JP3331939B2 - Vehicle state quantity estimation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle body which receives a wheel load, a lateral acceleration of a vehicle body, a longitudinal acceleration, a roll angle, a pitch angle, a yaw rate, a vehicle side slip angle, a cant road inclination angle, and a cross wind. The present invention relates to a vehicle state quantity estimating device that calculates a vehicle state quantity such as a force applied thereto.

[0002]

2. Description of the Related Art Conventionally, vehicle operation changes (roll, nose dive, squat) due to driving operations such as handling, brake braking, acceleration, etc., and traveling on a canted road are reduced, and an optimal alignment state is maintained. There has been proposed a vehicle stabilization technique for improving the steering stability of a vehicle and preventing the vehicle from spinning.

[0003] In such vehicle stabilization technology, wheel load,
Lateral acceleration, longitudinal acceleration, roll angle,
The vehicle detects state variables such as a pitch angle, a yaw rate, a side slip angle of a vehicle body, an inclination angle of a cant road, and a force applied to the vehicle body by a cross wind, and performs control for stabilizing the vehicle based on the vehicle state amount. .

[0004] For example, in the anti-roll control of the attitude control technology, when turning, the acceleration / deceleration acting in the lateral direction of the vehicle is detected by a G sensor (acceleration sensor) to determine the control amount. At the time of turning, the centroid moves due to centrifugal force, the vehicle height on the outer wheel side falls below the target vehicle height value, and the inner wheel side rises.In anti-roll control, the cylinder pressure of the suspension on the outer wheel side is increased and the inner wheel side By lowering the cylinder pressure, the force to roll the vehicle body due to the centrifugal force is canceled, and the vehicle posture in the lateral direction is made almost flat. In order to compensate for a control response delay at the beginning of turning, roll prediction is performed by predicting roll start from the relationship between the steering angle and the vehicle speed.

In the anti-dive and anti-squat controls, the longitudinal acceleration / deceleration acting on the vehicle is detected by a G sensor at the time of deceleration or acceleration, and the cylinder pressure of the front and rear suspensions is increased or decreased according to the detected value. By lowering, the vehicle posture in the front-rear direction is made almost flat.

As described above, in the conventional attitude control technique, the load movement is estimated based on the lateral acceleration / deceleration detected by the G sensor or the longitudinal acceleration / deceleration, and the control amount of the suspension is determined. When the vehicle is turning, it is necessary to control the braking / driving force acting on each wheel so that the spin does not occur beyond the cornering characteristics of the tire.

[0007] The yaw rate and the vehicle body acceleration in the lateral direction used in the vehicle stabilization control can be estimated as follows in a non-braking state.

[0008] First of all, the wheel speed _v wFR of the wheel speed _v wFL and the right wheel of the left wheel at the time of turning, represented by the following equation. In the following equation, the left turn is defined as a positive direction, and “・” is defined as a derivative with respect to time.

[0009]

(Equation 1) Here, T _rd is the tread interval between the left and right wheels, V is the vehicle speed (vehicle speed), and

[0010]

(Equation 2) Is the rotation speed around the center of gravity of the vehicle, that is, the yaw rate.

From equations (1) and (2), the yaw rate can be estimated by the following equation. In the following, 以下 is added to the physical quantity representing the estimated value.

[0012]

(Equation 3) In the steady circular turning state, the lateral acceleration G can be estimated by the following equation.

[0013]

(Equation 4)

[0014]

However, the above prior art has the following problems.

That is, when performing vehicle stabilization control such as vehicle attitude control or anti-spin control, it is important to estimate the wheel load applied to each wheel. It depends on the load distribution of the occupants and luggage, the state of the suspension, and the like. Therefore, G
Indirect estimation using acceleration detected by a sensor or the like makes the estimation calculation too complicated, and the estimated value may be inaccurate, so that good stabilization control cannot be easily performed. There is. Furthermore, when the road surface condition is different, the friction coefficient μ between the tire and the road surface is different, so that it is necessary to change the vehicle attitude control.

In the method of estimating the yaw rate and the lateral acceleration as in the above example, the slip speed Δv of each wheel due to the braking influences the speed change of the left and right wheels due to the turning of the vehicle during braking. Unless otherwise, the above equation has a problem that the yaw rate and the lateral acceleration cannot be accurately estimated. Such difficulty in estimation at the time of braking also occurs in other vehicle state quantities.

The present invention has been made in view of the above facts, and includes a wheel load of each wheel, a lateral acceleration of a vehicle body, a longitudinal acceleration, a roll angle, a pitch angle, a yaw rate, a vehicle body slip angle, and a cant road. The vehicle state variables required for vehicle stabilization control, such as the inclination angle of the vehicle and the force applied to the vehicle body due to cross wind, are determined by whether braking or non-braking and external conditions such as road surface conditions and load distribution. It is another object of the present invention to provide a vehicle state quantity estimating apparatus that can easily and accurately estimate.

[0018]

In order to achieve the above object, a first aspect of the present invention comprises a wheel speed detecting means for detecting a wheel speed, a gradient of braking / driving force with respect to a slip speed, or a physical quantity related to the gradient. Road μ gradient calculating means for calculating the road μ gradient, braking / driving force estimating means for estimating the braking / driving force acting on wheels, the wheel speed detected by the wheel speed detecting means, the road μ gradient calculating means Road surface calculating means for calculating a road surface state and a slip speed based on the road μ gradient calculated by the above and the braking / driving force estimated by the braking / driving force estimating means, and a relationship between the slip speed and the road μ gradient in the reference characteristic. Means for storing road conditions for each road surface state, and a road surface corresponding to the road surface state and the slip speed calculated by the road surface calculation means based on the relationship stored by the storage means. processing means for calculating the μ gradient, and wheel load calculating means for calculating the load on the wheel based on the road μ gradient calculated by the processing means and the road μ gradient calculated by the road μ gradient calculating means, It is configured to include.

Here, the reference characteristic is, as shown in FIG. 3, a braking / driving force F with respect to a slip speed (vehicle speed-wheel speed) measured in a specific vehicle behavior state.
(Either braking force or driving force), preferably in a quasi-static reference state where there is almost no change in wheel load (hereinafter, this example is treated), for each road surface state (for example, "dry" The change characteristics measured for the “road surface”, “snow road surface”, and “ice road surface” can be used. The braking / driving force F acting on the wheel as a reaction force from the road surface is represented by the following equation, where μ is the friction coefficient between the tire and the road surface, and W is the load (wheel load) applied to the wheel. You.

F = μW (5) In the reference characteristic shown in FIG. 3, the gradient α ₀ (road surface μ gradient) of the braking / driving force F with respect to the slip speed ΔV _i is uniquely determined according to the road surface condition. Therefore, in the storage means, the slip speed ΔV
Stored with a relationship between the gradient alpha _0, for example, in a table format for each road surface condition. That is, if at least the road surface condition and the slip speed are known, the road surface μ gradient α ₀ at this time can be obtained based on this relationship.

In the present invention, first, the wheel speed detecting means detects the wheel speed, and the road surface μ gradient calculating means calculates the road surface μ gradient which is the gradient of the braking / driving force with respect to the slip speed.
As will be described later, the road surface μ gradient can be obtained from the wheel speed reflecting the dynamic characteristics of the wheel motion. And
The braking / driving force estimating means estimates the braking / driving force acting on the wheels. For example, it can be approximately estimated from the wheel cylinder pressure during brake braking and from the accelerator opening during driving.

Then, the road surface calculating means calculates the wheel speed based on the wheel speed detected by the wheel speed detecting means, the road μ gradient calculated by the road surface μ gradient calculating means, and the braking / driving force estimated by the braking / driving force estimating means. , Road surface condition and slip speed are calculated. The operation principle of the road surface operation means will be described later.

Next, the processing means calculates a road surface μ corresponding to the road surface state and the slip speed calculated by the road surface calculation means based on the relationship between the slip speed and the road surface μ gradient for each road surface state stored by the storage means. Find the gradient. For example,
As the relationship, the storage means stores a table having, as data, a road surface μ gradient measured for each road surface state and each slip speed, and the processing means stores data corresponding to the calculated road surface state and slip speed in the table. To find the road surface μ gradient.

Here, when a vehicle behavior change such as a roll of the vehicle or running on a cant road occurs, a change in wheel load occurs due to the load movement. For example, when decelerating in the front-back direction,
A load shift occurs from the rear wheel to the front wheel.
Even at the same μ value, the braking / driving force increases for the front wheels and decreases for the rear wheels. When the wheel load increases, the braking / driving force F changes with respect to the slip speed as shown by a change characteristic 99 in FIG.

