JP2013007703A - Device for estimating tire ground contact length - Google Patents

Abstract

PROBLEM TO BE SOLVED: To stable estimate tire ground contact length without attaching an acceleration sensor to a wheel.SOLUTION: A tire ground contact length estimation device 60 includes: a reference self-aligning torque peak determination part 80 for determining whether or not reference self-aligning torque calculated as a value proportional to self-aligning torque on the basis of the self-aligning torque detected by a self-aligning torque detection part 62 and a wheel load change ratio estimated by a wheel load change ratio estimation part 61 reaches a peak; and a tire ground contact length estimation part 63 which when the reference self-aligning torque peak determination part 80 determines that the reference self-aligning torque reaches the peak, estimates tire ground contact length from a value obtained by dividing the self-aligning torque detected by the self-aligning torque detection part 62 by lateral force estimated by a tire lateral force estimation part 70.

Description

  The present invention relates to a technique for estimating a tire contact length.

As a technique for estimating the tire contact length, for example, there is a technique described in Patent Document 1.
In this prior art, an acceleration sensor is provided inside the tread of the tire in order to acquire acceleration data of the rolling tire. In this conventional technique, first, the peak value on the stepping side and the peak value on the kicking side of the tire contact surface are extracted from the time-series acceleration data in the vicinity of contact with the rolling tire. Moreover, in this prior art, the step-on side threshold value and the kick-out side threshold value for determining the contact length are set based on these peak values. Further, in this prior art, the tire contact length is calculated by obtaining the intersection point where the waveform of the acceleration measurement time-series data crosses the set stepping-side threshold value and the kicking-side threshold value.

JP 2010-32355 A

However, in the prior art, it is necessary to provide an acceleration sensor inside the tire tread in order to acquire acceleration data of the rolling tire. Although the acceleration information of the acceleration sensor is periodically transmitted to the vehicle as a radio signal, the vehicle cannot always receive all the information due to a change in the transmission state or the reception state. As a result, the technique disclosed in Patent Document 1 may not be able to stably estimate the contact length of the tire.
An object of the present invention is to stably estimate a tire contact length without providing an acceleration sensor on a wheel.

  In order to solve the above-mentioned problems, in one embodiment of the present invention, the self-aligning torque of the tire, the yaw rate of the vehicle, and the lateral acceleration of the vehicle are detected. In one embodiment of the present invention, the lateral force of the tire is estimated based on the detected vehicle yaw rate and vehicle lateral acceleration. In one embodiment of the present invention, the change ratio of the wheel load is calculated. Furthermore, in one embodiment of the present invention, the reference self-aligning torque is calculated based on the change ratio of the self-aligning torque and the wheel load. In one embodiment of the present invention, when it is determined that the calculated reference self-aligning torque has a peak, a value obtained by dividing the self-aligning torque by the lateral force is estimated as a tire contact length.

According to the present invention, when the reference self-aligning torque reaches a peak, the tire contact length is estimated by dividing the self-aligning torque by the tire lateral force. As a result, according to the present invention, the tire ground contact length can be calculated based on the self-aligning torque and the tire lateral force that can be acquired by the existing vehicle-mounted sensor.
Thus, according to the present invention, it is possible to stably estimate the tire ground contact length without providing an acceleration sensor on the wheel.

FIG. 6 is a characteristic diagram showing a relationship between normalized self-aligning torque M ^z / (l · μ · F _z ) and φ. It is a figure showing an example of composition of vehicles of a 1st embodiment. It is a block diagram which shows the structural example of a tire contact length estimation apparatus. It is a block diagram which shows the structural example of the tire lateral force estimation part for calculating the tire lateral force of each front wheel on either side. It is a block diagram which shows the structural example of a reference | standard self-aligning torque peak determination part. It is a figure explaining an example of the judgment processing by a peak judgment part. It is a block diagram which shows the structural example for performing tire contact length estimation accurately. It is a block diagram which shows the structural example of an estimation stop condition determination part. It is a block diagram which shows the structural example of the self-aligning torque sudden change determination part. It is a block diagram which shows the structural example of a noise level determination part. It is a block diagram which shows the structural example of a turning state determination part. It is a block diagram which shows the other structural example of an estimation stop condition determination part. It is a block diagram which shows the structural example of 2nd Embodiment. It is a characteristic diagram showing an example of the characteristic of the aligning torque model j _f. It is a characteristic view which shows the road surface friction coefficient acquired by this embodiment, and the road surface friction coefficient of the comparative example. It is a block diagram which shows the structural example of 3rd Embodiment. It is a characteristic view which shows an example of the relationship between tire pressure and tire contact length. It is a block diagram which shows the structural example of 4th Embodiment. It is a characteristic view which shows an example of the relationship between tire rolling resistance and tire contact length. It is a block diagram which shows the structural example of 5th Embodiment.

(Explanation of vehicle parameters and variables)
In the following description of the present embodiment, vehicle parameters and variables defined below are used.
m: vehicle weight _Iz : body yaw moment of inertia _hcg : height from vehicle center of gravity to ground contact surface
g: acceleration of gravity l: tire contact length l _t : half length of tread base l _f : length from vehicle center of gravity to front wheel l _r : length from vehicle center of gravity to rear wheel C _f : front wheel tire Cornering stiffness C _r : Tire cornering stiffness of rear wheel V: Vehicle speed α: Tire slip angle α _f : Tire slip angle of front wheel α _r : Tire slip angle of rear wheel β: Vehicle slip angle β ': Time differential of vehicle slip angle Value γ: Yaw rate γ ′: Time differential value of yaw rate G _x : Longitudinal acceleration G _y : Lateral acceleration μ: Road surface friction coefficient δ: Steering angle of front wheels F _y : Tire lateral force F _z : Wheel load M _z : Self-application Lining torque Here, the subscript “f” indicates the value of the front wheel, and the subscript “r” indicates the value of the rear wheel. Also, the subscript “L” used in the following description indicates the value of the left wheel, and the subscript “R” indicates the value of the right wheel.

(Principle of tire contact length estimation)
First, the principle of tire contact length estimation according to this embodiment will be described.
According to the filar tire model, the lateral force F _y and the self-aligning torque M _z are expressed by the following equations (1) and (2).
F _y / (μ · F _z ) = φ− (1/3) · φ ² + (1/27) · φ ³ (1)
M ^z / (l · μ · F _z ) = (1/6) · φ− (1/6) · φ ² + (1/18) · φ ³ − (1/162) · φ ^4. 2)

Here, φ is expressed by the following equation (3).
φ = (C _f / (μ · F _z )) · tan α (3)
When solving the equations (1) and (2) for the mu · _{F z,} the tire contact length l is expressed by the following equation (4).
l = (φ− (1/3) · φ ² + (1/27) · φ ³ ) / ((1/6) · φ− (1/6) · φ ² + (1/18) · φ ³ − (1/162) · φ ⁴ ) · M _z / F _y (4)

As shown in this equation (4), the coefficient by which M _z / F _y corresponding to the lateral force lever arm is multiplied is a function of φ. Here, according to the above equation (3), information of the road surface friction coefficient μ is required to calculate φ.
In the present embodiment, the tire contact length l is calculated as follows without using information on the road surface friction coefficient μ.

FIG. 1 is a diagram showing the relationship between the normalized self-aligning torque M ^z / (l · μ · F _z ) and φ shown in the equation (2).
The normalized self-aligning torque M ^z / (l · μ · F _z ) is an upward convex function as shown in FIG. Therefore, from the equation (2), φ at which the reference self-aligning torque takes an extreme value is expressed by the following equation (5).

φ = 3/4 (5)
Then, by substituting this φ into the equation (4), the following equation (6) can be obtained.
l = (296/27) · (M _z / F _y ) (6)
From the above, when the normalized self-aligning torque M ^z / (l · μ · F _z ) is a peak, φ can be regarded as 3/4, and therefore the tire contact length l can be expressed by the above equation (6). If the self-aligning torque M _z and the lateral force F _{y at} this time can be obtained, the tire contact length l can be estimated without using the road surface friction coefficient μ.

(First embodiment)
Next, a first embodiment will be described with reference to the drawings.
The vehicle according to the first embodiment is a vehicle including a tire contact length estimation device that adopts the above-described principle of tire contact length estimation.

(Constitution)
FIG. 2 is a diagram illustrating a vehicle configuration example of the first embodiment.
This vehicle has a steering angle sensor (or rotation angle sensor) 2, an accelerator pedal sensor 3, wheel speed sensors 4FL, 4FR, 4RL, 4RR, a yaw rate sensor 5, and an acceleration sensor 6, as shown in FIG. Yes.

