observer>
The linear observer calculation unit 17b calculates a vehicle body slip angle β _t using a two-wheel model and outputs it as β _t ^* _Observer .
FIG. 4 is a diagram showing a two-wheel model. In the figure, m is the vehicle weight, I is the yaw moment of inertia, G _y is the lateral acceleration, l _f and l _r are the distance between the vehicle center of gravity and the axle, F _f and F _r are the frictional forces acting on the tire, F _yf and F _yr are tire lateral forces, F _xf and F _xr are rolling resistances, F _yf 'and F _yr ' are cornering forces, F _xf 'F _xr ' is cornering resistance, β _t is the vehicle body slip angle, β _wf , beta _wr tire slip angle, Cp _f, Cp _r denotes the cornering power.

以下では、コーナリングフォースF_yf',F_yr'をタイヤ横力F_yf,F_yrとして扱う。 Here, when the tire slip angle is small, it can be considered that the cornering forces F _yf ′, F _yr ′ are equal to the tire lateral forces F _yf , F _yr (equation (4)).

Hereinafter, the cornering forces F _yf ′ and F _yr ′ are treated as tire lateral forces F _yf and F _yr .

Also, there the ratio of the cornering power Cp _f, Cp _r tire lip angle beta _wf, beta _wr is small tire characteristics tire lip angle when a linear beta _wf, beta _wr and the tire lateral force F _yf, F _yr It is a fixed value. The tire characteristics will be described in detail later.
Tire slip angle beta _wf, tire lateral force F _yf when beta _wr is small, F _yr is cornering power Cp _f, Cp _r a tire slip angle beta _wf, can be expressed by the product of the beta _wr (formula (5) ).

From the two-wheel model shown in FIG. 4, the formula for balancing the lateral force and moment is given by the following formulas (6) and (7).

行列Cにおけるa₁₁,a₁₂は行列Aの要素、行列Dにおけるb₁は行列Bの要素である。 When these equations (6) and (7) are rewritten into the state equation, the following equation (8) is obtained.

A ₁₁ and a ₁₂ in the matrix C are elements of the matrix A, and b ₁ in the matrix D is an element of the matrix B.

Based on this equation of state, a linear two-input observer that estimates the vehicle body slip angle β _t is designed in the linear observer calculation unit 17b.
FIG. 5 is a block diagram of a linear 2-input observer. The inputs to the observer are the yaw rate γ and the lateral acceleration G _y , and the observer gain L is set so as to be less susceptible to modeling errors and to estimate a stable vehicle body slip angle β _t .

As described above, when the tire slip angles β _wf and β _wr are small and the tire characteristics are linear, the relationship between the tire slip angles β _wf and β _wr and the tire lateral forces F _yf and F _yr is a fixed value. Using Cp, it can be expressed as in equation (5).

However, if the tire slip angles β _wf and β _wr increase and the tire characteristics become nonlinear, the difference between the fixed cornering power Cp and the tire characteristics increases, and the tire slip angle β _wf using the fixed cornering power Cp is increased. , β _wr and tire lateral force F _yf , F _yr can not be expressed.

Therefore, the linear observer calculation unit 17b corrects the cornering power Cp in the vehicle model using the cornering stiffness Cs of the cornering stiffness calculation unit 20 as shown in FIG. As a result, the vehicle body slip angle β _t can be estimated by the linear observer even in a portion where the tire characteristics are in a non-linear region.

<Atan method>
The Atan method calculation unit 17c geometrically calculates the vehicle body slip angle β _t from the direction of acceleration applied by the vehicle, and outputs it as β _t ^* _Atan .

6 is a diagram showing a state where the turning vehicle is traveling with the vehicle body slip angle beta _t.
At the time of turning, centrifugal force (hereinafter referred to as field force) acts on the vehicle body in the outward direction of the turning radius. The acceleration A _{y in the} turning radius direction due to this field force can be obtained by combining the lateral acceleration G _y and the longitudinal acceleration G _x . When the vehicle is turning with a vehicle body slip angle beta _t, the angle beta ₂ formed by the direction and the vehicle width direction of the acceleration A _y is equal to the vehicle body slip angle _{β t (β t = β 2} ) that Is geometrically apparent from FIG. The angle β ₂ can be expressed by the following equation (9) from FIG.