As shown in the ΔV-F characteristic (reference characteristic, change characteristic 99) of FIG.
It can be seen that up to the state immediately before (maximum braking / driving force), the change characteristic of the braking / driving force F = μW with respect to the slip speed has a substantially proportional relationship. Therefore, in this slip speed range, at a certain slip speed ΔV _i , the μ gradient α in the change characteristic 99 when the load moves is the μ gradient α ₀ in the reference characteristic.
Rather, it changes in proportion to the rate of change of the wheel load. That is, at a certain slip speed ΔV _i , assuming that the braking / driving force of the reference characteristic is W _f0 , and the braking / driving force of the change characteristic 99 during load movement is W _f ,

[0026]

(Equation 5) Holds. Of course, the equation (6) holds both when the wheel load increases and when the wheel load decreases.

When the vehicle stabilization control is actually performed, the peak
Tire lock and tire idling do not occur without exceeding μ
Equation (6) always holds because the braking / driving force is controlled as follows.
You can think of it. Therefore, the wheel load calculating means
Road surface gradient α obtained by the processing means₀And the road surface
based on the road surface μ gradient α calculated by the μ gradient calculation means.
Thus, for example, using equation (6), the braking / driving force W of the reference characteristic_f0
From the load W of the wheel _fCan be calculated.

The road surface μ gradient calculating means of the present invention
As a physical quantity relating to the gradient of the longitudinal force with respect to the slip speed, slip rate - may be calculated gradient G _d of the longitudinal force with respect to ((vehicle body speed wheel speed) / (vehicle speed)).

For example, the gradient of the braking / driving force with respect to the slip rate is converted into an amount equivalent to the gradient of the braking / driving force with respect to the slip speed according to the vehicle speed (or wheel speed).
Alternatively, the road surface calculating means calculates a slip rate and a vehicle speed instead of the slip speed, and the storage means stores the relationship between the slip rate and the gradient of the braking / driving force with respect to the slip rate for each road surface condition and each vehicle speed. To memorize. Then, a gradient G corresponding to the vehicle speed V, the road surface condition, and the slip ratio is processed by the processing means.
_d0 is calculated, and the wheel load calculating means may calculate the wheel load according to the following equation.

[0030]

(Equation 6) (Invention of claim 2) The invention of claim 2 is based on claim 1.
Based on the wheel load calculated by the wheel load calculating means, the vehicle body acceleration in the front-rear direction, the vehicle body acceleration in the lateral direction,
The vehicle further includes a vehicle state quantity calculating means for calculating at least one of the pitch angle, the roll angle, the inclination angle of the cant road, and the magnitude of the force received by the vehicle body due to the cross wind.

Hereinafter, a method for estimating the vehicle state quantity calculated by the vehicle state quantity calculation means will be individually described.

(1) Front-rear body acceleration (front-rear G) FIG. 9 shows a half-car model viewed from the side of the vehicle, showing the relationship between front-rear G and wheel load during braking / driving. I have. Each physical quantity in the model is as follows.

G: gravitational acceleration m: vehicle weight in units of 1 · g K _F : spring constant of suspension (F indicates front wheel) G _X : front and rear G h: height of center of gravity from road surface L: front wheel and rear horizontal distance lf between the wheel: horizontal distance lr between the front wheels and the center of gravity: horizontal distance between the rear wheels and the center of gravity phi: pitch angle W _F: 1 · g a wheel load of one unit (F is When the vehicle body decelerates (accelerates) during braking (driving) and a load shifts between the front and rear wheels, the following equation is obtained from the balance of the moment of force around the center of gravity in the half car model in FIG. Holds.

[0034] _{m · G X · h-W} F · g · L + m · g · lr = 0 (8) where the front wheels of the load W _F is the left front wheel load calculated by the wheel load computing unit W _fFL , The load on the right front wheel to W
If the _fFR, those obtained by _{W F = W fFL + W fFR} (9).

From this, the front and rear G (G _X ) can be calculated by the following equation (10) obtained by modifying the equation (8).

[0036]

(Equation 7) Note that (10) can be or represented using the wheel load W _R of the rear wheels calculated by the wheel load computing means, be represented by using the W _F and W _R.

(2) Estimation method of pitch angle φ At the time of braking (during driving), the pitch angle φ in FIG. 9 is calculated from the following equation based on the balance between the increase in the wheel load due to the load movement and the suspension spring against the load increase. Can be calculated.

Φ = tan ⁻¹ ((W _F −W _F0 ) / K _F / lf) (11) where K _F is the total value of the left and right front wheel suspension spring constants, W _F0
Is the total value of the static load of the left and right front wheels. Note that φ in Expression (11) can also be expressed using the total value of the suspension spring constant and the total value of the static load for the left and right rear wheels.

(3) Method of Estimating Lateral Body Acceleration (Lateral G) FIG. 10 shows a half car model viewed from the front and rear direction of the vehicle to show the relationship between the lateral G and the wheel load during turning. Have been. Each physical quantity in the model is as follows, and the same physical quantities as those in FIG. 9 are represented by the same reference numerals, and description thereof will be omitted.

[0040] T _rd: Tread K _L: spring constant (L is left wheel related) G _y of the suspension: lateral G tau: roll angle W _L: 1 · a (the L about left wheel) wheel load as one unit g lateral G Is preferably performed when it is determined that the vehicle is turning. Whether or not the vehicle is turning can be confirmed using, for example, the following equation.

V _FR ＶV _FL (12) where V _FR = V _WFR + Δv _FR (13) V _FL = V _WFL + Δv _FL (14) V _{WFR, WFL} : Wheel speeds of the right front wheel and the left front wheel Δv _{FR, FL} : It is the slip speed of the right front wheel and the left front wheel. If the formula (12) is satisfied, it is determined that the vehicle is turning,
The lateral G is estimated.

It is also possible to determine whether the vehicle is turning while V _RR ＶV _RL (15). Here, V _RR = V _WRR + Δv _RR (16) V _RL = V _WRL + Δv _RL (17) V _{WRR, WRL} : right rear wheel, left rear wheel speed Δv _{RR, RL} : right rear wheel, left rear wheel It is the slip speed.

When a load shift occurs between the left and right wheels during a turn, the following equation is established from the balance of the moment of the force around the center of gravity in the half car model of FIG.

[0044] _{m · G y · h-W} L · g · T rd + m · g · T rd / 2 = 0 (18) where the load W _L of the left wheel is a left front wheel, which is computed by the wheel load computing means _Is the load of W _fFL and the load of the left rear wheel is W
If the _fRL, those obtained by _{W L = W fFL + W fRL} (19).

From this, the horizontal G (G _y ) can be calculated by the following equation (20) obtained by modifying the equation (18).

[0046]

(Equation 8) Note that (20) can be or represented using the wheel load W _R of the right wheel calculated by the wheel load computing means, also be represented using the W _L and W _R.

(4) Method of Estimating Roll Angle τ At the time of turning, the roll angle τ in FIG. 10 can be calculated from the following equation based on the balance between the increment of the wheel load due to the load movement and the suspension spring against the increment of the load. it can.

Τ = tan ⁻¹ ((W _L −W _L0 ) / K _L / T _rd / 2) (21) where K _L is the total value of the suspension spring constants of the front and rear left wheels, W _L0
Is the total static load of the front and rear left wheels. Note that τ in the equation (21) can also be expressed using the total value of the suspension spring constant and the total value of the static load for the front and rear right wheels.

(5) Method of Estimating Cant Road Inclination Angle η FIG. 11 shows a half-car model viewed from the front and rear direction of the vehicle to show the relationship between the wheel load and the cant road inclination angle η during traveling on the cant road. It is shown. Each physical quantity in the model is the same as in FIG. 10, and is represented by the same reference numeral.

First, it is confirmed whether or not the vehicle is traveling straight, using the following equation. V _FR = V _FL (22) When the equation (22) holds, it can be determined that the vehicle is traveling straight because the same vehicle speed is used between the left and right wheels. Furthermore, when it is determined that there is no cross wind, it can be determined that the load movement between the left and right wheels is due to running on a canted road.

Here, the inclination angle η of the cant road is estimated by the following equation using the estimation equation of the lateral G of the equation (20).

[0052]

(Equation 9) Note that cos η ≒ 1.

[0053] (6) estimation method vehicle force F _w that received by crosswind even during straight and even if not in cant road, when the vehicle body is subjected to a lateral force by crosswind, between the left and right wheels Load transfer occurs. In half car model of Figure 11, in consideration of the balance about the center of gravity of the force of moment when the eta = 0, the same principle, the force F _w that received by crosswind, can be estimated by the following equation.