  The steering angle sensor 2 is attached to the rotating shaft 8 of the steering wheel 7. The steering angle sensor 2 detects the rotation of the rotating shaft 8 and outputs the detected steering angle signal of the steering wheel to the integrated controller 50. Further, the accelerator pedal sensor 3 outputs an accelerator opening signal AP0 obtained by detecting the accelerator opening to the integrated controller 50. The wheel speed sensors 4FL to 4RR are attached to the front, rear, left and right wheels 9FL, 9FR, 9RL, 9RR, respectively. The wheel speed sensors 4FL to 4RR output the detected wheel speeds of the wheels 9FL to 9RR to the integrated controller 50. The yaw rate sensor 5 detects the yaw rate of the vehicle, and outputs the detected yaw rate to the integrated controller 50. The acceleration sensor 6 detects the longitudinal acceleration and lateral acceleration of the vehicle, and outputs the detected longitudinal acceleration and lateral acceleration of the vehicle to the integrated controller 50.

The vehicle also includes a reaction force motor 11 and steering motors 12FL and 12FR.
The reaction force motor 11 is connected to the steering wheel 7. Then, the reaction force motor 11 applies a reaction force torque equivalent to a self-aligning torque generated by the tire during steering to the steering wheel 7. Thereby, the reaction torque equivalent to the self-aligning torque is transmitted to the driver. The reaction force motor 11 is controlled by the integrated controller 50.

  The steered motors 12FL and 12FR steer the front wheels 9FL and 9FR. Specifically, the steering motors 12FL and 12FR steer the front wheels 9FL and 9FR in phase by displacing a steering rack (not shown) in the vehicle width direction. The steered motors 12FL and 12FR output the output torque of the motors 12FL and 12FR and the rotation speed of the motor detected by a rotation position sensor (not shown) attached to the motor rotation shaft to the integrated controller 50. The steered motors 12FL and 12FR are controlled by the integrated controller 50.

In addition, the vehicle includes a drive motor 21 as a drive force generation source, a drive circuit 22 that drives and controls the drive motor 21, and a battery 23 that supplies power to the drive motor 21. For example, the battery 23 is a lithium ion battery.
The drive motor 21 has a power shaft connected to the left and right rear wheels 9RL and 9RR, and drives the left and right rear wheels 9RL and 9RR. The drive circuit 22 drives the drive motor 21 with the electric power from the battery 23 so that the output torque of the drive motor 21 matches the received torque command value.

  The integrated controller 50 controls each unit to be driven. The integrated controller 50 is a controller including a microcomputer and its peripheral circuits. For example, the integrated controller 50 includes a CPU, a ROM, a RAM, and the like as in a general ECU (Electronic Control Unit). The ROM stores one or more programs for realizing various processes. The CPU executes various processes according to one or more programs stored in the ROM.

  For example, the integrated controller 50 drives and controls the steering motors 12FL and 12FR based on the steering angle signal of the steering wheel 7 from the steering angle sensor 2, thereby displacing the steering rack in the vehicle width direction and the front wheels 9FL and 9FR. To the same phase. Further, the integrated controller 50 controls the rotational drive of the drive motor 21 by outputting a torque command value to the drive circuit 22.

In the configuration as described above, the integrated controller 50 has a tire contact length estimation device.
The tire contact length estimation device estimates the tire contact length based on the above-described principle of tire contact length estimation.
FIG. 3 is a block diagram illustrating a configuration example of the tire contact length estimation device 60.

As shown in FIG. 3, the tire contact length estimation device 60 includes a wheel load change ratio estimation unit 61, a self-aligning torque detection unit 62, a tire lateral force estimation unit 70, a reference self-aligning torque peak determination unit 80, and a tire. A contact length estimation unit 63 is provided.
The wheel load change ratio estimation unit 61 estimates the wheel load change ratio.
The wheel load change ratio r _Fz is defined by the following equation (7).

r _Fz = (F _z0 + ΔF _z ) / F _z0 (7)
Here, F _z0 represents the static wheel load of each wheel. ΔF _z represents a wheel load change amount that changes in accordance with wheel load acceleration / deceleration or turning.
The wheel load change ratio estimation unit 61 calculates the wheel load change ratio r _Fz defined by the equation (7) for each of the left and right front wheels. Here, the wheel load change amount, for example, by a formula derived from the balance of the static roll, pitch moment acting on the vehicle can be calculated based on the longitudinal acceleration G _x and the lateral acceleration G _y. That is, for example, the wheel load change amount of each of the left and right front wheels is calculated using the following formulas (8) and (9).

F _zL + ΔF _zL = m · (l _r · g−h _cg · G _x ) (l _t · g−h _cg · G _y ) / (2 · l _t · g · (l _f + l _r )) (8)
F _zR + ΔF _zR = m · (l _r · g−h _cg · G _x ) (l _t · g + h _cg · G _y ) / (2 · l _t · g · (l _f + l _r )) (9)
Here, the longitudinal acceleration G _x and the lateral acceleration G _y, may be used detected value by the built-in sensors to the existing anti-skid device, such as an acceleration sensor 6.

Here, it is not necessary to use both the longitudinal acceleration G _x (detection value of the longitudinal acceleration sensor) and the lateral acceleration G _y (detection value of the lateral acceleration sensor). That is, only one of them may be used. For example, when the tire ground contact length is estimated when the longitudinal acceleration is near zero, the term relating to G _x is omitted in the equations (8) and (9) without using the longitudinal acceleration G _x , and the wheel load The amount of change is calculated. Further, when estimating the tire contact length when the lateral acceleration is near zero, the terms related to G _y are omitted in the above equations (8) and (9) without using the lateral acceleration G _y , and the wheel load The amount of change is calculated.

Then, the wheel load change ratio estimation unit 61 outputs the estimated wheel load change ratio to the reference self-aligning torque peak determination unit 80.
The self-aligning torque detector 62 detects the self-aligning torque.
Specifically, the self-aligning torque detector 62 detects the self-aligning torque based on the command currents of the steering motors 12FL and 12FR in consideration of the motor torque constant and the suspension link ratio. Then, the self-aligning torque detection unit 62 outputs the detected self-aligning torque to the reference self-aligning torque peak determination unit 80 and the tire contact length estimation unit 63.

The tire lateral force estimation unit 70 estimates tire lateral force.
Specifically, the tire lateral force estimation unit 70 estimates the tire lateral force based on a detectable vehicle state. For example, the tire lateral force estimation unit 70 estimates the tire lateral force F _yf of the front wheels based on the yaw rate and the lateral acceleration according to the following equation (10) derived from the two-wheel model.

Here, a known two-wheel model can be used as the two-wheel model. For example, known two-wheel models include those described in “Motor Movement and Control, Second Edition,” Sankaido, Masato Abe, page 55, etc. Further, the yaw rate γ ′ is calculated, for example, by dividing the difference between the yaw rate detection value of the current processing step and the yaw rate detection value of the previous processing step by a preset sampling time. Also, the yaw rate γ and lateral acceleration G _y here, may be used detected value by the built-in sensors to the existing anti-skid device, such as a yaw rate sensor 5, an acceleration sensor 6.

In this embodiment, in order to independently calculate the tire ground contact lengths of the left and right front wheels, the tire lateral force F _yf of the front wheels is calculated for the left and right wheels as follows.
FIG. 4 is a block diagram illustrating a configuration example of the tire lateral force estimating unit 70 for calculating the tire lateral force of the left and right front wheels.
As shown in FIG. 4, the tire lateral force estimation unit 70 includes a steering angle detection unit 71, a vehicle speed detection unit 72, a yaw rate detection unit 73, a lateral acceleration detection unit 74, a wheel load change ratio estimation unit 75, and a front wheel slip angle estimation unit. 76, a left and right total lateral force estimating unit 77, and an optimization calculating unit 78.

The steering angle detection unit 71 detects the steering angle of the wheel. Then, the steering angle detection unit 71 outputs the detected steering angle of the wheel to the front wheel slip angle estimation unit 76.
The tire lateral force estimation unit 70 may use the steering angle detected by an existing vehicle control device such as a skid prevention device without providing the steering angle detection unit 71. Further, the tire lateral force estimation unit 70 may estimate the front wheel turning angle based on the detection signal of the steering angle sensor 2.

The vehicle speed detector 72 detects the vehicle speed. For example, the vehicle speed detection unit 72 detects the vehicle speed based on the wheel speed of the driven wheel. Then, the vehicle speed detection unit 72 outputs the detected vehicle speed to the front wheel slip angle estimation unit 76.
The tire lateral force estimation unit 70 may use the vehicle speed detected by an existing vehicle control device such as a skid prevention device without including the vehicle speed detection unit 72.