Although the above is a description of the case where the vehicle speed V _t is constant, the vehicle speed V _t changes the lateral acceleration G _y and longitudinal acceleration G _x is changed. Therefore, when the direction of the acceleration A _y is obtained from the lateral acceleration G _y and the longitudinal acceleration G _x, it may not coincide with the turning radius direction. Therefore, the acceleration A _y must seek in consideration of change in vehicle speed V _t.

FIG. 7 is a diagram illustrating a state where the speed of the vehicle that is turning changes. In FIG. 7, V _tx indicates the amount of change in the longitudinal speed, and V _ty indicates the amount of change in the lateral speed.
As shown in FIG. 7, when considering only the balance of force due to speed change and extracting only the result of turning, Equation (9) is expressed by Equation (10) below.

<Correction of direct integration method>
The above-mentioned direct integration method has a problem that measurement errors such as sensor noise and offset accumulate, and the estimated value drifts. Therefore, as shown in FIG. 3, and the correction of the direct integration method by the vehicle body slip angle beta _t ^* _Atan estimated by the vehicle body slip angle beta _t ^* _Observer and Atan method estimated by linear observer.

FIG. 8 is a graph showing the characteristics of the correction gains K1 and K2. In the Atan method, the vehicle body slip angle β _t ^* _Atan is calculated based on the lateral acceleration G _y , longitudinal acceleration G _x , and acceleration A _y in the turning radius direction due to the field force. Even in such a case, the vehicle body slip angle β _t ^* _Atan can be calculated with high accuracy.

Therefore, as shown in FIG. 8, when the tire slip angle beta _w and the lateral acceleration G _y is small tire characteristic becomes linear, the correction as correction allowance by linear observer of using a tire characteristic of the vehicle model is greater Increase the gain K1. Conversely, the correction gain K2 is set to be small so that the correction amount by the Atan method is small.

In the case the tire slip angle beta _w and the lateral acceleration G _y is large tire characteristic becomes nonlinear, the correction allowance by linear observer is set low and the correction gain K1 to be smaller. Conversely, the correction gain K2 is set large so that the correction allowance by the Atan method without using the tire characteristics is large.

(Tire slip angle calculator)
In the tire slip angle calculation unit 18, the steering angle δ detected by the steering angle sensor 1, the yaw rate γ detected by the yaw rate sensor 2, the vehicle body speed V _t calculated by the vehicle body speed calculation unit 16, and the vehicle body slip angle estimation unit 17 estimate. Based on the vehicle body slip angle β _t , the tire slip angle β _w is calculated. The tire slip angle calculation value is output to the cornering stiffness calculation unit 20. Specifically, tire slip angles β _wf and β _wr of the front and rear tires are calculated by the following equation (11).

(Tire lateral force calculation unit)
In the tire lateral force calculating section 19, based on the vehicle speed Vt of the yaw rate γ of the yaw rate sensor 2 detects the lateral acceleration Gy and body speed calculating section 16 lateral acceleration sensor 3 is detected by computing, calculates the tire lateral force F _y . The tire lateral force calculation value is output to the cornering stiffness calculation unit 20. Specifically, the tire lateral forces F _yf and F _yr of the front and rear tires are calculated by the following equation (12).

(Cornering stiffness calculation unit)
In cornering stiffness calculating section 20, based on the tire lateral force F _y in which the tire slip angle beta _w and the tire lateral force calculating section 19 tire slip angle calculating section 18 has calculated is calculated, it calculates a cornering stiffness Cs. The cornering stiffness calculation value is output to the vehicle body slip angle estimation unit 17, the stability factor calculation unit 21, and the vehicle behavior estimation unit 22.