[0054]

(Equation 10) Note that (22) is determined to be traveling straight satisfied, and when it is determined not to be in cant road, estimates the crosswind force F _w by (24). (Invention of claim 3) The invention of claim 3 comprises a wheel speed detecting means for detecting a wheel speed, a road μ gradient calculating means for calculating a road μ gradient which is a gradient of braking / driving force with respect to a slip speed, Calculating the slip speed based on the road / slope force calculated by the road / slope calculating means and the braking / driving force estimated by the braking / driving force estimating means. Slip speed calculating means, and the wheel speed of at least one of the front wheel and the rear wheel detected by the wheel speed detecting means and the slip speed of at least one of the front wheel and the rear wheel calculated by the slip speed calculating means. And a yaw rate calculating means for calculating a yaw rate based on the yaw rate.

According to a third aspect of the present invention, the slip speed calculating means calculates the road surface μ gradient and the estimated braking / driving force based on the same estimation method as the road surface calculating means of the first aspect, as described later. Calculate the slip speed.

The yaw rate calculating means calculates the wheel speed of at least one of the front wheel and the rear wheel detected by the wheel speed detecting means and the slip speed of at least one of the front wheel and the rear wheel calculated by the slip speed calculating means. The yaw rate is calculated based on the speed. For example, the yaw rate is calculated by the following equation.

[0057]

[Equation 11] Here, V _FL and V _FR are the vehicle speeds on the left front wheel side and the right front wheel side of the equations (13) and (14) in consideration of the decrease in the slip speed from the wheel speed during braking / driving, T _rd is shown in FIG.
Tread interval. Also, the yaw rate of equation (25) is
Wheel speeds V _RL , V _RR of the left rear wheel and the right rear wheel in equations (16) and (17)
, _Or by using all of V _FL , V _FR , V _RL , and V _RR .

As described above, in the present invention, the yaw rate is calculated in consideration of the change in the wheel speed accompanying the change in the slip speed caused by controlling the braking / driving force for each wheel. The yaw rate during braking / driving can be calculated more accurately than when calculating only from the wheel speeds that do not take into account the slip speed.

According to a fourth aspect of the present invention, in the third aspect of the present invention, there is further provided a side slip angle calculating means for calculating a body side slip angle on the basis of the lateral vehicle body acceleration and the yaw rate calculated by the yaw rate calculating means. It consists of.

According to the fourth aspect of the present invention, the side slip angle calculating means calculates the vehicle body slip angle β from the lateral body acceleration _Gy and the yaw rate calculated by the yaw rate calculating means, for example, by the following equation.

[0061]

(Equation 12) However, V _a is an average vehicle speed is represented by _{V a = (V FL + V} FR + V RL + V RR) / 4 (27).

It should be noted that the invention of claim 4 can be configured as an invention in which the yaw rate calculation means of claim 3 is added to the invention of claim 2 and, in this case, the lateral state calculated by the vehicle state quantity calculation means. G can be used in the calculation of equation (26). (Principle of Estimating the Road Surface Condition of the Road Surface Calculation Means of the Present Invention)
As shown in FIG. 4A, the road surface μ characteristics with respect to the slip speed of the friction coefficient μ between road surfaces are as follows: dry road surface (Dry), snow road surface (Snow), ice road surface (Ice). . . It shows different characteristics depending on the road surface condition. For example, FIG.
First, as shown in FIG. 4B, the road surface μ characteristic shown in FIG. 4A is represented for each road surface condition by a correlation between a braking force (drive force) _Pc , which is a wheel behavior amount, and a slip speed ΔV. Can be. Then, in FIG. 4A, not only the μ value but also the gradient G _d of the road surface μ with respect to the slip speed has a unique value depending on the road surface condition, and the relationship shown in FIG. 4B is used. FIG. 4 (c) shows the road surface μ characteristic of FIG.
As shown in, it can be expressed as required sufficient 3-dimensional characteristics for each road surface state represented by the longitudinal force and the vehicle speed and the road surface μ gradient G _d.

[0063] In FIG. 4 (c), the three-dimensional characteristics, represents the vehicle speed V is 20km / h, 40km / h, in relation to the braking force and the road surface μ gradient G _d for each case of 60 km / h From this figure, it can be seen that the relationship between these three wheel behavior amounts differs depending on the road surface condition. Conversely, if the mutual relationship between these three wheel behavior amounts is known, the road surface state can be determined. It should be noted that the road surface state can be calculated in the same manner even when the road μ gradient _Gd is the gradient of the friction coefficient μ (or braking / driving force) with respect to the slip speed.

In one embodiment of the present road surface calculating means, a mutual relationship between a plurality of wheel behavior quantities representing such road surface μ characteristics is provided for each road surface state. This correlation can be represented, for example, by a conversion table between input and output data (hereinafter, a first table). In the case of the characteristic shown in FIG. It is created in advance based on the data obtained by detecting the relationship between the road and the road μ gradient. In the case of calculating a road surface condition in the case of a certain value vehicle speed, it calculates the road surface condition and a longitudinal force Pc and the road surface μ gradient G _d.

For example, the vehicle speed and the braking force are shown in FIG.
In the case of conversion based on the correlation of
The converted values of the road μ gradient corresponding to the detected value P _c0 of 20 km / h and the braking force are G _d1 , G _d2 , and G _d3 for the dry road surface, the snow road surface, and the ice road surface, respectively. Then, the road surface state is calculated based on a comparison between the converted value of the wheel behavior amount for each road surface state obtained in this way and the detected value of the same amount as the wheel behavior amount. For example, in the case of the above example, the conversion values G _d1 , G _d2 , G _d3 of the road μ gradient for each road surface state are compared with the detected value G _{d0 of} the road μ gradient calculated by the road μ gradient calculating means. The road surface state is calculated and output based on the comparison result.

In this comparison, for example, the differences between the detected values G _d0 and the converted values G _d1 , G _d2 , G _d3 are respectively calculated, and the road surface state giving the smallest difference (absolute value) is calculated. For example, G
_When the difference between _d2 and G _d0 is the smallest, it is calculated that the road surface state is “snow road surface (Snow)”.

In the above method for estimating the road surface condition, the vehicle speed is used as one of the wheel behavior amounts. However, the present road surface calculating means estimates the vehicle speed and the slip speed from the road surface μ gradient, braking / driving force and wheel speed. Therefore, further includes a second table showing the correlation between the braking force P _c and the road μ gradient G _d for each slip speed. This second table can be created by the following procedure.

First, in each case where the road surface condition and the vehicle speed are variously changed as parameters, the change of the slip speed Δv with respect to the braking / driving force _Pc is obtained as shown in FIG. Next, in each case where the road surface condition and the vehicle speed were variously changed as parameters, the slip speed Δv
When the change of the road surface μ gradient G _d with respect to
(Dry) and FIG. 6B (Snow).

[0069] Then, FIG. 5, the relationship of FIG. 6, P _c for the same slip speed Δv for each vehicle speed, when determining the G _d, for each road surface state, FIG. 7 (a), the as (b) Relationship. P _{c for} each slip speed shown in FIG.
The relationship between the G _d and the second table.

Further, based on FIGS. 7 (a) and 7 (b), FIGS. 8 (a) to 8 (c) are shown for each vehicle speed (20 km / h, 40 km / h, 60 km / h).
Relationship between P _c and G _d as is shown for each road surface condition.
The relationship in FIG. 8 corresponds to the first table for each road surface condition.

In this embodiment, for example, the road surface μ gradient G _{d for} each slip speed obtained by converting the detected value of the braking force P _c by the second table for each slip speed, respectively.
A conversion value, calculates a slip speed based on a comparison between the detected value of the road surface μ gradient G _d. That is, the detected P _c
The relationship between the G _d and is calculated whether closest to the line of Δv 7 throat, obtains the Δv value were closest line as slip speed.

Since the slip speed has been calculated in this manner, in this embodiment, the vehicle speed is calculated from the calculated slip speed and the detected wheel speed. Then, as described above, the road surface calculating means can calculate the road surface condition based on the first table using the calculated vehicle speed, braking / driving force, and road surface μ gradient.

Based on the above estimation method, the road surface calculating means of the present invention can calculate the slip speed and the road surface state using the wheel speed, road surface μ gradient and braking / driving force. In the slip speed calculating means according to the third aspect, the slip speed can be calculated from only the second table using the road surface μ gradient and the braking / driving force. (Calculation Principle of Road Surface μ Gradient Calculating Means) In one aspect of the road surface μ gradient calculating means of the present invention, there is provided a micro excitation means for minutely exciting a braking force at a resonance frequency of a vibration system including a vehicle body, wheels, and a road surface. The ratio of the minute amplitude of the resonance frequency component of the wheel speed to the minute amplitude of the braking force when the braking force is minutely excited by the minute excitation means is calculated as a road surface μ gradient. Hereinafter, the operation principle will be described.