The yaw rate detector 73 detects the yaw rate of the vehicle. Then, the yaw rate detection unit 73 outputs the detected yaw rate of the vehicle to the left and right total lateral force estimation unit 77.
The tire lateral force estimation unit 70 may use the yaw rate detected by an existing vehicle control device such as a skid prevention device without including the yaw rate detection unit 73.
The lateral acceleration detector 74 detects the lateral acceleration of the vehicle. Then, the lateral acceleration detection unit 74 outputs the detected lateral acceleration of the vehicle to the left-right total lateral force estimation unit 77.

The tire lateral force estimating unit 70 may use the lateral acceleration detected by an existing vehicle control device such as a skid prevention device without including the lateral acceleration detecting unit 74.
The wheel load change ratio estimation unit 75 estimates the wheel load change ratio by the same processing as the wheel load change ratio estimation unit 61 shown in FIG. Then, the wheel load change ratio estimating unit 61 outputs the estimated wheel load change ratio to the optimization calculating unit 78.

The tire lateral force estimator 70 does not include the wheel load change ratio estimator 75, and the output of the wheel load change ratio estimator 61 included in the tire ground contact length estimator 60 shown in FIG. The configuration may be such that it is input to.
The front wheel slip angle estimator 76 estimates the front wheel slip angle based on the detected front wheel turning angle δ and the vehicle speed V.
Specifically, the front wheel slip angle estimation unit 76 estimates the front wheel slip angle based on the front wheel turning angle δ and the vehicle speed V using a two-wheel model represented by the following equations (11) and (12). To do.

Here, F _yf is the tire lateral force of the left and right wheels in the front wheels. F _yr is the total tire lateral force on the left and right wheels at the rear wheel. When the _fiara tire model is used, F _yf and F _yr can be calculated using the following equations (13) and (14), respectively.
F _yf / (μ · F _{zf 0} ) = φ _f − (1/3) · φ _f ² + (1/27) · φ _f ³ (13)
F _yr / (μ · F _zr0 ) = φ _r − (1/3) · φ _r ² + (1/27) · φ _r ³ (14)

Here, φ _f and φ _r are expressed by the following equations (15) and (16).
φ _f = C _f / (μ · F _zf ) · tan α _f (15)
φ _r = C _r / (μ · F _zr ) · tan α _r (16)
Here, the road surface friction coefficient μ is set to 0.9 or the like as a nominal value. Further, as a tire model, an experimentally acquired map may be used, or a model such as a magic formula may be used.

Then, the front wheel slip angle estimating unit 76 uses the following equations (17) and (18), and based on the vehicle slip angle β and the yaw rate γ described above, the tire slip angles α _f and α _{r of the} front wheels and the rear wheels are calculated. Is calculated. The front wheel slip angle estimator 76 then outputs the estimated front wheel tire slip angle α _f to the optimization calculator 78.
α _f = β + ((l _f · γ) / V) −δ (17)
α _r = β− (l _r · γ) / V (18)
Here, the front wheel turning angle detection value is substituted for δ, and the vehicle speed detection value is substituted for V.

The left / right total lateral force estimation unit 77 estimates the total left / right tire lateral force based on the yaw rate detection value and the lateral acceleration detection value, using the equation (10). Then, the left and right total lateral force estimating unit 77 outputs the estimated left and right total tire lateral force to the optimization calculating unit 78.
The optimization calculation unit 78 includes a front wheel slip angle estimated by the front wheel slip angle estimation unit 76, a left and right total tire lateral force estimated by the left and right total lateral force estimation unit 77, and a wheel load change ratio estimated by the wheel load change ratio estimation unit 75. Based on the above, the tire lateral force of each of the left and right wheels is estimated.

Specifically, first, the optimization calculation unit 78, using the following expression (19) and (20), left and right front wheels of the wheel load _F zL based on the wheel load change _ratio, and calculates the _{F zR.}
F _zL = r _FzL · F _zL0 (19)
_{F zR = r FzR · F zR0} ··· (20)
Here, F _zL0 and F _zR0 are the static wheel loads of the left and right front wheels, respectively. Assuming that the road surface friction coefficient μ of the front wheels is the same for the left and right wheels, the following equations (21) to (23) are established.

F _yL = f (α _f , μ, F _zL ) (21)
F _yR = f (α _f , μ, F _zR ) (22)
F _yf = F _yL + F _yR (23)
Here, F _yL and F _yR are lateral forces of the left and right front wheels. f (α _f , μ, F _z ) is a tire model, and an experimentally acquired map may be used, or an analysis model such as a filar model may be used.

Then, in the equations (21) to (23), the front wheel tire slip angle estimated by the front wheel slip angle estimating unit 76 is substituted into the tire slip angle α _f of the front wheel. Further, the left and right wheel load _{F zL,} assigns a value calculated by the in _{F zR} (19) and equation (20) below. Furthermore, the left and right total lateral force estimated by the left and right total lateral force estimating unit 77 is substituted into F _yf . Then, in the equations (21) to (23), the unknowns are μ, F _yL and F _yR .

Therefore, the optimization calculation unit 78 numerically solves the above equations for these unknowns using, for example, an optimization calculation algorithm, and estimates the tire lateral forces F _yL and F _yR of the left and right front wheels. Then, the optimization calculation unit 78 outputs the estimated tire lateral forces of the left and right wheels to the tire contact length estimation unit 63.
Returning to the description of FIG. 3, the reference self-aligning torque peak determination unit 80 determines the peak of the reference self-aligning torque.

FIG. 5 is a block diagram illustrating a configuration example of the reference self-aligning torque peak determination unit 80.
As shown in FIG. 5, the reference self-aligning torque peak determination unit 80 includes a contact length change ratio calculation unit 81, a reference self-aligning torque proportional amount calculation unit 82, and a peak determination unit 83.

The contact length change ratio calculation unit 81 calculates the contact length change ratio based on the wheel load change ratio.
Specifically, the contact length change ratio calculation unit 81 calculates the contact length change ratio r ₁ using the following equation (24). Then, the contact length change ratio calculation unit 81 outputs the calculated contact length change ratio to the reference self-aligning torque proportional amount calculation unit 82.
r _l = (l ₀ + Δl) / l ₀ (24)

Here, l ₀ represents the tire contact length when the vehicle is stationary. Δl represents the fluctuation amount of the tire ground contact length caused by the braking / driving force and the turning motion. For example, the contact length change ratio calculation unit 81 calculates the tire contact length variation amount _Δl based on the wheel load variation ratio r _Fz using a model of the following equation (25).
Δl = l ₀ · (r _Fz ^1/2 −1) (25)

Note that the estimation method of the variation Δl in the tire contact length is not limited to the equation (25). That is, for example, the fluctuation amount of the tire contact length may be estimated by experimentally acquiring the relationship between the change amount of the tire contact length and the wheel load change ratio and referring to a map created thereby.
The reference self-aligning torque proportional amount calculator 82 calculates a reference self-aligning torque proportional amount.

Specifically, the reference self-aligning torque proportional amount calculation unit 82 uses the following equation (26) to calculate the wheel load change ratio r _Fz , the contact length change ratio r _l, and the self-aligning torque M _z . A reference self-aligning torque proportional amount M _znp is calculated.
M _znp = M _z / (r _Fz · r _l ) (26)
The reference self-aligning torque proportional amount calculation unit 82 outputs the calculated reference self-aligning torque proportional amount to the peak determination unit 83.
The peak determination unit 83 determines whether or not the reference self-aligning torque proportional amount takes a peak.

FIG. 6 is a diagram illustrating an example of the determination process.
As shown in FIG. 6A, the peak determination unit 83 first obtains a reference self-aligning torque (reference self-aligning torque proportional amount) between N sample times from the processing step at the current time. Then, as shown in FIG. 6B, the peak determination unit 83 fits a straight line with respect to the reference self-aligning torque acquired during the N sample time from the processing step of the current time. And the peak determination part 83 determines with the peak of a reference | standard self-aligning torque, when the absolute value of the inclination of the fitted straight line is smaller than the preset threshold value.

  Here, as an example of straight line fitting, an algorithm such as a least square method can be cited. The preset threshold value is a value within a preset range from 0 so that the peak can be determined including the vicinity of the peak of the reference self-aligning torque. Needless to say, the peak itself can be determined on condition that the slope of the fitted straight line is zero.