The cornering stiffness Cs is obtained using the cornering stiffness map of FIG. The cornering stay Fune SMAP is to determine the cornering stiffness Cs based on the ratio of the tire lateral force F _y and the tire slip angle beta _w. In Example 1, the size and magnitude of the tire lateral force F _y of the tire lateral force F _y is regarded as approximately equal, are used tire lateral force F _y as the tire lateral force F _y. Further, actually cornering stay Fune SMAP is used and for equivalent cornering stiffness Cs _r of the two rear wheels and for equivalent cornering stiffness Cs _f of two front wheels.
Hereinafter, the cornering stiffness Cs and the cornering stiffness map will be described in detail.

<Tire characteristics>
Figure 10 is a diagram showing a tire characteristic curve showing the relationship between tire slip angle beta _w and the tire lateral force F _y when the road surface μ is "1.0". The tire characteristic curve is created based on tire characteristic values obtained from experimental data obtained by conducting a turning test of the vehicle in advance. Specifically, a tire model (Magic Formula) is set based on experimental data, and an equivalent characteristic curve of two wheels of the front tire and two wheels of the rear tire is created.

すなわち、タイヤスリップ角β_wが小さくタイヤ特性曲線が線形で場合には、固定値であるコーナリングパワーCpを用いて車両モデルを規定することができる。 In the tire slip angle beta _w is small range A shown in FIG. 10, the tire characteristic curve is linear. The cornering power Cp indicates the slope of the tire characteristic curve when the tire characteristic curve is linear. The cornering power Cp is a fixed value determined by the specifications of the vehicle and the tire, and the tire characteristics can be expressed by the following equation (13).

That is, when the tire characteristic curve small tire slip angle beta _w is a linear, it is possible to define the vehicle model by using the cornering power Cp is a fixed value.

In the range B where the tire slip angle β _w shown in FIG. 10 is large, the tire lateral force F _y is saturated and the tire characteristic curve becomes nonlinear. Therefore, tire characteristics cannot be expressed by the cornering power Cp which is the above-mentioned fixed value.
That is, when large tire characteristic curve tire slip angle beta _w there is nonlinear, it is impossible to precisely define the vehicle model by using the cornering power Cp is a fixed value.
Accordingly, in the first embodiment, the cornering power Cp of the vehicle model is corrected using the cornering stiffness Cs, and the tire characteristic curve is extended to a non-linear region.

<About cornering stiffness>
Next, the cornering stiffness Cs will be described.
Figure 11 is a diagram for explaining cornering stiffness Cs, a tire characteristic curve showing the relationship between tire slip angle beta _w and the tire lateral force F _y in time and Figure 10 also road μ is "1.0".
The cornering stiffness Cs a slope of the tire characteristic curve, that is obtained by partially differentiating the tire characteristic curve by the tire slip angle beta _w.

As shown in FIG. 11, a plurality of straight lines having different inclinations from the origin are drawn on the tire characteristic curve (FIG. 11 shows some straight lines a, b and c to be drawn). The slope of the tangent line of the tire characteristic curve at the intersection of the tire characteristic curve and each of the straight lines a, b, c (intersections within the circles in FIG. 11) and the inclination of each straight line a, b, c are obtained. When the tangent slope (cornering stiffness Cs) of the tire characteristic curve is taken along the vertical axis and the slopes of the curves a, b, c are taken along the horizontal axis, the cornering stiffness map shown in FIG. In the cornering stiffness map, the vertical axis represents the value obtained by partial differentiation of the tire characteristic curve with the tire slip angle β _w , and the ratio between the tire lateral force F _y and the tire slip angle β _w at the point where the partial differentiation was performed You may obtain | require by taking on a horizontal axis.