A vehicle equipped with a vehicle body having a weight W _v has a vehicle body speed ω
Referring to the model shown in FIG. 13 in which the vibration phenomenon at the wheels when traveling in _v , that is, the vibration phenomenon of the vibration system composed of the vehicle body, the wheels, and the road surface is equivalently modeled by the wheel rotation axis. To consider.

In the model of FIG. 13, the braking force is
The brake force acts on the road surface via the surface of the tread 115 of the tire in contact with the road surface. However, since this braking force actually acts on the vehicle body as a reaction (braking force) from the road surface, the equivalent of the vehicle weight _Wv in terms of the rotation axis is used. The model 117 is connected to a side opposite to the wheel 113 via a friction element 116 (road surface μ) between the tread of the tire and the road surface. This is the same as the large inertia under the wheels, that is, the weight of the vehicle body can be simulated by the mass on the side opposite to the wheels, as in the chassis dynamo device.

In FIG. 13, the inertia of the wheel 113 including the tire rim is represented by J _w , and the spring element 1 between the rim and the tread 115
14 is K, the wheel radius is R, the inertia of the tread 115 is J _t , and the friction element 1 between the tread 115 and the road surface is 1
Assuming that the friction coefficient of μ16 is μ and the inertia of the equivalent model 117 in terms of the rotation axis of the weight of the vehicle body is J _V , the transmission characteristic from the brake torque T _b ′ generated by the wheel cylinder pressure to the wheel speed ω _w is From the equation,

[0077]

(Equation 13) Becomes Note that s is an operator of Laplace transform.

When it is considered that the tread 115 and the vehicle equivalent model 117 are directly connected when the tire is gripping the road surface, the vehicle equivalent model 117 and the tread 115
And the inertia of the wheel 113 resonates. That is, this vibration system can be regarded as a wheel resonance system composed of wheels, a vehicle body, and a road surface. Resonant frequency Omega∞ wheel resonance system at this time, the transfer characteristic of equation (28), ω∞ = √ { (J w + J t + J v) K / J w (J t + J v)} / 2π (29 ). This state corresponds to the region A1 before reaching the peak μ in FIG.

On the other hand, when the friction coefficient μ of the tire approaches the peak μ, the friction coefficient μ of the tire surface hardly changes with respect to the slip speed, and the component accompanying the inertia vibration of the tread 115 is a vehicle equivalent model. 117 will not be affected.
That is, the tread 115 and the vehicle equivalent model 117 are equivalently provided.
Are separated, and the tread 115 and the wheel 113 resonate. The wheel resonance system at this time can be considered to be composed of wheels and a road surface, and its resonance frequency ω∞ ′ is equal to the value obtained by setting the vehicle equivalent inertia J _v to 0 in the equation (29). . In other words, the _{ω∞ '= √ {(J w} + J t) K / J w J t)} / 2π (30). This state corresponds to the region A2 near the peak μ in FIG.

[0080] Compared (29) and the expression (30), when the vehicle equivalent inertia J _v is assumed wheel inertia J _w, a larger tread inertia J _t, (30) the resonance of the wheel resonance system in the case of formula Frequency ω
∞ ′ is shifted to a higher frequency side than ω∞ according to the equation (29). Therefore, it is possible to determine the friction state between the wheel and the road surface based on the physical quantity reflecting the change in the resonance frequency of the wheel resonance system.

[0081] In the present invention, as a physical quantity that reflects such a change in the resonant frequency, the micro-gain G _d as follows to introduce as a wheel motion amount.

[0082] First, the micro excitation means, when small exciting the braking force at the wheel and the vehicle body and the road surface and the vibration system of the resonance frequency ω∞ consisting ((29)) (here, when the brake pressure P _b is small excitation ), The wheel speed ω _w also slightly vibrates around the average wheel speed at the resonance frequency ω∞. Here, the small amplitude of the resonance frequency ω∞ brake pressure P _b in this case P
_v , when the minute amplitude of the resonance frequency ω∞ of the wheel speed is ω _wv , the minute gain G _{d is given} by G _d = ω _wv / P _v (31) It should be noted that this small gain G _d is used as the brake pressure P _b
The ratio of the wheel speed ω _w with respect to (ω _w / P _b) of the resonance frequency ω
Considering the vibration component of ∞, G _d = ((ω _w / P _b ) | s = jω∞) (32)

Since the small gain G _d is a vibration component of the resonance frequency ω∞ of (ω _w / P _b ) as shown in the equation (32), when the friction state reaches a region near the peak μ, Since the resonance frequency shifts to ω∞ ′, it sharply decreases. In other words, it can be said that the minute gain _Gd is a wheel behavior amount that sensitively reflects the friction state that defines the road surface μ characteristic.

Next, it will be described that the small gain G _d is a physical quantity equivalent to the road surface μ gradient.

As described above, in the characteristics of FIG. 14, a functional relationship is established between the slip speed Δω and the friction coefficient μ between the wheel and the road surface, where the friction coefficient μ has a peak at a certain slip speed. .

When the brake pressure is minutely excited by the minute excitation means, the wheel speed is minutely excited, so that the slip speed also slightly oscillates around a certain slip speed. Here, consider a change in the friction coefficient μ with respect to the slip speed Δω when the vehicle slightly vibrates around a certain slip speed on the road surface having the characteristics shown in FIG.

At this time, the friction coefficient μ of the road surface can be approximated as μ = μ ₀ + αRΔω (33) That is, since the change in the slip speed due to the minute vibration is small, it can be approximated by a straight line having the slope αR.

[0088] Here, substituting tire and the braking torque T _b = μWR caused by friction coefficient between the road surface mu (33) equation, the _{T b = μWR = μ 0 WR} + αR 2 ΔωW (34). Here, W is a wheel load. Let both sides of equation (34) be Δ
Differentiating first with ω,

[0089]

[Equation 14] Get. Therefore, the braking torque gradient (dT
_b / Δω) was shown to be equal to αR ² W.

On the other hand, since the brake torque T _b ′ is proportional to the brake pressure P _b , the small gain G _d is determined by the ratio of the wheel speed ω _{w to} the brake torque T _b ′ (ω _w /
T _b ′) is proportional to the vibration component of the resonance frequency ω∞. Therefore, the small gain G _d
Is represented by the following equation.

[0091]

(Equation 15) In general, in equation (37), | A | = 0.012 << | B | = 0.1 (39). From equations (35) and (36),

[0092]

(Equation 16) Get. In other words, the gradient of the braking torque T _b (or braking force T _b / R) with respect to the slip speed Δω is small gain G _d
Is proportional to

[0093] Thus, micro-gain G _d is the road μ gradient (the braking torque gradient, the braking force gradient) is shown to be equivalent to a physical quantity, that the road surface condition can be accurately calculated based on the micro-gain G _d Understand. Moreover, since the minute gain G _d is a wheel behavior amount that accurately represents the friction state based on the resonance characteristics of the vibration system, according to the present invention, it is possible to more accurately calculate the road surface state and the slip speed. Become. It should be noted that a minute gain when the driving force is minutely excited is, of course, equivalent to a driving torque gradient or a driving force gradient.

[0094]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a vehicle state quantity estimation device according to the present invention will be described in detail with reference to the drawings. (First Embodiment) FIG. 1 is a block diagram showing the configuration of a vehicle state quantity estimating apparatus according to a first embodiment of the present invention. As shown in the drawing, the present vehicle state quantity estimating device includes a front left wheel (FL), a front right wheel (FR), and a rear left wheel (R
L) and the right rear wheel (RR) are provided with the following constituent blocks. That is, the wheel speed detecting section 65 for detecting a wheel speed V _w, the braking force detecting section 34 for detecting the braking force P _c applied to the wheel, based on the wheel speed V _w, the braking force small amplitude at the resonant frequency ω∞ Gain G _d (braking / driving torque gradient), which is the ratio of the wheel speed minute amplitude to
Fine gain calculation unit 36 for calculating a, based on the wheel speed V _w and the braking force P _c and fine gain G _d, and the vehicle speed V at a slip speed ΔV and the wheels of the road surface condition and the wheels of the vehicle is now traveling A road surface state calculating device 10 for calculating, based on the vehicle speed V, a road surface μ for converting the minute gain G _d into a road surface μ gradient α.
The vehicle includes a gradient calculation unit 90 and a wheel load detection unit 91 that calculates a wheel load W based on the slip speed ΔV, the road surface condition, and the road surface μ gradient α. Note that the physical quantities calculated and detected for each wheel are distinguished from the physical quantities of other wheels by adding subscripts of FL, FR, RL, and RR.