Here, the reason for determining the peak for the reference self-aligning torque proportional amount instead of the normalized self-aligning torque will be described.
That is, in order to estimate the tire contact length based on the principle of tire contact length estimation described above, whether or not the normalized self-aligning torque M _zn originally defined by the following equation (27) takes a peak or not? It is necessary to judge whether.

M _zn = M _z / (l · μ · F _z ) (27)
However, the denominator of the normalized self-aligning torque includes the tire contact length l to be estimated from now on and the road surface friction coefficient μ that is not easy to estimate.
For this reason, the reference self-aligning torque (reference self-aligning torque proportional amount) M _znp expressed by the following equation (28) is defined.

M _znp = M _z / (r _Fz · r _l ) (28)
On the other hand, the wheel load F _z is calculated based on the wheel load change ratio r _Fz and the static wheel load F _z0 using the following equation (29).
F _z = r _Fz · F _z0 (29)
The tire contact length l is calculated based on the tire contact length change ratio r _l and the static tire contact length l ₀ using the following equation (30).

l = r _{l·l 0} (30)
By substituting these equations (29) and (30) into the equation (28), the following equation (31) is derived.
M _znp = l ₀ · μ · F _z0 · M _z / (l · μ · F _z )
= L ₀ · μ · F _z0 · M _zn (31)

Here, if l ₀ · μ · F _z0 is _regarded as a proportional constant, the reference self-aligning torque M _znp is a value proportional to the normalized self-aligning torque M _zn . Therefore, the peak of the normalized self-aligning torque M _zn matches the peak of the reference self-aligning torque M _znp . Thus, by determining the peak for the reference self-aligning torque M _znp instead of the normalized self-aligning torque M _zn , the tire contact length can be estimated based on the above-described principle.

Then, the peak determination unit 83 outputs a peak determination result (peak determination signal) of the reference self-aligning torque to the tire contact length estimation unit 63.
Returning to the description of FIG. 3, when the reference self-aligning torque peak determination unit 80 determines that the reference self-aligning torque peak is a peak (including the vicinity of the peak), the tire contact length estimation unit 63 determines the self-aligning torque and The tire contact length is estimated from the tire lateral force. For example, when the tire force is described in a fiara model, the tire contact length estimation unit 63 can estimate the tire contact length l using the following equation (32).

l = (296/27) · (M _z / F _y ) (32)
Here, M _z / F _y represents the distance from the tire center of the point of application of the tire lateral force. In the above equation, the coefficient multiplied by the amount M _z / F _y may be the theoretical value 296/27, but may be a value obtained experimentally.
Next, a configuration example for more accurately performing the above-described tire contact length estimation will be described.

FIG. 7 is a block diagram illustrating a configuration example for more accurately estimating the tire contact length.
As shown in FIG. 7, the tire contact length estimation device 60 includes an estimated stop condition determination unit 90 in addition to the configuration shown in FIG. 3.
In this configuration example, the tire contact length estimation device 60 does not estimate the tire contact length while the estimated stop condition determination unit 90 outputs the calculation stop signal to the configuration (main body portion 60A) shown in FIG. I am doing so. For example, the tire contact length estimation device 60 stops the calculation of the reference self-aligning torque peak determination unit 80 and the tire contact length estimation unit 63 constituting the main body 60A, and stops the estimation of the tire contact length. Yes.

FIG. 8 is a block diagram illustrating a configuration example of the estimated stop condition determination unit 90.
As shown in FIG. 8, the estimated stop condition determination unit 90 includes a self-aligning torque sudden change determination unit 100, a noise level determination unit 110, and a turning state determination unit 120.
FIG. 9 is a block diagram illustrating a configuration example of the self-aligning torque sudden change determination unit 100.

As shown in FIG. 9, the self-aligning torque sudden change determination unit 100 includes a steering angle detection unit 101, a steering angular speed estimation unit 102, a self-aligning torque detection unit 103, a self-aligning torque change rate estimation unit 104, and a ratio calculation unit. 105, and a self-aligning torque change determination unit 106.
The steering angle detection unit 101 detects the steering angle of the steering wheel 7. For example, the steering angle detection unit 101 detects the steering angle with reference to a signal from the steering angle sensor 2. Then, the rudder angle detection unit 101 outputs the detected steering angle to the rudder angular velocity estimation unit 102.

Note that the self-aligning torque sudden change determination unit 100 does not include the rudder angle detection unit 101, and the output of the rudder angle detection unit 101 included in another configuration such as the tire lateral force estimation unit 70 illustrated in FIG. The configuration may be such that it is input to the estimation unit 102.
The rudder angular velocity estimation unit 102 estimates the rudder angular velocity. For example, the rudder angular velocity estimation unit 102 uses the following equation (33) to obtain the rudder angle detection value δ (k) obtained in the current processing step and the rudder angle detection value δ (k−1) obtained in the previous processing step. ) Is divided by a preset sample time T _s to calculate a steering angular velocity estimated value γ _δ . Then, the steering angular speed estimation unit 102 outputs the calculated steering angular speed estimation value γ _δ to the ratio calculation unit 105.

γ _δ = (δ (k) −δ (k−1)) / T _s (33)
Here, k indicates the current calculation time step.
Note that the tire ground contact length estimation device 60 may include a low-pass filter to which the estimated steering angular velocity value is input. That is, when the detected steering angle value includes noise, the tire contact length estimation device 60 performs filtering using a low-pass filter and outputs the result to the ratio calculation unit 105.

The self-aligning torque detector 103 detects the self-aligning torque by the same processing as the self-aligning change determination unit 62 shown in FIG. Then, the self-aligning torque detection unit 103 outputs the detected self-aligning torque to the self-aligning torque change rate estimation unit 104.
The self-aligning torque sudden change determination unit 100 does not include the self-aligning torque detection unit 103, and the output of the self-aligning torque detection unit 62 included in the tire contact length estimation device 60 shown in FIG. The configuration may be such that the torque change rate estimation unit 104 is input.

The self-aligning torque change rate estimation unit 104 estimates the self-aligning torque change rate based on the self-aligning torque.
For example, the self-aligning torque change rate estimation unit 104 uses the following equation (34) to calculate the self-aligning torque (M _z (k) obtained in the current processing step and the self-aligning obtained in the previous processing step. The difference from the torque (M _z (k−1)) is divided by a preset sample time T _s to calculate a self-aligning torque change rate estimated value γ _Mz . The calculated self-aligning torque change rate estimated value is output to the ratio calculation unit 105.

γ _Mz = (M _z (k) −M _z (k−1)) / T _s (34)
The ratio calculation unit 105 calculates the ratio between the estimated steering angular velocity value and the estimated self-aligning torque change rate. That is, the ratio calculation unit 105 calculates the ratio λ based on the steering angular velocity estimated value γ _δ and the self-aligning torque change rate estimated value γ _Mz using the following equation (35). Then, the ratio calculation unit 105 outputs the calculated ratio λ to the self-aligning torque change determination unit 106.

λ = γ _Mz / γ _δ (35)
The self-aligning torque change determination unit 106 outputs a calculation stop signal based on the ratio λ calculated by the ratio calculation unit 105.
Specifically, the self-aligning torque change determination unit 106 outputs a calculation stop signal when the absolute value of the ratio λ is larger than a preset threshold value. That is, when the absolute value of the ratio λ is larger than the threshold value, the self-aligning torque change determination unit 106 assumes that the change in the self-aligning torque is not a change in the rudder angle but a sudden change in the road surface friction coefficient. An operation stop signal is output so as not to estimate the tire contact length. That is, when the absolute value of the ratio λ is larger than the threshold value, the self-aligning torque change determination unit 106 outputs a calculation stop signal so as not to estimate the tire contact length, assuming that the road surface friction coefficient changes greatly.

The threshold value here is a value set so that the peak of the reference self-aligning torque can be appropriately obtained and the tire contact length can be obtained with high accuracy. For example, the threshold value is set by an experimental value, an empirical value, or a theoretical value. Is the value to be
FIG. 10 is a block diagram illustrating a configuration example of the noise level determination unit 110 illustrated in FIG.

As shown in FIG. 10, the noise level determination unit 110 includes a wheel load change ratio estimation unit 111, a contact length change ratio calculation unit 112, a self-aligning torque detection unit 113, a reference self-aligning torque calculation unit 114, and a high-frequency component detection. A unit 115, an integration unit 116, and a determination unit 117.
The wheel load change ratio estimation unit 111, the contact length change ratio calculation unit 112, the self-aligning torque detection unit 113, and the reference self-aligning torque calculation unit 114 perform the same processing as described above.