  FIG. 12 is a diagram showing tire characteristic curves having different road surface μ (FIG. 12A) and cornering stiffness maps having different road surface μ (FIG. 12B). In FIG. 12, the thick solid line indicates the tire characteristic curve and the cornering stiffness when the road surface μ is “1.0”, the dotted line indicates the road surface μ is “0.5”, and the thin solid line indicates the tire characteristic curve and the cornering stiffness when the road surface μ is “0.2”. In FIG. 12 (b), the thin solid line overlaps with the thick solid line, and the dotted line overlaps with the thick line, and a part of the thin solid line is not visible in the drawing.

  As shown in FIG. 12 (a), the inclination at the intersection of the straight line having a certain inclination and the tire characteristics is almost the same even when the road surface μ is different. Therefore, when cornering stiffness maps having different road surfaces μ are created, they almost coincide as shown in FIG. Therefore, the cornering stiffness map is not affected by the road surface μ, and information on the road surface μ is not required for the calculation of the cornering stiffness Cs.

  Further, since the tire characteristics change according to the wheel load change of the tire, the cornering stiffness calculation unit 20 corrects the cornering stiffness map according to the wheel load change of the tire.

For example, the wheel load when the longitudinal acceleration G _x and the lateral acceleration G _y are not generated is set as an initial value, and the ratio of the wheel load obtained from the longitudinal acceleration G _x and the lateral acceleration G _y to the initial value is set as the scale ratio, and the cornering stiffness is set. Correct the map.

  FIG. 13 is a cornering stiffness map corrected according to the wheel load. FIG. 13 shows the case where the ratio of the wheel load to the initial value is 0.6 times, 0.8 times, 1 time, and 1.2 times. By using the corrected cornering stiffness map shown in FIG. 13, more accurate cornering stiffness can be calculated.

(Stability factor calculation section)
The stability factor calculation unit 21 calculates a static margin SM based on the cornering stiffness Cs calculated by the cornering stiffness calculation unit 20 and outputs the static margin information to the vehicle behavior estimation unit 22.

The static margin SM is calculated by the following equation (14).

(Vehicle behavior estimation part)
The vehicle behavior estimation unit 22 estimates the vehicle behavior based on the cornering stiffness Cs calculated by the cornering stiffness calculation unit and the static margin SM calculated by the stability factor calculation unit 21. The estimated vehicle behavior information is output to the unstable behavior suppression assist command value calculation unit 23.

  The vehicle behavior is estimated by the value of the static margin SM. Specifically, understeer is estimated when the static margin SM> 0, normal steer when the static margin SM = 0, and oversteer when the static margin SM <0.

(Unstable behavior suppression assist command value calculator)
The unstable behavior suppression assist command value calculation unit 23 calculates an assist torque command value that suppresses the unstable behavior of the vehicle based on the vehicle behavior information estimated by the vehicle behavior estimation unit 22, and uses the calculated assist torque command value as the EPSECU6. Output to.

When the drift-out occurs, a state where the front wheel 11FL, a tire lateral force F _y in 11FR is saturated. This can be detected from the front wheel cornering stiffness Cs _f , and when it starts to decrease from the initial value, the occurrence of drift out can be expected thereafter.

Therefore, the reduction of the front wheel cornering stiffness Cs _f, the front wheels 11FL, a quantitative indication of the state where the tire lateral force F _y in 11FR approaches saturation, the control of heavy steering reaction force in accordance with the reduction of the front wheel cornering stiffness Cs _f I do. That is, control for reducing steering reaction force assist by the EPS motor 7 is performed. By increasing the steering reaction force, it is possible to prevent the driver from turning the steering wheel 9 too much, and the information effect that informs the driver that the front wheel grip has been lowered due to a change in the steering reaction force. Obtainable.

However, it is not always necessary to make the steering wheel 9 heavier when the front wheel cornering stiffness Cs _f decreases. For example, _{if the} rear wheel cornering stiffness Cs _r is reduced at the same time as the front wheel cornering stiffness Cs _f , depending on the situation, oversteer, not understeer, that is, spin may occur. At this time, since the counter steer cannot be applied quickly if the steering wheel is still heavy, there is a possibility that the vehicle behavior may be unstable due to the return.