Further, the present vehicle state quantity estimating apparatus includes a vehicle state quantity calculating section 92 connected to a road surface state calculating apparatus 10 for each wheel and a wheel load detecting section 91. The calculation unit calculates the detected wheel loads W _FL , W W
_FR , W _RL , W _RR and vehicle speed V _FL , V _FR , V _RL , V for each wheel
Based on the _RR , at least one of the lateral G, the front and rear G, the roll angle, the pitch angle, the inclination angle of the cant road, the force received by the cross wind, the yaw rate, and the side slip angle is calculated by the above-described corresponding equations.

As shown in FIG. 2, the road surface condition calculating device 10 calculates the wheel speed _Vw , the braking force _Pc, and the minute gain G as shown in FIG.
vehicle speed calculating unit 70 for calculating a vehicle speed V from a _d, and a road surface condition calculating unit 80 for calculating a road surface condition and a vehicle speed V calculated by the vehicle speed calculating unit 70 the braking force P _c and a small gain G _d Be composed.

[0097] The vehicle speed calculation unit 70 calculates Δv indicating the relationship between the braking force and the small gain for each fixed slip speed (0.1 m / s, 0.4 m / s, 0.8 m / s, ...). The conversion units 71, 72, 73,. . . . . It has.

The conversion units 71, 72, 73,. . . Is
When the detected braking force is input, a Δv table (FIG. 7) of a functional relationship as shown in FIG.
(A), (b)), the small gains G _d1 , G _d2 , G _d3,. . . . . (In FIG. 2, a ∧ indicating an estimation; hereinafter, ∧ is omitted) are converted and output.

[0099] A vehicle speed calculation unit 70, from the detection value G _d of the small gain is input to the device, the conversion unit 71,
73,. . . . . Gains G _d1 , G converted by
_d2 , _Gd3,. . . . Are subtracted to obtain the differences G _d −G _d1 , G _d −G _d2 , G _d −G _d3,. . . . Are respectively provided with differentiators 74, 75, and 76.

Further, the vehicle speed calculating section 70 includes a differentiator 74,
75, 76,. . . And a minimum value selection unit 77 for selecting a minimum value from the absolute values of the differences calculated by The minimum value selection unit 77 specifies the conversion unit that has output the minute gain giving the minimum value, and outputs the slip speed Δv corresponding to the conversion unit.

A vehicle speed calculator 78 is connected to the minimum value selector 77. The vehicle speed calculator 78, a slip speed Δv the minimum value selection unit 77 is output, and calculates the vehicle speed V by adding the wheel speed V _w of the wheel speed detecting unit 65 has detected.

As shown, the slip speed Δv output by the minimum value selector 77 and the vehicle speed V calculated by the vehicle speed calculator 78 are also output to the outside of the road surface arithmetic unit 10. Further, there is provided a divider 81 that outputs the slip speed ΔV by dividing the slip speed Δv output by the minimum value selecting unit 77 by the vehicle speed V calculated by the vehicle speed calculator 78. In addition, the road surface state calculation unit 80 calculates the slip speed Δv
(Or slip speed) slope of road surface μμμ / ∂Δ
v (in this embodiment, the micro-gain G _d), the braking force P _c, and ∂μ / ∂Δv table showing three wheel movements of the relationship of the vehicle body speed V dry road (Dry), snow road surface (Snow) and conversion sections 12, 14, and 16 for storing ice road surfaces (Ice), respectively.

When the vehicle speed and the braking force are input, the converters 12, 14, and 16 convert the data based on a functional relationship table (see FIG. 4C) stored therein as shown in FIG. The minute gains G _d1 , G _d2 ,
G _d3 (in FIG. 2, a ∧ indicating estimation, hereinafter, ∧ is omitted) is output.

The conversion unit can be configured as a ROM that stores all conversion values for input values in a predetermined range in advance and, when data is input, outputs a conversion value corresponding to the data value. . Further, it may be constituted by a neural network capable of learning a nonlinear function such as a ∂μ / ∂Δv table, an applied filter, or the like.

[0105] Further, the road surface condition calculating unit 80, from the detection value G _d of the small gain is input to the device, the conversion unit 12,1
Small gains G _d1 , G _d2 , G _d3 converted by 4 and 16
Are subtracted respectively to obtain the differences G _d −G _d1 , G _d
−G _d2 , G _d −G _d3
22.

Further, the road surface condition calculating section 80 is provided with a differentiator 1
A minimum value selector 24 is provided for selecting a minimum value from the absolute values of the differences calculated by 8, 20, and 22. The minimum value selection unit 24 specifies the conversion unit that has output the minute gain giving the minimum value, and outputs information on the road surface state corresponding to the conversion unit.

The road μ gradient calculating section 90 shown in FIG. 1 is provided with, for example, a table having data of the road μ gradient with respect to each value of the minute gain. Equivalent small gain G
_d is converted into a road surface μ gradient α equivalent to the gradient of the braking force with respect to the slip speed.

Further, the braking force detector 34 shown in FIGS.
Can be constituted by a pressure sensor or the like that detects the wheel cylinder pressure as a physical quantity approximately equal to the braking force _Pc , but a means for accurately estimating the braking force according to a wheel dynamic model as follows: It can also be configured as

[0109] That is, the wheel, and the brake torque T _B acting in the direction opposite to the rotational direction of the wheel to the wheel, and tire torque T _f by the braking force F acting in the rotational direction of the wheel as a frictional force to the wheel , Works. Brake torque T _B is derived from the braking force which acts to prevent rotation of the wheel relative to the wheel of the brake disc, the braking force F and the tire torque T _f is the friction coefficient between the wheel and the road surface When μ _B , the wheel radius is r, and the wheel load is W, it is expressed by the following equation.

[0110] _{F = μ B W T f =} F × r = μ B Wr Therefore, the equation of motion of the wheel,

[0111]

[Equation 17] Becomes Here, I is the moment of inertia of the wheel, and ω is the rotational speed (wheel speed) of the wheel.

[0112] detecting the wheel acceleration (d [omega / dt), by obtaining a brake torque T _B based on the wheel cylinder pressure applied to the brake disc, it is possible to estimate the braking force F on the basis of the (41) equation. Specifically, a dynamic model equivalent to the equation (41) in which the driving torque of the wheel obtained from the accelerator opening and the like and the braking force F as a disturbance acts on the wheel is configured as an observer. In this observer, (41)
The disturbance and rotation speed of the equivalent model are corrected for each control cycle so that the deviation between the rotation position obtained by the second-order integration of the equation and the rotation position actually detected is equal to 0, and the corrected disturbance is corrected. Estimate as braking force.

Also, the small gain calculating section 3 shown in FIGS.
6 shows a case where the brake pressure is slightly excited at the resonance frequency ω∞ (equation (29)) of the vibration system composed of the vehicle body, the wheels, and the road surface around the average brake pressure, as shown in detail in FIG. Wheel speed minute amplitude detector 40 for detecting the minute amplitude of the resonance frequency ω∞ (wheel speed minute amplitude ω _wv ) of the wheel speed V _w
If, fine gain G by dividing the brake pressure differential small-amplitude detector 42 for detecting a small amplitude P _v of the braking pressure of the resonance frequency Omega∞, the detected wheel speed micro amplitude omega _wv brake pressure differential small amplitude P _v and a divider 44 for outputting _d .

Of these, the wheel speed minute amplitude detecting section 40 can be realized as an arithmetic section for performing a filtering process for extracting a vibration component of the resonance frequency ω∞. For example, since the resonance frequency ω∞ of this vibration system is about 40 [Hz], one cycle is set to 24 [ms] and about 41.7 [Hz] in consideration of controllability. The output of the filter is subjected to full-wave rectification and DC smoothing to output a small amplitude of the wheel speed. Also, an integral multiple of the cycle, for example, 24 [ms] for one cycle and 48 [m] for two cycles
s] is continuously taken in, and 41.7 [H
z] can be realized by calculating the correlation with the unit sine wave and unit cosine wave.

[0115] Here, a description will be given small excitation means for applying a brake pressure differential small amplitude P _v of the resonance frequency around the mean braking pressure P _m. First, the part (valve control system) that converts the average brake pressure command and the minute excitation command into actual braking torque for the wheels is, as shown in FIG. 15, a master cylinder 48, a control valve 52, a wheel cylinder 56, and a reservoir 58. And an oil pump 60.