Note that the noise level determination unit 110 may use the output of the wheel load change ratio estimation unit 111 and the like of other configurations without including the wheel load change ratio estimation unit 111 and the like.
The high frequency component detector 115 extracts a high frequency component of the reference self-aligning torque. For example, as the high frequency component detection unit 115, a Butterworth filter designed so as to have a high-pass characteristic and a breakpoint frequency set appropriately is used. Accordingly, the high frequency component detection unit 115 extracts the high frequency component by filtering the reference self-aligning torque. Then, the high frequency component detector 115 outputs the extracted high frequency component of the reference self-aligning torque to the integrator 116.

The integrating unit 116 integrates the absolute value of the high frequency component in a certain interval. Specifically, the integration unit 116 calculates the integration value using the following equation (36). Then, the integration unit 116 outputs the calculated G _noise to the determination unit 117.

_{Here, G Ynoise} is the noise level, the output of the integrator 116. G _yhigh is high-frequency component. t ₀ is the current time. T is an integration interval and is a value set in advance.
The determination unit 117 determines that the noise level included in the reference self-aligning torque is high when the output of the integration unit 116 is larger than a preset threshold value, and outputs a calculation stop signal.

The threshold value here is a value set so that the peak of the reference self-aligning torque can be appropriately obtained and the tire contact length can be obtained with high accuracy. For example, the threshold value is set by an experimental value, an empirical value, or a theoretical value. Is the value to be
For example, if the output of the integration unit 116 calculated when the test drive is performed on the assumed road surface is set as a threshold value, the output of the integration unit 116 at the time of actual measurement is larger than the set threshold value. It can also be estimated that the road surface has a high noise level of the self-aligning torque. Therefore, if the determination unit 117 estimates that the road surface has a higher noise level of the self-aligning torque than the assumed road surface, the determination unit 117 determines that it is inappropriate for estimating the tire contact length, and outputs a calculation stop signal.

Alternatively, this threshold value may be set so that the estimation is not performed except when the noise signal ratio becomes high enough to find the peak of the reference self-aligning torque compared to the noise level.
FIG. 11 is a block diagram illustrating a configuration example of the turning state determination unit 120 illustrated in FIG.
As shown in FIG. 11, the turning state determination unit 120 includes a lateral acceleration detection unit 121, a low frequency component detection unit 122, and a lateral acceleration state determination unit 123.

The lateral acceleration detector 121 detects the lateral acceleration of the vehicle. For example, the lateral acceleration detection value may be a value detected by an existing skid prevention device. Then, the lateral acceleration detection unit 121 outputs the detected lateral acceleration to the low frequency component detection unit 122.
The low frequency component detection unit 122 extracts a low frequency component from the lateral acceleration. For example, a low-frequency component may be extracted by appropriately setting a corner frequency of a Butterworth filter having low-pass characteristics and filtering lateral acceleration. Then, the low frequency component detection unit 122 outputs the extracted low frequency component to the lateral acceleration state determination unit 123.

The lateral acceleration state determination unit 123 determines that the possibility that the reference self-aligning torque has a peak is extremely low when the absolute value of the lateral acceleration low frequency component from the low frequency component detection unit 122 is smaller than a preset threshold value. The operation stop signal is output.
The threshold value here is a value set so that the peak of the reference self-aligning torque can be appropriately obtained and the tire contact length can be obtained with high accuracy. For example, the threshold value is set by an experimental value, an empirical value, or a theoretical value. Is the value to be

(Operation etc.)
An example of the operation of the tire contact length estimation device 60 by the configuration and processing as described above will be described.
In the tire ground contact length estimation device 60, the wheel load change ratio estimation unit 111 estimates the wheel load change ratio and outputs the estimated wheel load change ratio to the reference self-aligning torque peak determination unit 80. Further, the self-aligning change determination unit 113 detects the self-aligning torque, and outputs the detected self-aligning torque to the reference self-aligning torque peak determination unit 80 and the tire contact length estimation unit 63. Further, the tire lateral force estimation unit 70 estimates the tire lateral force, and outputs the estimated tire lateral force to the tire contact length estimation unit 63.

Then, the reference self-aligning torque peak determination unit 80 determines the peak of the reference self-aligning torque based on the wheel load change ratio and the self-aligning torque, and outputs the determination result to the tire ground contact length estimation unit 63. .
Thereby, when the reference self-aligning torque is at a peak (including the vicinity of the peak), the tire contact length estimation unit 63 estimates the tire contact length based on the self-aligning torque and the tire lateral force.

Further, in the tire contact length estimation device 60, when the estimated stop condition determination unit 90 is included, the tire contact length estimation is stopped when the calculation stop signal is output from the estimated stop condition determination unit 90.
Here, in the present embodiment, the self-aligning torque detector 62 constitutes, for example, a self-aligning torque detector. The yaw rate detection unit 73 constitutes a yaw rate detection unit, for example. Further, the lateral acceleration detector 74 constitutes, for example, a lateral acceleration detector. Moreover, the acceleration sensor 6 comprises a longitudinal acceleration detection means, for example. Further, the tire lateral force estimating unit 70 constitutes, for example, tire lateral force estimating means. The wheel load change ratio estimation unit 61 constitutes, for example, a wheel load change ratio calculation unit. Further, the reference self-aligning torque peak determination unit 80 constitutes, for example, a reference self-aligning torque peak determination unit. Moreover, the tire contact length estimation part 63 comprises a tire contact length estimation means, for example. Moreover, the noise level determination part 110 comprises a noise level determination means, for example.

(Effect of 1st Embodiment)
The first embodiment has the following effects.
(1) The reference self-aligning torque peak determination unit 80 is based on the self-aligning torque detected by the self-aligning torque detection unit 62 and the wheel load change ratio estimated by the wheel load change ratio estimation unit 61. It is determined whether or not the torque is at a peak (including the vicinity of the peak). Then, when it is determined that the tire contact length estimation unit 63 is at the peak, the self-aligning torque detected by the self-aligning torque detection unit 62 is divided by the lateral force estimated by the tire lateral force estimation unit 70. The value is estimated as the tire contact length.

  Thereby, the tire ground contact length estimation device 60 estimates the tire ground contact length by dividing the self aligning torque by the lateral force of the tire when the reference self aligning torque reaches a peak. As a result, the tire contact length estimation device 60 can estimate the tire contact length based on the self-aligning torque and the tire lateral force that can be acquired by an existing vehicle-mounted sensor. That is, the tire ground contact length estimation device 60 can stably estimate the tire ground contact length using an existing in-vehicle sensor without providing an acceleration sensor on the wheel.

(2) In the reference self-aligning torque peak determination unit 80, the contact length change ratio calculation unit 81 uses a tire contact length change ratio model set in advance from the wheel load change ratio based on the wheel load change ratio. The tire contact length change ratio is calculated. Further, the reference self-aligning torque proportional amount calculation unit 82 calculates the reference self-aligning torque by dividing the self-aligning torque by the wheel load change ratio and the contact length change ratio.

  Here, the normalized self-aligning torque that is originally intended to detect the peak is defined by an amount obtained by dividing the self-aligning torque by the wheel load, the road surface friction coefficient, and the tire contact length. However, it is difficult to directly estimate the wheel load and the tire contact length estimated from the wheel load. On the other hand, in this embodiment, a reference self-aligning torque proportional amount obtained by dividing the self-aligning torque by the wheel load change ratio and the tire contact length change ratio is defined. At this time, the reference self-aligning torque proportional amount is a value proportional to the normalized self-aligning torque.

Accordingly, the reference self-aligning torque peak determination unit 80 determines the peak of the normalized self-aligning torque by determining the peak of the reference self-aligning torque proportional amount.
As a result, the tire contact length estimation device 60 can improve the estimation accuracy of the tire contact length.

(3) In the tire lateral force estimator 70, the left and right total lateral force estimator 77 estimates the total value of the tire lateral forces of the left and right wheels based on the vehicle yaw rate and the lateral acceleration of the vehicle from the vehicle motion characteristics. Further, the front wheel slip angle estimating unit 76 estimates the tire slip angle based on the steering angle and the vehicle speed from the vehicle motion characteristics. And the optimization calculating part 78 calculates the tire lateral force of each right and left wheel based on the estimated total value of the tire lateral force of the left and right wheels, the tire slip angle, and the wheel load change ratio.

As described above, the tire lateral force estimating unit 70 distributes the tire lateral force to both the left and right wheels and calculates the tire lateral force of each of the left and right wheels.
As a result, the tire contact length estimation device 60 can accurately estimate the tire contact length of each wheel of the vehicle.