Therefore, at the same time it monitors the front wheel cornering stiffness Cs _f, vehicle characteristics (oversteer / understeer tendency) is also monitored.

For example, the front wheels 11FL, and cornering stiffness Cs of 11FR is not reduced, when the vehicle is in understeer, by reducing the output of EPS motor 7 in accordance with the degree of reduction of the front wheel cornering stiffness Cs _f, the vehicle unstable It is possible to prevent drift-out which is a behavior. At this time, if an oversteer occurs unexpectedly and falls into the spin mode, countersteer can be prevented from being inhibited.

[Vehicle running state estimation processing]
FIG. 14 is a flowchart showing a flow of processing performed in the vehicle running state estimation device 8.

In step S1, the process proceeds to estimate the vehicle speed V _t to step S2.
In step S2, the process proceeds to step S3 to calculate the tire slip angle beta _w.
In step S3, the tire lateral force _Fy is calculated and the process proceeds to step S4.
In step S4, the cornering stiffness Cs is calculated, and the process proceeds to step S5 and step S7.

In step S5, the static margin SM is calculated and the process proceeds to step S6.
In step S6, it is determined whether or not the vehicle behavior has drifted out. When the vehicle has drifted out, the process proceeds to step S8, and when it has not drifted out, the process ends.
In step S7, the cornering power Cp in the vehicle model is updated by the cornering stiffness Cs, and the process proceeds to step S2.

  In the flowchart of FIG. 14, the process proceeds from step S1, step S2, step S3, step S4, step S7, and the cornering power Cp in the vehicle model is updated by the cornering stiffness Cs in step S7.

  When the vehicle behavior is drifting out, a transition is made to step S1, step S2, step S3, step S4, step S5, step S6, and step S8, and a command is output to reduce the steering assist force in step S8. .

  When the vehicle behavior has not drifted out, the process proceeds from step S1, step S2, step S3, step S4, step S5, step S6, and END.

[Action]
Next, the operation of the vehicle travel state estimation device 8 according to the first embodiment will be described.
In the first embodiment, the cornering power Cp of the vehicle model in the linear observer calculation unit 17 is corrected by the cornering stiffness Cs. Assume that the cornering stiffness calculation unit 20 calculates the cornering stiffness Cs based on the vehicle body slip angle estimated value β _t ^* _Observer calculated by the linear observer calculation unit 17. In this case, the estimated value calculated by the linear observer calculating unit 17 is used to further calculate the estimated value by the linear observer calculating unit 17, and the calculation results of the vehicle body slip angle estimated value β _t ^* _Observer and the cornering stiffness Cs diverge. And become unstable.

  On the other hand, in the Atan method calculation unit 17c, since the correction by the estimated value is not performed, a stable calculation result can be obtained.

Therefore, the divergence of the estimated vehicle slip angle value β _t ^* _Observer is suppressed by using the estimated vehicle slip angle value β _t ^* _Atan calculated by the Atan method calculating unit 17c.
Therefore suppressing the vehicle body slip angle estimation value beta _t ^* divergence, it is possible to suppress the divergence of cornering stiffness Cs to calculation using the vehicle body slip angle estimation value beta _t ^*.

Further, the divergence of the estimated vehicle body slip angle β _t ^* _Observer and the cornering stiffness Cs is caused by the accumulation of errors. Therefore, the Atan method computing unit 17c computes the estimated vehicle body slip angle β _t ^* _Atan without using the integral term as shown in equations (9) and (10).
Thus, with a higher operation accuracy of the vehicle body slip angle estimation value beta _t ^* and cornering stiffness Cs by the correction of the vehicle model according cornering stiffness Cs, it is possible to suppress the divergence of the vehicle body slip angle estimation value beta _t ^* and cornering stiffness Cs it can.