The brake pedal 46 is provided with a master cylinder 48 whose pressure is increased according to the depression force of the brake pedal 46.
Is connected to the pressure-intensifying valve 50 of the control valve 52. The control valve 52 is connected to a reservoir 58 as a low pressure source via a pressure reducing valve 54. Further, a wheel cylinder 56 for applying the brake pressure supplied by the control valve to the brake disc is connected to the control valve 52. The control valve 52 controls the opening and closing of the pressure increasing valve 50 and the pressure reducing valve 54 based on the input valve operation command.

The control valve 52 is connected to the pressure increasing valve 5
0 is controlled to open only the wheel cylinder 5
The hydraulic pressure (wheel cylinder pressure) of the master cylinder 48 (master cylinder pressure) is proportional to the pressure obtained when the driver depresses the brake pedal 46.
To rise. Conversely, when the pressure is controlled so as to open only the pressure reducing valve 54, the wheel cylinder pressure decreases to the pressure of the reservoir 58 (reservoir pressure), which is approximately atmospheric pressure. Further, when both valves are controlled to be closed, the wheel cylinder pressure is maintained.

The braking force (corresponding to the wheel cylinder pressure) applied to the brake disk by the wheel cylinder 56 includes a pressure increasing time during which the high hydraulic pressure of the master cylinder 48 is supplied, a pressure reducing time during which the low hydraulic pressure of the reservoir 58 is supplied,
And the ratio of the holding time during which the supply oil pressure is held, and the master cylinder pressure and the reservoir pressure detected by a pressure sensor or the like.

Therefore, a desired brake torque can be realized by controlling the increasing / decreasing time of the control valve 52 according to the master cylinder pressure. Further, the minute excitation of the brake pressure can be performed by performing the pressure increase / decrease control at a cycle corresponding to the resonance frequency simultaneously with the pressure increase / decrease control of the control valve 52 for realizing the average braking force.

As the specific contents of the control, as shown in FIG. 16, each mode of pressure increase and pressure reduction is switched every half cycle T / 2 of a minute excitation cycle (for example, 24 [ms]), and the valve is switched to the valve. decrease pressure command, the moment between the pressure boosting t _i of mode switching, and outputs the respective just time period the pressure increase, pressure reduction command decompression time t _r, the rest of the time, and outputs the held command. The average braking force, as well as determined by the ratio of the pressure increasing time t _i corresponding to the master cylinder pressure and the decompression time t _r, by switching the pressure-increasing-decreasing mode of the half period T / every 2 corresponding to the resonance frequency, the average A minute vibration is applied around the braking force.

Note that the brake pressure minute amplitude P_vIs the master
Cylinder pressure, pressure increase time t of the valve shown in FIG._iHead of
And decompression time t_rIs determined in a predetermined relationship by the length of
Therefore, the brake pressure minute amplitude detector 42 in FIG.
Star cylinder pressure, pressure increase time t _iAnd decompression time t_rFrom
Rake pressure small amplitude P_vAs a table to output
Can be

Next, the operation of the first embodiment will be described. The braking force P _c detected by the braking force detector 34 is
Each of the conversion units 71, 72, 73,. . . , Each conversion unit applies Δv
The input detection value is converted by a table, and minute gains G _d1 , G _d2 , G _d3,. . . . Are output.

The converted minute gains G _d1 , G _d2 ,
G _d3,. . . Together, fine gain G _d calculated by the micro-gain calculation section 36, the differentiator 74,75,7
6,. . . . Respectively, these differentiators provide the differences G _d −G _d1 , G _d −G _d2 , G _d −
G _d3,. . . . Are calculated and output to the minimum value selection unit 77.

The minimum value selection unit 77 selects the difference having the smallest absolute value from the calculated differences, and specifies the conversion unit that has given the selected minimum difference value. Then, the slip speed Δv corresponding to the specified conversion unit is output.
For example, when G _d −G _d2 becomes the minimum value, the slip speed is output as 0.4 m / s corresponding to the conversion unit 72.

Then, the vehicle speed calculator 78 calculates the slip speed Δv output from the minimum value selector 77 and the wheel speed detector 65
Vehicle speed V by but for adding the wheel speed V _w detected
Is calculated and output to the road surface state calculation unit 80.

The slip speed Δv output by the minimum value selector 77 and the vehicle speed V calculated by the vehicle speed calculator 78 are output to the outside of the apparatus, and the divider 81 calculates Δv /
V is calculated, and the calculation result is output as the slip speed ΔV.

Each of the conversion units 12, 1 of the road surface state calculation unit 80
Reference numerals 4 and 16 convert and output the minute gains G _d1 , G _d2 and G _d3 corresponding to the input vehicle speed V and braking force P _c , respectively, according to the ∂μ / ∂Δv tables respectively given.

[0128] with the micro gain transformed _{G d1, G d2, G d3} , fine gain G _d calculated by the micro-gain calculation section 36 and are input to the differentiator 18, 19 and 20, these differences vessel, the difference G _d -G _d1, G _d -
G _d2 , G _d −G _d3 , respectively, and the minimum value selection unit 24
Output to

The minimum value selection unit 24 selects the difference having the smallest absolute value from the calculated differences, and specifies the conversion unit that has given the selected minimum difference value. Then, the road surface state corresponding to the specified conversion unit is output as an information code in a predetermined format. The information code represented in this format corresponds to any of Dry, Snow, and Ice, and is represented in a format that can be recognized by the control unit of the vehicle to which the road surface state calculation device 10 is applied. For example, when G _d −G _d3 becomes the minimum value, an information code indicating that the road surface state is “Ice” is output.

Next, the wheel load detecting section 91 is provided with a table (not shown) showing the relationship between the slip speed and the road μ gradient in the reference characteristic of FIG. 3 measured without a load movement for each road surface condition. Then, based on a table corresponding to the calculated road surface condition, a road surface μ gradient α ′ corresponding to the calculated slip speed ΔV is obtained. Then, the road surface μ gradient α ′ and the road surface μ calculated by the road surface
The wheel load W is calculated from the equation (6) based on the gradient α and the static load W _f0 provided as data in advance.

The vehicle state quantity calculation unit 92 calculates each of the calculated values.
Wheel load W of wheel_FL, W_FR, W_RL, W _RRBased on the horizontal
G, front and rear G, roll angle, pitch angle, cant road inclination angle,
Calculate at least one of the forces received by the crosswind
((10), (11), (20), (21), (23), (24)
formula). When calculating the lateral G and the inclination angle of the cant road
Is the vehicle speed V for each wheel_FL, V_FR, V_RL, V_RROn the basis of the,
It is determined whether the vehicle is turning or traveling straight ahead.
When the vehicle is traveling straight ahead, the inclination angle of the cant road
Is calculated.

The vehicle speeds V _FL , V _FR , V _RL , V
Based on _RR , the yaw rate is calculated from equation (25) and the vehicle body slip angle is calculated from equation (26). (Second Embodiment) In a second embodiment, a wheel speed detection unit 6 is used instead of the small gain calculation unit 36 shown in FIGS.
Torque gradient estimation for estimating the gradient (braking torque gradient) of the braking torque (driving torque) with respect to the slip speed using an online identification method based on the time-series data of the wheel speed for each of the predetermined sample times τ detected by Step 5 Unit (not shown). The other configuration is the same as that of the first embodiment, and thus the same reference numerals are given and the detailed description is omitted. Hereinafter, the estimation principle of the torque gradient estimation unit according to the second embodiment will be described. (Principle of Estimating Gradient of Braking Torque or Driving Torque) The wheel motion of each wheel and the vehicle body motion are described by the following equations of motion. In the following, it is assumed that the number of wheels is four, but the present embodiment is not limited to this.

[0133]

(Equation 18) Where F _i ′ is the braking force generated on the i-th wheel, T _bi is the braking torque applied to the i-th wheel corresponding to the treading force, M is the vehicle mass, R _c is the effective radius of the wheel, and J is the wheel Inertia, v is the vehicle speed. In addition, * shows differentiation with respect to time.
In equations (42) and (43), F _i ′ is the slip speed (v /
R _c −ω _i ).

[0134] Here described, together with representative of the vehicle speed in equivalent body of the angular velocity omega _v, as the braking torque R _c F _i 'the slip speed of a linear function (gradient k _i, y intercept T _i).

V = R_cω_v (44) R_cF_i’(Ω_v−ω_i) = K_i× (ω_v−ω_i) + T_i (45) Furthermore, substituting equations (44) and (45) into equations (42) and (43),
Degree ω_iAnd body speed ω _vIs discretized every sample time τ
Time series data ω_i[k], ω_v[k] (k is sample time
Sample time in units of τ, k = 1,2, ...)
When expressed, the following equation is obtained.