(4) The reference self-aligning torque peak determination unit 80 acquires the reference self-aligning torque during the N sample time from the processing step at the current time. When the absolute value of the slope of the straight line fitted to the acquired reference self-aligning torque is smaller than a preset threshold value, the reference self-aligning torque peak determination unit 80 determines that the reference self-aligning torque has a peak (the vicinity of the peak). Is included).

Thereby, the reference self-aligning torque peak determination unit 80 can estimate the peak of the reference self-aligning torque calculated from the self-aligning torque with high accuracy even when the self-aligning torque includes noise. .
As a result, the tire contact length estimation device 60 can improve the estimation accuracy of the tire contact length estimated at the peak timing.

(5) The wheel load change ratio estimation unit 61 estimates the wheel load change ratio based on the lateral acceleration of the vehicle.
As a result, the tire ground contact length estimation device 60 can estimate the wheel load movement ratio during turning based on the lateral acceleration, and can accurately detect the peak of the reference self-aligning torque calculated from the wheel load movement ratio, for example. . Furthermore, the tire contact length estimation device 60 can improve the estimation accuracy of the tire contact length by dividing the self-aligning torque detection value by the tire lateral force.

(6) The wheel load change ratio estimation unit 61 estimates the wheel load change ratio based on the longitudinal acceleration of the vehicle.
As a result, the tire ground contact length estimation device 60 can estimate the wheel load movement ratio during turning based on the lateral acceleration, and can accurately detect the peak of the reference self-aligning torque calculated from the wheel load movement ratio, for example. . Furthermore, the tire contact length estimation device 60 can improve the estimation accuracy of the tire contact length by dividing the self-aligning torque detection value by the tire lateral force.

(7) The tire contact length estimation device 60 stops estimating the contact length of the tire when the self aligning torque sudden change determination unit 100 determines that the rate of change of the self aligning torque is greater than a preset threshold value. .
Thereby, the tire contact length estimation device 60 stops the estimation of the tire contact length when the ratio of the self-aligning torque change rate estimated value is larger than a preset threshold value, so that the road surface friction coefficient changes suddenly. The detection of the peak of the reference self-aligning torque is prevented.

As a result, the tire contact length estimation device 60 can detect the peak of the reference self-aligning torque with high accuracy, and the tire contact length estimation device 60 improves the estimation accuracy of the tire contact length estimated at the peak timing. be able to.
(8) If the tire ground contact length estimation device 60 determines that the noise level of the reference self-aligning torque is greater than a preset threshold by the noise level determination unit 110, the tire contact length estimation device 60 stops estimating the tire contact length.

As a result, the tire contact length estimation device 60 stops the estimation of the tire contact length when the noise intensity of the reference self-aligning torque estimated value is larger than a predetermined threshold value. The detection of the peak of the reference self-aligning torque is prevented.
As a result, the tire contact length estimation device 60 can detect the peak of the reference self-aligning torque with high accuracy, and the tire contact length estimation device 60 improves the estimation accuracy of the tire contact length estimated at the peak timing. be able to.

(9) If the tire contact length estimation device 60 determines that the absolute value of the lateral acceleration is smaller than a preset threshold value by the turning state determination unit 120, the tire contact length estimation device 60 stops estimating the tire contact length.
Accordingly, the tire contact length estimation device 60 may stop the estimation of the tire contact length when the absolute value of the lateral acceleration detection value is smaller than a preset threshold value, so that the reference self-aligning torque may have a peak. It is possible to prevent the tire contact length from being estimated in an extremely low scene.
As a result, the tire contact length estimation device 60 can be effectively utilized without wasting the calculation resources of the tire contact length, and the performance and cost of the microcomputer that performs the calculation can be reduced.

(Modification of this embodiment)
In the present embodiment, the estimated stop condition determination unit 90 is not limited to the configuration shown in FIG.
For example, FIG. 12 is a block diagram illustrating another configuration example of the estimated stop condition determination unit 90.
As shown in FIG. 12, the estimated stop condition determination unit 90 includes a braking detection unit 131 and a drive wheel detection unit 132.

When it is determined that the vehicle is being braked, the brake detection unit 131 outputs a calculation stop signal. Further, when the driving wheel detection unit 132 determines that the driving force is transmitted to the tire, the driving wheel detection unit 132 outputs a calculation stop signal.
With the configuration described above, the tire contact length estimation device 60 stops estimating the tire contact length when it is determined that the vehicle is being braked or when it is determined that the driving force is being transmitted to the tire.
In this case, the tire contact length estimation device 60 stops the estimation of the tire contact length during braking, thereby stopping the estimation of the tire contact length when the tire characteristics fluctuate due to the braking force. The estimation accuracy of can be improved.

Further, the tire contact length estimation device 60 is used when, for example, the self-aligning torque detection wheel is a driving wheel and the estimation of the tire contact length is stopped at the time of driving so that the tire characteristics fluctuate due to the driving force. The estimation of the tire contact length can be stopped. As a result, the tire contact length estimation device 60 can improve the estimation accuracy of the tire contact length.
Further, the present embodiment is not limited to the above-described electric vehicle, and may be a vehicle using an existing internal combustion engine as a power source as long as the vehicle can estimate or detect the self-aligning torque and the tire lateral force. good.

(Second Embodiment)
Next, a second embodiment will be described with reference to the drawings. The same components as those in the first embodiment will be described with the same reference numerals.

(Configuration etc.)
In the second embodiment, the tire ground contact length is estimated using the same principle with the same configuration as the first embodiment. In the second embodiment, the road surface friction coefficient is estimated based on the estimated tire contact length.

FIG. 13 is a block diagram illustrating a configuration example of the tire ground contact length estimation device 60 according to the second embodiment.
As shown in FIG. 13, the tire contact length estimation device 60 in the second embodiment includes a road surface friction coefficient estimation unit 141 in addition to the configuration of the tire contact length estimation device 60 in the first embodiment shown in FIG. is doing.

The road surface friction coefficient estimation unit 141 estimates the road surface friction coefficient as follows.
Assuming that the vehicle behavior is modeled using a two-wheel model and that the time variation of the road surface friction coefficient is piecewise constant, that is, the first-order derivative with respect to time of the road surface friction coefficient is zero, the vehicle of the following equation (37) The equation of motion holds.

Here, β + (l _f · γ / V) −δ represents the tire slip angle of the front wheel. β + (l _r · γ / V) −δ represents the tire slip angle of the rear wheel. Further, assuming that the total value of the self-aligning torque of the yaw rate and the left and right front wheels can be observed, the output y of the system is expressed by the following equation (38).

Here, j _f () is a self-aligning torque model. The actual total self-aligning torque of the left and right front wheels can be calculated based on, for example, the current of the assist motor provided in the electric power steering system and the steering torque detected by the steering torque sensor. When these mathematical expressions are expressed in the form of a general state equation, the following expression (39) is obtained.

Here, the state quantity is x = [β γ μ] ^T. The input is u = δ. When the state observer is configured based on this state equation, the following equation (40) can be obtained.

  Here, a value with “^” added to x (hereinafter referred to as “x ^”) and a value with “^” added to y represent a state quantity and an estimated output value, respectively. X ^ 'represents a differential value of the state quantity. In the state observer, the vehicle speed V that is a variable other than the state quantity can be calculated from the wheel speed, for example. The observer gain L is a value determined so that the observer has a stable pole.

FIG. 14 is a characteristic diagram showing an example of the characteristic of the self-aligning torque model j _f ().
As shown in FIG. 14, the characteristics of the self-aligning torque model j _f () in the equation (38) greatly change depending on the tire contact length l. For example, when a filar model is used as the self-aligning torque model, as shown in FIG. 14, the self-aligning torque changes in proportion to the tire contact length. Accordingly, in order to accurately estimate the road surface friction coefficient based on the self-aligning torque model, it is necessary to correctly set the tire contact length in the self-aligning torque model.

  Here, FIG. 15 is a characteristic diagram showing a road surface friction coefficient obtained by applying this embodiment and a road surface friction coefficient of a comparative example. Here, the road surface friction coefficient obtained by applying the present embodiment is a road surface friction coefficient estimated using a self-aligning torque model in which the tire contact length estimated by the present embodiment is set. Moreover, the road surface friction coefficient of the comparative example is a road surface friction coefficient estimated using a self-aligning torque model in which a tire contact length including an error of 20% is set.

As can be seen from FIG. 15, in this embodiment, the road surface friction coefficient estimation error can be greatly improved.
There are known techniques for estimating the road surface friction coefficient using a tire model such as a self-aligning torque model. For example, as a known technique, there is a technique as described in Japanese Patent No. 4213994.
In addition, the other configuration of the tire contact length estimation device 60 in the second embodiment is the same as the configuration of the tire contact length estimation device 60 in the first embodiment.