In Atan method computing unit 17c, obtains the acceleration A _y of the turning radial direction from the lateral acceleration G _y and longitudinal acceleration sensor 4 detects the longitudinal acceleration G _x lateral acceleration sensor 3 detects the turning radius direction and vehicle width direction The vehicle body slip angle was calculated from the angle formed by
Therefore, the vehicle body slip angle β _t can be estimated without using the past history.

In the Atan method, the vehicle body slip angle β _t ^* _Atan is calculated based on the lateral acceleration G _y , longitudinal acceleration G _x , and acceleration A _y in the turning radius direction due to the field force. Even in such a case, the vehicle body slip angle β _t ^* _Atan can be calculated with high accuracy.
So when the lateral acceleration G _y and tire slip angle beta _w is smaller tire characteristic is nonlinear, and to set a large correction gain K2 as correction allowance by Atan method using no tire characteristic increases.
Therefore, it is possible to estimate the vehicle body slip angle β _t with high accuracy in the region where the tire characteristics are nonlinear.

[effect]
Next, the effects of Example 1 are listed below.
(1) and the tire lateral force calculating section 19 for calculating a tire lateral force F _y, the linear observer calculating unit 17b and the tire slip angle calculating means 18 for estimating the tire slip angle beta _w using a vehicle model with cornering power Cp If a cornering stiffness calculating section 20 for calculating a cornering stiffness Cs from the tire lateral force F _y and the tire slip angle beta _w, and Atan method calculating section 17c for preventing the divergence of calculated values of cornering stiffness Cs in cornering stiffness calculating section 20 The linear observer calculation unit 17b corrects the cornering power Cp of the vehicle model by the cornering stiffness Cs.

Therefore, since correction is performed using the estimated vehicle slip angle value β _t ^* _Atan calculated by the Atan method calculating unit 17c, the divergence of the estimated vehicle slip angle value β _t ^* and the cornering stiffness Cs can be suppressed.

(2) The Atan method calculation unit 17c corrects the vehicle body slip angle β _t ^* estimated by the vehicle body slip angle estimation unit 17 by the vehicle body slip angle β _t ^* _Atan calculated without using an integral term.
Thus, with a higher operation accuracy of the vehicle body slip angle estimation value beta _t ^* and cornering stiffness Cs by the correction of the vehicle model according cornering stiffness Cs, it is possible to suppress the divergence of the vehicle body slip angle estimation value beta _t ^* and cornering stiffness Cs it can.

(3) and a lateral acceleration G longitudinal acceleration sensor 4 and the lateral acceleration sensor 3 for detecting a longitudinal acceleration G _{x of y} to detect a provided a turning radial Atan method computing unit 17c from the lateral acceleration G _y and longitudinal acceleration G _x The vehicle body slip angle β _t ^* _Atan is calculated from the angle β ₂ formed by the turning radius direction and the vehicle width direction.
Therefore, the vehicle body slip angle β _t can be estimated without using the past history.

(4) the lateral acceleration sensor 3 for detecting a lateral acceleration G _y is provided, the lateral acceleration G _y or, as the vehicle body slip angle beta _t is set large correction gain K2 of the slip angle calculated without using the more integral term increases did.
Therefore, it is possible to estimate the vehicle body slip angle β _t with high accuracy in the region where the tire characteristics are nonlinear.

[Other embodiments]
The best mode for carrying out the present invention has been described based on the first embodiment. However, the specific configuration of the present invention is not limited to the first embodiment and does not depart from the gist of the present invention. Any change in the design of the range is included in the present invention.

For example, although the correction is performed at the stage of the vehicle body slip angle β _t, the correction may be performed after the tire slip angle β _w is decomposed.

Further, as the divergence prevention means, the Atan method is used, but another method that does not use the integral term may be used.
Further, the cornering stiffness map is used for the equivalent cornering stiffness Cs _f of the two front wheels and the equivalent cornering stiffness Cs _r of the two rear wheels, but a four-wheel independent cornering stiffness map may be used.