[0136]

[Equation 19] Here, equations (46) and (47) are simultaneously set, and the equivalent angular velocity ω _v
When you erase,

[0137]

(Equation 20) Get.

By the way, assuming that a maximum braking torque of R _c Mg / 4 (g is a gravitational acceleration) is generated under the condition of a slip speed of 3 rad / s.

[0139]

(Equation 21) Get. Here, as a specific constant, τ = 0.005 (se
c), R _c = 0.3 (m) and M = 1000 (kg)
the (k _i) = 245. Therefore,

[0140]

(Equation 22) Equation (48) can be approximated as the following equation.

[0141]

(Equation 23) It is.

By rearranging in this manner, equation (49) can be described in a linear form with respect to unknown coefficients k _i and f _i , and the on-line parameter identification method can be applied to equation (49). As a result, the braking torque gradient k _i with respect to the slip speed can be estimated.

In addition, not only when the braking torque is applied but also when the driving torque is applied, the on-line system identification method is similarly applied to the equation (49) to obtain the driving speed corresponding to the slip speed. The gradient of the torque (hereinafter, referred to as “drive torque gradient”) can be obtained.

For example, by repeating the following step 1 and the following step 2 derived based on the least squares method which is one method of the online system identification method, the time series data of the detected wheel speed is obtained. The braking torque gradient or the driving torque gradient can be estimated from ω _i [k].

[0145]

(Equation 24) far.

The first element of the matrix φ _i [k] in the equation (50) is a physical quantity relating to the change in the wheel speed in one sample time, and the equation (51) is the change in the wheel speed in the one sample time. Is a physical quantity related to a change in one sample time. This means that equation (49) is the equation of motion for wheel (deceleration) motion, and it can be seen that the braking torque gradient is proportional to the characteristic root expressing the dynamic characteristics of wheel deceleration. .
That is, the identification of the braking torque gradient can be interpreted as identifying the characteristic root of the wheel (deceleration) motion.

[0147]

(Equation 25) The estimated value of θ _i is calculated from the recurrence formula, and the first element of the matrix of the estimated value of θ _i is extracted as the gradient of the estimated braking torque. Here, λ is a forgetting factor (for example, λ = 0.98) indicating the degree of removing past data, and “
^T "denotes the transpose of the matrix.

Note that θ _{i on} the left side of the equation (52) is a physical quantity representing the history of the physical quantity related to the change in the wheel speed and the history of the physical quantity related to the change in the wheel speed.

In the second embodiment, since the braking / driving torque gradient can be estimated only from the wheel speed without slightly exciting the braking / driving force, there is an advantage that the hardware can be simplified. (Third Embodiment) The braking torque gradient can also be estimated from the wheel deceleration and the brake torque. Therefore,
In the third embodiment, online identification is performed based on the time series data of the wheel deceleration and the time series data of the brake torque for each predetermined sample time τ, instead of the minute gain calculation unit 36 of FIGS. A braking torque gradient estimating unit for estimating a braking torque gradient using a technique is provided (not shown). The wheel deceleration may be calculated from the wheel speed detected by the wheel speed detector 65. Further, the brake torque can be estimated from the wheel cylinder pressure detected by a pressure sensor or the like.

The other configuration is the same as that of the first embodiment, and therefore the same reference numerals are given and the detailed description is omitted. Hereinafter, the estimation principle of the braking torque gradient estimation unit according to the third embodiment will be described. (Estimation Principle of Braking Torque Gradient) Equations (42) and (43) are converted into braking torque F _i (= F _i ′ · R _c ) and vehicle speed ω in terms of angular velocity.
_When expressed using _v (= v / R _c ),

[0151]

(Equation 26) Becomes

Further, from equation (53), the wheel deceleration y _i (= −dω _i / dt) of the _i -th wheel is given by

[0153]

[Equation 27] It is expressed as

Here, the slip speed of the i-th wheel (ω _v −ω
_i ) is replaced with x _i and rearranging equations (53) to (55) gives

[0155]

[Equation 28] Becomes

Here, it is assumed that the braking torque F _i of the i-th wheel is a nonlinear function of the slip speed (see FIG. 14).
A braking torque F (x _i ) near a certain slip speed x _i is approximated by a straight line as in the following equation. That is, the braking torque F
(X _i ) is the braking torque gradient k with respect to the slip speed x _i
A gradient model that changes linearly according to _i is applied.

[0157] In _{F i (x i) = k} i x i + μ i (58) where, relates the i wheel (i = 1, 2, 3, 4), the time series data of the slip speed x _i [j] , The time series data of the brake torque is T _bi [j], and the time series data of the wheel deceleration is y
_i [j] (j = 0, 1, 2,...). However, it is assumed that each time-series data is sampled at every predetermined sampling time τ.

When the equation (58) is substituted into the equations (56) and (57), and the obtained equation is discretized using the above time series data for each sampling time τ,

[0159]

(Equation 29) Becomes That is, x [j], y [j], _Tb [j]
Is a vector having a slip speed, a wheel deceleration, and a brake torque for each wheel in each component.

From equation (60), the wheel deceleration y [j + 1] after one sample is

[0161]

[Equation 30] Becomes

[0162] (60), (61) from the equation, K · (x [j + 1] -x [j]) = -J (y [j + 1] -y [j]) + T b [j + 1] -T b [j (62) is obtained.

In the equation (62), φ = x [j + 1] −x [j] (63) f = −J (y [j + 1] −y [j]) + T _b [j + 1] −T _b [j] ( 64), K ・ φ = f (65).

Here, considering the meaning of φ, it can be seen that it indicates a difference in slip speed between adjacent samples, that is, a physical quantity related to a change in slip speed.

The equations (59) and (60) are made simultaneous and (Kx [j] +
μ), and rearranging, from equation (63),

[0166]

(Equation 31) Is obtained.

The time series data of the braking torque is expressed as F
[J] (time-series data F _i [j] of the braking torque of the i-th wheel
As a vector) having the ingredients, and rearranging by discretizing the equation (55), F [j] = - Jy [j] + T b [j] (67) is obtained.

Then, when equation (67) is applied to equation (64), f = F [j + 1] -F [j] (68)

From equation (68), it can be seen that f indicates a difference in braking torque between adjacent samples, that is, a physical quantity related to a change in braking torque.

As described above, the motion state of the wheels expressed by the equations (53) to (55) is approximated by the gradient model of the equation (58) as shown by the equations (59) and (60), It is shown that the exercise state can be converted into the relation of equation (65). That is, the motion state of the wheel is determined by the gradient of the braking torque with respect to the slip speed, which is a parameter to be identified, and the physical quantity related to the change in the braking torque expressed by the equations (64) and (66) by the braking torque and the wheel deceleration. And a relationship between the physical quantities related to the change in the slip speed.

As a result, the number of identification parameters can be reduced to one, and the calculation accuracy can be improved and the calculation time can be shortened.

[0172] Here, the expression (65), indicating the i-th wheel, and _{k i · φ i = f i} (69). Here, f and φ in Expression (65) are set as f = [f ₁ f ₂ f ₃ f ₄ ] ^T φ = [φ ₁ φ ₂ φ ₃ φ ₄ ] ^T.

In this embodiment, f _{i of} the i-th wheel is based on time-series data y _i [j] of the wheel deceleration of the i-th wheel and time-series data T _bi [j] of the brake torque of the i-th wheel. , Φ _i
Is calculated from equations (64) and (66), and the calculated f _i and φ _i are calculated as (6
Each data obtained by substituting the 9), can be estimated calculating the braking torque gradient k _i of the i wheel by applying the online system identification techniques.

In the third embodiment, since the braking / driving torque gradient can be estimated from the wheel deceleration and the brake torque without slightly exciting the braking force (brake torque), the hardware can be simplified. is there.

The embodiments of the present invention have been described above. However, the present invention is not limited to the above-described examples, and can be arbitrarily and suitably changed without departing from the gist of the present invention.

For example, in the above embodiment, the road surface conditions are classified into three types: dry road surfaces, snowy road surfaces, and icy road surfaces.
It is also possible to calculate other types of road surfaces, for example, a road surface such as a sloping road, a sand road surface, a rainy road surface, etc., and to calculate more finely more than three types of road surface conditions.

In FIG. 2, the vehicle speed and the braking force are converted into minute gains by the respective tables, and the converted value of the minute gain is compared with the detected value.
Alternatively, it is also possible to calculate the road surface condition by converting the minute gain and the braking force into the braking force or the vehicle speed by using each table and comparing the converted value with the detected value of the same wheel behavior amount. The same applies to the Δv table.