(Effect of 2nd Embodiment)
The second embodiment has the following effects in addition to the effects of the first embodiment.
The road surface friction coefficient estimating unit 141 estimates a road surface friction coefficient corresponding to the tire contact length by using a self-aligning torque (tire model) that can specify the self-aligning torque from the tire contact length and the road surface friction coefficient. Yes.

As a result, the road surface friction coefficient estimating unit 141 can estimate the road surface friction coefficient based on the self-aligning torque model (tire model) with high accuracy by using the tire contact length estimated with high accuracy.
As a result, the road surface friction coefficient estimator 141 can estimate the road surface friction coefficient with high accuracy regardless of the movement of the wheel load depending on the occupant loading position or the change in tire characteristics due to the decrease in tire air pressure.

(Third embodiment)
Next, a third embodiment will be described with reference to the drawings. The same components as those in the first embodiment will be described with the same reference numerals.

(Configuration etc.)
In the third embodiment, the tire ground contact length is estimated using the same principle with the same configuration as the first embodiment. In the third embodiment, the tire air pressure is estimated based on the estimated tire contact length.

FIG. 16 is a block diagram illustrating a configuration example of the tire ground contact length estimation device 60 according to the third embodiment.
As shown in FIG. 16, the tire contact length estimation apparatus 60 in the third embodiment includes a tire air pressure estimation unit 142 in addition to the configuration of the tire contact length estimation apparatus 60 in the first embodiment shown in FIG. ing.

The tire pressure estimation unit 142 estimates the tire pressure based on the relationship between the tire pressure and the tire contact length.
FIG. 17 shows an example of the relationship between tire pressure and tire contact length.
As shown in FIG. 17, the tire air pressure has a characteristic that it becomes smaller as the tire contact length becomes larger. The tire air pressure monotonously decreases as the tire contact length increases.

The tire pressure estimation unit 142 estimates the tire pressure from the tire contact length based on such a relationship. For example, such a relationship is acquired by experiment to create a map, and the tire air pressure estimation unit 142 estimates the tire air pressure from the tire contact length with reference to the created map.
In addition, the other structure in the tire contact length estimation apparatus 60 in 3rd Embodiment is the same as that of the tire contact length estimation apparatus 60 in the said 1st Embodiment.

(Effect of the third embodiment)
The third embodiment has the following effects in addition to the effects of the first embodiment.
The tire air pressure estimation unit 142 estimates the tire air pressure based on the tire ground contact length.
Thereby, the tire pressure estimation unit 142 can estimate the tire pressure without directly providing the tire pressure sensor on the tire. For example, as a technique for estimating a tire pressure by directly providing a tire pressure sensor on a tire, there is a technique described in Japanese Patent Application Laid-Open No. 2010-254018.

Thus, according to the present embodiment, the cost of the tire air pressure sensor and the labor associated with the installation become unnecessary, so that a reduction in the vehicle cost can be expected.
Further, the tire pressure estimating unit 142 can estimate the tire pressure with high accuracy by using the tire contact length estimated with high accuracy.

(Fourth embodiment)
Next, a fourth embodiment will be described with reference to the drawings. The same components as those in the first embodiment will be described with the same reference numerals.

(Configuration etc.)
In the fourth embodiment, the tire ground contact length is estimated using the same principle with the same configuration as that of the first embodiment. In the fourth embodiment, the tire rolling resistance is estimated based on the estimated tire contact length.

FIG. 18 is a block diagram illustrating a configuration example of the tire ground contact length estimation device 60 according to the fourth embodiment.
As shown in FIG. 18, the tire contact length estimation device 60 in the fourth embodiment includes a tire rolling resistance estimation unit 151, an inter-vehicle distance in addition to the configuration of the tire contact length estimation device 60 in the first embodiment shown in FIG. 3. A distance detection unit 152, an acceleration / deceleration command value generation unit 153, a feedforward control unit 154, a feedback control unit 155, and a sum calculation unit 156 are provided.

The tire rolling resistance estimation unit 151 estimates the tire rolling resistance from the tire contact length.
FIG. 19 is a characteristic diagram showing an example of the relationship between tire rolling resistance and tire ground contact length.
As shown in FIG. 19, the tire rolling resistance has a characteristic of increasing as the tire contact length increases. The tire rolling resistance increases monotonously as the tire contact length increases.
The tire rolling resistance estimation unit 151 estimates the tire rolling resistance from the tire contact length based on such a relationship. For example, such a relationship is acquired by experiment to create a map, and the tire rolling resistance estimation unit 151 estimates the tire rolling resistance from the tire contact length with reference to the created map. Then, the tire rolling resistance estimation unit 151 outputs the estimated tire rolling resistance to the feedforward control unit 154.

  The inter-vehicle distance detector 152 detects the inter-vehicle distance. For example, the inter-vehicle distance detection unit 152 detects the distance between the preceding vehicle and the host vehicle with reference to a signal from a laser range finder provided in front of the vehicle. Then, the inter-vehicle distance detection unit 152 outputs the detected inter-vehicle distance to the acceleration / deceleration command value generation unit 153, the feedforward control unit 154, and the feedback control unit 155.

The acceleration / deceleration command value generation unit 153 is based on the deviation between the detected inter-vehicle distance and the target inter-vehicle distance detected by the inter-vehicle distance detection unit 152 and the opening degree of the accelerator pedal or the brake pedal operated by the driver. Is generated. Then, the acceleration / deceleration command value generation unit 153 outputs the generated acceleration / deceleration command value to the feedforward control unit 154.
The feedforward control unit 154 calculates a feedforward control signal based on the acceleration / deceleration command value, the inter-vehicle distance, and the tire rolling resistance torque.

Here, when a nominal dynamic model from the braking / driving torque T (s) to the vehicle longitudinal acceleration G _x (s) is D (s), the following equation (41) is derived.
G _x (s) = D (s) T (s) (41)
Thereby, the feedforward control part 154 is comprised based on the inverse model D ^<-1> (s) of a dynamic model like following (42) Formula. U _ff (s) in the following equation (42) is a feedforward control signal, and the feedforward control unit 154 outputs the calculated feedforward control signal to the sum calculation unit 156.

Here, R _es (s) (R is a value with “^” added thereto) is a tire rolling resistance torque estimated value. G _xcom (s) is an acceleration / deceleration command value. F _LPF (s) is a low-pass filter for making the system proper. In the second term, the tire rolling resistance torque estimated in advance is compensated, and thereby the responsiveness and accuracy of the vehicle longitudinal acceleration generated as a result of the control can be improved.

  The feedback control unit 155 calculates a feedback control signal based on the inter-vehicle distance. For example, the feedback control unit 155 calculates the feedback control signal using an integral controller so as to reduce the deviation between the detected inter-vehicle distance and the target inter-vehicle distance. Then, the feedback control unit 155 outputs the calculated feedback control signal to the sum calculation unit 156.

The sum calculator 156 calculates a braking / driving torque command value for controlling the braking / driving torque from the sum of the feedforward control signal and the feedback control signal.
In addition, the other structure in the tire contact length estimation apparatus 60 in 4th Embodiment is the same as that of the tire contact length estimation apparatus 60 in the said 1st Embodiment.
(Effect of 4th Embodiment)
The fourth embodiment has the following effects in addition to the effects of the first embodiment.

The tire rolling resistance estimation unit 151 estimates the tire rolling resistance based on the tire contact length.
Accordingly, the tire rolling resistance estimation unit 151 can estimate the tire rolling resistance with high accuracy by using the tire contact length estimated with high accuracy.
As a result, the driving torque can be corrected based on the estimated tire rolling resistance value with high accuracy as in the present embodiment. Further, according to the present embodiment, by using such tire rolling resistance estimation value, for example, the cruising distance of the electric vehicle can be estimated with high accuracy, and the driver can make a more reliable travel plan. become. In addition, according to the present embodiment, in the control for causing the host vehicle to follow the preceding vehicle while maintaining a constant inter-vehicle distance, the driving torque is corrected according to the estimated tire rolling resistance value, thereby achieving high responsiveness and accuracy. A target longitudinal acceleration can be generated.

(Fifth embodiment)
Next, a fifth embodiment will be described with reference to the drawings. The same components as those in the first embodiment will be described with the same reference numerals.

(Configuration etc.)
In the fifth embodiment, the tire contact length is estimated using the same principle with the same configuration as that of the first embodiment. In the fifth embodiment, the tire cornering power is estimated based on the estimated tire contact length.
FIG. 20 is a block diagram illustrating a configuration example of the tire ground contact length estimation device 60 according to the fifth embodiment.