  The lateral acceleration sensor 3 corresponds to the acceleration calculation means in claim 3 and the lateral acceleration detection means in claim 4. The longitudinal acceleration sensor 4 corresponds to the acceleration calculation means in claims 3 and 4, and linear observer calculation. The part 17b and the tire slip angle calculation part 188 correspond to the vehicle behavior calculation means of claims 1 to 4, and the Atan method calculation part 17c corresponds to the divergence prevention means of claims 1 to 4, and the tire lateral force The computing unit 19 corresponds to the lateral force equivalent physical quantity computing means of claims 1 to 4, and the cornering stiffness computing unit 20 corresponds to the cornering stiffness computing means of claims 1 to 4.

1 is a configuration diagram illustrating a schematic configuration of a vehicle according to a first embodiment. It is a control block diagram of the vehicle running state estimation device of the first embodiment. FIG. 3 is a control block diagram of a vehicle body slip angle estimation unit according to the first embodiment. 2 is a two-wheel model of Example 1. FIG. 3 is a block diagram of a linear 2-input observer according to the first embodiment. FIG. It is a figure which shows the state which the vehicle in turning of Example 1 is drive | working with a vehicle body slip angle. It is a figure which shows the state which the vehicle in turning of Example 1 is drive | working with a vehicle body slip angle. 6 is a graph illustrating correction gain characteristics according to the first exemplary embodiment. 2 is a cornering stiffness map according to the first embodiment. It is a figure which shows the tire characteristic curve of Example 1. FIG. It is a figure which shows the tire characteristic curve of Example 1. FIG. It is a figure which shows the tire characteristic curve and cornering stiffness map of Example 1. 2 is a cornering stiffness map corrected according to the wheel load of the first embodiment. It is a flowchart which shows the flow of the process performed in the vehicle running condition estimation apparatus of Example 1. FIG.

Explanation of symbols

3 Lateral acceleration sensor (acceleration calculation means, lateral acceleration detection means)
4 Longitudinal acceleration sensor (acceleration calculation means)
17 Vehicle body slip angle estimation unit 17b Linear observer calculation unit (vehicle behavior calculation means)
17c Atan method computing unit (divergence prevention means)
19 Tire lateral force calculation section (wheel lateral force equivalent physical quantity calculation means)
20 Cornering stiffness calculation unit (corner stiffness calculation means)

Claims (4)

A lateral force equivalent physical quantity calculating means for calculating a lateral force equivalent physical quantity of the wheel;
Vehicle behavior calculating means for calculating a tire slip angle based on a vehicle model;
Cornering stiffness calculation means for calculating cornering stiffness from the lateral force equivalent physical quantity of the wheel and the tire slip angle;
Divergence preventing means for preventing divergence of the calculated value of the cornering stiffness calculating means;
Provided,
The vehicle running state estimating device, wherein the vehicle behavior calculating means corrects the vehicle model by the cornering stiffness. In the vehicle travel state estimation device according to claim 1,
The vehicle behavior calculation means is means for estimating a vehicle body slip angle from the vehicle model and calculating the tire slip angle from the vehicle body slip angle,
The divergence prevention unit is a unit that corrects the vehicle body slip angle calculated by the vehicle behavior calculation unit using a vehicle body slip angle calculated without using an integral term. In the vehicle travel state estimation device according to claim 2,
An acceleration calculating means for calculating acceleration in two directions on the turning plane with respect to the vehicle;
The divergence prevention means is a means for obtaining a turning radius direction from the acceleration in the two directions and calculating a vehicle body slip angle from an angle formed by the turning radius direction and the vehicle width direction of the vehicle. Estimating device. In the vehicle travel state estimation device according to claim 2 or 3,
A lateral acceleration detecting means for detecting the lateral acceleration of the vehicle is provided;
The divergence prevention means sets a correction gain for a slip angle calculated without using the integral term as the lateral acceleration or the vehicle body slip angle increases, so as to increase the vehicle running state estimation device.

Images