Further, in the above embodiment, the vehicle state quantity at the time of braking is calculated. However, the present invention is not limited to this, and the vehicle state quantity can be similarly calculated at the time of driving. Can be extended to In this case, the braking force detection unit 34 is configured as a unit that detects the driving force. When the road surface μ gradient is detected from the minute gain, the driving force is minutely excited, and the minute gain is obtained from the ratio between the minute excitation amplitude of the driving force and the minute amplitude of the wheel speed.

[0179] In the road μ gradient calculating section 90, the gradient G _d of the longitudinal force with respect to the slip rate computed by the micro-gain calculation section 36, etc., directly, be configured to enter into the wheel load detection unit 91 Can be. In this case, the gradient G
_In order to compensate for the vehicle speed dependence of _d ,
Output the slip speed Δv and the vehicle speed V instead of the slip speed ΔV. Then, the wheel load detecting section 91, the relationship between the slip speed and the gradient G _d, stores for each respective road surface condition and vehicle speed, the input vehicle speed V, the road surface condition, a gradient corresponding to the slip speed Delta] v G _d0 Is obtained, and the wheel load is obtained from the relationship of equation (7).

[0180]

As described above, according to the first and second aspects of the present invention, the road surface μ gradient on the actual road surface, which directly reflects the change in wheel load, is calculated, and the reference characteristic is further calculated. The road surface gradient corresponding to the calculated road surface condition and slip speed, that is, the road surface μ gradient in the reference characteristic, is obtained based on the relationship between the slip speed and the road surface gradient in Because the wheel load is calculated based on the road μ gradient on the road surface, compared to the method of indirectly estimating the wheel load from the acceleration detected by the sensor, the difference between braking and non-braking is different. An excellent effect that the wheel load can be easily and accurately estimated irrespective of external conditions such as road surface conditions and load distribution.

Further, according to the invention of claim 2, based on the wheel load calculated by the wheel load calculating means, the longitudinal vehicle body acceleration, the lateral vehicle body acceleration, the pitch angle, the roll angle, and the Since at least one of the magnitude of the force applied to the vehicle body by the inclination angle and the cross wind is calculated, the effect that the vehicle stabilization control can be more accurately performed using the vehicle state quantity is obtained. .

Further, according to the third and fourth aspects of the present invention, the road surface μ gradient on the actual road surface, which directly reflects the change in the slip speed during braking / driving, and the braking force actually acting on the wheels. Since the slip speed is obtained from the driving force and the yaw rate is calculated from the slip speed and the detected wheel speed, the yaw rate can be calculated very accurately even during braking / driving. The effect is obtained.

Further, according to the present invention, the vehicle body side slip angle can be accurately calculated even during braking / driving, based on the vehicle body acceleration in the lateral direction and the yaw rate calculated by the yaw rate calculation means. The effect is obtained that the vehicle stabilization control can be performed more accurately.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a vehicle state quantity estimation device according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a road surface state calculation device according to the first embodiment of the present invention.

FIG. 3 shows slip speed ΔV and braking / driving force μ of a tire
FIG. 9 is a diagram showing a relationship with W.

FIGS. 4A and 4B are graphs showing a relationship between wheel behavior amounts representing road surface μ characteristics, wherein FIG. 4A shows a relationship between slip speed and road surface μ, FIG. 4B shows a relationship between slip speed and braking force, and FIG. ) shows the relationship between the braking force and the road surface μ gradient G _d for each vehicle speed.

FIG. 5 is a graph showing a relationship between a braking force _Pc and a slip speed Δv for each road surface condition and vehicle speed.

[6] A graph showing the relationship between the slip speed Δv and the road surface μ gradient G _d of each road surface condition and vehicle speed, (a) shows the dry road surface, (b) is a graph for snow road surface.

[7] The relationship between the braking force P _c and the road surface μ gradient G _d in the case of the slip speed Δv parameter a graph shown in each vehicle, (a) shows the dry road surface, (b) Snow It is a graph about a road surface.

[8] Based on FIG. 7, a graph of the obtained relation between the braking force P _c and the road surface μ gradient G _d on the vehicle speed and each road surface condition, (a) shows the vehicle speed V = 20km / h, (b ) Is the vehicle speed V =
40 km / h and (c) are graphs for the case where the vehicle speed V = 60 km / h.

FIG. 9 is a half-car model viewed from the lateral side of the vehicle body for explaining the relationship between front and rear G and wheel load during braking.

FIG. 10 is a half-car model for explaining the relationship between the lateral G and the wheel load when turning, as viewed from the front-rear direction of the vehicle body.

FIG. 11 is a half-car model viewed from the vehicle front-rear direction for explaining the relationship between the inclination angle of the cant road and the wheel load.

FIG. 12 is a block diagram illustrating a configuration of a small gain calculation unit according to the first embodiment of the present invention.

FIG. 13 is a diagram showing an equivalent model of a vibration system including wheels, a vehicle body, and a road surface.

FIG. 14 shows a change characteristic of the friction coefficient μ with respect to the slip speed, and shows that the change of μ around the center of the minute vibration can be approximated by a straight line in order to explain that the minute gain is equivalent to the braking torque gradient. FIG.

FIG. 15 is a block diagram illustrating a hardware configuration of a brake unit.

FIG. 16 is a diagram showing a command to a control valve in a case where a minute excitation of a brake pressure and a control of an average braking force are simultaneously performed.

[Explanation of symbols]

 Reference Signs List 10 road surface state calculating device 34 braking force detecting unit 36 minute gain calculating unit 65 wheel speed detecting unit 90 road surface μ gradient calculating unit 91 wheel load detecting unit 92 vehicle state amount calculating unit

──────────────────────────────────────────────────続 き Continuing on the front page (72) Koji Umeno, 41, Chuchu Yokomichi, Nagakute-cho, Aichi-gun, Aichi, Japan 1 Toyota Central Research Laboratory Co., Ltd. 41 No. 1 Toyoda Central Research Institute, Inc. Examiner Hiroyuki Tago (56) References JP-A-9-58449 (JP, A) JP-A-8-324414 (JP, A) JP-A 9-2220 ( JP, A) JP-A-9-58444 (JP, A) JP-A-7-186928 (JP, A) JP-A-4-22447 (JP, A) JP-A-61-94864 (JP, A) JP JP-A-7-112569 (JP, A) JP-A-62-255284 (JP, A) JP-A-8-150908 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) B60T 8 / 00; 8/32-8/96

Claims (4)

(57) [Claims] 1. A wheel speed detecting means for detecting a wheel speed; a road μ gradient calculating means for calculating a gradient of a braking / driving force with respect to a slip speed or a road μ gradient which is a physical quantity related to the gradient; Braking / driving force estimating means for estimating braking / driving force; a wheel speed detected by the wheel speed detecting means; a road μ gradient calculated by the road μ gradient calculating means; Road surface calculation means for calculating a road surface state and a slip speed based on a driving force; storage means for storing a relationship between a slip speed and a road surface μ gradient in reference characteristics for each road surface state; Processing means for obtaining a road surface μ gradient corresponding to the road surface state and the slip speed calculated by the road surface calculation means based on the relationship; Road surface gradient and the road surface μ
A wheel load calculating means for calculating a load on the wheel based on the road μ gradient calculated by the gradient calculating means. 2. A vehicle body acceleration in a front-rear direction, a vehicle body acceleration in a lateral direction, a pitch angle, a roll angle, a cant road inclination angle, based on a wheel load calculated by the wheel load calculating means.
The vehicle state quantity estimating device according to claim 1, further comprising: vehicle state quantity calculating means for calculating at least one of the magnitude of the force applied to the vehicle body by the crosswind. 3. A wheel speed detecting means for detecting a wheel speed, a road μ gradient calculating means for calculating a road μ gradient which is a gradient of a braking / driving force with respect to a slip speed, and a braking / driving force acting on a wheel. Driving force estimating means, slip speed calculating means for calculating a slip speed based on the road μ gradient calculated by the road μ gradient calculating means and the braking / driving force estimated by the braking / driving force estimating means, the wheel Yaw rate calculation for calculating a yaw rate based on at least one of the front wheel and rear wheel speeds detected by the speed detection means and at least one of the front and rear wheel slip speeds calculated by the slip speed calculation means Means, and a vehicle state quantity estimation device. 4. The vehicle state quantity estimating apparatus according to claim 3, further comprising: a vehicle slip angle calculating means for calculating a vehicle body slip angle based on a vehicle body acceleration in a lateral direction and a yaw rate calculated by the yaw rate calculating means.