As shown in FIG. 20, the tire contact length estimation apparatus 60 in the fifth embodiment includes a tire cornering power estimation unit 161 in addition to the configuration of the tire contact length estimation apparatus 60 in the first embodiment shown in FIG. is doing.
The tire cornering power estimation unit 161 estimates the tire cornering power from the tire contact length.
The tire cornering power estimation unit 161 estimates tire cornering power based on the following principle.

  According to the brush tire model ("Tire and Vehicle dynamics", Butterworth Heinemann, Hans Pacejka, p99, etc.), the tire slip angle at which the tire lateral force is saturated is inversely proportional to the square of the tire contact length. Therefore, the tire cornering power, which is the initial inclination of the tire lateral force with respect to the tire slip angle, is proportional to the square of the tire contact length. The following equation (43) shows such a relationship.

C _p = (l / l ₀ ) ² · C _p0 (43)
Here, l ₀ is a reference tire contact length. C _p0 is the tire cornering power when the tire contact length is the reference value ₁₀ .
As shown in the equation (43), the tire cornering power C _p increases as the tire ground contact length l increases because the ground contact area in the brush model increases and the tire lateral force acting on the tire microelements increases. This is because the area to be integrated increases.

In addition, 5th Embodiment is not limited to using such a brush tire model, Refer to the map created by acquiring the relationship between tire contact length and tire cornering power experimentally. Thus, the tire cornering power may be estimated.
In addition, the other structure in the tire contact length estimation apparatus 60 in 5th Embodiment is the same as that of the tire contact length estimation apparatus 60 in the said 1st Embodiment.

(Effect of 5th Embodiment)
The fifth embodiment has the following effects in addition to the effects of the first embodiment.
The tire cornering power estimation unit 161 estimates the tire cornering power based on the tire contact length.

Accordingly, the tire cornering power estimation unit 161 can estimate the tire cornering power with high accuracy by using the tire contact length estimated with high accuracy.
As a result, in the present embodiment, by using the tire cornering power, for example, the skid prevention control can be more smoothly performed. That is, according to this embodiment, even when the tire ground contact length varies due to factors such as a change in the occupant loading position and a decrease in tire air pressure, the vehicle model in the skid prevention control can be corrected to the dynamics of the actual vehicle motion. As a result, in the present embodiment, fluctuations in control design values such as the amount of overshoot and settling time can be suppressed regardless of these factors.

As a modification of the present embodiment, the tire contact length estimation device 60 includes a road surface friction coefficient estimation unit 141, a tire air pressure estimation unit 142, a tire rolling resistance estimation unit 151, and a tire described in the second to fifth embodiments. A combination of at least two of the cornering power estimation units 161 may be provided.
In this case, when the tire contact length estimation device 60 includes at least the tire rolling resistance estimation unit 151, the inter-vehicle distance detection unit 152, the acceleration / deceleration command value generation unit 153, and the feedforward control unit 154 are provided as in the fourth embodiment. Further, a feedback control unit 155 and a sum calculation unit 156 may be provided.

  60 tire contact length estimation device, 61 wheel load change ratio estimation unit, 62 self-aligning torque detection unit, 63 tire contact length estimation unit, 70 tire lateral force estimation unit, 73 yaw rate detection unit, 74 lateral acceleration detection unit, 80 reference Self-aligning torque peak determination unit

Claims (15)

Self-aligning torque detecting means for detecting the self-aligning torque of the tire;
Yaw rate detection means for detecting the yaw rate of the vehicle;
Lateral acceleration detecting means for detecting the lateral acceleration of the vehicle;
Tire lateral force estimation means for estimating the lateral force of the tire based on the yaw rate detected by the yaw rate detection means and the lateral acceleration detected by the lateral acceleration detection means;
A wheel load change ratio calculating means for calculating a wheel load change ratio that is a ratio of a wheel load change amount based on the lateral acceleration detected by the lateral acceleration detecting means with respect to the static wheel load;
A reference self-aligning torque calculated as a value proportional to the self-aligning torque based on the self-aligning torque detected by the self-aligning torque detecting means and the wheel load change ratio calculating means calculated by the wheel load change ratio calculating means. A reference self-aligning torque peak determination means for determining whether or not is a peak;
When the reference self-aligning torque peak determining means determines that the reference self-aligning torque is at a peak, the lateral force estimated by the tire lateral force estimating means is the self-aligning torque detected by the self-aligning torque detecting means. Tire contact length estimation means for estimating the tire contact length from the value divided by the force;
A tire ground contact length estimation device comprising: The reference self-aligning torque peak determining means is
Based on the wheel load change ratio calculated by the wheel load change ratio calculation means using a tire contact length change ratio model set in advance from the wheel load change ratio, the tire contact length relative to the tire contact length is calculated. A contact length change ratio estimating means for estimating a change ratio of the contact length of the tire, which is a ratio of
A value obtained by dividing the self-aligning torque detected by the self-aligning torque detecting means by the wheel load change ratio calculated by the wheel load change ratio calculating means and the contact length change ratio estimated by the contact length change ratio estimating means. A reference self-aligning torque estimating means for estimating a reference self-aligning torque from:
The tire contact length estimation device according to claim 1, comprising: The tire lateral force estimating means includes
Based on the yaw rate detected by the yaw rate detection means and the lateral acceleration detected by the lateral acceleration detection means, a tire lateral force total value estimation means for estimating a total value of tire lateral forces of left and right wheels from vehicle motion characteristics;
Tire slip angle estimating means for estimating a tire slip angle based on a steering angle and a vehicle speed from vehicle motion characteristics;
The total tire lateral force value of the left and right wheels estimated by the tire lateral force total value estimating means, the tire slip angle estimated by the tire slip angle estimating means, and the wheel load change calculated by the wheel load change ratio calculating means Left and right wheel tire lateral force estimating means for estimating the left wheel tire lateral force and the right wheel tire lateral force, respectively, based on the ratio;
The tire ground contact length estimation device according to claim 1 or 2, characterized by comprising:   The reference self-aligning torque peak determining means linearly approximates the history of the reference self-aligning torque in a preset section, and when the absolute value of the slope of the straight line approximation is smaller than a preset threshold, The tire contact length estimation apparatus according to any one of claims 1 to 3, wherein the alignment torque is determined to be a peak.   The tire ground contact length estimation device according to claim 3 or 4, wherein the wheel load change ratio calculation means calculates a wheel load change ratio based on a lateral acceleration of the vehicle. The longitudinal load detecting means for detecting longitudinal acceleration of the vehicle is further provided. The wheel load change ratio calculating means calculates the wheel load change ratio based on the detected longitudinal acceleration of the vehicle. The tire ground contact length estimation device described.   The tire contact length estimation device according to any one of claims 1 to 6, wherein the tire contact length estimation unit stops estimating the contact length of the tire when it is determined that the vehicle is braking. .   The tire contact length estimation means stops the estimation of the contact length of the tire when it is determined that the driving force is transmitted to the tire. Tire contact length estimation device.   The tire contact length estimation means stops estimating the contact length of a tire when the rate of change of the self-aligning torque detected by the self-aligning torque detection means is greater than a preset threshold value. The tire ground contact length estimation device according to any one of claims 1 to 8. Noise level determination means for determining a noise level of the reference self-aligning torque;
The tire contact length estimation means determines that the noise level is high when the noise level is larger than a preset threshold value, and stops estimating the contact length of the tire. The tire ground contact length estimation apparatus described in any one of the above.   11. The tire contact length estimation means stops the tire contact length estimation when the absolute value of the lateral acceleration is smaller than a preset threshold value. 11. The tire ground contact length estimation device described in 1.   Road surface friction coefficient estimation that estimates the road surface friction coefficient corresponding to the tire contact length estimated by the tire contact length estimation means by using a tire model capable of specifying self-aligning torque from the tire contact length and the road surface friction coefficient The tire ground contact length estimation device according to any one of claims 1 to 11, further comprising means.   The tire ground contact according to any one of claims 1 to 12, further comprising an air pressure estimating means for estimating an air pressure of the tire based on a tire ground contact length calculated by the tire ground contact length estimating means. Long estimation device.   The tire rolling resistance estimation means for estimating the rolling resistance of the tire based on the tire contact length calculated by the tire contact length estimation means is provided. Tire contact length estimation device.   The tire cornering power estimation means for estimating the tire cornering power based on the tire contact length calculated by the tire contact length estimation means is provided. Tire contact length estimation device.

Images